int FeaturesMiner::findBlackBorder(int orientation) {
if (image == NULL) {
return -1;
}
int sum;
unsigned int val;
if (orientation == HORIZONTAL) {
for (int i = 0; i < width; i++) {
val = sum = 0;
for (int j = 0; j < height; j++) {
pixGetPixel(image, i, j, &val);
sum += val;
}
if ((sum) >= 200 * height) {
return i;
}
}
} else if (orientation == VERTICAL) {
for (int i = 0; i < height; i++) {
val = sum = 0;
for (int j = 0; j < width; j++) {
pixGetPixel(image, j, i, &val);
sum += val;
}
if ((sum) >= 200 * width) {
return i;
}
}
}
return -1;
}
/*!
* pixMeanInRectangle()
*
* Input: pix (8 bpp)
* box (region to compute mean value)
* pixma (mean accumulator)
* &val (<return> mean value
* Return: 0 if OK, 1 on error
*
* Notes:
* (1) This function is intended to be used for many rectangles
* on the same image. It can find the mean within a
* rectangle in O(1), independent of the size of the rectangle.
*/
l_int32
pixMeanInRectangle(PIX *pixs,
BOX *box,
PIX *pixma,
l_float32 *pval)
{
l_int32 w, h, bx, by, bw, bh;
l_uint32 val00, val01, val10, val11;
l_float32 norm;
BOX *boxc;
PROCNAME("pixMeanInRectangle");
if (!pval)
return ERROR_INT("&val not defined", procName, 1);
*pval = 0.0;
if (!pixs || pixGetDepth(pixs) != 8)
return ERROR_INT("pixs not defined", procName, 1);
if (!box)
return ERROR_INT("box not defined", procName, 1);
if (!pixma)
return ERROR_INT("pixma not defined", procName, 1);
/* Clip rectangle to image */
pixGetDimensions(pixs, &w, &h, NULL);
boxc = boxClipToRectangle(box, w, h);
boxGetGeometry(boxc, &bx, &by, &bw, &bh);
boxDestroy(&boxc);
if (bw == 0 || bh == 0)
return ERROR_INT("no pixels in box", procName, 1);
/* Use up to 4 points in the accumulator */
norm = 1.0 / (bw * bh);
if (bx > 0 && by > 0) {
pixGetPixel(pixma, bx + bw - 1, by + bh - 1, &val11);
pixGetPixel(pixma, bx + bw - 1, by - 1, &val10);
pixGetPixel(pixma, bx - 1, by + bh - 1, &val01);
pixGetPixel(pixma, bx - 1, by - 1, &val00);
*pval = norm * (val11 - val01 + val00 - val10);
}
else if (by > 0) { /* bx == 0 */
pixGetPixel(pixma, bw - 1, by + bh - 1, &val11);
pixGetPixel(pixma, bw - 1, by - 1, &val10);
*pval = norm * (val11 - val10);
}
else if (bx > 0) { /* by == 0 */
pixGetPixel(pixma, bx + bw - 1, bh - 1, &val11);
pixGetPixel(pixma, bx - 1, bh - 1, &val01);
*pval = norm * (val11 - val01);
}
else { /* bx == 0 && by == 0 */
pixGetPixel(pixma, bw - 1, bh - 1, &val11);
*pval = norm * val11;
}
return 0;
}
// Helper to compute edge offsets for all the blobs on the list.
// See coutln.h for an explanation of edge offsets.
void BLOBNBOX::ComputeEdgeOffsets(Pix* thresholds, Pix* grey,
BLOBNBOX_LIST* blobs) {
int grey_height = 0;
int thr_height = 0;
int scale_factor = 1;
if (thresholds != NULL && grey != NULL) {
grey_height = pixGetHeight(grey);
thr_height = pixGetHeight(thresholds);
scale_factor =
IntCastRounded(static_cast<double>(grey_height) / thr_height);
}
BLOBNBOX_IT blob_it(blobs);
for (blob_it.mark_cycle_pt(); !blob_it.cycled_list(); blob_it.forward()) {
BLOBNBOX* blob = blob_it.data();
if (blob->cblob() != NULL) {
// Get the threshold that applies to this blob.
l_uint32 threshold = 128;
if (thresholds != NULL && grey != NULL) {
const TBOX& box = blob->cblob()->bounding_box();
// Transform the coordinates if required.
TPOINT pt((box.left() + box.right()) / 2,
(box.top() + box.bottom()) / 2);
pixGetPixel(thresholds, pt.x / scale_factor,
thr_height - 1 - pt.y / scale_factor, &threshold);
}
blob->cblob()->ComputeEdgeOffsets(threshold, grey);
}
}
}
/*!
* kernelCreateFromPix()
*
* Input: pix
* cy, cx (origin of kernel)
* Return: kernel, or null on error
*
* Notes:
* (1) The origin must be positive and within the dimensions of the pix.
*/
L_KERNEL *
kernelCreateFromPix(PIX *pix,
l_int32 cy,
l_int32 cx)
{
l_int32 i, j, w, h, d;
l_uint32 val;
L_KERNEL *kel;
PROCNAME("kernelCreateFromPix");
if (!pix)
return (L_KERNEL *)ERROR_PTR("pix not defined", procName, NULL);
pixGetDimensions(pix, &w, &h, &d);
if (d != 8)
return (L_KERNEL *)ERROR_PTR("pix not 8 bpp", procName, NULL);
if (cy < 0 || cx < 0 || cy >= h || cx >= w)
return (L_KERNEL *)ERROR_PTR("(cy, cx) invalid", procName, NULL);
kel = kernelCreate(h, w);
kernelSetOrigin(kel, cy, cx);
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
pixGetPixel(pix, j, i, &val);
kernelSetElement(kel, i, j, (l_float32)val);
}
}
return kel;
}
main(int argc,
char **argv)
{
l_int32 i, j, w, h, same, width, height, cx, cy;
l_uint32 val;
PIX *pixs, *pixse, *pixd1, *pixd2;
SEL *sel;
static char mainName[] = "rasterop_reg";
if (argc != 1)
return ERROR_INT(" Syntax: rasterop_reg", mainName, 1);
pixs = pixRead("feyn.tif");
for (width = 1; width <= 25; width += 3) {
for (height = 1; height <= 25; height += 4) {
cx = width / 2;
cy = height / 2;
/* Dilate using an actual sel */
sel = selCreateBrick(height, width, cy, cx, SEL_HIT);
pixd1 = pixDilate(NULL, pixs, sel);
/* Dilate using a pix as a sel */
pixse = pixCreate(width, height, 1);
pixSetAll(pixse);
pixd2 = pixCopy(NULL, pixs);
w = pixGetWidth(pixs);
h = pixGetHeight(pixs);
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
pixGetPixel(pixs, j, i, &val);
if (val)
pixRasterop(pixd2, j - cx, i - cy, width, height,
PIX_SRC | PIX_DST, pixse, 0, 0);
}
}
pixEqual(pixd1, pixd2, &same);
if (same == 1)
fprintf(stderr, "Correct for (%d,%d)\n", width, height);
else {
fprintf(stderr, "Error: results are different!\n");
fprintf(stderr, "SE: width = %d, height = %d\n", width, height);
pixWrite("/tmp/junkout1", pixd1, IFF_PNG);
pixWrite("/tmp/junkout2", pixd2, IFF_PNG);
return 1;
}
pixDestroy(&pixse);
pixDestroy(&pixd1);
pixDestroy(&pixd2);
selDestroy(&sel);
}
}
pixDestroy(&pixs);
return 0;
}
/*!
* pixOtsuThreshOnBackgroundNorm()
*
* Input: pixs (8 bpp grayscale; not colormapped)
* pixim (<optional> 1 bpp 'image' mask; can be null)
* sx, sy (tile size in pixels)
* thresh (threshold for determining foreground)
* mincount (min threshold on counts in a tile)
* bgval (target bg val; typ. > 128)
* smoothx (half-width of block convolution kernel width)
* smoothy (half-width of block convolution kernel height)
* scorefract (fraction of the max Otsu score; typ. 0.1)
* &thresh (<optional return> threshold value that was
* used on the normalized image)
* Return: pixd (1 bpp thresholded image), or null on error
*
* Notes:
* (1) This does background normalization followed by Otsu
* thresholding. Otsu binarization attempts to split the
* image into two roughly equal sets of pixels, and it does
* a very poor job when there are large amounts of dark
* background. By doing a background normalization first,
* to get the background near 255, we remove this problem.
* Then we use a modified Otsu to estimate the best global
* threshold on the normalized image.
* (2) See pixBackgroundNorm() for meaning and typical values
* of input parameters. For a start, you can try:
* sx, sy = 10, 15
* thresh = 100
* mincount = 50
* bgval = 255
* smoothx, smoothy = 2
*/
PIX *
pixOtsuThreshOnBackgroundNorm(PIX *pixs,
PIX *pixim,
l_int32 sx,
l_int32 sy,
l_int32 thresh,
l_int32 mincount,
l_int32 bgval,
l_int32 smoothx,
l_int32 smoothy,
l_float32 scorefract,
l_int32 *pthresh)
{
l_int32 w, h;
l_uint32 val;
PIX *pixn, *pixt, *pixd;
PROCNAME("pixOtsuThreshOnBackgroundNorm");
if (pthresh) *pthresh = 0;
if (!pixs || pixGetDepth(pixs) != 8)
return (PIX *)ERROR_PTR("pixs undefined or not 8 bpp", procName, NULL);
if (pixGetColormap(pixs))
return (PIX *)ERROR_PTR("pixs is colormapped", procName, NULL);
if (sx < 4 || sy < 4)
return (PIX *)ERROR_PTR("sx and sy must be >= 4", procName, NULL);
if (mincount > sx * sy) {
L_WARNING("mincount too large for tile size\n", procName);
mincount = (sx * sy) / 3;
}
pixn = pixBackgroundNorm(pixs, pixim, NULL, sx, sy, thresh,
mincount, bgval, smoothx, smoothy);
if (!pixn)
return (PIX *)ERROR_PTR("pixn not made", procName, NULL);
/* Just use 1 tile for a global threshold, which is stored
* as a single pixel in pixt. */
pixGetDimensions(pixn, &w, &h, NULL);
pixOtsuAdaptiveThreshold(pixn, w, h, 0, 0, scorefract, &pixt, &pixd);
pixDestroy(&pixn);
if (pixt && pthresh) {
pixGetPixel(pixt, 0, 0, &val);
*pthresh = val;
}
pixDestroy(&pixt);
if (!pixd)
return (PIX *)ERROR_PTR("pixd not made", procName, NULL);
else
return pixd;
}
main(int argc,
char **argv)
{
char *filein;
l_int32 i, j, w, h, d, sampling;
l_float32 rank, rval;
l_uint32 val;
NUMA *na, *nah, *nar, *nav;
PIX *pix;
static char mainName[] = "numaranktest";
if (argc != 3)
exit(ERROR_INT(" Syntax: numaranktest filein sampling", mainName, 1));
filein = argv[1];
sampling = atoi(argv[2]);
if ((pix = pixRead(filein)) == NULL)
exit(ERROR_INT("pix not made", mainName, 1));
pixGetDimensions(pix, &w, &h, &d);
if (d != 8)
return ERROR_INT("d != 8 bpp", mainName, 1);
na = numaCreate(0);
for (i = 0; i < h; i += sampling) {
for (j = 0; j < w; j += sampling) {
pixGetPixel(pix, j, i, &val);
numaAddNumber(na, val);
}
}
nah = numaMakeHistogramClipped(na, BIN_SIZE, 255);
nar = numaCreate(0);
for (rval = 0.0; rval < 256.0; rval += 2.56) {
numaHistogramGetRankFromVal(nah, rval, &rank);
numaAddNumber(nar, rank);
}
gplotSimple1(nar, GPLOT_X11, "/tmp/junkroot1", "rank vs val");
nav = numaCreate(0);
for (rank = 0.0; rank <= 1.0; rank += 0.01) {
numaHistogramGetValFromRank(nah, rank, &rval);
numaAddNumber(nav, rval);
}
gplotSimple1(nav, GPLOT_X11, "/tmp/junkroot2", "val vs rank");
pixDestroy(&pix);
numaDestroy(&na);
numaDestroy(&nah);
numaDestroy(&nar);
numaDestroy(&nav);
return 0;
}
/*!
* localSearchForBackground()
*
* Input: &x, &y (starting position for search; return found position)
* maxrad (max distance to search from starting location)
* Return: 0 if bg pixel found; 1 if not found
*/
static l_int32
localSearchForBackground(PIX *pix,
l_int32 *px,
l_int32 *py,
l_int32 maxrad)
{
l_int32 x, y, w, h, r, i, j;
l_uint32 val;
x = *px;
y = *py;
pixGetPixel(pix, x, y, &val);
if (val == 0) return 0;
/* For each value of r, restrict the search to the boundary
* pixels in a square centered on (x,y), clipping to the
* image boundaries if necessary. */
pixGetDimensions(pix, &w, &h, NULL);
for (r = 1; r < maxrad; r++) {
for (i = -r; i <= r; i++) {
if (y + i < 0 || y + i >= h)
continue;
for (j = -r; j <= r; j++) {
if (x + j < 0 || x + j >= w)
continue;
if (L_ABS(i) != r && L_ABS(j) != r) /* not on "r ring" */
continue;
pixGetPixel(pix, x + j, y + i, &val);
if (val == 0) {
*px = x + j;
*py = y + i;
return 0;
}
}
}
}
return 1;
}
int main(){
int i,j,k, n,pixl, count=0;
PIX *pixs;
glob_t *flist;
FILE *file=fopen("train_file.trn","w");
for(n=0;n<10;n++){
flist=list_files(n);
if(flist->gl_pathc==0) {
fprintf(stderr,"No files in directory %d... Exiting...\n",n);
exit(1);
}
for(i=0;i<flist->gl_pathc;i++){
pixs=loadimage(flist->gl_pathv[i]);
if(pixs==NULL){
fprintf(stderr,"Pix error... Exiting...\n");
exit(1);
}
for(j=0;j<pixGetHeight(pixs);j++){
for(k=0;k<pixGetWidth(pixs);k++){
pixGetPixel(pixs,k,j,&pixl);
fprintf(file,"%d",pixl); /* funny transformation of rows/columns. TODO: works with intels, but others??? */
}
fprintf(file,"\n");
}
count++;
fprintf(file," %d\n",n);
}
globfree(flist);
free(flist);
}
fclose(file);
// *file=fopen("train_file.trn","r");
FILE *file1=fopen("train_file_header.trn","w");
insert_header(file1,count);
fclose(file1);
return (0);
}
/*!
* adjacentOnPixelInRaster()
*
* Input: pixs (1 bpp)
* x, y (current pixel)
* xa, ya (adjacent ON pixel, found by simple CCW search)
* Return: 1 if a pixel is found; 0 otherwise or on error
*
* Notes:
* (1) Search is in 4-connected directions first; then on diagonals.
* This allows traversal along a 4-connected boundary.
*/
l_int32
adjacentOnPixelInRaster(PIX *pixs,
l_int32 x,
l_int32 y,
l_int32 *pxa,
l_int32 *pya)
{
l_int32 w, h, i, xa, ya, found;
l_int32 xdel[] = {-1, 0, 1, 0, -1, 1, 1, -1};
l_int32 ydel[] = {0, 1, 0, -1, 1, 1, -1, -1};
l_uint32 val;
PROCNAME("adjacentOnPixelInRaster");
if (!pixs)
return ERROR_INT("pixs not defined", procName, 0);
if (pixGetDepth(pixs) != 1)
return ERROR_INT("pixs not 1 bpp", procName, 0);
w = pixGetWidth(pixs);
h = pixGetHeight(pixs);
found = 0;
for (i = 0; i < 8; i++) {
xa = x + xdel[i];
ya = y + ydel[i];
if (xa < 0 || xa >= w || ya < 0 || ya >= h)
continue;
pixGetPixel(pixs, xa, ya, &val);
if (val == 1) {
found = 1;
*pxa = xa;
*pya = ya;
break;
}
}
return found;
}
char* _process_frame_white_basic(struct lib_hardsubx_ctx *ctx, AVFrame *frame, int width, int height, int index)
{
//printf("frame : %04d\n", index);
PIX *im;
PIX *edge_im;
PIX *lum_im;
PIX *feat_im;
char *subtitle_text=NULL;
im = pixCreate(width,height,32);
lum_im = pixCreate(width,height,32);
feat_im = pixCreate(width,height,32);
int i,j;
for(i=(3*height)/4;i<height;i++)
{
for(j=0;j<width;j++)
{
int p=j*3+i*frame->linesize[0];
int r=frame->data[0][p];
int g=frame->data[0][p+1];
int b=frame->data[0][p+2];
pixSetRGBPixel(im,j,i,r,g,b);
float L,A,B;
rgb_to_lab((float)r,(float)g,(float)b,&L,&A,&B);
if(L > ctx->lum_thresh)
pixSetRGBPixel(lum_im,j,i,255,255,255);
else
pixSetRGBPixel(lum_im,j,i,0,0,0);
}
}
//Handle the edge image
edge_im = pixCreate(width,height,8);
edge_im = pixConvertRGBToGray(im,0.0,0.0,0.0);
edge_im = pixSobelEdgeFilter(edge_im, L_VERTICAL_EDGES);
edge_im = pixDilateGray(edge_im, 21, 11);
edge_im = pixThresholdToBinary(edge_im,50);
for(i=3*(height/4);i<height;i++)
{
for(j=0;j<width;j++)
{
unsigned int p1,p2,p3;
pixGetPixel(edge_im,j,i,&p1);
// pixGetPixel(pixd,j,i,&p2);
pixGetPixel(lum_im,j,i,&p3);
if(p1==0&&p3>0)
pixSetRGBPixel(feat_im,j,i,255,255,255);
else
pixSetRGBPixel(feat_im,j,i,0,0,0);
}
}
if(ctx->detect_italics)
{
ctx->ocr_mode = HARDSUBX_OCRMODE_WORD;
}
// TESSERACT OCR FOR THE FRAME HERE
switch(ctx->ocr_mode)
{
case HARDSUBX_OCRMODE_WORD:
if(ctx->conf_thresh > 0)
subtitle_text = get_ocr_text_wordwise_threshold(ctx, lum_im, ctx->conf_thresh);
else
subtitle_text = get_ocr_text_wordwise(ctx, lum_im);
break;
case HARDSUBX_OCRMODE_LETTER:
if(ctx->conf_thresh > 0)
subtitle_text = get_ocr_text_letterwise_threshold(ctx, lum_im, ctx->conf_thresh);
else
subtitle_text = get_ocr_text_letterwise(ctx, lum_im);
break;
case HARDSUBX_OCRMODE_FRAME:
if(ctx->conf_thresh > 0)
subtitle_text = get_ocr_text_simple_threshold(ctx, lum_im, ctx->conf_thresh);
else
subtitle_text = get_ocr_text_simple(ctx, lum_im);
break;
default:
fatal(EXIT_MALFORMED_PARAMETER,"Invalid OCR Mode");
}
pixDestroy(&lum_im);
pixDestroy(&im);
pixDestroy(&edge_im);
pixDestroy(&feat_im);
return subtitle_text;
}
char *_process_frame_color_basic(struct lib_hardsubx_ctx *ctx, AVFrame *frame, int width, int height, int index)
{
char *subtitle_text=NULL;
PIX *im;
im = pixCreate(width,height,32);
PIX *hue_im = pixCreate(width,height,32);
int i,j;
for(i=0;i<height;i++)
{
for(j=0;j<width;j++)
{
int p=j*3+i*frame->linesize[0];
int r=frame->data[0][p];
int g=frame->data[0][p+1];
int b=frame->data[0][p+2];
pixSetRGBPixel(im,j,i,r,g,b);
float H,S,V;
rgb_to_hsv((float)r,(float)g,(float)b,&H,&S,&V);
if(abs(H-ctx->hue)<20)
{
pixSetRGBPixel(hue_im,j,i,r,g,b);
}
}
}
PIX *edge_im = pixCreate(width,height,8),*edge_im_2 = pixCreate(width,height,8);
edge_im = pixConvertRGBToGray(im,0.0,0.0,0.0);
edge_im = pixSobelEdgeFilter(edge_im, L_VERTICAL_EDGES);
edge_im = pixDilateGray(edge_im, 21, 1);
edge_im = pixThresholdToBinary(edge_im,50);
PIX *pixd = pixCreate(width,height,1);
pixSauvolaBinarize(pixConvertRGBToGray(hue_im,0.0,0.0,0.0), 15, 0.3, 1, NULL, NULL, NULL, &pixd);
edge_im_2 = pixConvertRGBToGray(hue_im,0.0,0.0,0.0);
edge_im_2 = pixDilateGray(edge_im_2, 5, 5);
PIX *feat_im = pixCreate(width,height,32);
for(i=3*(height/4);i<height;i++)
{
for(j=0;j<width;j++)
{
unsigned int p1,p2,p3,p4;
pixGetPixel(edge_im,j,i,&p1);
pixGetPixel(pixd,j,i,&p2);
// pixGetPixel(hue_im,j,i,&p3);
pixGetPixel(edge_im_2,j,i,&p4);
if(p1==0&&p2==0&&p4>0)//if(p4>0&&p1==0)//if(p2==0&&p1==0&&p3>0)
{
pixSetRGBPixel(feat_im,j,i,255,255,255);
}
}
}
if(ctx->detect_italics)
{
ctx->ocr_mode = HARDSUBX_OCRMODE_WORD;
}
// TESSERACT OCR FOR THE FRAME HERE
switch(ctx->ocr_mode)
{
case HARDSUBX_OCRMODE_WORD:
if(ctx->conf_thresh > 0)
subtitle_text = get_ocr_text_wordwise_threshold(ctx, feat_im, ctx->conf_thresh);
else
subtitle_text = get_ocr_text_wordwise(ctx, feat_im);
break;
case HARDSUBX_OCRMODE_LETTER:
if(ctx->conf_thresh > 0)
subtitle_text = get_ocr_text_letterwise_threshold(ctx, feat_im, ctx->conf_thresh);
else
subtitle_text = get_ocr_text_letterwise(ctx, feat_im);
break;
case HARDSUBX_OCRMODE_FRAME:
if(ctx->conf_thresh > 0)
subtitle_text = get_ocr_text_simple_threshold(ctx, feat_im, ctx->conf_thresh);
else
subtitle_text = get_ocr_text_simple(ctx, feat_im);
break;
default:
fatal(EXIT_MALFORMED_PARAMETER,"Invalid OCR Mode");
}
pixDestroy(&feat_im);
pixDestroy(&im);
pixDestroy(&edge_im);
pixDestroy(&hue_im);
return subtitle_text;
}
/*!
* pixMaskedThreshOnBackgroundNorm()
*
* Input: pixs (8 bpp grayscale; not colormapped)
* pixim (<optional> 1 bpp 'image' mask; can be null)
* sx, sy (tile size in pixels)
* thresh (threshold for determining foreground)
* mincount (min threshold on counts in a tile)
* smoothx (half-width of block convolution kernel width)
* smoothy (half-width of block convolution kernel height)
* scorefract (fraction of the max Otsu score; typ. ~ 0.1)
* &thresh (<optional return> threshold value that was
* used on the normalized image)
* Return: pixd (1 bpp thresholded image), or null on error
*
* Notes:
* (1) This begins with a standard background normalization.
* Additionally, there is a flexible background norm, that
* will adapt to a rapidly varying background, and this
* puts white pixels in the background near regions with
* significant foreground. The white pixels are turned into
* a 1 bpp selection mask by binarization followed by dilation.
* Otsu thresholding is performed on the input image to get an
* estimate of the threshold in the non-mask regions.
* The background normalized image is thresholded with two
* different values, and the result is combined using
* the selection mask.
* (2) Note that the numbers 255 (for bgval target) and 190 (for
* thresholding on pixn) are tied together, and explicitly
* defined in this function.
* (3) See pixBackgroundNorm() for meaning and typical values
* of input parameters. For a start, you can try:
* sx, sy = 10, 15
* thresh = 100
* mincount = 50
* smoothx, smoothy = 2
*/
PIX *
pixMaskedThreshOnBackgroundNorm(PIX *pixs,
PIX *pixim,
l_int32 sx,
l_int32 sy,
l_int32 thresh,
l_int32 mincount,
l_int32 smoothx,
l_int32 smoothy,
l_float32 scorefract,
l_int32 *pthresh)
{
l_int32 w, h;
l_uint32 val;
PIX *pixn, *pixm, *pixd, *pixt1, *pixt2, *pixt3, *pixt4;
PROCNAME("pixMaskedThreshOnBackgroundNorm");
if (pthresh) *pthresh = 0;
if (!pixs || pixGetDepth(pixs) != 8)
return (PIX *)ERROR_PTR("pixs undefined or not 8 bpp", procName, NULL);
if (pixGetColormap(pixs))
return (PIX *)ERROR_PTR("pixs is colormapped", procName, NULL);
if (sx < 4 || sy < 4)
return (PIX *)ERROR_PTR("sx and sy must be >= 4", procName, NULL);
if (mincount > sx * sy) {
L_WARNING("mincount too large for tile size\n", procName);
mincount = (sx * sy) / 3;
}
/* Standard background normalization */
pixn = pixBackgroundNorm(pixs, pixim, NULL, sx, sy, thresh,
mincount, 255, smoothx, smoothy);
if (!pixn)
return (PIX *)ERROR_PTR("pixn not made", procName, NULL);
/* Special background normalization for adaptation to quickly
* varying background. Threshold on the very light parts,
* which tend to be near significant edges, and dilate to
* form a mask over regions that are typically text. The
* dilation size is chosen to cover the text completely,
* except for very thick fonts. */
pixt1 = pixBackgroundNormFlex(pixs, 7, 7, 1, 1, 20);
pixt2 = pixThresholdToBinary(pixt1, 240);
pixInvert(pixt2, pixt2);
pixm = pixMorphSequence(pixt2, "d21.21", 0);
pixDestroy(&pixt1);
pixDestroy(&pixt2);
/* Use Otsu to get a global threshold estimate for the image,
* which is stored as a single pixel in pixt3. */
pixGetDimensions(pixs, &w, &h, NULL);
pixOtsuAdaptiveThreshold(pixs, w, h, 0, 0, scorefract, &pixt3, NULL);
if (pixt3 && pthresh) {
pixGetPixel(pixt3, 0, 0, &val);
*pthresh = val;
}
pixDestroy(&pixt3);
/* Threshold the background normalized images differentially,
* using a high value correlated with the background normalization
* for the part of the image under the mask (i.e., near the
* darker, thicker foreground), and a value that depends on the Otsu
* threshold for the rest of the image. This gives a solid
* (high) thresholding for the foreground parts of the image,
* while allowing the background and light foreground to be
* reasonably well cleaned using a threshold adapted to the
* input image. */
pixd = pixThresholdToBinary(pixn, val + 30); /* for bg and light fg */
pixt4 = pixThresholdToBinary(pixn, 190); /* for heavier fg */
pixCombineMasked(pixd, pixt4, pixm);
pixDestroy(&pixt4);
pixDestroy(&pixm);
pixDestroy(&pixn);
if (!pixd)
return (PIX *)ERROR_PTR("pixd not made", procName, NULL);
else
return pixd;
}
int main(int argc,
char **argv)
{
l_int32 i, j, w, h, same, width, height, cx, cy;
l_uint32 val;
BOX *box;
PIX *pix0, *pixs, *pixse, *pixd1, *pixd2;
SEL *sel;
L_REGPARAMS *rp;
if (regTestSetup(argc, argv, &rp))
return 1;
pix0 = pixRead("feyn-fract.tif");
box = boxCreate(293, 37, pixGetWidth(pix0) - 691, pixGetHeight(pix0) -145);
pixs = pixClipRectangle(pix0, box, NULL);
boxDestroy(&box);
if (rp->display) pixDisplay(pixs, 100, 100);
/* Test 63 different sizes */
for (width = 1; width <= 25; width += 3) { /* 9 values */
for (height = 1; height <= 25; height += 4) { /* 7 values */
cx = width / 2;
cy = height / 2;
/* Dilate using an actual sel */
sel = selCreateBrick(height, width, cy, cx, SEL_HIT);
pixd1 = pixDilate(NULL, pixs, sel);
/* Dilate using a pix as a sel */
pixse = pixCreate(width, height, 1);
pixSetAll(pixse);
pixd2 = pixCopy(NULL, pixs);
w = pixGetWidth(pixs);
h = pixGetHeight(pixs);
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
pixGetPixel(pixs, j, i, &val);
if (val)
pixRasterop(pixd2, j - cx, i - cy, width, height,
PIX_SRC | PIX_DST, pixse, 0, 0);
}
}
pixEqual(pixd1, pixd2, &same);
regTestCompareValues(rp, 1, same, 0.0); /* 0 - 62 */
if (same == 0) {
fprintf(stderr,
"Results differ for SE (width,height) = (%d,%d)\n",
width, height);
}
pixDestroy(&pixse);
pixDestroy(&pixd1);
pixDestroy(&pixd2);
selDestroy(&sel);
}
}
pixDestroy(&pix0);
pixDestroy(&pixs);
return regTestCleanup(rp);
}
/*!
* pixOtsuAdaptiveThreshold()
*
* Input: pixs (8 bpp)
* sx, sy (desired tile dimensions; actual size may vary)
* smoothx, smoothy (half-width of convolution kernel applied to
* threshold array: use 0 for no smoothing)
* scorefract (fraction of the max Otsu score; typ. 0.1;
* use 0.0 for standard Otsu)
* &pixth (<optional return> array of threshold values
* found for each tile)
* &pixd (<optional return> thresholded input pixs, based on
* the threshold array)
* Return: 0 if OK, 1 on error
*
* Notes:
* (1) The Otsu method finds a single global threshold for an image.
* This function allows a locally adapted threshold to be
* found for each tile into which the image is broken up.
* (2) The array of threshold values, one for each tile, constitutes
* a highly downscaled image. This array is optionally
* smoothed using a convolution. The full width and height of the
* convolution kernel are (2 * @smoothx + 1) and (2 * @smoothy + 1).
* (3) The minimum tile dimension allowed is 16. If such small
* tiles are used, it is recommended to use smoothing, because
* without smoothing, each small tile determines the splitting
* threshold independently. A tile that is entirely in the
* image bg will then hallucinate fg, resulting in a very noisy
* binarization. The smoothing should be large enough that no
* tile is only influenced by one type (fg or bg) of pixels,
* because it will force a split of its pixels.
* (4) To get a single global threshold for the entire image, use
* input values of @sx and @sy that are larger than the image.
* For this situation, the smoothing parameters are ignored.
* (5) The threshold values partition the image pixels into two classes:
* one whose values are less than the threshold and another
* whose values are greater than or equal to the threshold.
* This is the same use of 'threshold' as in pixThresholdToBinary().
* (6) The scorefract is the fraction of the maximum Otsu score, which
* is used to determine the range over which the histogram minimum
* is searched. See numaSplitDistribution() for details on the
* underlying method of choosing a threshold.
* (7) This uses enables a modified version of the Otsu criterion for
* splitting the distribution of pixels in each tile into a
* fg and bg part. The modification consists of searching for
* a minimum in the histogram over a range of pixel values where
* the Otsu score is within a defined fraction, @scorefract,
* of the max score. To get the original Otsu algorithm, set
* @scorefract == 0.
*/
l_int32
pixOtsuAdaptiveThreshold(PIX *pixs,
l_int32 sx,
l_int32 sy,
l_int32 smoothx,
l_int32 smoothy,
l_float32 scorefract,
PIX **ppixth,
PIX **ppixd)
{
l_int32 w, h, nx, ny, i, j, thresh;
l_uint32 val;
PIX *pixt, *pixb, *pixthresh, *pixth, *pixd;
PIXTILING *pt;
PROCNAME("pixOtsuAdaptiveThreshold");
if (!ppixth && !ppixd){
return ERROR_INT("neither &pixth nor &pixd defined", procName, 1);
LOGE("neither &pixth nor &pixd defined");
}
if (ppixth) *ppixth = NULL;
if (ppixd) *ppixd = NULL;
if (!pixs || pixGetDepth(pixs) != 8){
return ERROR_INT("pixs not defined or not 8 bpp", procName, 1);
LOGE("pixs not defined or not 8 bpp");
}
if (sx < 16 || sy < 16){
return ERROR_INT("sx and sy must be >= 16", procName, 1);
LOGE("sx and sy must be >= 16");
}
/* Compute the threshold array for the tiles */
pixGetDimensions(pixs, &w, &h, NULL);
nx = L_MAX(1, w / sx);
ny = L_MAX(1, h / sy);
smoothx = L_MIN(smoothx, (nx - 1) / 2);
smoothy = L_MIN(smoothy, (ny - 1) / 2);
pt = pixTilingCreate(pixs, nx, ny, 0, 0, 0, 0);
pixthresh = pixCreate(nx, ny, 8);
for (i = 0; i < ny; i++) {
for (j = 0; j < nx; j++) {
pixt = pixTilingGetTile(pt, i, j);
pixSplitDistributionFgBg(pixt, scorefract, 1, &thresh,
NULL, NULL, 0);
pixSetPixel(pixthresh, j, i, thresh); /* see note (4) */
pixDestroy(&pixt);
}
}
/* Optionally smooth the threshold array */
if (smoothx > 0 || smoothy > 0)
pixth = pixBlockconv(pixthresh, smoothx, smoothy);
else
pixth = pixClone(pixthresh);
pixDestroy(&pixthresh);
/* Optionally apply the threshold array to binarize pixs */
if (ppixd) {
pixd = pixCreate(w, h, 1);
for (i = 0; i < ny; i++) {
for (j = 0; j < nx; j++) {
pixt = pixTilingGetTile(pt, i, j);
pixGetPixel(pixth, j, i, &val);
pixb = pixThresholdToBinary(pixt, val);
pixTilingPaintTile(pixd, i, j, pixb, pt);
pixDestroy(&pixt);
pixDestroy(&pixb);
}
}
*ppixd = pixd;
}
if (ppixth)
*ppixth = pixth;
else
pixDestroy(&pixth);
pixTilingDestroy(&pt);
return 0;
}
int main(int argc,
char **argv)
{
l_int32 w, h, ystart, yend, y, ymax, ymid, i, window, sum1, sum2, rankx;
l_uint32 uval;
l_float32 ave, rankval, maxvar, variance, norm, conf, angle, radangle;
NUMA *na1;
PIX *pix1, *pix2, *pix3, *pix4, *pix5, *pix6, *pix7;
PIXA *pixa;
static char mainName[] = "findbinding";
if (argc != 1)
return ERROR_INT(" Syntax: findbinding", mainName, 1);
lept_mkdir("lept/binding");
pixa = pixaCreate(0);
pix1 = pixRead("binding-example.45.jpg");
pix2 = pixConvertTo8(pix1, 0);
/* Find the skew angle */
pix3 = pixConvertTo1(pix2, 150);
pixFindSkewSweepAndSearch(pix3, &angle, &conf, 2, 2, 7.0, 1.0, 0.01);
fprintf(stderr, "angle = %f, conf = %f\n", angle, conf);
/* Deskew, bringing in black pixels at the edges */
if (L_ABS(angle) < 0.1 || conf < 1.5) {
pix4 = pixClone(pix2);
} else {
radangle = 3.1416 * angle / 180.0;
pix4 = pixRotate(pix2, radangle, L_ROTATE_AREA_MAP,
L_BRING_IN_BLACK, 0, 0);
}
/* Rotate 90 degrees to make binding horizontal */
pix5 = pixRotateOrth(pix4, 1);
/* Sort pixels in each row by their gray value.
* Dark pixels on the left, light ones on the right. */
pix6 = pixRankRowTransform(pix5);
pixDisplay(pix5, 0, 0);
pixDisplay(pix6, 550, 0);
pixaAddPix(pixa, pix4, L_COPY);
pixaAddPix(pixa, pix5, L_COPY);
pixaAddPix(pixa, pix6, L_COPY);
/* Make an a priori estimate of the y-interval within which the
* binding will be found. The search will be done in this interval. */
pixGetDimensions(pix6, &w, &h, NULL);
ystart = 0.25 * h;
yend = 0.75 * h;
/* Choose a very light rank value; close to white, which
* corresponds to a column in pix6 near the right side. */
rankval = 0.98;
rankx = (l_int32)(w * rankval);
/* Investigate variance in a small window (vertical, size = 5)
* of the pixels in that column. These are the %rankval
* pixels in each raster of pix6. Find the y-location of
* maximum variance. */
window = 5;
norm = 1.0 / window;
maxvar = 0.0;
na1 = numaCreate(0);
numaSetParameters(na1, ystart, 1);
for (y = ystart; y <= yend; y++) {
sum1 = sum2 = 0;
for (i = 0; i < window; i++) {
pixGetPixel(pix6, rankx, y + i, &uval);
sum1 += uval;
sum2 += uval * uval;
}
ave = norm * sum1;
variance = norm * sum2 - ave * ave;
numaAddNumber(na1, variance);
ymid = y + window / 2;
if (variance > maxvar) {
maxvar = variance;
ymax = ymid;
}
}
/* Plot the windowed variance as a function of the y-value
* of the window location */
fprintf(stderr, "maxvar = %f, ymax = %d\n", maxvar, ymax);
gplotSimple1(na1, GPLOT_PNG, "/tmp/lept/binding/root", NULL);
pix7 = pixRead("/tmp/lept/binding/root.png");
pixDisplay(pix7, 0, 800);
pixaAddPix(pixa, pix7, L_COPY);
/* Superimpose the variance plot over the image.
* The variance peak is at the binding. */
pixRenderPlotFromNumaGen(&pix5, na1, L_VERTICAL_LINE, 3, w - 120, 100, 1,
0x0000ff00);
pixDisplay(pix5, 1050, 0);
pixaAddPix(pixa, pix5, L_COPY);
/* Bundle the results up in a pdf */
fprintf(stderr, "Writing pdf output file: /tmp/lept/binding/binding.pdf\n");
pixaConvertToPdf(pixa, 45, 1.0, 0, 0, "Binding locator",
"/tmp/lept/binding/binding.pdf");
pixDestroy(&pix1);
pixDestroy(&pix2);
pixDestroy(&pix3);
pixDestroy(&pix4);
pixDestroy(&pix5);
pixDestroy(&pix6);
pixDestroy(&pix7);
pixaDestroy(&pixa);
numaDestroy(&na1);
return 0;
}
int main(int argc,
char **argv) {
char *filein, *fileout;
l_int32 i;
l_uint32 val;
l_float32 size;
PIX *pixs, *pixd, *pixm, *pixmi, *pixt1, *pixt2, *pixt3;
static char mainName[] = "seedfilltest";
if (argc != 3)
return ERROR_INT(" Syntax: seedfilltest filein fileout", mainName, 1);
filein = argv[1];
fileout = argv[2];
pixd = NULL;
if ((pixm = pixRead(filein)) == NULL)
return ERROR_INT("pixm not made", mainName, 1);
pixmi = pixInvert(NULL, pixm);
size = pixGetWidth(pixm) * pixGetHeight(pixm);
pixs = pixCreateTemplate(pixm);
for (i = 0; i < 100; i++) {
pixGetPixel(pixm, XS + 5 * i, YS + 5 * i, &val);
if (val == 0) break;
}
if (i == 100)
return ERROR_INT("no seed pixel found", mainName, 1);
pixSetPixel(pixs, XS + 5 * i, YS + 5 * i, 1);
#if 0
/* hole filling; use "hole-filler.png" */
pixt1 = pixHDome(pixmi, 100, 4);
pixt2 = pixThresholdToBinary(pixt1, 10);
/* pixInvert(pixt1, pixt1); */
pixDisplay(pixt1, 100, 500);
pixDisplay(pixt2, 600, 500);
pixt3 = pixHolesByFilling(pixt2, 4);
pixDilateBrick(pixt3, pixt3, 7, 7);
pixd = pixConvertTo8(pixt3, FALSE);
pixDisplay(pixd, 0, 100);
pixSeedfillGray(pixd, pixmi, CONNECTIVITY);
pixInvert(pixd, pixd);
pixDisplay(pixmi, 500, 100);
pixDisplay(pixd, 1000, 100);
pixWrite("/tmp/junkpixm.png", pixmi, IFF_PNG);
pixWrite("/tmp/junkpixd.png", pixd, IFF_PNG);
#endif
#if 0
/* hole filling; use "hole-filler.png" */
pixt1 = pixThresholdToBinary(pixm, 110);
pixInvert(pixt1, pixt1);
pixDisplay(pixt1, 100, 500);
pixt2 = pixHolesByFilling(pixt1, 4);
pixd = pixConvertTo8(pixt2, FALSE);
pixDisplay(pixd, 0, 100);
pixSeedfillGray(pixd, pixmi, CONNECTIVITY);
pixInvert(pixd, pixd);
pixDisplay(pixmi, 500, 100);
pixDisplay(pixd, 1000, 100);
pixWrite("/tmp/junkpixm.png", pixmi, IFF_PNG);
pixWrite("/tmp/junkpixd.png", pixd, IFF_PNG);
#endif
#if 0
/* hole filling; use "hole-filler.png" */
pixd = pixInvert(NULL, pixm);
pixAddConstantGray(pixd, -50);
pixDisplay(pixd, 0, 100);
/* pixt1 = pixThresholdToBinary(pixd, 20);
pixDisplayWithTitle(pixt1, 600, 600, "pixt1", DFLAG); */
pixSeedfillGray(pixd, pixmi, CONNECTIVITY);
/* pixInvert(pixd, pixd); */
pixDisplay(pixmi, 500, 100);
pixDisplay(pixd, 1000, 100);
pixWrite("/tmp/junkpixm.png", pixmi, IFF_PNG);
pixWrite("/tmp/junkpixd.png", pixd, IFF_PNG);
#endif
#if 0
/* test in-place seedfill for speed */
pixd = pixClone(pixs);
startTimer();
pixSeedfillBinary(pixs, pixs, pixmi, CONNECTIVITY);
fprintf(stderr, "Filling rate: %7.4f Mpix/sec\n",
(size/1000000.) / stopTimer());
pixWrite(fileout, pixd, IFF_PNG);
pixOr(pixd, pixd, pixm);
pixWrite("/tmp/junkout1.png", pixd, IFF_PNG);
#endif
#if 0
/* test seedfill to dest for speed */
pixd = pixCreateTemplate(pixm);
startTimer();
for (i = 0; i < NTIMES; i++) {
pixSeedfillBinary(pixd, pixs, pixmi, CONNECTIVITY);
}
fprintf(stderr, "Filling rate: %7.4f Mpix/sec\n",
(size/1000000.) * NTIMES / stopTimer());
pixWrite(fileout, pixd, IFF_PNG);
pixOr(pixd, pixd, pixm);
pixWrite("/tmp/junkout1.png", pixd, IFF_PNG);
#endif
/* use same connectivity to compare with the result of the
* slow parallel operation */
#if 1
pixDestroy(&pixd);
pixd = pixSeedfillMorph(pixs, pixmi, 100, CONNECTIVITY);
pixOr(pixd, pixd, pixm);
pixWrite("/tmp/junkout2.png", pixd, IFF_PNG);
#endif
pixDestroy(&pixs);
pixDestroy(&pixm);
pixDestroy(&pixmi);
pixDestroy(&pixd);
return 0;
}
int main(int argc,
char **argv)
{
l_int32 i, j, k, w, h, w2, w4, w8, w16, w32, wpl;
l_int32 count1, count2, count3;
l_uint32 val32, val1, val2;
l_uint32 *data1, *line1, *data2, *line2;
void **lines1, **linet1, **linet2;
PIX *pixs, *pix1, *pix2;
L_REGPARAMS *rp;
if (regTestSetup(argc, argv, &rp))
return 1;
pixs = pixRead("feyn-fract.tif");
pixGetDimensions(pixs, &w, &h, NULL);
data1 = pixGetData(pixs);
wpl = pixGetWpl(pixs);
lines1 = pixGetLinePtrs(pixs, NULL);
/* Get timing for the 3 different methods */
startTimer();
for (k = 0; k < 10; k++) {
count1 = 0;
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
if (GET_DATA_BIT(lines1[i], j))
count1++;
}
}
}
fprintf(stderr, "Time with line ptrs = %5.3f sec, count1 = %d\n",
stopTimer(), count1);
startTimer();
for (k = 0; k < 10; k++) {
count2 = 0;
for (i = 0; i < h; i++) {
line1 = data1 + i * wpl;
for (j = 0; j < w; j++) {
if (l_getDataBit(line1, j))
count2++;
}
}
}
fprintf(stderr, "Time with l_get* = %5.3f sec, count2 = %d\n",
stopTimer(), count2);
startTimer();
for (k = 0; k < 10; k++) {
count3 = 0;
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
pixGetPixel(pixs, j, i, &val32);
count3 += val32;
}
}
}
fprintf(stderr, "Time with pixGetPixel() = %5.3f sec, count3 = %d\n",
stopTimer(), count3);
pix1 = pixCreateTemplate(pixs);
linet1 = pixGetLinePtrs(pix1, NULL);
pix2 = pixCreateTemplate(pixs);
data2 = pixGetData(pix2);
linet2 = pixGetLinePtrs(pix2, NULL);
/* ------------------------------------------------- */
/* Test different methods for 1 bpp */
/* ------------------------------------------------- */
count1 = 0;
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
val1 = GET_DATA_BIT(lines1[i], j);
count1 += val1;
if (val1) SET_DATA_BIT(linet1[i], j);
}
}
count2 = 0;
for (i = 0; i < h; i++) {
line1 = data1 + i * wpl;
line2 = data2 + i * wpl;
for (j = 0; j < w; j++) {
val2 = l_getDataBit(line1, j);
count2 += val2;
if (val2) l_setDataBit(line2, j);
}
}
CompareResults(pixs, pix1, pix2, count1, count2, "1 bpp", rp);
/* ------------------------------------------------- */
/* Test different methods for 2 bpp */
/* ------------------------------------------------- */
count1 = 0;
w2 = w / 2;
for (i = 0; i < h; i++) {
for (j = 0; j < w2; j++) {
val1 = GET_DATA_DIBIT(lines1[i], j);
count1 += val1;
val1 += 0xbbbbbbbc;
SET_DATA_DIBIT(linet1[i], j, val1);
}
}
count2 = 0;
for (i = 0; i < h; i++) {
line1 = data1 + i * wpl;
line2 = data2 + i * wpl;
for (j = 0; j < w2; j++) {
val2 = l_getDataDibit(line1, j);
count2 += val2;
val2 += 0xbbbbbbbc;
l_setDataDibit(line2, j, val2);
}
}
CompareResults(pixs, pix1, pix2, count1, count2, "2 bpp", rp);
/* ------------------------------------------------- */
/* Test different methods for 4 bpp */
/* ------------------------------------------------- */
count1 = 0;
w4 = w / 4;
for (i = 0; i < h; i++) {
for (j = 0; j < w4; j++) {
val1 = GET_DATA_QBIT(lines1[i], j);
count1 += val1;
val1 += 0xbbbbbbb0;
SET_DATA_QBIT(linet1[i], j, val1);
}
}
count2 = 0;
for (i = 0; i < h; i++) {
line1 = data1 + i * wpl;
line2 = data2 + i * wpl;
for (j = 0; j < w4; j++) {
val2 = l_getDataQbit(line1, j);
count2 += val2;
val2 += 0xbbbbbbb0;
l_setDataQbit(line2, j, val2);
}
}
CompareResults(pixs, pix1, pix2, count1, count2, "4 bpp", rp);
/* ------------------------------------------------- */
/* Test different methods for 8 bpp */
/* ------------------------------------------------- */
count1 = 0;
w8 = w / 8;
for (i = 0; i < h; i++) {
for (j = 0; j < w8; j++) {
val1 = GET_DATA_BYTE(lines1[i], j);
count1 += val1;
val1 += 0xbbbbbb00;
SET_DATA_BYTE(linet1[i], j, val1);
}
}
count2 = 0;
for (i = 0; i < h; i++) {
line1 = data1 + i * wpl;
line2 = data2 + i * wpl;
for (j = 0; j < w8; j++) {
val2 = l_getDataByte(line1, j);
count2 += val2;
val2 += 0xbbbbbb00;
l_setDataByte(line2, j, val2);
}
}
CompareResults(pixs, pix1, pix2, count1, count2, "8 bpp", rp);
/* ------------------------------------------------- */
/* Test different methods for 16 bpp */
/* ------------------------------------------------- */
count1 = 0;
w16 = w / 16;
for (i = 0; i < h; i++) {
for (j = 0; j < w16; j++) {
val1 = GET_DATA_TWO_BYTES(lines1[i], j);
count1 += val1;
val1 += 0xbbbb0000;
SET_DATA_TWO_BYTES(linet1[i], j, val1);
}
}
count2 = 0;
for (i = 0; i < h; i++) {
line1 = data1 + i * wpl;
line2 = data2 + i * wpl;
for (j = 0; j < w16; j++) {
val2 = l_getDataTwoBytes(line1, j);
count2 += val2;
val2 += 0xbbbb0000;
l_setDataTwoBytes(line2, j, val2);
}
}
CompareResults(pixs, pix1, pix2, count1, count2, "16 bpp", rp);
/* ------------------------------------------------- */
/* Test different methods for 32 bpp */
/* ------------------------------------------------- */
count1 = 0;
w32 = w / 32;
for (i = 0; i < h; i++) {
for (j = 0; j < w32; j++) {
val1 = GET_DATA_FOUR_BYTES(lines1[i], j);
count1 += val1 & 0xfff;
SET_DATA_FOUR_BYTES(linet1[i], j, val1);
}
}
count2 = 0;
for (i = 0; i < h; i++) {
line1 = data1 + i * wpl;
line2 = data2 + i * wpl;
for (j = 0; j < w32; j++) {
val2 = l_getDataFourBytes(line1, j);
count2 += val2 & 0xfff;
l_setDataFourBytes(line2, j, val2);
}
}
CompareResults(pixs, pix1, pix2, count1, count2, "32 bpp", rp);
pixDestroy(&pixs);
pixDestroy(&pix1);
pixDestroy(&pix2);
lept_free(lines1);
lept_free(linet1);
lept_free(linet2);
return regTestCleanup(rp);
}
void _display_frame(struct lib_hardsubx_ctx *ctx, AVFrame *frame, int width, int height, int timestamp)
{
// Debug: Display the frame after processing
PIX *im;
im = pixCreate(width,height,32);
PIX *hue_im = pixCreate(width,height,32);
int i,j;
for(i=0;i<height;i++)
{
for(j=0;j<width;j++)
{
int p=j*3+i*frame->linesize[0];
int r=frame->data[0][p];
int g=frame->data[0][p+1];
int b=frame->data[0][p+2];
pixSetRGBPixel(im,j,i,r,g,b);
float H,S,V;
rgb_to_hsv((float)r,(float)g,(float)b,&H,&S,&V);
if(abs(H-ctx->hue)<20)
{
pixSetRGBPixel(hue_im,j,i,r,g,b);
}
}
}
PIX *edge_im = pixCreate(width,height,8),*edge_im_2 = pixCreate(width,height,8);
edge_im = pixConvertRGBToGray(im,0.0,0.0,0.0);
edge_im = pixSobelEdgeFilter(edge_im, L_VERTICAL_EDGES);
edge_im = pixDilateGray(edge_im, 21, 1);
edge_im = pixThresholdToBinary(edge_im,50);
PIX *pixd = pixCreate(width,height,1);
pixSauvolaBinarize(pixConvertRGBToGray(hue_im,0.0,0.0,0.0), 15, 0.3, 1, NULL, NULL, NULL, &pixd);
edge_im_2 = pixConvertRGBToGray(hue_im,0.0,0.0,0.0);
edge_im_2 = pixDilateGray(edge_im_2, 5, 5);
PIX *feat_im = pixCreate(width,height,32);
for(i=3*(height/4);i<height;i++)
{
for(j=0;j<width;j++)
{
unsigned int p1,p2,p3,p4;
pixGetPixel(edge_im,j,i,&p1);
pixGetPixel(pixd,j,i,&p2);
// pixGetPixel(hue_im,j,i,&p3);
pixGetPixel(edge_im_2,j,i,&p4);
if(p1==0&&p2==0&&p4>0)//if(p4>0&&p1==0)//if(p2==0&&p1==0&&p3>0)
{
pixSetRGBPixel(feat_im,j,i,255,255,255);
}
}
}
char *txt=NULL;
// txt = get_ocr_text_simple(ctx, feat_im);
// txt=get_ocr_text_wordwise_threshold(ctx, feat_im, ctx->conf_thresh);
// if(txt != NULL)printf("%s\n", txt);
pixDestroy(&im);
pixDestroy(&edge_im);
pixDestroy(&feat_im);
pixDestroy(&edge_im_2);
pixDestroy(&pixd);
}
/*!
* pixSearchBinaryMaze()
*
* Input: pixs (1 bpp, maze)
* xi, yi (beginning point; use same initial point
* that was used to generate the maze)
* xf, yf (end point, or close to it)
* &ppixd (<optional return> maze with path illustrated, or
* if no path possible, the part of the maze
* that was searched)
* Return: pta (shortest path), or null if either no path
* exists or on error
*
* Notes:
* (1) Because of the overhead in calling pixGetPixel() and
* pixSetPixel(), we have used raster line pointers and the
* GET_DATA* and SET_DATA* macros for many of the pix accesses.
* (2) Commentary:
* The goal is to find the shortest path between beginning and
* end points, without going through walls, and there are many
* ways to solve this problem.
* We use a queue to implement a breadth-first search. Two auxiliary
* "image" data structures can be used: one to mark the visited
* pixels and one to give the direction to the parent for each
* visited pixels. The first structure is used to avoid putting
* pixels on the queue more than once, and the second is used
* for retracing back to the origin, like the breadcrumbs in
* Hansel and Gretel. Each pixel taken off the queue is destroyed
* after it is used to locate the allowed neighbors. In fact,
* only one distance image is required, if you initialize it
* to some value that signifies "not yet visited." (We use
* a binary image for marking visited pixels because it is clearer.)
* This method for a simple search of a binary maze is implemented in
* searchBinaryMaze().
* An alternative method would store the (manhattan) distance
* from the start point with each pixel on the queue. The children
* of each pixel get a distance one larger than the parent. These
* values can be stored in an auxiliary distance map image
* that is constructed simultaneously with the search. Once the
* end point is reached, the distance map is used to backtrack
* along a minimum path. There may be several equal length
* minimum paths, any one of which can be chosen this way.
*/
PTA *
pixSearchBinaryMaze(PIX *pixs,
l_int32 xi,
l_int32 yi,
l_int32 xf,
l_int32 yf,
PIX **ppixd)
{
l_int32 i, j, x, y, w, h, d, found;
l_uint32 val, rpixel, gpixel, bpixel;
void **lines1, **linem1, **linep8, **lined32;
MAZEEL *el, *elp;
PIX *pixd; /* the shortest path written on the maze image */
PIX *pixm; /* for bookkeeping, to indicate pixels already visited */
PIX *pixp; /* for bookkeeping, to indicate direction to parent */
L_QUEUE *lq;
PTA *pta;
PROCNAME("pixSearchBinaryMaze");
if (ppixd) *ppixd = NULL;
if (!pixs)
return (PTA *)ERROR_PTR("pixs not defined", procName, NULL);
pixGetDimensions(pixs, &w, &h, &d);
if (d != 1)
return (PTA *)ERROR_PTR("pixs not 1 bpp", procName, NULL);
if (xi <= 0 || xi >= w)
return (PTA *)ERROR_PTR("xi not valid", procName, NULL);
if (yi <= 0 || yi >= h)
return (PTA *)ERROR_PTR("yi not valid", procName, NULL);
pixGetPixel(pixs, xi, yi, &val);
if (val != 0)
return (PTA *)ERROR_PTR("(xi,yi) not bg pixel", procName, NULL);
pixd = NULL;
pta = NULL;
/* Find a bg pixel near input point (xf, yf) */
localSearchForBackground(pixs, &xf, &yf, 5);
#if DEBUG_MAZE
fprintf(stderr, "(xi, yi) = (%d, %d), (xf, yf) = (%d, %d)\n",
xi, yi, xf, yf);
#endif /* DEBUG_MAZE */
pixm = pixCreate(w, h, 1); /* initialized to OFF */
pixp = pixCreate(w, h, 8); /* direction to parent stored as enum val */
lines1 = pixGetLinePtrs(pixs, NULL);
linem1 = pixGetLinePtrs(pixm, NULL);
linep8 = pixGetLinePtrs(pixp, NULL);
lq = lqueueCreate(0);
/* Prime the queue with the first pixel; it is OFF */
el = mazeelCreate(xi, yi, 0); /* don't need direction here */
pixSetPixel(pixm, xi, yi, 1); /* mark visited */
lqueueAdd(lq, el);
/* Fill up the pix storing directions to parents,
* stopping when we hit the point (xf, yf) */
found = FALSE;
while (lqueueGetCount(lq) > 0) {
elp = (MAZEEL *)lqueueRemove(lq);
x = elp->x;
y = elp->y;
if (x == xf && y == yf) {
found = TRUE;
FREE(elp);
break;
}
if (x > 0) { /* check to west */
val = GET_DATA_BIT(linem1[y], x - 1);
if (val == 0) { /* not yet visited */
SET_DATA_BIT(linem1[y], x - 1); /* mark visited */
val = GET_DATA_BIT(lines1[y], x - 1);
if (val == 0) { /* bg, not a wall */
SET_DATA_BYTE(linep8[y], x - 1, DIR_EAST); /* parent E */
el = mazeelCreate(x - 1, y, 0);
lqueueAdd(lq, el);
}
}
}
if (y > 0) { /* check north */
val = GET_DATA_BIT(linem1[y - 1], x);
if (val == 0) { /* not yet visited */
SET_DATA_BIT(linem1[y - 1], x); /* mark visited */
val = GET_DATA_BIT(lines1[y - 1], x);
if (val == 0) { /* bg, not a wall */
SET_DATA_BYTE(linep8[y - 1], x, DIR_SOUTH); /* parent S */
el = mazeelCreate(x, y - 1, 0);
lqueueAdd(lq, el);
}
}
}
if (x < w - 1) { /* check east */
val = GET_DATA_BIT(linem1[y], x + 1);
if (val == 0) { /* not yet visited */
SET_DATA_BIT(linem1[y], x + 1); /* mark visited */
val = GET_DATA_BIT(lines1[y], x + 1);
if (val == 0) { /* bg, not a wall */
SET_DATA_BYTE(linep8[y], x + 1, DIR_WEST); /* parent W */
el = mazeelCreate(x + 1, y, 0);
lqueueAdd(lq, el);
}
}
}
if (y < h - 1) { /* check south */
val = GET_DATA_BIT(linem1[y + 1], x);
if (val == 0) { /* not yet visited */
SET_DATA_BIT(linem1[y + 1], x); /* mark visited */
val = GET_DATA_BIT(lines1[y + 1], x);
if (val == 0) { /* bg, not a wall */
SET_DATA_BYTE(linep8[y + 1], x, DIR_NORTH); /* parent N */
el = mazeelCreate(x, y + 1, 0);
lqueueAdd(lq, el);
}
}
}
FREE(elp);
}
lqueueDestroy(&lq, TRUE);
pixDestroy(&pixm);
FREE(linem1);
if (ppixd) {
pixd = pixUnpackBinary(pixs, 32, 1);
*ppixd = pixd;
}
composeRGBPixel(255, 0, 0, &rpixel); /* start point */
composeRGBPixel(0, 255, 0, &gpixel);
composeRGBPixel(0, 0, 255, &bpixel); /* end point */
if (!found) {
L_INFO(" No path found", procName);
if (pixd) { /* paint all visited locations */
lined32 = pixGetLinePtrs(pixd, NULL);
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
val = GET_DATA_BYTE(linep8[i], j);
if (val != 0 && pixd)
SET_DATA_FOUR_BYTES(lined32[i], j, gpixel);
}
}
FREE(lined32);
}
}
else { /* write path onto pixd */
L_INFO(" Path found", procName);
pta = ptaCreate(0);
x = xf;
y = yf;
while (1) {
ptaAddPt(pta, x, y);
if (x == xi && y == yi)
break;
if (pixd)
pixSetPixel(pixd, x, y, gpixel);
pixGetPixel(pixp, x, y, &val);
if (val == DIR_NORTH)
y--;
else if (val == DIR_SOUTH)
y++;
else if (val == DIR_EAST)
x++;
else if (val == DIR_WEST)
x--;
}
}
if (pixd) {
pixSetPixel(pixd, xi, yi, rpixel);
pixSetPixel(pixd, xf, yf, bpixel);
}
pixDestroy(&pixp);
FREE(lines1);
FREE(linep8);
return pta;
}
/*!
* generateBinaryMaze()
*
* Input: w, h (size of maze)
* xi, yi (initial location)
* wallps (probability that a pixel to the side is ON)
* ranis (ratio of prob that pixel in forward direction
* is a wall to the probability that pixel in
* side directions is a wall)
* Return: pix, or null on error
*
* Notes:
* (1) We have two input probability factors that determine the
* density of walls and average length of straight passages.
* When ranis < 1.0, you are more likely to generate a wall
* to the side than going forward. Enter 0.0 for either if
* you want to use the default values.
* (2) This is a type of percolation problem, and exhibits
* different phases for different parameters wallps and ranis.
* For larger values of these parameters, regions in the maze
* are not explored because the maze generator walls them
* off and cannot get through. The boundary between the
* two phases in this two-dimensional parameter space goes
* near these values:
* wallps ranis
* 0.35 1.00
* 0.40 0.85
* 0.45 0.70
* 0.50 0.50
* 0.55 0.40
* 0.60 0.30
* 0.65 0.25
* 0.70 0.19
* 0.75 0.15
* 0.80 0.11
* (3) Because here is a considerable amount of overhead in calling
* pixGetPixel() and pixSetPixel(), this function can be sped
* up with little effort using raster line pointers and the
* GET_DATA* and SET_DATA* macros.
*/
PIX *
generateBinaryMaze(l_int32 w,
l_int32 h,
l_int32 xi,
l_int32 yi,
l_float32 wallps,
l_float32 ranis)
{
l_int32 x, y, dir;
l_uint32 val;
l_float32 frand, wallpf, testp;
MAZEEL *el, *elp;
PIX *pixd; /* the destination maze */
PIX *pixm; /* for bookkeeping, to indicate pixels already visited */
L_QUEUE *lq;
/* On Windows, seeding is apparently necessary to get decent mazes.
* Windows rand() returns a value up to 2^15 - 1, whereas unix
* rand() returns a value up to 2^31 - 1. Therefore the generated
* mazes will differ on the two platforms. */
#ifdef _WIN32
srand(28*333);
#endif /* _WIN32 */
if (w < MIN_MAZE_WIDTH)
w = MIN_MAZE_WIDTH;
if (h < MIN_MAZE_HEIGHT)
h = MIN_MAZE_HEIGHT;
if (xi <= 0 || xi >= w)
xi = w / 6;
if (yi <= 0 || yi >= h)
yi = h / 5;
if (wallps < 0.05 || wallps > 0.95)
wallps = DEFAULT_WALL_PROBABILITY;
if (ranis < 0.05 || ranis > 1.0)
ranis = DEFAULT_ANISOTROPY_RATIO;
wallpf = wallps * ranis;
#if DEBUG_MAZE
fprintf(stderr, "(w, h) = (%d, %d), (xi, yi) = (%d, %d)\n", w, h, xi, yi);
fprintf(stderr, "Using: prob(wall) = %7.4f, anisotropy factor = %7.4f\n",
wallps, ranis);
#endif /* DEBUG_MAZE */
/* These are initialized to OFF */
pixd = pixCreate(w, h, 1);
pixm = pixCreate(w, h, 1);
lq = lqueueCreate(0);
/* Prime the queue with the first pixel; it is OFF */
el = mazeelCreate(xi, yi, START_LOC);
pixSetPixel(pixm, xi, yi, 1); /* mark visited */
lqueueAdd(lq, el);
/* While we're at it ... */
while (lqueueGetCount(lq) > 0) {
elp = (MAZEEL *)lqueueRemove(lq);
x = elp->x;
y = elp->y;
dir = elp->dir;
if (x > 0) { /* check west */
pixGetPixel(pixm, x - 1, y, &val);
if (val == 0) { /* not yet visited */
pixSetPixel(pixm, x - 1, y, 1); /* mark visited */
frand = (l_float32)rand() / (l_float32)RAND_MAX;
testp = wallps;
if (dir == DIR_WEST)
testp = wallpf;
if (frand <= testp) { /* make it a wall */
pixSetPixel(pixd, x - 1, y, 1);
}
else { /* not a wall */
el = mazeelCreate(x - 1, y, DIR_WEST);
lqueueAdd(lq, el);
}
}
}
if (y > 0) { /* check north */
pixGetPixel(pixm, x, y - 1, &val);
if (val == 0) { /* not yet visited */
pixSetPixel(pixm, x, y - 1, 1); /* mark visited */
frand = (l_float32)rand() / (l_float32)RAND_MAX;
testp = wallps;
if (dir == DIR_NORTH)
testp = wallpf;
if (frand <= testp) { /* make it a wall */
pixSetPixel(pixd, x, y - 1, 1);
}
else { /* not a wall */
el = mazeelCreate(x, y - 1, DIR_NORTH);
lqueueAdd(lq, el);
}
}
}
if (x < w - 1) { /* check east */
pixGetPixel(pixm, x + 1, y, &val);
if (val == 0) { /* not yet visited */
pixSetPixel(pixm, x + 1, y, 1); /* mark visited */
frand = (l_float32)rand() / (l_float32)RAND_MAX;
testp = wallps;
if (dir == DIR_EAST)
testp = wallpf;
if (frand <= testp) { /* make it a wall */
pixSetPixel(pixd, x + 1, y, 1);
}
else { /* not a wall */
el = mazeelCreate(x + 1, y, DIR_EAST);
lqueueAdd(lq, el);
}
}
}
if (y < h - 1) { /* check south */
pixGetPixel(pixm, x, y + 1, &val);
if (val == 0) { /* not yet visited */
pixSetPixel(pixm, x, y + 1, 1); /* mark visited */
frand = (l_float32)rand() / (l_float32)RAND_MAX;
testp = wallps;
if (dir == DIR_SOUTH)
testp = wallpf;
if (frand <= testp) { /* make it a wall */
pixSetPixel(pixd, x, y + 1, 1);
}
else { /* not a wall */
el = mazeelCreate(x, y + 1, DIR_SOUTH);
lqueueAdd(lq, el);
}
}
}
FREE(elp);
}
lqueueDestroy(&lq, TRUE);
pixDestroy(&pixm);
return pixd;
}
/*!
* wshedApply()
*
* Input: wshed (generated from wshedCreate())
* Return: 0 if OK, 1 on error
*
* Iportant note:
* (1) This is buggy. It seems to locate watersheds that are
* duplicates. The watershed extraction after complete fill
* grabs some regions belonging to existing watersheds.
* See prog/watershedtest.c for testing.
*/
l_int32
wshedApply(L_WSHED *wshed) {
char two_new_watersheds[] = "Two new watersheds";
char seed_absorbed_into_seeded_basin[] = "Seed absorbed into seeded basin";
char one_new_watershed_label[] = "One new watershed (label)";
char one_new_watershed_index[] = "One new watershed (index)";
char minima_absorbed_into_seeded_basin[] =
"Minima absorbed into seeded basin";
char minima_absorbed_by_filler_or_another[] =
"Minima absorbed by filler or another";
l_int32 nseeds, nother, nboth, arraysize;
l_int32 i, j, val, x, y, w, h, index, mindepth;
l_int32 imin, imax, jmin, jmax, cindex, clabel, nindex;
l_int32 hindex, hlabel, hmin, hmax, minhindex, maxhindex;
l_int32 *lut;
l_uint32 ulabel, uval;
void **lines8, **linelab32;
NUMA *nalut, *nalevels, *nash, *namh, *nasi;
NUMA **links;
L_HEAP *lh;
PIX *pixmin, *pixsd;
PIXA *pixad;
L_STACK *rstack;
PTA *ptas, *ptao;
PROCNAME("wshedApply");
if (!wshed)
return ERROR_INT("wshed not defined", procName, 1);
/* ------------------------------------------------------------ *
* Initialize priority queue and pixlab with seeds and minima *
* ------------------------------------------------------------ */
lh = lheapCreate(0, L_SORT_INCREASING); /* remove lowest values first */
rstack = lstackCreate(0); /* for reusing the WSPixels */
pixGetDimensions(wshed->pixs, &w, &h, NULL);
lines8 = wshed->lines8; /* wshed owns this */
linelab32 = wshed->linelab32; /* ditto */
/* Identify seed (marker) pixels, 1 for each c.c. in pixm */
pixSelectMinInConnComp(wshed->pixs, wshed->pixm, &ptas, &nash);
pixsd = pixGenerateFromPta(ptas, w, h);
nseeds = ptaGetCount(ptas);
for (i = 0; i < nseeds; i++) {
ptaGetIPt(ptas, i, &x, &y);
uval = GET_DATA_BYTE(lines8[y], x);
pushWSPixel(lh, rstack, (l_int32) uval, x, y, i);
}
wshed->ptas = ptas;
nasi = numaMakeConstant(1, nseeds); /* indicator array */
wshed->nasi = nasi;
wshed->nash = nash;
wshed->nseeds = nseeds;
/* Identify minima that are not seeds. Use these 4 steps:
* (1) Get the local minima, which can have components
* of arbitrary size. This will be a clipping mask.
* (2) Get the image of the actual seeds (pixsd)
* (3) Remove all elements of the clipping mask that have a seed.
* (4) Shrink each of the remaining elements of the minima mask
* to a single pixel. */
pixLocalExtrema(wshed->pixs, 200, 0, &pixmin, NULL);
pixRemoveSeededComponents(pixmin, pixsd, pixmin, 8, 2);
pixSelectMinInConnComp(wshed->pixs, pixmin, &ptao, &namh);
nother = ptaGetCount(ptao);
for (i = 0; i < nother; i++) {
ptaGetIPt(ptao, i, &x, &y);
uval = GET_DATA_BYTE(lines8[y], x);
pushWSPixel(lh, rstack, (l_int32) uval, x, y, nseeds + i);
}
wshed->namh = namh;
/* ------------------------------------------------------------ *
* Initialize merging lookup tables *
* ------------------------------------------------------------ */
/* nalut should always give the current after-merging index.
* links are effectively backpointers: they are numas associated with
* a dest index of all indices in nalut that point to that index. */
mindepth = wshed->mindepth;
nboth = nseeds + nother;
arraysize = 2 * nboth;
wshed->arraysize = arraysize;
nalut = numaMakeSequence(0, 1, arraysize);
lut = numaGetIArray(nalut);
wshed->lut = lut; /* wshed owns this */
links = (NUMA **) CALLOC(arraysize, sizeof(NUMA * ));
wshed->links = links; /* wshed owns this */
nindex = nseeds + nother; /* the next unused index value */
/* ------------------------------------------------------------ *
* Fill the basins, using the priority queue *
* ------------------------------------------------------------ */
pixad = pixaCreate(nseeds);
wshed->pixad = pixad; /* wshed owns this */
nalevels = numaCreate(nseeds);
wshed->nalevels = nalevels; /* wshed owns this */
L_INFO("nseeds = %d, nother = %d\n", procName, nseeds, nother);
while (lheapGetCount(lh) > 0) {
popWSPixel(lh, rstack, &val, &x, &y, &index);
/* fprintf(stderr, "x = %d, y = %d, index = %d\n", x, y, index); */
ulabel = GET_DATA_FOUR_BYTES(linelab32[y], x);
if (ulabel == MAX_LABEL_VALUE)
clabel = ulabel;
else
clabel = lut[ulabel];
cindex = lut[index];
if (clabel == cindex) continue; /* have already seen this one */
if (clabel == MAX_LABEL_VALUE) { /* new one; assign index and try to
* propagate to all neighbors */
SET_DATA_FOUR_BYTES(linelab32[y], x, cindex);
imin = L_MAX(0, y - 1);
imax = L_MIN(h - 1, y + 1);
jmin = L_MAX(0, x - 1);
jmax = L_MIN(w - 1, x + 1);
for (i = imin; i <= imax; i++) {
for (j = jmin; j <= jmax; j++) {
if (i == y && j == x) continue;
uval = GET_DATA_BYTE(lines8[i], j);
pushWSPixel(lh, rstack, (l_int32) uval, j, i, cindex);
}
}
} else { /* pixel is already labeled (differently); must resolve */
/* If both indices are seeds, check if the min height is
* greater than mindepth. If so, we have two new watersheds;
* locate them and assign to both regions a new index
* for further waterfill. If not, absorb the shallower
* watershed into the deeper one and continue filling it. */
pixGetPixel(pixsd, x, y, &uval);
if (clabel < nseeds && cindex < nseeds) {
wshedGetHeight(wshed, val, clabel, &hlabel);
wshedGetHeight(wshed, val, cindex, &hindex);
hmin = L_MIN(hlabel, hindex);
hmax = L_MAX(hlabel, hindex);
if (hmin == hmax) {
hmin = hlabel;
hmax = hindex;
}
if (wshed->debug) {
fprintf(stderr, "clabel,hlabel = %d,%d\n", clabel, hlabel);
fprintf(stderr, "hmin = %d, hmax = %d\n", hmin, hmax);
fprintf(stderr, "cindex,hindex = %d,%d\n", cindex, hindex);
if (hmin < mindepth)
fprintf(stderr, "Too shallow!\n");
}
if (hmin >= mindepth) {
debugWshedMerge(wshed, two_new_watersheds,
x, y, clabel, cindex);
wshedSaveBasin(wshed, cindex, val - 1);
wshedSaveBasin(wshed, clabel, val - 1);
numaSetValue(nasi, cindex, 0);
numaSetValue(nasi, clabel, 0);
if (wshed->debug) fprintf(stderr, "nindex = %d\n", nindex);
debugPrintLUT(lut, nindex, wshed->debug);
mergeLookup(wshed, clabel, nindex);
debugPrintLUT(lut, nindex, wshed->debug);
mergeLookup(wshed, cindex, nindex);
debugPrintLUT(lut, nindex, wshed->debug);
nindex++;
} else /* extraneous seed within seeded basin; absorb */ {
debugWshedMerge(wshed, seed_absorbed_into_seeded_basin,
x, y, clabel, cindex);
}
maxhindex = clabel; /* TODO: is this part of above 'else'? */
minhindex = cindex;
if (hindex > hlabel) {
maxhindex = cindex;
minhindex = clabel;
}
mergeLookup(wshed, minhindex, maxhindex);
} else if (clabel < nseeds && cindex >= nboth) {
/* If one index is a seed and the other is a merge of
* 2 watersheds, generate a single watershed. */
debugWshedMerge(wshed, one_new_watershed_label,
x, y, clabel, cindex);
wshedSaveBasin(wshed, clabel, val - 1);
numaSetValue(nasi, clabel, 0);
mergeLookup(wshed, clabel, cindex);
} else if (cindex < nseeds && clabel >= nboth) {
debugWshedMerge(wshed, one_new_watershed_index,
x, y, clabel, cindex);
wshedSaveBasin(wshed, cindex, val - 1);
numaSetValue(nasi, cindex, 0);
mergeLookup(wshed, cindex, clabel);
} else if (clabel < nseeds) { /* cindex from minima; absorb */
/* If one index is a seed and the other is from a minimum,
* merge the minimum wshed into the seed wshed. */
debugWshedMerge(wshed, minima_absorbed_into_seeded_basin,
x, y, clabel, cindex);
mergeLookup(wshed, cindex, clabel);
} else if (cindex < nseeds) { /* clabel from minima; absorb */
debugWshedMerge(wshed, minima_absorbed_into_seeded_basin,
x, y, clabel, cindex);
mergeLookup(wshed, clabel, cindex);
} else { /* If neither index is a seed, just merge */
debugWshedMerge(wshed, minima_absorbed_by_filler_or_another,
x, y, clabel, cindex);
mergeLookup(wshed, clabel, cindex);
}
}
}
#if 0
/* Use the indicator array to save any watersheds that fill
* to the maximum value. This seems to screw things up! */
for (i = 0; i < nseeds; i++) {
numaGetIValue(nasi, i, &ival);
if (ival == 1) {
wshedSaveBasin(wshed, lut[i], val - 1);
numaSetValue(nasi, i, 0);
}
}
#endif
numaDestroy(&nalut);
pixDestroy(&pixmin);
pixDestroy(&pixsd);
ptaDestroy(&ptao);
lheapDestroy(&lh, TRUE);
lstackDestroy(&rstack, TRUE);
return 0;
}
char* _process_frame_tickertext(struct lib_hardsubx_ctx *ctx, AVFrame *frame, int width, int height, int index)
{
PIX *im;
PIX *edge_im;
PIX *lum_im;
PIX *feat_im;
char *subtitle_text=NULL;
im = pixCreate(width,height,32);
lum_im = pixCreate(width,height,32);
feat_im = pixCreate(width,height,32);
int i,j;
for(i=(92*height)/100;i<height;i++)
{
for(j=0;j<width;j++)
{
int p=j*3+i*frame->linesize[0];
int r=frame->data[0][p];
int g=frame->data[0][p+1];
int b=frame->data[0][p+2];
pixSetRGBPixel(im,j,i,r,g,b);
float L,A,B;
rgb_to_lab((float)r,(float)g,(float)b,&L,&A,&B);
if(L > ctx->lum_thresh)
pixSetRGBPixel(lum_im,j,i,255,255,255);
else
pixSetRGBPixel(lum_im,j,i,0,0,0);
}
}
//Handle the edge image
edge_im = pixCreate(width,height,8);
edge_im = pixConvertRGBToGray(im,0.0,0.0,0.0);
edge_im = pixSobelEdgeFilter(edge_im, L_VERTICAL_EDGES);
edge_im = pixDilateGray(edge_im, 21, 11);
edge_im = pixThresholdToBinary(edge_im,50);
for(i=92*(height/100);i<height;i++)
{
for(j=0;j<width;j++)
{
unsigned int p1,p2,p3;
pixGetPixel(edge_im,j,i,&p1);
// pixGetPixel(pixd,j,i,&p2);
pixGetPixel(lum_im,j,i,&p3);
if(p1==0&&p3>0)
pixSetRGBPixel(feat_im,j,i,255,255,255);
else
pixSetRGBPixel(feat_im,j,i,0,0,0);
}
}
// Tesseract OCR for the ticker text here
subtitle_text = get_ocr_text_simple(ctx, lum_im);
char write_path[100];
sprintf(write_path,"./lum_im%04d.jpg",index);
pixWrite(write_path,lum_im,IFF_JFIF_JPEG);
sprintf(write_path,"./im%04d.jpg",index);
pixWrite(write_path,im,IFF_JFIF_JPEG);
pixDestroy(&lum_im);
pixDestroy(&im);
pixDestroy(&edge_im);
pixDestroy(&feat_im);
return subtitle_text;
}
/*!
* pixSearchGrayMaze()
*
* Input: pixs (1 bpp, maze)
* xi, yi (beginning point; use same initial point
* that was used to generate the maze)
* xf, yf (end point, or close to it)
* &ppixd (<optional return> maze with path illustrated, or
* if no path possible, the part of the maze
* that was searched)
* Return: pta (shortest path), or null if either no path
* exists or on error
*
* Commentary:
* Consider first a slight generalization of the binary maze
* search problem. Suppose that you can go through walls,
* but the cost is higher (say, an increment of 3 to go into
* a wall pixel rather than 1)? You're still trying to find
* the shortest path. One way to do this is with an ordered
* queue, and a simple way to visualize an ordered queue is as
* a set of stacks, each stack being marked with the distance
* of each pixel in the stack from the start. We place the
* start pixel in stack 0, pop it, and process its 4 children.
* Each pixel is given a distance that is incremented from that
* of its parent (0 in this case), depending on if it is a wall
* pixel or not. That value may be recorded on a distance map,
* according to the algorithm below. For children of the first
* pixel, those not on a wall go in stack 1, and wall
* children go in stack 3. Stack 0 being emptied, the process
* then continues with pixels being popped from stack 1.
* Here is the algorithm for each child pixel. The pixel's
* distance value, were it to be placed on a stack, is compared
* with the value for it that is on the distance map. There
* are three possible cases:
* (1) If the pixel has not yet been registered, it is pushed
* on its stack and the distance is written to the map.
* (2) If it has previously been registered with a higher distance,
* the distance on the map is relaxed to that of the
* current pixel, which is then placed on its stack.
* (3) If it has previously been registered with an equal
* or lower value, the pixel is discarded.
* The pixels are popped and processed successively from
* stack 1, and when stack 1 is empty, popping starts on stack 2.
* This continues until the destination pixel is popped off
* a stack. The minimum path is then derived from the distance map,
* going back from the end point as before. This is just Dijkstra's
* algorithm for a directed graph; here, the underlying graph
* (consisting of the pixels and four edges connecting each pixel
* to its 4-neighbor) is a special case of a directed graph, where
* each edge is bi-directional. The implementation of this generalized
* maze search is left as an exercise to the reader.
*
* Let's generalize a bit further. Suppose the "maze" is just
* a grayscale image -- think of it as an elevation map. The cost
* of moving on this surface depends on the height, or the gradient,
* or whatever you want. All that is required is that the cost
* is specified and non-negative on each link between adjacent
* pixels. Now the problem becomes: find the least cost path
* moving on this surface between two specified end points.
* For example, if the cost across an edge between two pixels
* depends on the "gradient", you can use:
* cost = 1 + L_ABS(deltaV)
* where deltaV is the difference in value between two adjacent
* pixels. If the costs are all integers, we can still use an array
* of stacks to avoid ordering the queue (e.g., by using a heap sort.)
* This is a neat problem, because you don't even have to build a
* maze -- you can can use it on any grayscale image!
*
* Rather than using an array of stacks, a more practical
* approach is to implement with a priority queue, which is
* a queue that is sorted so that the elements with the largest
* (or smallest) key values always come off first. The
* priority queue is efficiently implemented as a heap, and
* this is how we do it. Suppose you run the algorithm
* using a priority queue, doing the bookkeeping with an
* auxiliary image data structure that saves the distance of
* each pixel put on the queue as before, according to the method
* described above. We implement it as a 2-way choice by
* initializing the distance array to a large value and putting
* a pixel on the queue if its distance is less than the value
* found on the array. When you finally pop the end pixel from
* the queue, you're done, and you can trace the path backward,
* either always going downhill or using an auxiliary image to
* give you the direction to go at each step. This is implemented
* here in searchGrayMaze().
*
* Do we really have to use a sorted queue? Can we solve this
* generalized maze with an unsorted queue of pixels? (Or even
* an unsorted stack, doing a depth-first search (DFS)?)
* Consider a different algorithm for this generalized maze, where
* we travel again breadth first, but this time use a single,
* unsorted queue. An auxiliary image is used as before to
* store the distances and to determine if pixels get pushed
* on the stack or dropped. As before, we must allow pixels
* to be revisited, with relaxation of the distance if a shorter
* path arrives later. As a result, we will in general have
* multiple instances of the same pixel on the stack with different
* distances. However, because the queue is not ordered, some of
* these pixels will be popped when another instance with a lower
* distance is still on the stack. Here, we're just popping them
* in the order they go on, rather than setting up a priority
* based on minimum distance. Thus, unlike the priority queue,
* when a pixel is popped we have to check the distance map to
* see if a pixel with a lower distance has been put on the queue,
* and, if so, we discard the pixel we just popped. So the
* "while" loop looks like this:
* - pop a pixel from the queue
* - check its distance against the distance stored in the
* distance map; if larger, discard
* - otherwise, for each of its neighbors:
* - compute its distance from the start pixel
* - compare this distance with that on the distance map:
* - if the distance map value higher, relax the distance
* and push the pixel on the queue
* - if the distance map value is lower, discard the pixel
*
* How does this loop terminate? Before, with an ordered queue,
* it terminates when you pop the end pixel. But with an unordered
* queue (or stack), the first time you hit the end pixel, the
* distance is not guaranteed to be correct, because the pixels
* along the shortest path may not have yet been visited and relaxed.
* Because the shortest path can theoretically go anywhere,
* we must keep going. How do we know when to stop? Dijkstra
* uses an ordered queue to systematically remove nodes from
* further consideration. (Each time a pixel is popped, we're
* done with it; it's "finalized" in the Dijkstra sense because
* we know the shortest path to it.) However, with an unordered
* queue, the brute force answer is: stop when the queue
* (or stack) is empty, because then every pixel in the image
* has been assigned its minimum "distance" from the start pixel.
*
* This is similar to the situation when you use a stack for the
* simpler uniform-step problem: with breadth-first search (BFS)
* the pixels on the queue are automatically ordered, so you are
* done when you locate the end pixel as a neighbor of a popped pixel;
* whereas depth-first search (DFS), using a stack, requires,
* in general, a search of every accessible pixel. Further, if
* a pixel is revisited with a smaller distance, that distance is
* recorded and the pixel is put on the stack again.
*
* But surely, you ask, can't we stop sooner? What if the
* start and end pixels are very close to each other?
* OK, suppose they are, and you have very high walls and a
* long snaking level path that is actually the minimum cost.
* That long path can wind back and forth across the entire
* maze many times before ending up at the end point, which
* could be just over a wall from the start. With the unordered
* queue, you very quickly get a high distance for the end
* pixel, which will be relaxed to the minimum distance only
* after all the pixels of the path have been visited and placed
* on the queue, multiple times for many of them. So that's the
* price for not ordering the queue!
*/
PTA *
pixSearchGrayMaze(PIX *pixs,
l_int32 xi,
l_int32 yi,
l_int32 xf,
l_int32 yf,
PIX **ppixd)
{
l_int32 x, y, w, h, d;
l_uint32 val, valr, vals, rpixel, gpixel, bpixel;
void **lines8, **liner32, **linep8;
l_int32 cost, dist, distparent, sival, sivals;
MAZEEL *el, *elp;
PIX *pixd; /* optionally plot the path on this RGB version of pixs */
PIX *pixr; /* for bookkeeping, to indicate the minimum distance */
/* to pixels already visited */
PIX *pixp; /* for bookkeeping, to indicate direction to parent */
L_HEAP *lh;
PTA *pta;
PROCNAME("pixSearchGrayMaze");
if (ppixd) *ppixd = NULL;
if (!pixs)
return (PTA *)ERROR_PTR("pixs not defined", procName, NULL);
pixGetDimensions(pixs, &w, &h, &d);
if (d != 8)
return (PTA *)ERROR_PTR("pixs not 8 bpp", procName, NULL);
if (xi <= 0 || xi >= w)
return (PTA *)ERROR_PTR("xi not valid", procName, NULL);
if (yi <= 0 || yi >= h)
return (PTA *)ERROR_PTR("yi not valid", procName, NULL);
pixd = NULL;
pta = NULL;
pixr = pixCreate(w, h, 32);
pixSetAll(pixr); /* initialize to max value */
pixp = pixCreate(w, h, 8); /* direction to parent stored as enum val */
lines8 = pixGetLinePtrs(pixs, NULL);
linep8 = pixGetLinePtrs(pixp, NULL);
liner32 = pixGetLinePtrs(pixr, NULL);
lh = lheapCreate(0, L_SORT_INCREASING); /* always remove closest pixels */
/* Prime the heap with the first pixel */
pixGetPixel(pixs, xi, yi, &val);
el = mazeelCreate(xi, yi, 0); /* don't need direction here */
el->distance = 0;
pixGetPixel(pixs, xi, yi, &val);
el->val = val;
pixSetPixel(pixr, xi, yi, 0); /* distance is 0 */
lheapAdd(lh, el);
/* Breadth-first search with priority queue (implemented by
a heap), labeling direction to parents in pixp and minimum
distance to visited pixels in pixr. Stop when we pull
the destination point (xf, yf) off the queue. */
while (lheapGetCount(lh) > 0) {
elp = (MAZEEL *)lheapRemove(lh);
if (!elp)
return (PTA *)ERROR_PTR("heap broken!!", procName, NULL);
x = elp->x;
y = elp->y;
if (x == xf && y == yf) { /* exit condition */
FREE(elp);
break;
}
distparent = (l_int32)elp->distance;
val = elp->val;
sival = val;
if (x > 0) { /* check to west */
vals = GET_DATA_BYTE(lines8[y], x - 1);
valr = GET_DATA_FOUR_BYTES(liner32[y], x - 1);
sivals = (l_int32)vals;
cost = 1 + L_ABS(sivals - sival); /* cost to move to this pixel */
dist = distparent + cost;
if (dist < valr) { /* shortest path so far to this pixel */
SET_DATA_FOUR_BYTES(liner32[y], x - 1, dist); /* new dist */
SET_DATA_BYTE(linep8[y], x - 1, DIR_EAST); /* parent to E */
el = mazeelCreate(x - 1, y, 0);
el->val = vals;
el->distance = dist;
lheapAdd(lh, el);
}
}
if (y > 0) { /* check north */
vals = GET_DATA_BYTE(lines8[y - 1], x);
valr = GET_DATA_FOUR_BYTES(liner32[y - 1], x);
sivals = (l_int32)vals;
cost = 1 + L_ABS(sivals - sival); /* cost to move to this pixel */
dist = distparent + cost;
if (dist < valr) { /* shortest path so far to this pixel */
SET_DATA_FOUR_BYTES(liner32[y - 1], x, dist); /* new dist */
SET_DATA_BYTE(linep8[y - 1], x, DIR_SOUTH); /* parent to S */
el = mazeelCreate(x, y - 1, 0);
el->val = vals;
el->distance = dist;
lheapAdd(lh, el);
}
}
if (x < w - 1) { /* check east */
vals = GET_DATA_BYTE(lines8[y], x + 1);
valr = GET_DATA_FOUR_BYTES(liner32[y], x + 1);
sivals = (l_int32)vals;
cost = 1 + L_ABS(sivals - sival); /* cost to move to this pixel */
dist = distparent + cost;
if (dist < valr) { /* shortest path so far to this pixel */
SET_DATA_FOUR_BYTES(liner32[y], x + 1, dist); /* new dist */
SET_DATA_BYTE(linep8[y], x + 1, DIR_WEST); /* parent to W */
el = mazeelCreate(x + 1, y, 0);
el->val = vals;
el->distance = dist;
lheapAdd(lh, el);
}
}
if (y < h - 1) { /* check south */
vals = GET_DATA_BYTE(lines8[y + 1], x);
valr = GET_DATA_FOUR_BYTES(liner32[y + 1], x);
sivals = (l_int32)vals;
cost = 1 + L_ABS(sivals - sival); /* cost to move to this pixel */
dist = distparent + cost;
if (dist < valr) { /* shortest path so far to this pixel */
SET_DATA_FOUR_BYTES(liner32[y + 1], x, dist); /* new dist */
SET_DATA_BYTE(linep8[y + 1], x, DIR_NORTH); /* parent to N */
el = mazeelCreate(x, y + 1, 0);
el->val = vals;
el->distance = dist;
lheapAdd(lh, el);
}
}
FREE(elp);
}
lheapDestroy(&lh, TRUE);
if (ppixd) {
pixd = pixConvert8To32(pixs);
*ppixd = pixd;
}
composeRGBPixel(255, 0, 0, &rpixel); /* start point */
composeRGBPixel(0, 255, 0, &gpixel);
composeRGBPixel(0, 0, 255, &bpixel); /* end point */
x = xf;
y = yf;
pta = ptaCreate(0);
while (1) { /* write path onto pixd */
ptaAddPt(pta, x, y);
if (x == xi && y == yi)
break;
if (pixd)
pixSetPixel(pixd, x, y, gpixel);
pixGetPixel(pixp, x, y, &val);
if (val == DIR_NORTH)
y--;
else if (val == DIR_SOUTH)
y++;
else if (val == DIR_EAST)
x++;
else if (val == DIR_WEST)
x--;
pixGetPixel(pixr, x, y, &val);
#if DEBUG_PATH
fprintf(stderr, "(x,y) = (%d, %d); dist = %d\n", x, y, val);
#endif /* DEBUG_PATH */
}
if (pixd) {
pixSetPixel(pixd, xi, yi, rpixel);
pixSetPixel(pixd, xf, yf, bpixel);
}
pixDestroy(&pixp);
pixDestroy(&pixr);
FREE(lines8);
FREE(linep8);
FREE(liner32);
return pta;
}
int main(int argc,
char **argv) {
char *infile;
l_int32 w, d, threshval, ival, newval;
l_uint32 val;
PIX *pixs, *pixg, *pixg2;
PIX *pix1, *pix2;
PIXA *pixa;
static char mainName[] = "binarize_set";
if (argc != 2)
return ERROR_INT(" Syntax: binarize_set infile", mainName, 1);
infile = argv[1];
pixa = pixaCreate(5);
pixs = pixRead(infile);
pixGetDimensions(pixs, &w, NULL, &d);
pixSaveTiled(pixs, pixa, 1.0, 1, 50, 32);
pixDisplay(pixs, 100, 0);
#if ALL
/* 1. Standard background normalization with a global threshold. */
pixg = pixConvertTo8(pixs, 0);
pix1 = pixBackgroundNorm(pixg, NULL, NULL, 10, 15, 100, 50, 255, 2, 2);
pix2 = pixThresholdToBinary(pix1, 160);
pixWrite("/tmp/binar1.png", pix2, IFF_PNG);
pixDisplay(pix2, 100, 0);
pixSaveTiled(pix2, pixa, 1.0, 1, 50, 32);
pixDestroy(&pixg);
pixDestroy(&pix1);
pixDestroy(&pix2);
#endif
#if ALL
/* 2. Background normalization followed by Otsu thresholding. Otsu
* binarization attempts to split the image into two roughly equal
* sets of pixels, and it does a very poor job when there are large
* amounts of dark background. By doing a background normalization
* first (to get the background near 255), we remove this problem.
* Then we use a modified Otsu to estimate the best global
* threshold on the normalized image. */
pixg = pixConvertTo8(pixs, 0);
pix1 = pixOtsuThreshOnBackgroundNorm(pixg, NULL, 10, 15, 100,
50, 255, 2, 2, 0.10, &threshval);
fprintf(stderr, "thresh val = %d\n", threshval);
pixSaveTiled(pix1, pixa, 1.0, 1, 50, 32);
pixWrite("/tmp/binar2.png", pix1, IFF_PNG);
pixDisplay(pix1, 100, 200);
pixDestroy(&pixg);
pixDestroy(&pix1);
#endif
#if ALL
/* 3. Background normalization with Otsu threshold estimation and
* masking for threshold selection. */
pixg = pixConvertTo8(pixs, 0);
pix1 = pixMaskedThreshOnBackgroundNorm(pixg, NULL, 10, 15, 100,
50, 2, 2, 0.10, &threshval);
fprintf(stderr, "thresh val = %d\n", threshval);
pixSaveTiled(pix1, pixa, 1.0, 1, 50, 32);
pixWrite("/tmp/binar3.png", pix1, IFF_PNG);
pixDisplay(pix1, 100, 400);
pixDestroy(&pixg);
pixDestroy(&pix1);
#endif
#if ALL
/* 4. Background normalization followed by Sauvola binarization */
if (d == 32)
pixg = pixConvertRGBToGray(pixs, 0.2, 0.7, 0.1);
else
pixg = pixConvertTo8(pixs, 0);
pixg2 = pixContrastNorm(NULL, pixg, 20, 20, 130, 2, 2);
pixSauvolaBinarizeTiled(pixg2, 25, 0.40, 1, 1, NULL, &pix1);
pixSaveTiled(pix1, pixa, 1.0, 1, 50, 32);
pixWrite("/tmp/binar4.png", pix1, IFF_PNG);
pixDisplay(pix1, 100, 600);
pixDestroy(&pixg);
pixDestroy(&pixg2);
pixDestroy(&pix1);
#endif
#if ALL
/* 5. Contrast normalization followed by background normalization, and
* thresholding. */
if (d == 32)
pixg = pixConvertRGBToGray(pixs, 0.2, 0.7, 0.1);
else
pixg = pixConvertTo8(pixs, 0);
pixOtsuAdaptiveThreshold(pixg, 5000, 5000, 0, 0, 0.1, &pix1, NULL);
pixGetPixel(pix1, 0, 0, &val);
ival = (l_int32) val;
newval = ival + (l_int32)(0.6 * (110 - ival));
fprintf(stderr, "th1 = %d, th2 = %d\n", ival, newval);
pixDestroy(&pix1);
pixContrastNorm(pixg, pixg, 50, 50, 130, 2, 2);
pixg2 = pixBackgroundNorm(pixg, NULL, NULL, 20, 20, 70, 40, 200, 2, 2);
ival = L_MIN(ival, 110);
pix1 = pixThresholdToBinary(pixg2, ival);
pixSaveTiled(pix1, pixa, 1.0, 1, 50, 32);
pixWrite("/tmp/binar5.png", pix1, IFF_PNG);
pixDisplay(pix1, 100, 800);
pixDestroy(&pixg);
pixDestroy(&pixg2);
pixDestroy(&pix1);
#endif
pix1 = pixaDisplayTiledInRows(pixa, 32, w + 100, 1.0, 0, 30, 2);
pixWrite("/tmp/binar6.png", pix1, IFF_PNG);
pixDisplay(pix1, 1000, 0);
pixDestroy(&pix1);
pixaDestroy(&pixa);
pixDestroy(&pixs);
return 0;
}
main(int argc,
char **argv)
{
l_int32 x, y, i, j, k, w, h, w2, w4, w8, w16, w32, wpl, nerrors;
l_int32 count1, count2, count3, ret, val1, val2;
l_uint32 val32;
l_uint32 *data, *line, *line1, *line2, *data1, *data2;
void **lines1, **linet1, **linet2;
PIX *pixs, *pixt1, *pixt2;
static char mainName[] = "lowaccess_reg";
pixs = pixRead("feyn.tif"); /* width divisible by 16 */
pixGetDimensions(pixs, &w, &h, NULL);
data = pixGetData(pixs);
wpl = pixGetWpl(pixs);
lines1 = pixGetLinePtrs(pixs, NULL);
/* Get timing for the 3 different methods */
startTimer();
for (k = 0; k < 10; k++) {
count1 = 0;
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
if (GET_DATA_BIT(lines1[i], j))
count1++;
}
}
}
fprintf(stderr, "Time with line ptrs = %5.3f sec, count1 = %d\n",
stopTimer(), count1);
startTimer();
for (k = 0; k < 10; k++) {
count2 = 0;
for (i = 0; i < h; i++) {
line = data + i * wpl;
for (j = 0; j < w; j++) {
if (l_getDataBit(line, j))
count2++;
}
}
}
fprintf(stderr, "Time with l_get* = %5.3f sec, count2 = %d\n",
stopTimer(), count2);
startTimer();
for (k = 0; k < 10; k++) {
count3 = 0;
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
pixGetPixel(pixs, j, i, &val32);
count3 += val32;
}
}
}
fprintf(stderr, "Time with pixGetPixel() = %5.3f sec, count3 = %d\n",
stopTimer(), count3);
pixt1 = pixCreateTemplate(pixs);
data1 = pixGetData(pixt1);
linet1 = pixGetLinePtrs(pixt1, NULL);
pixt2 = pixCreateTemplate(pixs);
data2 = pixGetData(pixt2);
linet2 = pixGetLinePtrs(pixt2, NULL);
nerrors = 0;
/* Test different methods for 1 bpp */
count1 = 0;
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
val1 = GET_DATA_BIT(lines1[i], j);
count1 += val1;
if (val1) SET_DATA_BIT(linet1[i], j);
}
}
count2 = 0;
for (i = 0; i < h; i++) {
line = data + i * wpl;
line2 = data2 + i * wpl;
for (j = 0; j < w; j++) {
val2 = l_getDataBit(line, j);
count2 += val2;
if (val2) l_setDataBit(line2, j);
}
}
ret = compareResults(pixs, pixt1, pixt2, count1, count2, "1 bpp");
nerrors += ret;
/* Test different methods for 2 bpp */
count1 = 0;
w2 = w / 2;
for (i = 0; i < h; i++) {
for (j = 0; j < w2; j++) {
val1 = GET_DATA_DIBIT(lines1[i], j);
count1 += val1;
val1 += 0xbbbbbbbc;
SET_DATA_DIBIT(linet1[i], j, val1);
}
}
count2 = 0;
for (i = 0; i < h; i++) {
line = data + i * wpl;
line2 = data2 + i * wpl;
for (j = 0; j < w2; j++) {
val2 = l_getDataDibit(line, j);
count2 += val2;
val2 += 0xbbbbbbbc;
l_setDataDibit(line2, j, val2);
}
}
ret = compareResults(pixs, pixt1, pixt2, count1, count2, "2 bpp");
nerrors += ret;
/* Test different methods for 4 bpp */
count1 = 0;
w4 = w / 4;
for (i = 0; i < h; i++) {
for (j = 0; j < w4; j++) {
val1 = GET_DATA_QBIT(lines1[i], j);
count1 += val1;
val1 += 0xbbbbbbb0;
SET_DATA_QBIT(linet1[i], j, val1);
}
}
count2 = 0;
for (i = 0; i < h; i++) {
line = data + i * wpl;
line2 = data2 + i * wpl;
for (j = 0; j < w4; j++) {
val2 = l_getDataQbit(line, j);
count2 += val2;
val2 += 0xbbbbbbb0;
l_setDataQbit(line2, j, val2);
}
}
ret = compareResults(pixs, pixt1, pixt2, count1, count2, "4 bpp");
nerrors += ret;
/* Test different methods for 8 bpp */
count1 = 0;
w8 = w / 8;
for (i = 0; i < h; i++) {
for (j = 0; j < w8; j++) {
val1 = GET_DATA_BYTE(lines1[i], j);
count1 += val1;
val1 += 0xbbbbbb00;
SET_DATA_BYTE(linet1[i], j, val1);
}
}
count2 = 0;
for (i = 0; i < h; i++) {
line = data + i * wpl;
line2 = data2 + i * wpl;
for (j = 0; j < w8; j++) {
val2 = l_getDataByte(line, j);
count2 += val2;
val2 += 0xbbbbbb00;
l_setDataByte(line2, j, val2);
}
}
ret = compareResults(pixs, pixt1, pixt2, count1, count2, "8 bpp");
nerrors += ret;
/* Test different methods for 16 bpp */
count1 = 0;
w16 = w / 16;
for (i = 0; i < h; i++) {
for (j = 0; j < w16; j++) {
val1 = GET_DATA_TWO_BYTES(lines1[i], j);
count1 += val1;
val1 += 0xbbbb0000;
SET_DATA_TWO_BYTES(linet1[i], j, val1);
}
}
count2 = 0;
for (i = 0; i < h; i++) {
line = data + i * wpl;
line2 = data2 + i * wpl;
for (j = 0; j < w16; j++) {
val2 = l_getDataTwoBytes(line, j);
count2 += val2;
val2 += 0xbbbb0000;
l_setDataTwoBytes(line2, j, val2);
}
}
ret = compareResults(pixs, pixt1, pixt2, count1, count2, "16 bpp");
nerrors += ret;
/* Test different methods for 32 bpp */
count1 = 0;
w32 = w / 32;
for (i = 0; i < h; i++) {
for (j = 0; j < w32; j++) {
val1 = GET_DATA_FOUR_BYTES(lines1[i], j);
count1 += val1 & 0xfff;
SET_DATA_FOUR_BYTES(linet1[i], j, val1);
}
}
count2 = 0;
for (i = 0; i < h; i++) {
line = data + i * wpl;
line2 = data2 + i * wpl;
for (j = 0; j < w32; j++) {
val2 = l_getDataFourBytes(line, j);
count2 += val2 & 0xfff;
l_setDataFourBytes(line2, j, val2);
}
}
ret = compareResults(pixs, pixt1, pixt2, count1, count2, "32 bpp");
nerrors += ret;
if (!nerrors)
fprintf(stderr, "**** No errors ****\n");
else
fprintf(stderr, "**** %d errors found! ****\n", nerrors);
pixDestroy(&pixs);
pixDestroy(&pixt1);
pixDestroy(&pixt2);
lept_free(lines1);
lept_free(linet1);
lept_free(linet2);
return 0;
}
/*!
* pixVarianceInRectangle()
*
* Input: pix (8 bpp)
* box (region to compute variance and/or root variance)
* pix_ma (mean accumulator)
* dpix_msa (mean square accumulator)
* &var (<optional return> variance)
* &rvar (<optional return> root variance)
* Return: 0 if OK, 1 on error
*
* Notes:
* (1) This function is intended to be used for many rectangles
* on the same image. It can find the variance and/or the
* square root of the variance within a rectangle in O(1),
* independent of the size of the rectangle.
*/
l_int32
pixVarianceInRectangle(PIX *pixs,
BOX *box,
PIX *pix_ma,
DPIX *dpix_msa,
l_float32 *pvar,
l_float32 *prvar) {
l_int32 w, h, bx, by, bw, bh;
l_uint32 val00, val01, val10, val11;
l_float64 dval00, dval01, dval10, dval11, mval, msval, var, norm;
BOX *boxc;
PROCNAME("pixVarianceInRectangle");
if (!pvar && !prvar)
return ERROR_INT("neither &var nor &rvar defined", procName, 1);
if (pvar) *pvar = 0.0;
if (prvar) *prvar = 0.0;
if (!pixs || pixGetDepth(pixs) != 8)
return ERROR_INT("pixs not defined", procName, 1);
if (!box)
return ERROR_INT("box not defined", procName, 1);
if (!pix_ma)
return ERROR_INT("pix_ma not defined", procName, 1);
if (!dpix_msa)
return ERROR_INT("dpix_msa not defined", procName, 1);
/* Clip rectangle to image */
pixGetDimensions(pixs, &w, &h, NULL);
boxc = boxClipToRectangle(box, w, h);
boxGetGeometry(boxc, &bx, &by, &bw, &bh);
boxDestroy(&boxc);
if (bw == 0 || bh == 0)
return ERROR_INT("no pixels in box", procName, 1);
/* Use up to 4 points in the accumulators */
norm = 1.0 / (bw * bh);
if (bx > 0 && by > 0) {
pixGetPixel(pix_ma, bx + bw - 1, by + bh - 1, &val11);
pixGetPixel(pix_ma, bx + bw - 1, by - 1, &val10);
pixGetPixel(pix_ma, bx - 1, by + bh - 1, &val01);
pixGetPixel(pix_ma, bx - 1, by - 1, &val00);
dpixGetPixel(dpix_msa, bx + bw - 1, by + bh - 1, &dval11);
dpixGetPixel(dpix_msa, bx + bw - 1, by - 1, &dval10);
dpixGetPixel(dpix_msa, bx - 1, by + bh - 1, &dval01);
dpixGetPixel(dpix_msa, bx - 1, by - 1, &dval00);
mval = norm * (val11 - val01 + val00 - val10);
msval = norm * (dval11 - dval01 + dval00 - dval10);
var = (msval - mval * mval);
if (pvar) *pvar = (l_float32) var;
if (prvar) *prvar = (l_float32)(sqrt(var));
} else if (by > 0) { /* bx == 0 */
pixGetPixel(pix_ma, bw - 1, by + bh - 1, &val11);
pixGetPixel(pix_ma, bw - 1, by - 1, &val10);
dpixGetPixel(dpix_msa, bw - 1, by + bh - 1, &dval11);
dpixGetPixel(dpix_msa, bw - 1, by - 1, &dval10);
mval = norm * (val11 - val10);
msval = norm * (dval11 - dval10);
var = (msval - mval * mval);
if (pvar) *pvar = (l_float32) var;
if (prvar) *prvar = (l_float32)(sqrt(var));
} else if (bx > 0) { /* by == 0 */
pixGetPixel(pix_ma, bx + bw - 1, bh - 1, &val11);
pixGetPixel(pix_ma, bx - 1, bh - 1, &val01);
dpixGetPixel(dpix_msa, bx + bw - 1, bh - 1, &dval11);
dpixGetPixel(dpix_msa, bx - 1, bh - 1, &dval01);
mval = norm * (val11 - val01);
msval = norm * (dval11 - dval01);
var = (msval - mval * mval);
if (pvar) *pvar = (l_float32) var;
if (prvar) *prvar = (l_float32)(sqrt(var));
} else { /* bx == 0 && by == 0 */
pixGetPixel(pix_ma, bw - 1, bh - 1, &val11);
dpixGetPixel(dpix_msa, bw - 1, bh - 1, &dval11);
mval = norm * val11;
msval = norm * dval11;
var = (msval - mval * mval);
if (pvar) *pvar = (l_float32) var;
if (prvar) *prvar = (l_float32)(sqrt(var));
}
return 0;
}
/*!
* pixGenerateSelRandom()
*
* Input: pix (1 bpp, typically small, to be used as a pattern)
* hitfract (fraction of allowable fg pixels that are hits)
* missfract (fraction of allowable bg pixels that are misses)
* distance (min distance from boundary pixel; use 0 for default)
* toppix (number of extra pixels of bg added above)
* botpix (number of extra pixels of bg added below)
* leftpix (number of extra pixels of bg added to left)
* rightpix (number of extra pixels of bg added to right)
* &pixe (<optional return> input pix expanded by extra pixels)
* Return: sel (hit-miss for input pattern), or null on error
*
* Notes:
* (1) Either of hitfract and missfract can be zero. If both are zero,
* the sel would be empty, and NULL is returned.
* (2) No elements are selected that are less than 'distance' pixels away
* from a boundary pixel of the same color. This makes the
* match much more robust to edge noise. Valid inputs of
* 'distance' are 0, 1, 2, 3 and 4. If distance is either 0 or
* greater than 4, we reset it to the default value.
* (3) The 4 numbers for adding rectangles of pixels outside the fg
* can be use if the pattern is expected to be surrounded by bg
* (white) pixels. On the other hand, if the pattern may be near
* other fg (black) components on some sides, use 0 for those sides.
* (4) The input pix, as extended by the extra pixels on selected sides,
* can optionally be returned. For debugging, call
* pixDisplayHitMissSel() to visualize the hit-miss sel superimposed
* on the generating bitmap.
*/
SEL *
pixGenerateSelRandom(PIX *pixs,
l_float32 hitfract,
l_float32 missfract,
l_int32 distance,
l_int32 toppix,
l_int32 botpix,
l_int32 leftpix,
l_int32 rightpix,
PIX **ppixe)
{
l_int32 ws, hs, w, h, x, y, i, j, thresh;
l_uint32 val;
PIX *pixt1, *pixt2, *pixfg, *pixbg;
SEL *seld, *sel;
PROCNAME("pixGenerateSelRandom");
if (ppixe) *ppixe = NULL;
if (!pixs)
return (SEL *)ERROR_PTR("pixs not defined", procName, NULL);
if (pixGetDepth(pixs) != 1)
return (SEL *)ERROR_PTR("pixs not 1 bpp", procName, NULL);
if (hitfract <= 0.0 && missfract <= 0.0)
return (SEL *)ERROR_PTR("no hits or misses", procName, NULL);
if (hitfract > 1.0 || missfract > 1.0)
return (SEL *)ERROR_PTR("fraction can't be > 1.0", procName, NULL);
if (distance <= 0)
distance = DEFAULT_DISTANCE_TO_BOUNDARY;
if (distance > MAX_DISTANCE_TO_BOUNDARY) {
L_WARNING("distance too large; setting to max value", procName);
distance = MAX_DISTANCE_TO_BOUNDARY;
}
/* Locate the foreground */
pixClipToForeground(pixs, &pixt1, NULL);
if (!pixt1)
return (SEL *)ERROR_PTR("pixt1 not made", procName, NULL);
ws = pixGetWidth(pixt1);
hs = pixGetHeight(pixt1);
w = ws;
h = hs;
/* Crop out a region including the foreground, and add pixels
* on sides depending on the side flags */
if (toppix || botpix || leftpix || rightpix) {
x = y = 0;
if (toppix) {
h += toppix;
y = toppix;
}
if (botpix)
h += botpix;
if (leftpix) {
w += leftpix;
x = leftpix;
}
if (rightpix)
w += rightpix;
pixt2 = pixCreate(w, h, 1);
pixRasterop(pixt2, x, y, ws, hs, PIX_SRC, pixt1, 0, 0);
}
else
pixt2 = pixClone(pixt1);
if (ppixe)
*ppixe = pixClone(pixt2);
pixDestroy(&pixt1);
/* Identify fg and bg pixels that are at least 'distance' pixels
* away from the boundary pixels in their set */
seld = selCreateBrick(2 * distance + 1, 2 * distance + 1,
distance, distance, SEL_HIT);
pixfg = pixErode(NULL, pixt2, seld);
pixbg = pixDilate(NULL, pixt2, seld);
pixInvert(pixbg, pixbg);
selDestroy(&seld);
pixDestroy(&pixt2);
/* Generate the sel from a random selection of these points */
sel = selCreateBrick(h, w, h / 2, w / 2, SEL_DONT_CARE);
if (hitfract > 0.0) {
thresh = (l_int32)(hitfract * (l_float64)RAND_MAX);
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
pixGetPixel(pixfg, j, i, &val);
if (val) {
if (rand() < thresh)
selSetElement(sel, i, j, SEL_HIT);
}
}
}
}
if (missfract > 0.0) {
thresh = (l_int32)(missfract * (l_float64)RAND_MAX);
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
pixGetPixel(pixbg, j, i, &val);
if (val) {
if (rand() < thresh)
selSetElement(sel, i, j, SEL_MISS);
}
}
}
}
pixDestroy(&pixfg);
pixDestroy(&pixbg);
return sel;
}
main(int argc,
char **argv)
{
char *filein;
l_int32 count;
CCBORDA *ccba, *ccba2;
PIX *pixs, *pixd, *pixd2, *pixd3;
PIX *pixt, *pixc, *pixc2;
static char mainName[] = "ccbordtest";
if (argc != 2)
exit(ERROR_INT(" Syntax: ccbordtest filein", mainName, 1));
filein = argv[1];
if ((pixs = pixRead(filein)) == NULL)
exit(ERROR_INT("pixs not made", mainName, 1));
fprintf(stderr, "Get border representation...");
startTimer();
ccba = pixGetAllCCBorders(pixs);
fprintf(stderr, "%6.3f sec\n", stopTimer());
#if 0
/* get global locs directly and display borders */
fprintf(stderr, "Convert from local to global locs...");
startTimer();
ccbaGenerateGlobalLocs(ccba);
fprintf(stderr, "%6.3f sec\n", stopTimer());
fprintf(stderr, "Display border representation...");
startTimer();
pixd = ccbaDisplayBorder(ccba);
fprintf(stderr, "%6.3f sec\n", stopTimer());
pixWrite("/tmp/junkborder1.png", pixd, IFF_PNG);
#else
/* get step chain code, then global coords, and display borders */
fprintf(stderr, "Get step chain code...");
startTimer();
ccbaGenerateStepChains(ccba);
fprintf(stderr, "%6.3f sec\n", stopTimer());
fprintf(stderr, "Convert from step chain to global locs...");
startTimer();
ccbaStepChainsToPixCoords(ccba, CCB_GLOBAL_COORDS);
fprintf(stderr, "%6.3f sec\n", stopTimer());
fprintf(stderr, "Display border representation...");
startTimer();
pixd = ccbaDisplayBorder(ccba);
fprintf(stderr, "%6.3f sec\n", stopTimer());
pixWrite("/tmp/junkborder1.png", pixd, IFF_PNG);
#endif
/* check if border pixels are in original set */
fprintf(stderr, "Check if border pixels are in original set ...\n");
pixt = pixSubtract(NULL, pixd, pixs);
pixCountPixels(pixt, &count, NULL);
if (count == 0)
fprintf(stderr, " all border pixels are in original set\n");
else
fprintf(stderr, " %d border pixels are not in original set\n", count);
pixDestroy(&pixt);
/* display image */
fprintf(stderr, "Reconstruct image ...");
startTimer();
/* pixc = ccbaDisplayImage1(ccba); */
pixc = ccbaDisplayImage2(ccba);
fprintf(stderr, "%6.3f sec\n", stopTimer());
pixWrite("/tmp/junkrecon1.png", pixc, IFF_PNG);
/* check with original to see if correct */
fprintf(stderr, "Check with original to see if correct ...\n");
pixXor(pixc, pixc, pixs);
pixCountPixels(pixc, &count, NULL);
if (count == 0)
fprintf(stderr, " perfect direct recon\n");
else {
l_int32 w, h, i, j;
l_uint32 val;
fprintf(stderr, " %d pixels in error in recon\n", count);
#if 1
w = pixGetWidth(pixc);
h = pixGetHeight(pixc);
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
pixGetPixel(pixc, j, i, &val);
if (val == 1)
fprintf(stderr, "bad pixel at (%d, %d)\n", j, i);
}
}
pixWrite("/tmp/junkbadpixels.png", pixc, IFF_PNG);
#endif
}
/*----------------------------------------------------------*
* write to file (compressed) and read back *
*----------------------------------------------------------*/
fprintf(stderr, "Write serialized step data...");
startTimer();
ccbaWrite("/tmp/junkstepout", ccba);
fprintf(stderr, "%6.3f sec\n", stopTimer());
fprintf(stderr, "Read serialized step data...");
startTimer();
ccba2 = ccbaRead("/tmp/junkstepout");
fprintf(stderr, "%6.3f sec\n", stopTimer());
/* display the border pixels again */
fprintf(stderr, "Convert from step chain to global locs...");
startTimer();
ccbaStepChainsToPixCoords(ccba2, CCB_GLOBAL_COORDS);
fprintf(stderr, "%6.3f sec\n", stopTimer());
fprintf(stderr, "Display border representation...");
startTimer();
pixd2 = ccbaDisplayBorder(ccba2);
fprintf(stderr, "%6.3f sec\n", stopTimer());
pixWrite("/tmp/junkborder2.png", pixd2, IFF_PNG);
/* check if border pixels are same as first time */
pixXor(pixd2, pixd2, pixd);
pixCountPixels(pixd2, &count, NULL);
if (count == 0)
fprintf(stderr, " perfect w/r border recon\n");
else {
l_int32 w, h, i, j, val;
fprintf(stderr, " %d pixels in error in w/r recon\n", count);
}
pixDestroy(&pixd2);
/* display image again */
fprintf(stderr, "Convert from step chain to local coords...");
startTimer();
ccbaStepChainsToPixCoords(ccba2, CCB_LOCAL_COORDS);
fprintf(stderr, "%6.3f sec\n", stopTimer());
fprintf(stderr, "Reconstruct image from file ...");
startTimer();
/* pixc2 = ccbaDisplayImage1(ccba2); */
pixc2 = ccbaDisplayImage2(ccba2);
fprintf(stderr, "%6.3f sec\n", stopTimer());
pixWrite("/tmp/junkrecon2.png", pixc2, IFF_PNG);
/* check with original to see if correct */
fprintf(stderr, "Check with original to see if correct ...\n");
pixXor(pixc2, pixc2, pixs);
pixCountPixels(pixc2, &count, NULL);
if (count == 0)
fprintf(stderr, " perfect image recon\n");
else {
l_int32 w, h, i, j;
l_uint32 val;
fprintf(stderr, " %d pixels in error in image recon\n", count);
#if 1
w = pixGetWidth(pixc2);
h = pixGetHeight(pixc2);
for (i = 0; i < h; i++) {
for (j = 0; j < w; j++) {
pixGetPixel(pixc2, j, i, &val);
if (val == 1)
fprintf(stderr, "bad pixel at (%d, %d)\n", j, i);
}
}
pixWrite("/tmp/junkbadpixels2.png", pixc2, IFF_PNG);
#endif
}
/*----------------------------------------------------------*
* make, display and check single path border for svg *
*----------------------------------------------------------*/
/* make local single path border for svg */
fprintf(stderr, "Make local single path borders for svg ...");
startTimer();
ccbaGenerateSinglePath(ccba);
fprintf(stderr, "%6.3f sec\n", stopTimer());
/* generate global single path border */
fprintf(stderr, "Generate global single path borders ...");
startTimer();
ccbaGenerateSPGlobalLocs(ccba, CCB_SAVE_TURNING_PTS);
fprintf(stderr, "%6.3f sec\n", stopTimer());
/* display border pixels from single path */
fprintf(stderr, "Display border from single path...");
startTimer();
pixd3 = ccbaDisplaySPBorder(ccba);
fprintf(stderr, "%6.3f sec\n", stopTimer());
pixWrite("/tmp/junkborder3.png", pixd3, IFF_PNG);
/* check if border pixels are in original set */
fprintf(stderr, "Check if border pixels are in original set ...\n");
pixt = pixSubtract(NULL, pixd3, pixs);
pixCountPixels(pixt, &count, NULL);
if (count == 0)
fprintf(stderr, " all border pixels are in original set\n");
else
fprintf(stderr, " %d border pixels are not in original set\n", count);
pixDestroy(&pixt);
pixDestroy(&pixd3);
/* output in svg file format */
fprintf(stderr, "Write output in svg file format ...\n");
startTimer();
ccbaWriteSVG("/tmp/junksvg", ccba);
fprintf(stderr, "%6.3f sec\n", stopTimer());
ccbaDestroy(&ccba2);
ccbaDestroy(&ccba);
pixDestroy(&pixs);
pixDestroy(&pixd);
pixDestroy(&pixc);
pixDestroy(&pixc2);
return 0;
}
/*!
* pixGetRunsOnLine()
*
* Input: pixs (1 bpp)
* x1, y1, x2, y2
* Return: numa, or null on error
*
* Notes:
* (1) Action: this function uses the bresenham algorithm to compute
* the pixels along the specified line. It returns a Numa of the
* runlengths of the fg (black) and bg (white) runs, always
* starting with a white run.
* (2) If the first pixel on the line is black, the length of the
* first returned run (which is white) is 0.
*/
NUMA *
pixGetRunsOnLine(PIX *pixs,
l_int32 x1,
l_int32 y1,
l_int32 x2,
l_int32 y2)
{
l_int32 w, h, x, y, npts;
l_int32 i, runlen, preval;
l_uint32 val;
NUMA *numa;
PTA *pta;
PROCNAME("pixGetRunsOnLine");
if (!pixs)
return (NUMA *)ERROR_PTR("pixs not defined", procName, NULL);
if (pixGetDepth(pixs) != 1)
return (NUMA *)ERROR_PTR("pixs not 1 bpp", procName, NULL);
w = pixGetWidth(pixs);
h = pixGetHeight(pixs);
if (x1 < 0 || x1 >= w)
return (NUMA *)ERROR_PTR("x1 not valid", procName, NULL);
if (x2 < 0 || x2 >= w)
return (NUMA *)ERROR_PTR("x2 not valid", procName, NULL);
if (y1 < 0 || y1 >= h)
return (NUMA *)ERROR_PTR("y1 not valid", procName, NULL);
if (y2 < 0 || y2 >= h)
return (NUMA *)ERROR_PTR("y2 not valid", procName, NULL);
if ((pta = generatePtaLine(x1, y1, x2, y2)) == NULL)
return (NUMA *)ERROR_PTR("pta not made", procName, NULL);
if ((npts = ptaGetCount(pta)) == 0)
return (NUMA *)ERROR_PTR("pta has no pts", procName, NULL);
if ((numa = numaCreate(0)) == NULL)
return (NUMA *)ERROR_PTR("numa not made", procName, NULL);
for (i = 0; i < npts; i++) {
ptaGetIPt(pta, i, &x, &y);
pixGetPixel(pixs, x, y, &val);
if (i == 0) {
if (val == 1) { /* black pixel; append white run of size 0 */
numaAddNumber(numa, 0);
}
preval = val;
runlen = 1;
continue;
}
if (val == preval) { /* extend current run */
preval = val;
runlen++;
}
else { /* end previous run */
numaAddNumber(numa, runlen);
preval = val;
runlen = 1;
}
}
numaAddNumber(numa, runlen); /* append last run */
ptaDestroy(&pta);
return numa;
}
