/*
* pixDebugFlipDetect()
*
* Input: filename (for output debug file)
* pixs (input to pix*Detect)
* pixhm (hit-miss result from ascenders or descenders)
* enable (1 to enable this function; 0 to disable)
* Return: void
*/
static void
pixDebugFlipDetect(const char *filename,
PIX *pixs,
PIX *pixhm,
l_int32 enable)
{
PIX *pixt, *pixthm;
if (!enable) return;
/* Display with red dot at counted locations */
pixt = pixConvert1To4Cmap(pixs);
pixthm = pixMorphSequence(pixhm, "d5.5", 0);
pixSetMaskedCmap(pixt, pixthm, 0, 0, 255, 0, 0);
pixWrite(filename, pixt, IFF_PNG);
pixDestroy(&pixthm);
pixDestroy(&pixt);
return;
}
/*!
* pixFindPageForeground()
*
* Input: pixs (full resolution (any type or depth)
* threshold (for binarization; typically about 128)
* mindist (min distance of text from border to allow
* cleaning near border; at 2x reduction, this
* should be larger than 50; typically about 70)
* erasedist (when conditions are satisfied, erase anything
* within this distance of the edge;
* typically 30 at 2x reduction)
* pagenum (use for debugging when called repeatedly; labels
* debug images that are assembled into pdfdir)
* showmorph (set to a negative integer to show steps in
* generating masks; this is typically used
* for debugging region extraction)
* display (set to 1 to display mask and selected region
* for debugging a single page)
* pdfdir (subdirectory of /tmp where images showing the
* result are placed when called repeatedly; use
* null if no output requested)
* Return: box (region including foreground, with some pixel noise
* removed), or null if not found
*
* Notes:
* (1) This doesn't simply crop to the fg. It attempts to remove
* pixel noise and junk at the edge of the image before cropping.
* The input @threshold is used if pixs is not 1 bpp.
* (2) There are several debugging options, determined by the
* last 4 arguments.
* (3) If you want pdf output of results when called repeatedly,
* the pagenum arg labels the images written, which go into
* /tmp/<pdfdir>/<pagenum>.png. In that case,
* you would clean out the /tmp directory before calling this
* function on each page:
* lept_rmdir(pdfdir);
* lept_mkdir(pdfdir);
*/
BOX *
pixFindPageForeground(PIX *pixs,
l_int32 threshold,
l_int32 mindist,
l_int32 erasedist,
l_int32 pagenum,
l_int32 showmorph,
l_int32 display,
const char *pdfdir)
{
char buf[64];
l_int32 flag, nbox, intersects;
l_int32 w, h, bx, by, bw, bh, left, right, top, bottom;
PIX *pixb, *pixb2, *pixseed, *pixsf, *pixm, *pix1, *pixg2;
BOX *box, *boxfg, *boxin, *boxd;
BOXA *ba1, *ba2;
PROCNAME("pixFindPageForeground");
if (!pixs)
return (BOX *)ERROR_PTR("pixs not defined", procName, NULL);
/* Binarize, downscale by 0.5, remove the noise to generate a seed,
* and do a seedfill back from the seed into those 8-connected
* components of the binarized image for which there was at least
* one seed pixel. Also clear out any components that are within
* 10 pixels of the edge at 2x reduction. */
flag = (showmorph) ? -1 : 0; /* if showmorph == -1, write intermediate
* images to /tmp/seq_output_1.pdf */
pixb = pixConvertTo1(pixs, threshold);
pixb2 = pixScale(pixb, 0.5, 0.5);
pixseed = pixMorphSequence(pixb2, "o1.2 + c9.9 + o3.5", flag);
pixsf = pixSeedfillBinary(NULL, pixseed, pixb2, 8);
pixSetOrClearBorder(pixsf, 10, 10, 10, 10, PIX_SET);
pixm = pixRemoveBorderConnComps(pixsf, 8);
if (display) pixDisplay(pixm, 100, 100);
/* Now, where is the main block of text? We want to remove noise near
* the edge of the image, but to do that, we have to be convinced that
* (1) there is noise and (2) it is far enough from the text block
* and close enough to the edge. For each edge, if the block
* is more than mindist from that edge, then clean 'erasedist'
* pixels from the edge. */
pix1 = pixMorphSequence(pixm, "c50.50", flag - 1);
ba1 = pixConnComp(pix1, NULL, 8);
ba2 = boxaSort(ba1, L_SORT_BY_AREA, L_SORT_DECREASING, NULL);
pixGetDimensions(pix1, &w, &h, NULL);
nbox = boxaGetCount(ba2);
if (nbox > 1) {
box = boxaGetBox(ba2, 0, L_CLONE);
boxGetGeometry(box, &bx, &by, &bw, &bh);
left = (bx > mindist) ? erasedist : 0;
right = (w - bx - bw > mindist) ? erasedist : 0;
top = (by > mindist) ? erasedist : 0;
bottom = (h - by - bh > mindist) ? erasedist : 0;
pixSetOrClearBorder(pixm, left, right, top, bottom, PIX_CLR);
boxDestroy(&box);
}
pixDestroy(&pix1);
boxaDestroy(&ba1);
boxaDestroy(&ba2);
/* Locate the foreground region; don't bother cropping */
pixClipToForeground(pixm, NULL, &boxfg);
/* Sanity check the fg region. Make sure it's not confined
* to a thin boundary on the left and right sides of the image,
* in which case it is likely to be noise. */
if (boxfg) {
boxin = boxCreate(0.1 * w, 0, 0.8 * w, h);
boxIntersects(boxfg, boxin, &intersects);
if (!intersects) {
L_INFO("found only noise on page %d\n", procName, pagenum);
boxDestroy(&boxfg);
}
boxDestroy(&boxin);
}
boxd = NULL;
if (!boxfg) {
L_INFO("no fg region found for page %d\n", procName, pagenum);
} else {
boxAdjustSides(boxfg, boxfg, -2, 2, -2, 2); /* tiny expansion */
boxd = boxTransform(boxfg, 0, 0, 2.0, 2.0);
/* Write image showing box for this page. This is to be
* bundled up into a pdf of all the pages, which can be
* generated by convertFilesToPdf() */
if (pdfdir) {
pixg2 = pixConvert1To4Cmap(pixb);
pixRenderBoxArb(pixg2, boxd, 3, 255, 0, 0);
snprintf(buf, sizeof(buf), "/tmp/%s/%05d.png", pdfdir, pagenum);
if (display) pixDisplay(pixg2, 700, 100);
pixWrite(buf, pixg2, IFF_PNG);
pixDestroy(&pixg2);
}
}
pixDestroy(&pixb);
pixDestroy(&pixb2);
pixDestroy(&pixseed);
pixDestroy(&pixsf);
pixDestroy(&pixm);
boxDestroy(&boxfg);
return boxd;
}
