static char triangleInRectangleTest(CvSeq* c, struct Triangle* t) {
if( cvPointPolygonTest( c, cvPointTo32f( t->pt[0]), 0) > 0 ) {
if ( cvPointPolygonTest( c, cvPointTo32f( t->pt[1]), 0 ) > 0 ){
if ( cvPointPolygonTest( c, cvPointTo32f( t->pt[2] ), 0 ) > 0 ){
return 1;
}
}
} else
return 0;
}
/*
* Determines whether the point is inside a contour, outside, or lies on an edge (or coinsides with a vertex).
* @overload point_polygon_test(point, measure_dist)
* @param point [CvPoint2D32f] Point tested against the contour
* @param measure_dist [Boolean] If true, the method estimates the signed distance from the point to
* the nearest contour edge. Otherwise, the function only checks if the point is inside a contour or not.
* @return [Number] When measure_dist = false, the return value is +1, -1 and 0, respectively.
* When measure_dist = true, it is a signed distance between the point and the nearest contour edge.
* @opencv_func cvPointPolygonTest
*/
VALUE
rb_point_polygon_test(VALUE self, VALUE point, VALUE measure_dist)
{
int measure_dist_flag;
if (measure_dist == Qtrue)
measure_dist_flag = 1;
else if (measure_dist == Qfalse)
measure_dist_flag = 0;
else
measure_dist_flag = NUM2INT(measure_dist);
double dist = Qnil;
try {
dist = cvPointPolygonTest(CVARR(self), VALUE_TO_CVPOINT2D32F(point), measure_dist_flag);
}
catch (cv::Exception& e) {
raise_cverror(e);
}
/* cvPointPolygonTest returns 100, -100 or 0 when measure_dist = 0 */
if ((!measure_dist_flag) && ((int)dist) != 0)
dist = (dist > 0) ? 1 : -1;
return rb_float_new(dist);
}
CvScalar contour_average(CvContour* contour, IplImage* image)
{
CvRect rect = ((CvContour*)contour)->rect;
CvScalar average = cvScalarAll(0);
int count = 0;
for(int x = rect.x; x < (rect.x+rect.width); x++) {
for(int y = rect.y; y < (rect.y+rect.height); y++) {
if(cvPointPolygonTest(contour, cvPointTo32f(cvPoint(x,y)),0) == 100) {
CvScalar s = cvGet2D(image,y,x);
average.val[0] += s.val[0];
average.val[1] += s.val[1];
average.val[2] += s.val[2];
count++;
}
}
}
for(int i = 0; i < 3; i++){
average.val[i] /= count;
}
return average;
}
double BlobGetXYInside::operator()(Blob &blob)
{
if( blob.GetExternalContour()->GetContourPoints() )
{
return cvPointPolygonTest( blob.GetExternalContour()->GetContourPoints(), m_p,0) >= 0;
}
return 0;
}
/*
* Calculates distance between a point and the nearest contour edgex
* @overload measure_distance(point)
* @param point [CvPoint2D32f] Point tested against the contour
* @return Signed distance between the point and the nearest contour edge
* @opencv_func cvPointPolygonTest
*/
VALUE
rb_measure_distance(VALUE self, VALUE point)
{
double distance = 0;
try {
distance = cvPointPolygonTest(CVARR(self), VALUE_TO_CVPOINT2D32F(point), 1);
}
catch (cv::Exception& e) {
raise_cverror(e);
}
return rb_float_new(distance);
}
/*
* Performs a point-in-contour test.
* The method determines whether the point is inside a contour, outside,
* or lies on an edge (or coincides with a vertex).
* @overload in?(point)
* @param point [CvPoint2D32f] Point tested against the contour
* @return [Boolean] If the point is inside, returns true. If outside, returns false.
* If lies on an edge, returns nil.
* @opencv_func cvPointPolygonTest
*/
VALUE
rb_in_q(VALUE self, VALUE point)
{
double n = 0;
try {
n = cvPointPolygonTest(CVARR(self), VALUE_TO_CVPOINT2D32F(point), 0);
}
catch (cv::Exception& e) {
raise_cverror(e);
}
return n == 0 ? Qnil : n > 0 ? Qtrue : Qfalse;
}
void lmprojinv(double *p, double *x, int m, int n, void *data)
{
double totdist = 0;
int iter = seq2->total<LMRAWCONTOURN?1:(seq2->total/LMRAWCONTOURN);
for(int i = 0;i < n;i++){
CvPoint*pt = (CvPoint*)cvGetSeqElem(seq2,i*iter);
double d = p[6]*pt->x + p[7]*pt->y + p[8];
double px = (p[0]*pt->x + p[1]*pt->y + p[2])/d;
double py = (p[3]*pt->x + p[4]*pt->y + p[5])/d;
double dist = abs(cvPointPolygonTest(seq1, cvPoint2D32f(px,py), 1));
x[i] = dist;
totdist+= dist;
}
}
void lmproj(double *p, double *x, int m, int n, void *data)
{
for(int i = 0;i < seq1->total;i++){
CvPoint*pt = (CvPoint*)cvGetSeqElem(seq1,i);
double d = p[6]*pt->x + p[7]*pt->y + 1.0;
double px = (p[0]*pt->x + p[1]*pt->y + p[2])/d;
double py = (p[3]*pt->x + p[4]*pt->y + p[5])/d;
double dist=0.0;
if (px>=0&&px<IMG_WIDTH&&py>=0&&py<IMG_HEIGHT)
{
dist = abs(cvPointPolygonTest(seq2, cvPoint2D32f(px,py), 1));
}
x[i] = dist;
}
}
/*
This method will calculate the convex hull of source landmarks
and populate the pointsInsideHull vector with these points coordinates
*/
void PAW::populatePointsInsideHull(){
//calc scrLandmarks convex hull
CvPoint* pointsHull = (CvPoint*)malloc( nLandmarks * sizeof(pointsHull[0]));
int* hull = (int*)malloc( nLandmarks * sizeof(hull[0]));
CvMat pointMat = cvMat( 1, nLandmarks, CV_32SC2, pointsHull );
CvMat hullMat = cvMat( 1, nLandmarks, CV_32SC1, hull );
for(int i = 0; i < nLandmarks; i++ )
{
pointsHull[i] = cvPoint(srcLandmarks.at<int>(i,0),srcLandmarks.at<int>(i,1));
}
cvConvexHull2( &pointMat, &hullMat, CV_CLOCKWISE, 0 );
int hullcount = hullMat.cols;
CvPoint* pointsHullFinal = (CvPoint*)malloc( hullcount * sizeof(pointsHullFinal[0]));
for(int i = 0; i < hullcount; i++ ){
int ptIndex = hull[i];
CvPoint pt = cvPoint( srcLandmarks.at<int>(ptIndex,0),
srcLandmarks.at<int>(ptIndex,1));
pointsHullFinal[i] = pt;
}
CvMat hullMatPoints = cvMat( 1, hullcount, CV_32SC2, pointsHullFinal);
//check if point belongs
for (int j=0;j<baseImageHeight;j++){
for(int i=0;i< baseImageWidth;i++){
double distance = cvPointPolygonTest(&hullMatPoints,cvPoint2D32f(i,j),1);
if(distance >=0){
pointsInsideHull.push_back(cvPoint(i,j));
}
}
}
}
bool CPolygonalRegion::IsPointInner(double x, double y) const {
if (GetSize() <= 2) return false;
else return (cvPointPolygonTest(_pPointsOuter, cvPoint2D32f(x,y), 0) >= 0);
}
Widget::Widget(QWidget *parent) :
QWidget(parent),
ui(new Ui::Widget)
{
ui->setupUi(this);
IplImage *image,*hsv,*hue,*saturation,*value;//图像空间
// cv::Mat image = cv::imread("/home/xiangang/qt/qt_example/yuzhi/pic.bmp" ,CV_LOAD_IMAGE_GRAYSCALE);
image = cvLoadImage("picture.bmp");
// image = cvLoadImage("2.jpg");
cvShowImage("image",image);
hsv = cvCreateImage(cvGetSize(image),8,3);//给hsv色系的图像申请空间
hue = cvCreateImage(cvGetSize(image),8,1); //色调
saturation = cvCreateImage(cvGetSize(image),8,1);//饱和度
value = cvCreateImage(cvGetSize(image),8,1);//亮度
cvNamedWindow("image",CV_WINDOW_AUTOSIZE);//用于显示图像的窗口
// cvNamedWindow("hsv",CV_WINDOW_AUTOSIZE);
// cvNamedWindow("hue",CV_WINDOW_AUTOSIZE);
// cvNamedWindow("saturation",CV_WINDOW_AUTOSIZE);
// cvNamedWindow("value",CV_WINDOW_AUTOSIZE);
cvCvtColor(image,hsv,CV_BGR2HSV);//将RGB色系转为HSV色系
qDebug("rgb to hsv");
cvSplit(hsv, hue, 0, 0, 0 );//分离三个通道
cvSplit(hsv, 0, saturation, 0, 0 );
cvSplit(hsv, 0, 0, value, 0 );
cvShowImage("hue",hue);
cvShowImage("saturation",saturation);
cvShowImage("value",value);
qDebug("show pic!");
int th1 = 140;
cv::Mat global1;
cv::threshold(cv::Mat(hue,false),global1,th1,360,CV_THRESH_BINARY_INV);
int th3 = 100;
cv::Mat global3;
cv::threshold(cv::Mat(saturation,false),global3,th3,255,CV_THRESH_BINARY_INV);
int th2 = 80;
cv::Mat global2;
cv::threshold(cv::Mat(value,false),global2,th2,255,CV_THRESH_BINARY_INV);
// int blockSize = 2;
// int constValue = 10;
// cv::Mat local;
// cv::adaptiveThreshold(cv::Mat( hue,false), local, 255, CV_ADAPTIVE_THRESH_MEAN_C, CV_THRESH_BINARY_INV, blockSize, constValue);
cv::imwrite("hue.jpg",cv::Mat(hue,false));
cv::imwrite("global1.jpg", global1);
cv::imwrite("global2.jpg", global2);
// cv::imwrite("local.jpg", local);
////cv::imshow("h分量otsu全局阈值", hue);
////cvShowImage("h分量otsu全局阈值", hue);
cv::imshow("h分量全局阈值", global1);
cv::imshow("v分量全局阈值", global2);
cv::imshow("s分量全局阈值", global3);
// cv::imshow("h分量局部阈值", local);
//otsu
ImageBinarization(hue);
IplImage *img = cvLoadImage("hue.jpg"); //加载图片
IplImage *imgBuf = cvCreateImage(cvGetSize(img), IPL_DEPTH_8U, 3); //通道为3
cvConvert(img, imgBuf); //将源图片转化为位深度为8，通道为3的图片
for (int i = 0; i < img->height; i++)
{
for(int j = 0; j < img->width; j++)
{
if((((uchar *)(img->imageData + i * img->widthStep ))[j * img->nChannels + 0] > 200)/*&&
(((uchar *)(img->imageData + i * img->widthStep ))[j * img->nChannels + 0] < 100)*/)
{
((uchar *)(img->imageData + i * img->widthStep ))[j * img->nChannels + 0] = 234; //改变该像素B的颜色分量
((uchar *)(img->imageData + i * img->widthStep ))[j * img->nChannels + 1] = 200; //改变该像素G的颜色分量
((uchar *)(img->imageData + i * img->widthStep ))[j * img->nChannels + 2] = 150; //改变该像素R的颜色分量
}
}
}
cv::imshow("h分量测试",cv::Mat(img,false));
for (int i = 0; i <(global2.rows)/2; i++)
{
for(int j = 0; j < global2.cols; j++)
{
for(int n = 0; n < global2.channels(); n++ )
{
if(global2.at<uchar>(i,j*global2.channels()+n) > 125)
{
global1.at<uchar>(i,j*global1.channels()+n) = 255;
// local.at<uchar>(i,j*local.channels()+n) = 255;
}
}
}
}
cv::imshow("合并h分量和v分量",global1);
cv::imwrite("success.jpg",global1);
IplImage *img3 = cvLoadImage("success.jpg",0);
if(img3 == NULL)
{
qDebug("img load failed!\n");
// return 0;
}
IplImage *img_erode = cvCreateImage(cvGetSize(img), 8, 1);
IplImage *img_dilate = cvCreateImage(cvGetSize(img), 8, 1);
cvErode( img3,img_erode, NULL,1); //腐蚀
cvDilate( img_erode,img_dilate, NULL,3); //膨胀
//// cvErode( img_dilate,img_erode, NULL,1); //腐蚀
cvShowImage("腐蚀",img_erode);
cvShowImage("膨胀",img_dilate);
cv::imwrite("dilate_success.jpg",cv::Mat(img_dilate,false));
/*****************提取轮廓*****************/
// IplImage* src = NULL;
// IplImage* img = NULL;
IplImage* dst = NULL;
CvMemStorage* storage = cvCreateMemStorage (0);;
CvSeq* contour = 0;
double contour_area_temp = 0, contour_area_max = 0, contour_area_2 = 0;
CvSeq* area_max_contour = 0;
CvSeq* area_2_contour = 0;
int contours = 0;
CvScalar external_color;
CvScalar hole_color;
//寻找轮廓 CV_RETR_EXTERNAL CV_RETR_LIST
contours = cvFindContours (img_dilate, storage, &contour, sizeof(CvContour), CV_RETR_LIST, CV_CHAIN_APPROX_NONE);
qDebug("contours:%d",contours);
//创建新的图像显示轮廓
dst = cvCreateImage (cvGetSize(image), image->depth, image->nChannels);
cvCopy(image,dst);
for (;contour != 0; contour = contour->h_next)
{
contour_area_temp = fabs(cvContourArea(contour,CV_WHOLE_SEQ));
if(contour_area_temp > contour_area_max)
{
contour_area_2 = contour_area_max;
contour_area_max = contour_area_temp;
area_2_contour = area_max_contour;
area_max_contour = contour;
}
}
external_color = CV_RGB(rand()&255, rand()&255, rand()&255);
hole_color = CV_RGB(rand()&255, rand()&255, rand()&255);
//画轮廓
cvDrawContours (dst, area_2_contour, external_color, hole_color, 0, 2, 8);
cvNamedWindow ("Contour", 1);
cvShowImage ("Contour", dst);
for (int i = 0; i < image->height; i++)
{
for(int j = 0; j < image->width; j++)
{
//-1 0 1分别代表轮廓外，轮廓上，轮廓里面
if(-1 == (cvPointPolygonTest(area_2_contour,cv::Point2f(j,i),0)))
{
((uchar *)(image->imageData + i * image->widthStep ))[j * image->nChannels + 0] = 234; //改变该像素B的颜色分量
((uchar *)(image->imageData + i * image->widthStep ))[j * image->nChannels + 1] = 200; //改变该像素G的颜色分量
((uchar *)(image->imageData + i * image->widthStep ))[j * image->nChannels + 2] = 150; //改变该像素R的颜色分量
}
}
}
cv::imshow("final",cv::Mat(image,false));
///*****************************************/
//// cvWrite("dilate_success.jpg",img_dilate);
//// cv::waitKey(0); /// 等待用户按键。如果是ESC健则退出等待过程。
//// while(true)
//// {
//// int c;
//// c = cv::waitKey( 20 );
//// if( (char)c == 27 )
//// { break; }
//// }
zhifangtu(value);
cvReleaseMemStorage (&storage);
// cvReleaseImage (&src);
cvReleaseImage (&img);
cvReleaseImage (&dst);
}
void FrameProcessor::DetectAndDrawQuads(IplImage* img, vector <Square>& cubes, CvCapture* capture)
{
float angle = 0.0f;
CvSeq* contours;
CvSeq* result;
CvMemStorage *storage = cvCreateMemStorage(0);
//IplImage* ret = cvCreateImage(cvGetSize(img), 8, 3);
IplImage* temp = cvCreateImage(cvGetSize(img), 8, 1);
cvCvtColor(img, temp, CV_BGR2GRAY);
IplImage* Img = cvCreateImage (cvGetSize (img), 8, 1);
cvThreshold(temp, Img, 0, 255, CV_THRESH_BINARY_INV | CV_THRESH_OTSU);
cvFindContours(Img, storage, &contours, sizeof(CvContour), CV_RETR_LIST, CV_CHAIN_APPROX_SIMPLE, cvPoint(0,0));
CvSeq* contours1 = contours;
while(contours)
{
result = cvApproxPoly (contours, sizeof (CvContour), storage, CV_POLY_APPROX_DP, cvContourPerimeter (contours) * 0.02, 0);
if((result->total==4) && (fabs(cvContourArea(result, CV_WHOLE_SEQ)) > 30) && cvCheckContourConvexity (result))
{
int countTriang = 0;
CvSeq* contours2 = contours1;
while(contours2){
CvSeq* result2 = cvApproxPoly(contours2, sizeof(CvContour), storage, CV_POLY_APPROX_DP, cvContourPerimeter(contours2)*0.02, 0);
if((result2->total==3) && (fabs(cvContourArea(result2, CV_WHOLE_SEQ)) > 50) && cvCheckContourConvexity(result2))
{
CvPoint *pt[3];
for(int i=0;i<3;i++)
pt[i] = (CvPoint*)cvGetSeqElem(result2, i);
CvPoint ptCent[1];
ptCent[0].x = (pt[0]->x + pt[1]->x + pt[2]->x)/3;
ptCent[0].y = (pt[0]->y + pt[1]->y + pt[2]->y)/3;
//printf("(%d, %d) (%d,%d) (%d,%d)\n", pt[0]->x, pt[0]->y, pt[1]->x, pt[1]->y, pt[2]->x, pt[2]->y);
//cvCircle(ret, ptCent[0], 5, cvScalar(255));
CvPoint2D32f triang;
triang.x = ptCent[0].x;
triang.y = ptCent[0].y;
if(cvPointPolygonTest(result, triang, 0) > 0){
countTriang++;
//cvLine(ret, *pt[0], *pt[1], cvScalar(30, 50, 50));
//cvLine(ret, *pt[1], *pt[2], cvScalar(30, 50, 50));
//cvLine(ret, *pt[2], *pt[0], cvScalar(30, 50, 50));
angle = GetAngle(pt);
}
}
contours2 = contours2->h_next;
}
if(countTriang == 1){
CvPoint *pt[4];
for(int i=0;i<4;i++)
pt[i] = (CvPoint*)cvGetSeqElem(result, i);
//cvLine(ret, *pt[0], *pt[1], cvScalar(255));
//cvLine(ret, *pt[1], *pt[2], cvScalar(255));
//cvLine(ret, *pt[2], *pt[3], cvScalar(255));
//cvLine(ret, *pt[3], *pt[0], cvScalar(255));
CvSeq* contours3 = contours1;
int countSquare = 0;
while(contours3){
CvSeq* result3 = cvApproxPoly(contours3, sizeof(CvContour), storage, CV_POLY_APPROX_DP, cvContourPerimeter(contours3)*0.02, 0);
if((result3->total==4) && (fabs(cvContourArea(result3, CV_WHOLE_SEQ)) > 50) && cvCheckContourConvexity(result3))
{
CvPoint *pt[4];
for(int i=0;i<4;i++)
pt[i] = (CvPoint*)cvGetSeqElem(result3, i);
CvPoint ptCent[1];
ptCent[0].x = (pt[0]->x + pt[1]->x + pt[2]->x + pt[3]->x)/4;
ptCent[0].y = (pt[0]->y + pt[1]->y + pt[2]->y + pt[3]->y)/4;
//cvCircle(ret, ptCent[0], 5, cvScalar(255));
CvPoint2D32f square;
square.x = ptCent[0].x;
square.y = ptCent[0].y;
//cout << cvPointPolygonTest(result, triang, 0);
if(cvPointPolygonTest(result, square, 0) > 0){
countSquare++;
//cvLine(ret, *pt[0], *pt[1], cvScalar(90, 50, 50));
//cvLine(ret, *pt[1], *pt[2], cvScalar(90, 50, 50));
//cvLine(ret, *pt[2], *pt[3], cvScalar(90, 50, 50));
//cvLine(ret, *pt[3], *pt[0], cvScalar(90, 50, 50));
}
}
contours3 = contours3->h_next;
}
countSquare--;
CvPoint *pt1[4];
for(int i=0;i<4;i++)
pt1[i] = (CvPoint*)cvGetSeqElem(result, i);
CvPoint ptCent[1];
ptCent[0].x = (pt1[0]->x + pt1[1]->x + pt1[2]->x + pt1[3]->x)/4;
ptCent[0].y = (pt1[0]->y + pt1[1]->y + pt1[2]->y + pt1[3]->y)/4;
//printf("%d %d\n", ptCent[0].x, ptCent[0].y);
//cvCircle(ret, ptCent[0], 5, cvScalar(255));
CvPoint pC[1];
pC[0].x = ptCent[0].x;
pC[0].y = ptCent[0].y;
int width = (int) cvGetCaptureProperty (capture, CV_CAP_PROP_FRAME_WIDTH);
int height = (int) cvGetCaptureProperty (capture, CV_CAP_PROP_FRAME_HEIGHT);
pC[0].x = width - ptCent[0].x;
pC[0].x = (pC[0].x * 100) / width;
pC[0].y = (ptCent[0].y * 100) / height;
Square test3 (countSquare, pC [0], abs(pt1[0]->x - pC[0].x) * 1.5, abs(pt1[0]->x - pC[0].x) * 1.5, angle);
cubes.push_back(test3);
}
}
contours = contours->h_next;
}
cvReleaseImage(&temp);
//cvReleaseImage(&ret);
cvReleaseImage(&Img);
cvReleaseMemStorage(&storage);
}