Thresholding in Android using opencv -

not sure if right way ask, please help. have image of dented car. have process , highlight dents , return number of dents. able reasonably following result:

the matlab code is:

  img2=rgb2gray(i1);   imshow(img2);   img3=imtophat(img2,strel('disk',15));   img4=imadjust(img3);   layer=img4(:,:,1);   img5=layer>100 & layer<250;   img6=imfill(img5,'holes');   img7=bwareaopen(img6,5);   [l,ans]=bwlabeln(img7);   imshow(img7);   i=imread(i1);   ians=cardentidentification(i); 

however, when try using opencv, this:

with following code:

   imgproc.cvtcolor(source, middle, imgproc.color_rgb2gray);     imgproc.equalizehist(middle, middle);     imgproc.threshold(middle, middle, 150, 255, imgproc.thresh_otsu); 

please tell me how can obtain better results in opencv, , how count dents? tried findcontour() gives large number. tried on other images well, i'm not getting proper results. please help.

so matlab site, imtophat - top-hat filtering computes morphological opening of image (using imopen) , subtracts result original image.

you in opencv following steps:

step 1: disk structuring element

kernel = cv2.getstructuringelement(cv2.morph_ellipse, (15, 15)) 

step 2: compute opening of image , subtract result original image

tophat = cv2.morphologyex(v, cv2.morph_tophat, kernel) 

this gives following result -

step 3 - manually threshold or use otsu -

ret, thresh = cv2.threshold(tophat, 17, 255, 0) 

which gives following image -

since op wants code in java, here probable code in java:

private mat tophat(mat image) { mat element  = imgproc.getstructuringelement(imgproc.morph_ellipse, new size(15, 15), new point (0, 0)); mat dst = new mat; imgproc.morphologyex(image, dst, imgproc.morph_tophat, element, new point(0, 0));  return dst; } 

make sure on gray scale image (cvtype.8uc1) , can threshold suitably.