c++ - OpenCV binary adaptive threshold OCR -

i need convert images binary ocr.

here functions using:

mat binarize(mat & img, mat& res, float blocksize, bool inverse) {     img.convertto(img,cv_32fc1,1.0/255.0);     calcblockmeanvariance(img,res, blocksize, inverse);     res=1.0-res;     res=img+res;     if (inverse) {         cv::threshold(res,res,0.85,1,cv::thresh_binary_inv);     } else {         cv::threshold(res,res,0.85,1,cv::thresh_binary);     }     cv::resize(res,res,cv::size(res.cols/2,res.rows/2));      return res; } 

where calcblockmeanvariance:

void calcblockmeanvariance(mat& img,mat& res,float blockside, bool inverse) //21 blockside - parameter (set greater larger font on image) {     mat i;     img.convertto(i,cv_32fc1);     res=mat::zeros(img.rows/blockside,img.cols/blockside,cv_32fc1);     mat inpaintmask;     mat patch;     mat smallimg;     scalar m,s;      for(int i=0;i<img.rows-blockside;i+=blockside)     {         (int j=0;j<img.cols-blockside;j+=blockside)         {             patch=i(range(i,i+blockside+1),range(j,j+blockside+1));             cv::meanstddev(patch,m,s);             if(s[0]>0.01) // thresholding parameter (set smaller lower contrast image)             {                 res.at<float>(i/blockside,j/blockside)=m[0];             }else             {                 res.at<float>(i/blockside,j/blockside)=0;             }         }     }      cv::resize(i,smallimg,res.size());      if (inverse) {         cv::threshold(res,inpaintmask,0.02,1.0,cv::thresh_binary_inv);     } else {         cv::threshold(res,inpaintmask,0.02,1.0,cv::thresh_binary);     }       mat inpainted;     smallimg.convertto(smallimg,cv_8uc1,255);      inpaintmask.convertto(inpaintmask,cv_8uc1);     inpaint(smallimg, inpaintmask, inpainted, 5, inpaint_telea);      cv::resize(inpainted,res,img.size());     res.convertto(res,cv_32fc1,1.0/255.0);  } 

when passing in 1 calcblockmeanvariance blockside result, have tried raise blockside results in worse results.

before:

after:

can suggest different method converting image binary prep ocr?

thanks.

i think can thresholding using otsu method. can apply on whole image or on blocks of image. did following steps:

• thresholding using otsu method on desired input.
• closing result.

---

python code

image = cv2.imread('image4.png', cv2.imread_grayscale) # reading image if image none:     print 'can not find image!'     exit(-1) # thresholding image using ostu method ret, thresh = cv2.threshold(image, 0, 255, cv2.thresh_binary_inv | cv2.thresh_otsu)  # applying closing operation using ellipse kernel n = 3 kernel = cv2.getstructuringelement(cv2.morph_ellipse, (n, n)) thresh = cv2.morphologyex(thresh, cv2.morph_close, kernel) # showing result cv2.imshow('thresh', thresh) cv2.waitkey(0) cv2.destroyallwindows() 

---

explanation

in first part read input image using imread , checked image opened correctly!.

image = cv2.imread('image4.png', cv2.imread_grayscale) # reading image if image none:     print 'can not find image!'     exit(-1) 

now thresholding image otsu method feeding thresh method thresh_binary_inv | thresh_otsu argument. otsu method works base on optimization problem finding best value thresholding. provided range of possible value threshold value giving lower bound 0 , upper bound 255.

ret, thresh = cv2.threshold(image, 0, 255, cv2.thresh_binary_inv | cv2.thresh_otsu) 

and closing operation done removing black holes in image using ellipse kernel.

kernel = cv2.getstructuringelement(cv2.morph_ellipse, (n, n)) thresh = cv2.morphologyex(thresh, cv2.morph_close, kernel) 

---

result