Fill / Fill OpenCV C ++ Letters

Question

I am trying OCR OCR in the following image, and as you can see, many of the letters are not complete. They miss parts that cause some bad results from Tesseract.

I tried to undermine the image using the following code, but it doesn't seem to do much:

Mat eroded;
double element_size = 25;
RNG rng(12345);
Mat element = getStructuringElement( cv::MORPH_ELLIPSE,cv::Size( 2*element_size + 1, 2*element_size+1 ),cv::Point( element_size, element_size ) );
erode(src, eroded, element);

Does anyone know how I can sharpen this text so that the letters become consistent?

Answer 1

Remembering digital image processing, I think Laplacian-Gaussian or Sobel filtering . /, (.. , ). , OCR, , , . . , , (.. , ), .

Answer 2

, , , , , . ( api opencv) , . . ( , ). , .

, , .

Answer 3

OCR , , ? .

Think so. You have an image and some parts of this image are missing. And you are trying to figure out what these parts are. This is not a problem of sharpness. This is a huge problem. Therefore, I think you should get the best binary image from the original image to perform OCR.

Good luck.