c++ - Cropping out an image from an existing image -

i crop out image existing image. i've taken image , applied monochrome on threshold 98% using imagemagick (is doable in opencv?)

the resulting image this:

now image crop out image final image looks this:

question how can in opencv? note, reason want crop image can use this answer part of text. if there no need crop out new image , instead concentrate on black part of image begin with, great.

if text @ top , @ bottom regions want crop out, if @ same location solution easy: just set roi ignores areas:

#include <cv.h> #include <highgui.h>  int main(int argc, char* argv[]) {     cv::mat img = cv::imread(argv[1]);     if (img.empty())     {         std::cout << "!!! imread() failed open target image" << std::endl;         return -1;             }      /* set region of interest */      int offset_x = 129;     int offset_y = 129;      cv::rect roi;     roi.x = offset_x;     roi.y = offset_y;     roi.width = img.size().width - (offset_x*2);     roi.height = img.size().height - (offset_y*2);      /* crop original image defined roi */      cv::mat crop = img(roi);     cv::imshow("crop", crop);     cv::waitkey(0);      cv::imwrite("noises_cropped.png", crop);      return 0; } 

output image:

if position of black rectangle, area of interest, not present on fixed location might want check out another approach: use rectangle detection technique:

on output above, area interested 2nd largest rectangle in image.

on side note, if plan isolate text later, simple cv::erode() remove noises in image left white box & text. technique remove noises use cv::medianblur().you can explore cv::morphologyex() trick:

cv::mat kernel = cv::getstructuringelement(cv::morph_ellipse, cv::size(7, 7), cv::point(3, 3)); cv::morphologyex(src, src, cv::morph_ellipse, kernel);     

a proper solution might combination of these 3. i've demonstrated little bit of on extract hand bones x-ray image.