检查cv :: mat的所有元素是正数还是等于0

Question

I have a matrix cv::mat A . Is there an easy way to check if all elements of this matrix are positive or equal to 0, and return a True boolean if this condition is true?

Answer 1

@francesco answer is nice in terms of C++, but not really effective nor functional tho.

More effective method is using OpenCV built-in operation called checkRange() , it also provides position of first invalid element and support for multi-channel array.

Example usage:

cv::Point *pos;
bool is_ok = cv::checkRange( A, pos, 0, DBL_MAX );

---

As @MarkSetchell suggested in comments, you might also find minimum value in matrix and see if it is non-negative. There is one OpenCV operation on array called minMaxLoc() which unfortunately doesn't work with multi-channel arrays, so you would have to reinterper whole array as single channel, first.

double min;
cv::minMaxLoc( A.reshape(1), &min );

---

To make francescos solution work, you have to specify cv::Mat type to iterators, eg. for CV_8U type of matrix

bool is_all_positive = (std::find_if(m.begin<uchar>(), m.end<uchar>(),
     [](auto x) { return (x < 0); }) == m.end<uchar>());

As I previously said, this is not effective approach and would be (in my opinion) more useable than checkRange() just in case we would like to compare values of specific channels - eg checking only for values of blue color in BGR image

bool is_all_positive = (std::find_if(m.begin<cv::Vec3b>(), m.end<cv::Vec3b>(),
     [](auto x) { return (x[0] < 0); }) == m.end<cv::Vec3b>());

Answer 2

To have an efficient method, rather than using procedures that iterate over all elements of the matrix, you should use a method that stops as soon as a condition is found. You can use std::find_if

#include <algorithm>

bool is_all_positive = (std::find_if(A.begin(), A.end(),
    [](auto x) { return (x < 0); }) == A.end());