Merge pull request #25553 from asmorkalov:as/HAL_min_max_idx

Fix HAL interface for hal_ni_minMaxIdx #25553

Fixes https://github.com/opencv/opencv/issues/25540

The original implementation call HAL with the same parameters independently from amount of channels. The patch uses HAL correctly for the case cn > 1.

### Pull Request Readiness Checklist

See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request

- [x] I agree to contribute to the project under Apache 2 License.
- [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [x] The PR is proposed to the proper branch
- [x] There is a reference to the original bug report and related work
- [ ] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
      Patch to opencv_extra has the same branch name.
- [ ] The feature is well documented and sample code can be built with the project CMake

modules/core/include/opencv2/core.hpp

@@ -912,11 +912,8 @@ CV_EXPORTS_W void reduceArgMax(InputArray src, OutputArray dst, int axis, bool l
 
 The function cv::minMaxIdx finds the minimum and maximum element values and their positions. The
 extremums are searched across the whole array or, if mask is not an empty array, in the specified
-array region. The function does not work with multi-channel arrays. If you need to find minimum or
+array region. In case of a sparse matrix, the minimum is found among non-zero elements
-maximum elements across all the channels, use Mat::reshape first to reinterpret the array as
+only. Multi-channel input is supported without mask and extremums indexes (should be nullptr).
-single-channel. Or you may extract the particular channel using either extractImageCOI, or
-mixChannels, or split. In case of a sparse matrix, the minimum is found among non-zero elements
-only.
 @note When minIdx is not NULL, it must have at least 2 elements (as well as maxIdx), even if src is
 a single-row or single-column matrix. In OpenCV (following MATLAB) each array has at least 2
 dimensions, i.e. single-column matrix is Mx1 matrix (and therefore minIdx/maxIdx will be

modules/core/src/minmax.cpp

@@ -1510,9 +1510,19 @@ void cv::minMaxIdx(InputArray _src, double* minVal,
 
     Mat src = _src.getMat(), mask = _mask.getMat();
 
+    int _minIdx, _maxIdx;
+    int* min_offset = (cn == 1) ? minIdx : &_minIdx;
+    int* max_offset = (cn == 1) ? maxIdx : &_maxIdx;
     if (src.dims <= 2)
-        CALL_HAL(minMaxIdx, cv_hal_minMaxIdx, src.data, src.step, src.cols, src.rows, src.depth(), minVal, maxVal,
+    {
-                 minIdx, maxIdx, mask.data);
+        CALL_HAL(minMaxIdx, cv_hal_minMaxIdx, src.data, src.step, src.cols*cn, src.rows, src.depth(),
+                    minVal, maxVal, min_offset, max_offset, mask.data);
+    }
+    else if (src.isContinuous())
+    {
+        CALL_HAL(minMaxIdx, cv_hal_minMaxIdx, src.data, 0, (int)src.total()*cn, 1, src.depth(),
+                 minVal, maxVal, min_offset, max_offset, mask.data);
+    }
 
     CV_OVX_RUN(!ovx::skipSmallImages<VX_KERNEL_MINMAXLOC>(src.cols, src.rows),
                openvx_minMaxIdx(src, minVal, maxVal, minIdx, maxIdx, mask))