moved rectStdDev to gpuarithm

modules/gpuarithm/include/opencv2/gpuarithm.hpp

@@ -274,6 +274,11 @@ CV_EXPORTS void reduce(const GpuMat& mtx, GpuMat& vec, int dim, int reduceOp, in
 //! applies fixed threshold to the image
 CV_EXPORTS double threshold(const GpuMat& src, GpuMat& dst, double thresh, double maxval, int type, Stream& stream = Stream::Null());
 
+//! computes the standard deviation of integral images
+//! supports only CV_32SC1 source type and CV_32FC1 sqr type
+//! output will have CV_32FC1 type
+CV_EXPORTS void rectStdDev(const GpuMat& src, const GpuMat& sqr, GpuMat& dst, const Rect& rect, Stream& stream = Stream::Null());
+
 }} // namespace cv { namespace gpu {
 
 #endif /* __OPENCV_GPUARITHM_HPP__ */

modules/gpuarithm/src/matrix_reductions.cpp

@@ -69,6 +69,7 @@ void cv::gpu::minMaxLoc(const GpuMat&, double*, double*, Point*, Point*, const G
 int cv::gpu::countNonZero(const GpuMat&) { throw_no_cuda(); return 0; }
 int cv::gpu::countNonZero(const GpuMat&, GpuMat&) { throw_no_cuda(); return 0; }
 void cv::gpu::reduce(const GpuMat&, GpuMat&, int, int, int, Stream&) { throw_no_cuda(); }
+void cv::gpu::rectStdDev(const GpuMat&, const GpuMat&, GpuMat&, const Rect&, Stream&) { throw_no_cuda(); }
 
 #else
 
@@ -696,4 +697,34 @@ void cv::gpu::reduce(const GpuMat& src, GpuMat& dst, int dim, int reduceOp, int
     }
 }
 
+//////////////////////////////////////////////////////////////////////////////
+// rectStdDev
+
+void cv::gpu::rectStdDev(const GpuMat& src, const GpuMat& sqr, GpuMat& dst, const Rect& rect, Stream& s)
+{
+    CV_Assert(src.type() == CV_32SC1 && sqr.type() == CV_64FC1);
+
+    dst.create(src.size(), CV_32FC1);
+
+    NppiSize sz;
+    sz.width = src.cols;
+    sz.height = src.rows;
+
+    NppiRect nppRect;
+    nppRect.height = rect.height;
+    nppRect.width = rect.width;
+    nppRect.x = rect.x;
+    nppRect.y = rect.y;
+
+    cudaStream_t stream = StreamAccessor::getStream(s);
+
+    NppStreamHandler h(stream);
+
+    nppSafeCall( nppiRectStdDev_32s32f_C1R(src.ptr<Npp32s>(), static_cast<int>(src.step), sqr.ptr<Npp64f>(), static_cast<int>(sqr.step),
+                dst.ptr<Npp32f>(), static_cast<int>(dst.step), sz, nppRect) );
+
+    if (stream == 0)
+        cudaSafeCall( cudaDeviceSynchronize() );
+}
+
 #endif