diff --git a/modules/gpu/include/opencv2/gpu/gpu.hpp b/modules/gpu/include/opencv2/gpu/gpu.hpp
index 1af408881b..ff6f279612 100644
--- a/modules/gpu/include/opencv2/gpu/gpu.hpp
+++ b/modules/gpu/include/opencv2/gpu/gpu.hpp
@@ -351,25 +351,25 @@ namespace cv
         //! supports CV_8UC1, CV_8UC4, CV_32SC1, CV_32FC1 types
         CV_EXPORTS void add(const GpuMat& a, const GpuMat& b, GpuMat& c);
         //! adds scalar to a matrix (c = a + s)
-        //! supports only CV_32FC1 type
+        //! supports CV_32FC1 and CV_32FC2 type
         CV_EXPORTS void add(const GpuMat& a, const Scalar& sc, GpuMat& c);
         //! subtracts one matrix from another (c = a - b)
         //! supports CV_8UC1, CV_8UC4, CV_32SC1, CV_32FC1 types
 		CV_EXPORTS void subtract(const GpuMat& a, const GpuMat& b, GpuMat& c);
         //! subtracts scalar from a matrix (c = a - s)
-        //! supports only CV_32FC1 type
+        //! supports CV_32FC1 and CV_32FC2 type
         CV_EXPORTS void subtract(const GpuMat& a, const Scalar& sc, GpuMat& c);
         //! computes element-wise product of the two arrays (c = a * b)
         //! supports CV_8UC1, CV_8UC4, CV_32SC1, CV_32FC1 types
 		CV_EXPORTS void multiply(const GpuMat& a, const GpuMat& b, GpuMat& c);
         //! multiplies matrix to a scalar (c = a * s)
-        //! supports only CV_32FC1 type
+        //! supports CV_32FC1 and CV_32FC2 type
         CV_EXPORTS void multiply(const GpuMat& a, const Scalar& sc, GpuMat& c);
         //! computes element-wise quotient of the two arrays (c = a / b)
         //! supports CV_8UC1, CV_8UC4, CV_32SC1, CV_32FC1 types
 		CV_EXPORTS void divide(const GpuMat& a, const GpuMat& b, GpuMat& c);
         //! computes element-wise quotient of matrix and scalar (c = a / s)
-        //! supports only CV_32FC1 type
+        //! supports CV_32FC1 and CV_32FC2 type
         CV_EXPORTS void divide(const GpuMat& a, const Scalar& sc, GpuMat& c);
 
         //! transposes the matrix
@@ -452,13 +452,20 @@ namespace cv
         //! supports only CV_32FC1 type
         CV_EXPORTS void log(const GpuMat& a, GpuMat& b);
 
-        //! computes magnitude (magnitude(i)) of each (x(i), y(i)) vector
+        //! computes magnitude of each (x(i), y(i)) vector
         //! supports only CV_32FC1 type
         CV_EXPORTS void magnitude(const GpuMat& x, const GpuMat& y, GpuMat& magnitude);
-        //! computes magnitude (magnitude(i)) of complex (x(i).re, x(i).im) vector
+        //! computes magnitude of complex (x(i).re, x(i).im) vector
         //! supports only CV_32FC2 type
         CV_EXPORTS void magnitude(const GpuMat& x, GpuMat& magnitude);
 
+        //! computes squared magnitude of each (x(i), y(i)) vector
+        //! supports only CV_32FC1 type
+        CV_EXPORTS void magnitudeSqr(const GpuMat& x, const GpuMat& y, GpuMat& magnitude);
+        //! computes squared magnitude of complex (x(i).re, x(i).im) vector
+        //! supports only CV_32FC2 type
+        CV_EXPORTS void magnitudeSqr(const GpuMat& x, GpuMat& magnitude);
+
         ////////////////////////////// Image processing //////////////////////////////
 
         //! DST[x,y] = SRC[xmap[x,y],ymap[x,y]] with bilinear interpolation.
@@ -529,6 +536,11 @@ namespace cv
         //! supports only CV_32FC1 source type
         CV_EXPORTS void integral(GpuMat& src, GpuMat& sum, GpuMat& sqsum);
 
+        //! 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);
+
         //! applies Canny edge detector and produces the edge map
         //! supprots only CV_8UC1 source type
         //! disabled until fix crash
@@ -718,14 +730,14 @@ namespace cv
         //! Output hist[i] will have one row and histSize[i] cols and CV_32SC1 type.
         CV_EXPORTS void histEven(const GpuMat& src, GpuMat hist[4], int histSize[4], int lowerLevel[4], int upperLevel[4]);
         //! Calculates histogram with bins determined by levels array.
-        //! levels must have one row and CV_32SC1 type.
-        //! Supports CV_8UC1, CV_16UC1 and CV_16SC1 source types.
+        //! levels must have one row and CV_32SC1 type if source has integer type or CV_32FC1 otherwise.
+        //! Supports CV_8UC1, CV_16UC1, CV_16SC1 and CV_32FC1 source types.
         //! Output hist will have one row and (levels.cols-1) cols and CV_32SC1 type.
         CV_EXPORTS void histRange(const GpuMat& src, GpuMat& hist, const GpuMat& levels);
         //! Calculates histogram with bins determined by levels array.
-        //! All levels must have one row and CV_32SC1 type.
+        //! All levels must have one row and CV_32SC1 type if source has integer type or CV_32FC1 otherwise.
         //! All channels of source are processed separately.
-        //! Supports CV_8UC4, CV_16UC4 and CV_16SC4 source types.
+        //! Supports CV_8UC4, CV_16UC4, CV_16SC4 and CV_32FC4 source types.
         //! Output hist[i] will have one row and (levels[i].cols-1) cols and CV_32SC1 type.
         CV_EXPORTS void histRange(const GpuMat& src, GpuMat hist[4], const GpuMat levels[4]);
 
diff --git a/modules/gpu/src/arithm.cpp b/modules/gpu/src/arithm.cpp
index e5eef84fb8..3d66a78115 100644
--- a/modules/gpu/src/arithm.cpp
+++ b/modules/gpu/src/arithm.cpp
@@ -71,6 +71,8 @@ void cv::gpu::exp(const GpuMat&, GpuMat&) { throw_nogpu(); }
 void cv::gpu::log(const GpuMat&, GpuMat&) { throw_nogpu(); }
 void cv::gpu::magnitude(const GpuMat&, const GpuMat&, GpuMat&) { throw_nogpu(); }
 void cv::gpu::magnitude(const GpuMat&, GpuMat&) { throw_nogpu(); }
+void cv::gpu::magnitudeSqr(const GpuMat&, const GpuMat&, GpuMat&) { throw_nogpu(); }
+void cv::gpu::magnitudeSqr(const GpuMat&, GpuMat&) { throw_nogpu(); }
 
 #else /* !defined (HAVE_CUDA) */
 
@@ -127,22 +129,49 @@ namespace
         }
 	}
 
-    typedef NppStatus (*npp_arithm_scalar_32f_t)(const Npp32f *pSrc, int nSrcStep, Npp32f nValue, Npp32f *pDst, 
-                                                 int nDstStep, NppiSize oSizeROI);
+    template<int SCN> struct NppArithmScalarFunc;
+    template<> struct NppArithmScalarFunc<1>
+    {
+        typedef NppStatus (*func_ptr)(const Npp32f *pSrc, int nSrcStep, Npp32f nValue, Npp32f *pDst, 
+                                      int nDstStep, NppiSize oSizeROI);
+    };
+    template<> struct NppArithmScalarFunc<2>
+    {        
+        typedef NppStatus (*func_ptr)(const Npp32fc *pSrc, int nSrcStep, Npp32fc nValue, Npp32fc *pDst,
+                                      int nDstStep, NppiSize oSizeROI);
+    };
 
-    void nppArithmCaller(const GpuMat& src1, const Scalar& sc, GpuMat& dst, 
-					     npp_arithm_scalar_32f_t npp_func)
-	{
-        CV_Assert(src1.type() == CV_32FC1);
+    template<int SCN, typename NppArithmScalarFunc<SCN>::func_ptr func> struct NppArithmScalar;
+    template<typename NppArithmScalarFunc<1>::func_ptr func> struct NppArithmScalar<1, func>
+    {
+        static void calc(const GpuMat& src, const Scalar& sc, GpuMat& dst)
+	    {
+            dst.create(src.size(), src.type());
 
-        dst.create(src1.size(), src1.type());
+		    NppiSize sz;
+		    sz.width  = src.cols;
+		    sz.height = src.rows;
 
-		NppiSize sz;
-		sz.width  = src1.cols;
-		sz.height = src1.rows;
+		    nppSafeCall( func(src.ptr<Npp32f>(), src.step, (Npp32f)sc[0], dst.ptr<Npp32f>(), dst.step, sz) );
+	    }
+    };
+    template<typename NppArithmScalarFunc<2>::func_ptr func> struct NppArithmScalar<2, func>
+    {
+        static void calc(const GpuMat& src, const Scalar& sc, GpuMat& dst)
+	    {
+            dst.create(src.size(), src.type());
 
-		nppSafeCall( npp_func(src1.ptr<Npp32f>(), src1.step, (Npp32f)sc[0], dst.ptr<Npp32f>(), dst.step, sz) );
-	}
+		    NppiSize sz;
+		    sz.width  = src.cols;
+		    sz.height = src.rows;
+
+            Npp32fc nValue;
+            nValue.re = (Npp32f)sc[0];
+            nValue.im = (Npp32f)sc[1];
+
+		    nppSafeCall( func(src.ptr<Npp32fc>(), src.step, nValue, dst.ptr<Npp32fc>(), dst.step, sz) );
+	    }
+    };
 }
 
 void cv::gpu::add(const GpuMat& src1, const GpuMat& src2, GpuMat& dst)
@@ -167,22 +196,42 @@ void cv::gpu::divide(const GpuMat& src1, const GpuMat& src2, GpuMat& dst)
 
 void cv::gpu::add(const GpuMat& src, const Scalar& sc, GpuMat& dst)
 {
-    nppArithmCaller(src, sc, dst, nppiAddC_32f_C1R);
+    typedef void (*caller_t)(const GpuMat& src, const Scalar& sc, GpuMat& dst);
+    static const caller_t callers[] = {NppArithmScalar<1, nppiAddC_32f_C1R>::calc, NppArithmScalar<2, nppiAddC_32fc_C1R>::calc};
+
+    CV_Assert(src.type() == CV_32FC1 || src.type() == CV_32FC2);
+
+    callers[src.channels()](src, sc, dst);
 }
 
 void cv::gpu::subtract(const GpuMat& src, const Scalar& sc, GpuMat& dst)
 {
-    nppArithmCaller(src, sc, dst, nppiSubC_32f_C1R);
+    typedef void (*caller_t)(const GpuMat& src, const Scalar& sc, GpuMat& dst);
+    static const caller_t callers[] = {NppArithmScalar<1, nppiSubC_32f_C1R>::calc, NppArithmScalar<2, nppiSubC_32fc_C1R>::calc};
+
+    CV_Assert(src.type() == CV_32FC1 || src.type() == CV_32FC2);
+
+    callers[src.channels()](src, sc, dst);
 }
 
 void cv::gpu::multiply(const GpuMat& src, const Scalar& sc, GpuMat& dst)
 {
-    nppArithmCaller(src, sc, dst, nppiMulC_32f_C1R);
+    typedef void (*caller_t)(const GpuMat& src, const Scalar& sc, GpuMat& dst);
+    static const caller_t callers[] = {NppArithmScalar<1, nppiMulC_32f_C1R>::calc, NppArithmScalar<2, nppiMulC_32fc_C1R>::calc};
+
+    CV_Assert(src.type() == CV_32FC1 || src.type() == CV_32FC2);
+
+    callers[src.channels()](src, sc, dst);
 }
 
 void cv::gpu::divide(const GpuMat& src, const Scalar& sc, GpuMat& dst)
 {
-    nppArithmCaller(src, sc, dst, nppiDivC_32f_C1R);
+    typedef void (*caller_t)(const GpuMat& src, const Scalar& sc, GpuMat& dst);
+    static const caller_t callers[] = {NppArithmScalar<1, nppiDivC_32f_C1R>::calc, NppArithmScalar<2, nppiDivC_32fc_C1R>::calc};
+
+    CV_Assert(src.type() == CV_32FC1 || src.type() == CV_32FC2);
+
+    callers[src.channels()](src, sc, dst);
 }
 
 ////////////////////////////////////////////////////////////////////////
@@ -589,4 +638,29 @@ void cv::gpu::magnitude(const GpuMat& src1, const GpuMat& src2, GpuMat& dst)
     cv::gpu::magnitude(src, dst);
 }
 
+void cv::gpu::magnitudeSqr(const GpuMat& src, GpuMat& dst)
+{
+    CV_Assert(src.type() == CV_32FC2);
+
+    dst.create(src.size(), CV_32FC1);
+
+    NppiSize sz;
+    sz.width = src.cols;
+    sz.height = src.rows;
+
+    nppSafeCall( nppiMagnitudeSqr_32fc32f_C1R(src.ptr<Npp32fc>(), src.step, dst.ptr<Npp32f>(), dst.step, sz) );
+}
+
+void cv::gpu::magnitudeSqr(const GpuMat& src1, const GpuMat& src2, GpuMat& dst)
+{
+    CV_DbgAssert(src1.type() == src2.type() && src1.size() == src2.size());
+    CV_Assert(src1.type() == CV_32FC1);
+
+    GpuMat src(src1.size(), CV_32FC2);
+    GpuMat srcs[] = {src1, src2};
+    cv::gpu::merge(srcs, 2, src);
+
+    cv::gpu::magnitudeSqr(src, dst);
+}
+
 #endif /* !defined (HAVE_CUDA) */
\ No newline at end of file
diff --git a/modules/gpu/src/imgproc_gpu.cpp b/modules/gpu/src/imgproc_gpu.cpp
index 14df108832..73b44498e0 100644
--- a/modules/gpu/src/imgproc_gpu.cpp
+++ b/modules/gpu/src/imgproc_gpu.cpp
@@ -63,6 +63,7 @@ void cv::gpu::warpAffine(const GpuMat&, GpuMat&, const Mat&, Size, int) { throw_
 void cv::gpu::warpPerspective(const GpuMat&, GpuMat&, const Mat&, Size, int) { throw_nogpu(); }
 void cv::gpu::rotate(const GpuMat&, GpuMat&, Size, double, double, double, int) { throw_nogpu(); }
 void cv::gpu::integral(GpuMat&, GpuMat&, GpuMat&) { throw_nogpu(); }
+void cv::gpu::rectStdDev(const GpuMat&, const GpuMat&, GpuMat&, const Rect&) { throw_nogpu(); }
 void cv::gpu::Canny(const GpuMat&, GpuMat&, double, double, int) { throw_nogpu(); }
 void cv::gpu::evenLevels(GpuMat&, int, int, int) { throw_nogpu(); }
 void cv::gpu::histEven(const GpuMat&, GpuMat&, int, int, int) { throw_nogpu(); }
@@ -970,6 +971,26 @@ void cv::gpu::integral(GpuMat& src, GpuMat& sum, GpuMat& sqsum)
         sum.step, sqsum.ptr<Npp32f>(), sqsum.step, sz, 0, 0.0f, h) );
 }
 
+void cv::gpu::rectStdDev(const GpuMat& src, const GpuMat& sqr, GpuMat& dst, const Rect& rect)
+{
+    CV_Assert(src.type() == CV_32SC1 && sqr.type() == CV_32FC1);
+
+    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;
+
+    nppSafeCall( nppiRectStdDev_32s32f_C1R(src.ptr<Npp32s>(), src.step, sqr.ptr<Npp32f>(), sqr.step,
+		dst.ptr<Npp32f>(), dst.step, sz, nppRect) );
+}
+
 ////////////////////////////////////////////////////////////////////////
 // Canny
 
@@ -1009,6 +1030,7 @@ namespace
     template<> struct NPPTypeTraits<CV_8U>  { typedef Npp8u npp_type; };
     template<> struct NPPTypeTraits<CV_16U> { typedef Npp16u npp_type; };
     template<> struct NPPTypeTraits<CV_16S> { typedef Npp16s npp_type; };
+    template<> struct NPPTypeTraits<CV_32F> { typedef Npp32f npp_type; };
     
     typedef NppStatus (*get_buf_size_c1_t)(NppiSize oSizeROI, int nLevels, int* hpBufferSize);
     typedef NppStatus (*get_buf_size_c4_t)(NppiSize oSizeROI, int nLevels[], int* hpBufferSize);
@@ -1082,26 +1104,50 @@ namespace
     template<int SDEPTH> struct NppHistogramRangeFuncC1
     {
         typedef typename NPPTypeTraits<SDEPTH>::npp_type src_t;
+        typedef Npp32s level_t;
+        enum {LEVEL_TYPE_CODE=CV_32SC1};
 
         typedef NppStatus (*func_ptr)(const src_t* pSrc, int nSrcStep, NppiSize oSizeROI, Npp32s* pHist, 
             const Npp32s* pLevels, int nLevels, Npp8u* pBuffer);
     };
+    template<> struct NppHistogramRangeFuncC1<CV_32F>
+    {
+        typedef Npp32f src_t;
+        typedef Npp32f level_t;
+        enum {LEVEL_TYPE_CODE=CV_32FC1};
+
+        typedef NppStatus (*func_ptr)(const Npp32f* pSrc, int nSrcStep, NppiSize oSizeROI, Npp32s* pHist, 
+            const Npp32f* pLevels, int nLevels, Npp8u* pBuffer);
+    };
     template<int SDEPTH> struct NppHistogramRangeFuncC4
     {
         typedef typename NPPTypeTraits<SDEPTH>::npp_type src_t;
+        typedef Npp32s level_t;
+        enum {LEVEL_TYPE_CODE=CV_32SC1};
 
         typedef NppStatus (*func_ptr)(const src_t* pSrc, int nSrcStep, NppiSize oSizeROI, Npp32s* pHist[4], 
             const Npp32s* pLevels[4], int nLevels[4], Npp8u* pBuffer);
     };
+    template<> struct NppHistogramRangeFuncC4<CV_32F>
+    {
+        typedef Npp32f src_t;
+        typedef Npp32f level_t;
+        enum {LEVEL_TYPE_CODE=CV_32FC1};
+
+        typedef NppStatus (*func_ptr)(const Npp32f* pSrc, int nSrcStep, NppiSize oSizeROI, Npp32s* pHist[4], 
+            const Npp32f* pLevels[4], int nLevels[4], Npp8u* pBuffer);
+    };
         
     template<int SDEPTH, typename NppHistogramRangeFuncC1<SDEPTH>::func_ptr func, get_buf_size_c1_t get_buf_size> 
     struct NppHistogramRangeC1
     { 
         typedef typename NppHistogramRangeFuncC1<SDEPTH>::src_t src_t;
+        typedef typename NppHistogramRangeFuncC1<SDEPTH>::level_t level_t;
+        enum {LEVEL_TYPE_CODE=NppHistogramRangeFuncC1<SDEPTH>::LEVEL_TYPE_CODE};
 
         static void hist(const GpuMat& src, GpuMat& hist, const GpuMat& levels)
         {            
-            CV_Assert(levels.type() == CV_32SC1 && levels.rows == 1);
+            CV_Assert(levels.type() == LEVEL_TYPE_CODE && levels.rows == 1);
 
             hist.create(1, levels.cols - 1, CV_32S);
 
@@ -1114,20 +1160,22 @@ namespace
 
             get_buf_size(sz, levels.cols, &buf_size);
             buffer.create(1, buf_size, CV_8U);
-            nppSafeCall( func(src.ptr<src_t>(), src.step, sz, hist.ptr<Npp32s>(), levels.ptr<Npp32s>(), levels.cols, buffer.ptr<Npp8u>()) );
+            nppSafeCall( func(src.ptr<src_t>(), src.step, sz, hist.ptr<Npp32s>(), levels.ptr<level_t>(), levels.cols, buffer.ptr<Npp8u>()) );
         }
     }; 
     template<int SDEPTH, typename NppHistogramRangeFuncC4<SDEPTH>::func_ptr func, get_buf_size_c4_t get_buf_size> 
     struct NppHistogramRangeC4
     { 
         typedef typename NppHistogramRangeFuncC4<SDEPTH>::src_t src_t;
+        typedef typename NppHistogramRangeFuncC1<SDEPTH>::level_t level_t;
+        enum {LEVEL_TYPE_CODE=NppHistogramRangeFuncC1<SDEPTH>::LEVEL_TYPE_CODE};
 
         static void hist(const GpuMat& src, GpuMat hist[4], const GpuMat levels[4])
         {
-            CV_Assert(levels[0].type() == CV_32SC1 && levels[0].rows == 1);
-            CV_Assert(levels[1].type() == CV_32SC1 && levels[1].rows == 1);
-            CV_Assert(levels[2].type() == CV_32SC1 && levels[2].rows == 1);
-            CV_Assert(levels[3].type() == CV_32SC1 && levels[3].rows == 1);
+            CV_Assert(levels[0].type() == LEVEL_TYPE_CODE && levels[0].rows == 1);
+            CV_Assert(levels[1].type() == LEVEL_TYPE_CODE && levels[1].rows == 1);
+            CV_Assert(levels[2].type() == LEVEL_TYPE_CODE && levels[2].rows == 1);
+            CV_Assert(levels[3].type() == LEVEL_TYPE_CODE && levels[3].rows == 1);
 
             hist[0].create(1, levels[0].cols - 1, CV_32S);
             hist[1].create(1, levels[1].cols - 1, CV_32S);
@@ -1136,7 +1184,7 @@ namespace
 
             Npp32s* pHist[] = {hist[0].ptr<Npp32s>(), hist[1].ptr<Npp32s>(), hist[2].ptr<Npp32s>(), hist[3].ptr<Npp32s>()};
             int nLevels[] = {levels[0].cols, levels[1].cols, levels[2].cols, levels[3].cols};
-            const Npp32s* pLevels[] = {levels[0].ptr<Npp32s>(), levels[1].ptr<Npp32s>(), levels[2].ptr<Npp32s>(), levels[3].ptr<Npp32s>()};
+            const level_t* pLevels[] = {levels[0].ptr<level_t>(), levels[1].ptr<level_t>(), levels[2].ptr<level_t>(), levels[3].ptr<level_t>()};
 
             NppiSize sz;
             sz.width = src.cols;
@@ -1193,7 +1241,7 @@ void cv::gpu::histEven(const GpuMat& src, GpuMat hist[4], int histSize[4], int l
 
 void cv::gpu::histRange(const GpuMat& src, GpuMat& hist, const GpuMat& levels)
 {
-    CV_Assert(src.type() == CV_8UC1 || src.type() == CV_16UC1 || src.type() == CV_16SC1);
+    CV_Assert(src.type() == CV_8UC1 || src.type() == CV_16UC1 || src.type() == CV_16SC1 || src.type() == CV_32FC1);
 
     typedef void (*hist_t)(const GpuMat& src, GpuMat& hist, const GpuMat& levels);
     static const hist_t hist_callers[] = 
@@ -1201,7 +1249,9 @@ void cv::gpu::histRange(const GpuMat& src, GpuMat& hist, const GpuMat& levels)
         NppHistogramRangeC1<CV_8U , nppiHistogramRange_8u_C1R , nppiHistogramRangeGetBufferSize_8u_C1R >::hist,
         0,
         NppHistogramRangeC1<CV_16U, nppiHistogramRange_16u_C1R, nppiHistogramRangeGetBufferSize_16u_C1R>::hist,
-        NppHistogramRangeC1<CV_16S, nppiHistogramRange_16s_C1R, nppiHistogramRangeGetBufferSize_16s_C1R>::hist
+        NppHistogramRangeC1<CV_16S, nppiHistogramRange_16s_C1R, nppiHistogramRangeGetBufferSize_16s_C1R>::hist,
+        0,
+        NppHistogramRangeC1<CV_32F, nppiHistogramRange_32f_C1R, nppiHistogramRangeGetBufferSize_32f_C1R>::hist
     };
 
     hist_callers[src.depth()](src, hist, levels);
@@ -1209,7 +1259,7 @@ void cv::gpu::histRange(const GpuMat& src, GpuMat& hist, const GpuMat& levels)
 
 void cv::gpu::histRange(const GpuMat& src, GpuMat hist[4], const GpuMat levels[4])
 {
-    CV_Assert(src.type() == CV_8UC4 || src.type() == CV_16UC4 || src.type() == CV_16SC4);
+    CV_Assert(src.type() == CV_8UC4 || src.type() == CV_16UC4 || src.type() == CV_16SC4 || src.type() == CV_32FC4);
 
     typedef void (*hist_t)(const GpuMat& src, GpuMat hist[4], const GpuMat levels[4]);
     static const hist_t hist_callers[] = 
@@ -1217,7 +1267,9 @@ void cv::gpu::histRange(const GpuMat& src, GpuMat hist[4], const GpuMat levels[4
         NppHistogramRangeC4<CV_8U , nppiHistogramRange_8u_C4R , nppiHistogramRangeGetBufferSize_8u_C4R >::hist,
         0,
         NppHistogramRangeC4<CV_16U, nppiHistogramRange_16u_C4R, nppiHistogramRangeGetBufferSize_16u_C4R>::hist,
-        NppHistogramRangeC4<CV_16S, nppiHistogramRange_16s_C4R, nppiHistogramRangeGetBufferSize_16s_C4R>::hist
+        NppHistogramRangeC4<CV_16S, nppiHistogramRange_16s_C4R, nppiHistogramRangeGetBufferSize_16s_C4R>::hist,
+        0,
+        NppHistogramRangeC4<CV_32F, nppiHistogramRange_32f_C4R, nppiHistogramRangeGetBufferSize_32f_C4R>::hist
     };
 
     hist_callers[src.depth()](src, hist, levels);
diff --git a/tests/gpu/src/arithm.cpp b/tests/gpu/src/arithm.cpp
index e906288c70..67e2de70e8 100644
--- a/tests/gpu/src/arithm.cpp
+++ b/tests/gpu/src/arithm.cpp
@@ -43,9 +43,6 @@
 #include <cmath>
 #include <limits>
 #include "gputest.hpp"
-#include "opencv2/core/core.hpp"
-#include "opencv2/imgproc/imgproc.hpp"
-#include "opencv2/highgui/highgui.hpp"
 
 using namespace cv;
 using namespace std;
diff --git a/tests/gpu/src/filters.cpp b/tests/gpu/src/filters.cpp
index 1505da659a..7c151ec06c 100644
--- a/tests/gpu/src/filters.cpp
+++ b/tests/gpu/src/filters.cpp
@@ -43,8 +43,6 @@
 #include <cmath>
 #include <limits>
 #include "gputest.hpp"
-#include "opencv2/imgproc/imgproc.hpp"
-#include "opencv2/highgui/highgui.hpp"
 
 using namespace cv;
 using namespace std;
diff --git a/tests/gpu/src/gputest.hpp b/tests/gpu/src/gputest.hpp
index 57afd493a0..d1edf80d3b 100644
--- a/tests/gpu/src/gputest.hpp
+++ b/tests/gpu/src/gputest.hpp
@@ -48,7 +48,9 @@
 #undef max
 #endif
 
-#include "opencv2/gpu/gpu.hpp"
+#include <opencv2/gpu/gpu.hpp>
+#include <opencv2/highgui/highgui.hpp>
+#include <opencv2/imgproc/imgproc.hpp>
 #include "cxts.h"
 
 /****************************************************************************************/
diff --git a/tests/gpu/src/imgproc_gpu.cpp b/tests/gpu/src/imgproc_gpu.cpp
index fae1f93e5c..3c1b289f33 100644
--- a/tests/gpu/src/imgproc_gpu.cpp
+++ b/tests/gpu/src/imgproc_gpu.cpp
@@ -43,9 +43,6 @@
 #include <cmath>
 #include <limits>
 #include "gputest.hpp"
-#include "opencv2/core/core.hpp"
-#include "opencv2/imgproc/imgproc.hpp"
-#include "opencv2/highgui/highgui.hpp"
 
 using namespace cv;
 using namespace std;
diff --git a/tests/gpu/src/meanshift.cpp b/tests/gpu/src/meanshift.cpp
index 3ac3790202..e46ee86000 100644
--- a/tests/gpu/src/meanshift.cpp
+++ b/tests/gpu/src/meanshift.cpp
@@ -40,9 +40,6 @@
 //M*/
 
 #include "gputest.hpp"
-#include <opencv2/opencv.hpp>
-#include <opencv2/gpu/gpu.hpp>
-
 #include <iostream>
 #include <string>
 
diff --git a/tests/gpu/src/mssegmentation.cpp b/tests/gpu/src/mssegmentation.cpp
index 478de07529..77d8447a11 100644
--- a/tests/gpu/src/mssegmentation.cpp
+++ b/tests/gpu/src/mssegmentation.cpp
@@ -39,8 +39,6 @@
 //
 //M*/
 
-#include <opencv2/opencv.hpp>
-#include <opencv2/gpu/gpu.hpp>
 #include <iostream>
 #include <string>
 #include <iosfwd>
diff --git a/tests/gpu/src/operator_async_call.cpp b/tests/gpu/src/operator_async_call.cpp
index 2a8b19a93f..b63d399485 100644
--- a/tests/gpu/src/operator_async_call.cpp
+++ b/tests/gpu/src/operator_async_call.cpp
@@ -40,9 +40,6 @@
 //M*/
 
 #include "gputest.hpp"
-#include "highgui.h"
-#include "cv.h"
-
 #include <string>
 #include <iostream>
 #include <fstream>
diff --git a/tests/gpu/src/operator_copy_to.cpp b/tests/gpu/src/operator_copy_to.cpp
index 56a6f115c8..21ce7eb536 100644
--- a/tests/gpu/src/operator_copy_to.cpp
+++ b/tests/gpu/src/operator_copy_to.cpp
@@ -40,9 +40,6 @@
 //M*/
 
 #include "gputest.hpp"
-#include "highgui.h"
-#include "cv.h"
-
 #include <string>
 #include <iostream>
 #include <fstream>
diff --git a/tests/gpu/src/split_merge.cpp b/tests/gpu/src/split_merge.cpp
index 8df7bacda9..12f922e845 100644
--- a/tests/gpu/src/split_merge.cpp
+++ b/tests/gpu/src/split_merge.cpp
@@ -41,8 +41,6 @@
 //M*/
 
 #include "gputest.hpp"
-#include <opencv2/opencv.hpp>
-#include <opencv2/gpu/gpu.hpp>
 
 #include <iostream>
 #include <string>
diff --git a/tests/gpu/src/stereo_bm.cpp b/tests/gpu/src/stereo_bm.cpp
index b9211e0476..5bf1bb9610 100644
--- a/tests/gpu/src/stereo_bm.cpp
+++ b/tests/gpu/src/stereo_bm.cpp
@@ -40,9 +40,6 @@
 //M*/
 
 #include "gputest.hpp"
-#include <opencv2/opencv.hpp>
-#include <opencv2/gpu/gpu.hpp>
-
 #include <iostream>
 #include <string>
 
diff --git a/tests/gpu/src/stereo_bm_async.cpp b/tests/gpu/src/stereo_bm_async.cpp
index 0ffc3f4aa6..422cf89510 100644
--- a/tests/gpu/src/stereo_bm_async.cpp
+++ b/tests/gpu/src/stereo_bm_async.cpp
@@ -40,9 +40,6 @@
 //M*/
 
 #include "gputest.hpp"
-#include "highgui.h"
-#include "cv.h"
-
 
 using namespace cv;
 using namespace std;
diff --git a/tests/gpu/src/stereo_bp.cpp b/tests/gpu/src/stereo_bp.cpp
index 601c78feef..8849fbdad4 100644
--- a/tests/gpu/src/stereo_bp.cpp
+++ b/tests/gpu/src/stereo_bp.cpp
@@ -40,13 +40,9 @@
 //M*/
 
 #include "gputest.hpp"
-#include <opencv2/opencv.hpp>
-#include <opencv2/gpu/gpu.hpp>
-
 #include <iostream>
 #include <string>
 
-
 struct CV_GpuStereoBPTest : public CvTest
 {
     CV_GpuStereoBPTest() : CvTest( "GPU-StereoBP", "StereoBP" ){}
diff --git a/tests/gpu/src/stereo_csbp.cpp b/tests/gpu/src/stereo_csbp.cpp
index c0cacb732f..6ea4327ff3 100644
--- a/tests/gpu/src/stereo_csbp.cpp
+++ b/tests/gpu/src/stereo_csbp.cpp
@@ -40,9 +40,6 @@
 //M*/
 
 #include "gputest.hpp"
-#include <opencv2/opencv.hpp>
-#include <opencv2/gpu/gpu.hpp>
-
 #include <iostream>
 #include <string>