From 32598639244ada2b5095baf08118a2ec80c6bb37 Mon Sep 17 00:00:00 2001
From: Yuantao Feng <yuantao.feng@outlook.com>
Date: Fri, 20 Jun 2025 17:00:37 +0800
Subject: [PATCH] Merge pull request #27368 from
 fengyuentau:4x/imgproc/CreateHanningWindow-simd

imgproc: vectorize cv::createHanningWindow #27368

### 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
- [ ] 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/imgproc/src/phasecorr.cpp | 47 ++++++++++++++++++++++++++++---
 1 file changed, 43 insertions(+), 4 deletions(-)

diff --git a/modules/imgproc/src/phasecorr.cpp b/modules/imgproc/src/phasecorr.cpp
index d2f88420be..3fbe62883d 100644
--- a/modules/imgproc/src/phasecorr.cpp
+++ b/modules/imgproc/src/phasecorr.cpp
@@ -38,6 +38,7 @@
 
 #include "precomp.hpp"
 #include <vector>
+#include "opencv2/core/hal/intrin.hpp"
 
 namespace cv
 {
@@ -614,8 +615,27 @@ void cv::createHanningWindow(OutputArray _dst, cv::Size winSize, int type)
     double* const wc = _wc.data();
 
     double coeff0 = 2.0 * CV_PI / (double)(cols - 1), coeff1 = 2.0 * CV_PI / (double)(rows - 1);
-    for(int j = 0; j < cols; j++)
-        wc[j] = 0.5 * (1.0 - cos(coeff0 * j));
+    int c = 0;
+#if CV_SIMD_64F || CV_SIMD_SCALABLE_64F
+    const int nlanes32 = VTraits<v_float32>::vlanes();
+    const int nlanes64 = VTraits<v_float64>::vlanes();
+    const int max_nlanes = VTraits<v_float64>::max_nlanes;
+    std::array<double, max_nlanes> index;
+    std::iota(index.data(), index.data()+max_nlanes, 0.f);
+    v_float64 vindex = vx_load(index.data());
+    v_float64 delta = vx_setall_f64(VTraits<v_float64>::vlanes());
+    v_float64 vcoeff0 = vx_setall_f64(coeff0);
+    v_float64 one = vx_setall_f64(1.f);
+    v_float64 half = vx_setall_f64(0.5f);
+    for (; c <= cols - nlanes64; c += nlanes64)
+    {
+        v_float64 v = v_mul(half, v_sub(one, v_cos(v_mul(vcoeff0, vindex))));
+        vx_store(wc + c, v);
+        vindex = v_add(vindex, delta);
+    }
+#endif
+    for(; c < cols; c++)
+        wc[c] = 0.5 * (1.0 - cos(coeff0 * c));
 
     if(dst.depth() == CV_32F)
     {
@@ -623,7 +643,17 @@ void cv::createHanningWindow(OutputArray _dst, cv::Size winSize, int type)
         {
             float* dstData = dst.ptr<float>(i);
             double wr = 0.5 * (1.0 - cos(coeff1 * i));
-            for(int j = 0; j < cols; j++)
+            int j = 0;
+#if CV_SIMD_64F || CV_SIMD_SCALABLE_64F
+            v_float64 vwr = vx_setall_f64(wr);
+            for (; j <= cols - nlanes32; j += nlanes32)
+            {
+                v_float64 v0 = v_mul(vwr, vx_load(wc + j));
+                v_float64 v1 = v_mul(vwr, vx_load(wc + j + nlanes64));
+                vx_store(dstData + j, v_cvt_f32(v0, v1));
+            }
+#endif
+            for(; j < cols; j++)
                 dstData[j] = (float)(wr * wc[j]);
         }
     }
@@ -633,7 +663,16 @@ void cv::createHanningWindow(OutputArray _dst, cv::Size winSize, int type)
         {
             double* dstData = dst.ptr<double>(i);
             double wr = 0.5 * (1.0 - cos(coeff1 * i));
-            for(int j = 0; j < cols; j++)
+            int j = 0;
+#if CV_SIMD_64F || CV_SIMD_SCALABLE_64F
+            v_float64 vwr = vx_setall_f64(wr);
+            for (; j <= cols - nlanes64; j += nlanes64)
+            {
+                v_float64 v = v_mul(vwr, vx_load(wc + j));
+                vx_store(dstData + j, v);
+            }
+#endif
+            for(; j < cols; j++)
                 dstData[j] = wr * wc[j];
         }
     }