diff --git a/复杂算子优化系统2.cpp b/复杂算子优化系统2.cpp
new file mode 100644
index 0000000..15eb4ea
--- /dev/null
+++ b/复杂算子优化系统2.cpp
@@ -0,0 +1,98 @@
+#include <stdio.h>
+#include <time.h>
+
+// 应用可分离的3×3 GaussianBlur到给定的二维数组src上，并将结果存到二维数组dst中
+void applySeparableGaussianBlur(float src[][5], float dst[][5], int h, int w, float kx[3], float ky[3]) {
+    float buf[3][3];  // 用于存储行内卷积的中间结果
+
+    // 宏定义，用于循环利用 buf 数组的3个行缓冲区，确保索引在0~2之间
+#define INDEX_MOD(i) ((i) % 3)
+
+    // 先进行行方向的卷积
+    for (int i = 1; i < h - 1; i++) {
+        for (int j = 1; j < w - 1; j++) {
+            buf[INDEX_MOD(i)][j] = kx[0] * src[i - 1][j] + kx[1] * src[i][j] + kx[2] * src[i + 1][j];
+        }
+    }
+
+    // 再进行列方向的卷积
+    for (int j = 1; j < w - 1; j++) {
+        for (int i = 1; i < h - 1; i++) {
+            dst[i][j] = ky[0] * buf[INDEX_MOD(i - 1)][j] + ky[1] * buf[INDEX_MOD(i)][j] + ky[2] * buf[INDEX_MOD(i + 1)][j];
+        }
+    }
+#undef INDEX_MOD
+}
+
+// 原始的应用3×3 GaussianBlur函数（与步骤1中的类似，方便对比）
+void applyGaussianBlur(float src[][5], float dst[][5], int h, int w, float kernel[3][3]) {
+    for (int i = 1; i < h - 1; i++) {
+        for (int j = 1; j < w - 1; j++) {
+            float sum = 0.0;
+            for (int m = 0; m < 3; m++) {
+                for (int n = 0; n < 3; n++) {
+                    sum += kernel[m][n] * src[i - 1 + m][j - 1 + n];
+                }
+            }
+            dst[i][j] = sum;
+        }
+    }
+}
+
+int main() {
+    // 输入图像矩阵（这里示例一个5×5的矩阵）
+    float image[5][5] = {
+        {1, 2, 3, 4, 5},
+        {6, 7, 8, 9, 10},
+        {11, 12, 13, 14, 15},
+        {16, 17, 18, 19, 20},
+        {21, 22, 23, 24, 25}
+    };
+    // 常用的kernel_size=3的kernel
+    float kernel[3][3] = {
+        {1.0 / 16, 2.0 / 16, 1.0 / 16},
+        {2.0 / 16, 4.0 / 16, 2.0 / 16},
+        {1.0 / 16, 2.0 / 16, 1.0 / 16}
+    };
+    float result[5][5];
+    float result_optimized[5][5];
+    // 拆分kernel矩阵为行方向和列方向两个向量
+    float kx[3] = {kernel[0][0], kernel[0][1], kernel[0][2]};
+    float ky[3] = {kernel[0][0], kernel[1][0], kernel[2][0]};
+
+    // 记录原始版本开始时间
+    clock_t start_time_original = clock();
+    applyGaussianBlur(image, result, 5, 5, kernel);
+    // 记录原始版本结束时间
+    clock_t end_time_original = clock();
+    double elapsed_time_original = (double)(end_time_original - start_time_original) / CLOCKS_PER_SEC;
+
+    // 记录优化版本开始时间
+    clock_t start_time_optimized = clock();
+    applySeparableGaussianBlur(image, result_optimized, 5, 5, kx, ky);
+    // 记录优化版本结束时间
+    clock_t end_time_optimized = clock();
+    double elapsed_time_optimized = (double)(end_time_optimized - start_time_optimized) / CLOCKS_PER_SEC;
+
+    // 输出原始版本结果
+    printf("Original Gaussian Blur Result Matrix:\n");
+    for (int i = 0; i < 5; i++) {
+        for (int j = 0; j < 5; j++) {
+            printf("%.2f ", result[i][j]);
+        }
+        printf("\n");
+    }
+    printf("Original Elapsed time: %.6f seconds\n", elapsed_time_original);
+
+    // 输出优化版本结果
+    printf("Optimized Gaussian Blur Result Matrix:\n");
+    for (int i = 0; i < 5; i++) {
+        for (int j = 0; j < 5; j++) {
+            printf("%.2f ", result_optimized[i][j]);
+        }
+        printf("\n");
+    }
+    printf("Optimized Elapsed time: %.6f seconds\n", elapsed_time_optimized);
+
+    return 0;
+}
\ No newline at end of file