unsigned int winsize = (1 + (((int)ceil(3 * sigma)) * 2)); //窗的大小
int *gaussKern = buildGaussKern(winsize, sigma); //构建高斯核，计算高斯系数
winsize *= 1; //3改为1，高斯窗的宽度变为原来的1/3
unsigned int halfsize = winsize / 2; //窗的边到中心的距离
unsigned char *tmpBuffer = (unsigned char*)malloc(width * height* sizeof(unsigned char)); //开辟新的内存存储处理高斯模糊后的数据
for (unsigned int h = 0; h < height; h++) //外层循环，图像的高度
{
unsigned int rowWidth = h * width; //当前行的宽度为图像的高度乘以每行图像的数据所占的宽度。因为是按行存储的数组。
for (unsigned int w = 0; w < width; w++) //w+=channels，可以修改为w++，因为是单通道数据，而不是三通道数据
{
unsigned int rowR = 0; //存储r分量的数据
int * gaussKernPtr = gaussKern;//将高斯系数赋值给gaussKernPtr
int whalfsize = w + width - halfsize;
unsigned int curPos = rowWidth + w; //当前位置
for (unsigned int k = 1; k < winsize;k++) // k += channels修改为k++
{
unsigned int pos = rowWidth + ((k + whalfsize) % width);
int fkern = *gaussKernPtr++;
rowR += (pixels[pos] * fkern); //当前像素值乘以高斯系数，rowR这了泛指单通道的当前像素点高斯处理后的数
}
tmpBuffer[curPos] = ((unsigned char)(rowR >> 8)); //除以256
}
}
halfsize = winsize / 2;
for (unsigned int w = 0; w < width; w++)
{
for (unsigned int h = 0; h < height; h++)
{
unsigned int col_all = 0;
int hhalfsize = h + height - halfsize;
for (unsigned int k = 0; k < winsize; k++)
{
col_all += tmpBuffer[((k + hhalfsize) % height)* width + w] * gaussKern[k];
}
pixelsout[h * width + w] = (unsigned char)(col_all >> 8);
}
}
free(tmpBuffer);
free(gaussKern);
}
int _tmain(int argc, _TCHAR* argv[])
{
const char* imagename = "C:\\Users\\Administrator.IES7LSEJAZ1GGRL\\Desktop\\PureGaussian-master\\GaussianBlur\\GaussianBlur\\InputName.bmp";
//从文件中读入图像
Mat img = imread(imagename);
Mat dst = imread(imagename);
Mat gray_img;
Mat gray_dst;
cvtColor(img, gray_img, CV_BGR2GRAY);
cvtColor(dst, gray_dst, CV_BGR2GRAY);
//如果读入图像失败
if(img.empty())
{
fprintf(stderr, "Can not load image %s\n", imagename);
return -1;
}
LARGE_INTEGER m_nFreq;
LARGE_INTEGER m_nBeginTime;
LARGE_INTEGER nEndTime;
QueryPerformanceFrequency(&m_nFreq); // 获取时钟周期
QueryPerformanceCounter(&m_nBeginTime); // 获取时钟计数
GaussBlur1D(gray_img.data,gray_dst.data,gray_img.cols,gray_img.rows,2);
QueryPerformanceCounter(&nEndTime);
cout << (nEndTime.QuadPart-m_nBeginTime.QuadPart)*100/m_nFreq.QuadPart << endl;
//显示图像
imshow("原图像",gray_img);
imshow("模糊图像", gray_dst);
//此函数等待按键，按键盘任意键就返回
waitKey();
return 0;
}
算法实现效果：sigma=2.0
