opencv 进行图像的花屏检测（模糊检测）

参考:

https://www.pyimagesearch.com/2015/09/07/blur-detection-with-opencv/

https://www.cnblogs.com/arkenstone/p/7900978.html

先对图像用拉普拉斯算子进行滤波，然后求取得到的结果图像的方差，如果方差小于一定值则图片视为模糊。利用python很好实现：

img2gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)  # 将图片压缩为单通道的灰度图
score = cv2.Laplacian(img2gray, cv2.CV_64F).var()

C++实现如下：

bool isImageBlurry(cv::Mat& img)
{
    cv::Mat matImageGray;
    // converting image's color space (RGB) to grayscale
    cv::cvtColor(img, matImageGray, CV_BGR2GRAY);
    cv::Mat dst, abs_dst;
        cv::Laplacian(matImageGray, dst, CV_16S, );
        cv::convertScaleAbs( dst, abs_dst );
    cv::Mat tmp_m, tmp_sd;
    double m = , sd = ;
    int threshold = ;
    cv::meanStdDev(dst, tmp_m, tmp_sd);
    m = tmp_m.at<double>(,);
    sd = tmp_sd.at<double>(,);
    std::cout << "StdDev: " << sd * sd << std::endl;
    return ((sd * sd) <= threshold);
}