python下的opencv和c++下的opencv的图片类型差异

Question

python下的opencv和c++下的opencv都使用imread方法打开，发现图片打印出来的bgr像素不一样？？？为什么？
python下打印mat bgr像素方法是：

im = cv2.imread(args.image_path, cv2.IMREAD_COLOR)

im_height, im_width, im_channel = im.shape
for imh in range(0, im_height - 1):
    for imw in range(0, im_width - 1):
        imb, img, imr = im[imw, imh]
        print ("w ", imw, "h ", imh, " b ", imb, " g ", img, " r ", imr)

c++下打印mat bgr像素方法是：

cv::Mat input_img = cv::imread(imgPath, cv::IMREAD_COLOR);
int nr= input_img.rows; // number of rows height
int nc= input_img.cols; // number of columns width
for (int j=0; j<nc; j++) {
    for (int i = 0; i<nr; i++) {
        int b = input_img_origin.at<cv::Vec3b>(j, i)[0];
        int g = input_img_origin.at<cv::Vec3b>(j, i)[1];
        int r = input_img_origin.at<cv::Vec3b>(j, i)[2];
        printf("input image j %d, i %d\n", j, i);
        printf("r %d", r);
        printf("    g %d", g);
        printf("    b %d\n", b);
    } // end of row
}       

Answer 1

擦，版本不一样，一个2.，一个3.，真是巨坑。。。