Function ImagePreprocess

samples/cpp/run_model/classification.cpp:38–72

change OpenCV Mat to NCHW format fp32 data

// change OpenCV Mat to NCHW format fp32 data
int32_t ImagePreprocess(const Mat& src_img, float* in_data) {
    const int32_t height = src_img.rows;
    const int32_t width = src_img.cols;
    const int32_t channels = src_img.channels();

    // convert data from bgr/gray to rgb
    Mat rgb_img;
    if (channels == 3) {
        cvtColor(src_img, rgb_img, COLOR_BGR2RGB);
    } else if (channels == 1) {
        cvtColor(src_img, rgb_img, COLOR_GRAY2RGB);
    } else {
        fprintf(stderr, "unsupported channel num: %d\n", channels);
        return -1;
    }

    // split 3 channel to change HWC to CHW
    vector<Mat> rgb_channels(3);
    split(rgb_img, rgb_channels);

    // by this constructor, when cv::Mat r_channel_fp32 changed, in_data will also change
    Mat r_channel_fp32(height, width, CV_32FC1, in_data + 0 * height * width);
    Mat g_channel_fp32(height, width, CV_32FC1, in_data + 1 * height * width);
    Mat b_channel_fp32(height, width, CV_32FC1, in_data + 2 * height * width);
    vector<Mat> rgb_channels_fp32{r_channel_fp32, g_channel_fp32, b_channel_fp32};

    // convert uint8 to fp32, y = (x - mean) / std
    const float mean[3] = {0, 0, 0}; // change mean & std according to your dataset & training param
    const float std[3] = {255.0f, 255.0f, 255.0f};
    for (uint32_t i = 0; i < rgb_channels.size(); ++i) {
        rgb_channels[i].convertTo(rgb_channels_fp32[i], CV_32FC1, 1.0f / std[i], -mean[i] / std[i]);
    }

    return 0;
}

// get classification result from network output
int32_t GetClassificationResult(const float* scores, const int32_t size) {