Usage of OpenCV C++ API to perform objection detection using MobileNet and SSD

demo.cpp

#include <opencv2/dnn.hpp>
#include <opencv2/imgproc.hpp>
#include <opencv2/highgui.hpp>
#include <opencv2/core/utils/trace.hpp>
using namespace cv;
using namespace cv::dnn;
#include <fstream>
#include <iostream>
#include <cstdlib>
using namespace std;

string CLASSES[] = {"background", "aeroplane", "bicycle", "bird", "boat",
	"bottle", "bus", "car", "cat", "chair", "cow", "diningtable",
	"dog", "horse", "motorbike", "person", "pottedplant", "sheep",
	"sofa", "train", "tvmonitor"};


int main(int argc, char **argv)
{
    CV_TRACE_FUNCTION();
    String modelTxt = "MobileNetSSD_deploy.prototxt";
    String modelBin = "MobileNetSSD_deploy.caffemodel";
    
    String imageFile = (argc > 1) ? argv[1] : "space_shuttle.jpg";
    Net net = dnn::readNetFromCaffe(modelTxt, modelBin);
    if (net.empty())
    {
        std::cerr << "Can't load network by using the following files: " << std::endl;
        std::cerr << "prototxt:   " << modelTxt << std::endl;
        std::cerr << "caffemodel: " << modelBin << std::endl;
        exit(-1);
    }
    Mat img = imread(imageFile);
    if (img.empty())
    {
        std::cerr << "Can't read image from the file: " << imageFile << std::endl;
        exit(-1);
    }
    
    Mat img2;
    resize(img, img2, Size(300,300));
    Mat inputBlob = blobFromImage(img2, 0.007843, Size(300,300), Scalar(127.5, 127.5, 127.5), false);

    net.setInput(inputBlob, "data");
    Mat detection = net.forward("detection_out");
    Mat detectionMat(detection.size[2], detection.size[3], CV_32F, detection.ptr<float>());

    ostringstream ss;
    float confidenceThreshold = 0.2;
    for (int i = 0; i < detectionMat.rows; i++)
    {
        float confidence = detectionMat.at<float>(i, 2);

        if (confidence > confidenceThreshold)
        {
            int idx = static_cast<int>(detectionMat.at<float>(i, 1));
            int xLeftBottom = static_cast<int>(detectionMat.at<float>(i, 3) * img.cols);
            int yLeftBottom = static_cast<int>(detectionMat.at<float>(i, 4) * img.rows);
            int xRightTop = static_cast<int>(detectionMat.at<float>(i, 5) * img.cols);
            int yRightTop = static_cast<int>(detectionMat.at<float>(i, 6) * img.rows);

            Rect object((int)xLeftBottom, (int)yLeftBottom,
                        (int)(xRightTop - xLeftBottom),
                        (int)(yRightTop - yLeftBottom));

            rectangle(img, object, Scalar(0, 255, 0), 2);

            cout << CLASSES[idx] << ": " << confidence << endl;

            ss.str("");
            ss << confidence;
            String conf(ss.str());
            String label = CLASSES[idx] + ": " + conf;
            int baseLine = 0;
            Size labelSize = getTextSize(label, FONT_HERSHEY_SIMPLEX, 0.5, 1, &baseLine);
            putText(img, label, Point(xLeftBottom, yLeftBottom),
                    FONT_HERSHEY_SIMPLEX, 0.5, Scalar(0,0,0));
        }
    }
    imshow("detections", img);
    waitKey();

    return 0;
}

I got an error like this:
"terminate called after throwing an instance of 'cv::Exception'
what(): OpenCV(4.1.2-pre) opencv/modules/dnn/src/caffe/caffe_io.cpp:1121: error: (-2:Unspecified error) FAILED: fs.is_open(). Can't open "MobileNetSSD_deploy.prototxt" in function 'ReadProtoFromTextFile'"

I just changed line 21: I added '.txt' at "MobileNetSSD_deploy.prototxt". (like this) "MobileNetSSD_deploy.prototxt.txt"

What is the purpose of L41? resize(img, img2, Size(300,300));... from what I understand the MobileNetSSD model already does this, doing it here just adds additional overhead. Testing shows it is not needed and causes quite a performance hit when using CPU.

My code works, but I don't have any reasonable outputs. I modified the code slightly to output detections before the image postprocessing happens. This is what I get:

Found 1 detection(s):
ID: -431602080 Confidence: -431602080

What could cause the problem?

I don't know: "Does exactly this code works?", but I took a few lines code from here and everything works nice. Thanks for help=)

All Right. It works thanks!

It works! great! thanks a lot

good! It can work!