How to use opencv featuredetector on small images

I use opencv 3 in Java, and I try to find small images (such as 25) on other images × 25 pixels) However, the feature detector detects that the (0,0) size mat is on a small image

Mat smallImage = ...

    FeatureDetector detector = FeatureDetector.create(FeatureDetector.ORB);
    DescriptorExtractor descriptor = DescriptorExtractor.create(DescriptorExtractor.ORB);
    DescriptorMatcher matcher = DescriptorMatcher.create(DescriptorMatcher.BRUTEFORCE_HAMMING);

    Mat descriptoRSSmall = new Mat();
    MatOfKeyPoint keyPointsSmall = new MatOfKeyPoint();

    detector.detect(smallImage,keyPointsSmall);
    descriptor.compute(smallImage,keyPointsSmall,descriptoRSSmall);

Here, I get the size of keypointssmall and descriptorssmall to zero, and make sure the detection doesn't work

But if I'm at 150 × Try this function on a larger image of 150 pixels Any suggestions? thank you.

I add samples here We have this source image:

And let it say that we have a template for the P letter, so we need to detect the P on the source image

Well, scaling the image to a higher resolution doesn't work for me That will take time and resources Ideally, it should be rotational scale invariant But a simple solution without rotation and scaling is OK

Solutions other than opencv are unacceptable to me (for example, using Tesseract)

Solution

Key point detection for text recognition is not the best solution because you will get many similar functions, and if the template is very small, the sliding window will not produce enough detected functions

Fortunately, opencv 3 contains a text detection / recognition module in the contrib Repository: link. One example is taken from here, and many other modules can be found here:

/*
 * cropped_word_recognition.cpp
 *
 * A demo program of text recognition in a given cropped word.
 * Shows the use of the OCRBeamSearchDecoder class API using the provided default classifier.
 *
 * Created on: Jul 9,2015
 *     Author: Lluis Gomez i Bigorda <lgomez AT cvc.uab.es>
 */

#include "opencv2/text.hpp"
#include "opencv2/core/utility.hpp"
#include "opencv2/highgui.hpp"
#include "opencv2/imgproc.hpp"

#include <iostream>

using namespace std;
using namespace cv;
using namespace cv::text;

int main(int argc,char* argv[])
{

    cout << endl << argv[0] << endl << endl;
    cout << "A demo program of Scene Text Character Recognition: " << endl;
    cout << "Shows the use of the OCRBeamSearchDecoder::ClassifierCallback class using the Single Layer CNN character classifier described in:" << endl;
    cout << "Coates,Adam,et al. \"Text detection and character recognition in scene images with unsupervised feature learning.\" ICDAR 2011." << endl << endl;

    Mat image;
    if(argc>1)
        image  = imread(argv[1]);
    else
    {
        cout << "    Usage: " << argv[0] << " <input_image>" << endl;
        cout << "           the input image must contain a single character (e.g. scenetext_char01.jpg)." << endl << endl;
        return(0);
    }

    string vocabulary = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789"; // must have the same order as the clasifier output classes

    Ptr<OCRHMMDecoder::ClassifierCallback> ocr = loadOCRHMMClassifierCNN("OCRBeamSearch_CNN_model_data.xml.gz");

    double t_r = (double)getTickCount();
    vector<int> out_classes;
    vector<double> out_confidences;

    ocr->eval(image,out_classes,out_confidences);

    cout << "OCR output = \"" << vocabulary[out_classes[0]] << "\" with confidence "
         << out_confidences[0] << ". Evaluated in "
         << ((double)getTickCount() - t_r)*1000/getTickFrequency() << " ms." << endl << endl;

    return 0;
}