如何在openCv中训练两个大小不同的对象

Question

I try to train the licence plate car and licence plate motorcycle with OpenCv version 2.4.10.

I have a 235 images positives and a 300 negative images.

For positive images the licence plate car the real size is 40 cm x 13 cm and licence plate motorcycle have real size 20 cm x 17 cm. When tried to train the vectors separately the software works fine, but when I tried to train utilizing the two plates, the software doesn't work.

Here is the utilized code of training.

This code works with a licence plate car, but doesn't works with licence plate motorcycle

createsamples -info c:\harrkit\HAARKITV10\povitivos\info.txt -vec vetor\veiculos_64x24.vec -num 235 -w 64 -h 24
trainCascade -data cascade -vec c:\harrkit\HAARKITV10\vetor\veiculos_64x24.vec -bg c:\harrkit\HAARKITV10\negativos\negativos.txt -numPos 235 -numNeg 200 -numStages 12  -featureTyp HAAR -minHitRate 0.999 -maxFalseAlarmRate 0.5 -w 64 -h 24

// detectMultiScale code

if (mAbsolutePlateSize == 0) {
                int height = mGray.rows();
                if (Math.round(height * mRelativePlateSize) > 0) {
                    mAbsolutePlateSize = Math.round(height * mRelativePlateSize);
                }
                mNativeDetector.setMinFaceSize(mAbsolutePlateSize);
            }

            MatOfRect mMatPlacas = new MatOfRect();

            if (mDetectorType == JAVA_DETECTOR) {
                if (mJavaDetector != null)
                    mJavaDetector.detectMultiScale(mGray, mMatPlacas, 1.1, 2, 2, new Size(mAbsolutePlateSize, mAbsolutePlateSize), new Size()); 
            } else if (mDetectorType == NATIVE_DETECTOR) {
                if (mNativeDetector != null)
                    mNativeDetector.detect(mGray, mMatPlacas);
            } else {
                Log.e(TAG, "Detection method is not selected!");
            }

Answer 1

Since these two types of plates are of different aspect ratios, you can try to train a different classifier for each type. On paper, that is more accurate.

Answer 2

Your problem may come from a missing minus sign before h in createsamples

Try the following:

createsamples -info c:\harrkit\HAARKITV10\povitivos\info.txt -vec vetor\veiculos_64x24.vec -num 235 -w 64 -h 24

It should work because createsamples a collection of positives with fix width and height (64x64).

Answer 3

You must change your "h" and "w" parameters to train the second type of your objects , those parameters can cause problem especially if they don't match with your object's size. Also , can you put your console, we must take a look on your errors to figure out the origin of your problem!