Opencv Traincascade

1. Create a list of positive images:

find data/Positive_Images -iname "*.jpg" > positives.txt

2. Create a list of negative images:

find data/Negative_Images -iname "*.jpg" > negatives.txt

Create samples first way

3. Create positive samples with the bin/createsamples.pl script and save them to the ./samples folder:

perl bin/createsamples.pl positives.txt negatives.txt samples 1000 "opencv_createsamples -bgcolor 0 -bgthresh 0 -maxxangle 1.1 -maxyangle 1.1 maxzangle 0.5 -maxidev 40 -w 80 -h 40"

Pay attention to the number of created samples!!!

4. Use the compiled executable mergevec to merge the samples in ./samples into one file:

find ./samples -name '*.vec' > samples.txt

./mergevec samples.txt samples.vec

Create samples second way

3. Compile and run rectangle.cpp to create positives.dat

4. Create positive samples with opencv_createsamples

./opencv_createsamples -info positives.dat -vec positives.vec -bg negatives.txt -num 1000 -bgcolor 0 -bgthresh 0 -maxxangle 1.1 -maxyangle 1.1 maxzangle 0.5 -maxidev 40 -w 80 -h 40

Pay attention to the number of created samples!!!

5. Start training the classifier with opencv_traincascade, which comes with OpenCV, and save the results to ./classifier:

./opencv_traincascade -data classifier -vec samples.vec -bg negatives.txt -numStages 20 -minHitRate 0.999 -maxFalseAlarmRate 0.5 -numPos 227 -numNeg 3019 -w 80 -h 40 -mode ALL -precalcValBufSize 1024 -precalcIdxBufSize 1024

Notes:

1. Compile mergevec.cpp

cp src/mergevec.cpp ~/opencv-2.4.6.1/apps/haartraining

cd ~/opencv-2.4.6.1/apps/haartraining

g++ pkg-config --libs --cflags opencv -I. -o mergevec mergevec.cpp cvboost.cpp cvcommon.cpp cvsamples.cpp cvhaarclassifier.cpp cvhaartraining.cpp -lopencv_core -lopencv_calib3d -lopencv_imgproc -lopencv_highgui -lopencv_objdetect

mv mergevec ~/folder of opencv_traincascade