Haar Cascade Training

Gathering image from ImageNet (from urls )

python3 gathering_pics_from_urls.py

Remove ugly images. Make sure u have already placed ugly images in directory 'uglies'

python3 remove_ugly_imgs.py

Creating background text

python3 creating_bgtxt.py

Creating positive images

python3 creating_pos_image.py

Make sure creating_pos_image works properly

python3 check_info_lst.py

Create positive vectors

-info -num -w <recommend using 20> -h <recommend using 20>

Note: copy info.lst into info/

opencv_createsamples -info info/info.lst -num 1000 -w 50 -h 50 \
-vec positives.vec

start training

-data <directory to store Haar classifier -bg -numPos <max is value of -num> -numNeg <half of -numPos> empty the folder data/

opencv_traincascade -data data -vec positives.vec  -bg bg.txt \
-numPos 1000 -numNeg 600 -numStages 20 -w 50 -h 50 \
-featureType HAAR \
-precalcValBufSize 1024 -precalcIdxBufSize 1024 \
-minHitRate 0.999 -maxFalseAlarmRate 0.2

#####refs:http://answers.opencv.org/question/64431/number-of-stages-or-maxfalsealarmrate/

A minHitRate:

is the parameter that ensures us that our positive training data yields at least a decent detection output. We do not want to lower this value to much. For example a value of 0.8 would mean that 20% of our positive object training data can be misclassified, which would be a disaster. Using a rate of 1% misclassification is a common value used in research.

A maxFalseAlarmRate:

is used to define how much features need to be added. Actually we want each weak classifier to have a very good hit rate on the positives, and then to allow them to remove negative windows, as fast as possible, but doing better then random guessing. 0.5 means you apply a random guess, better than that means you successfully remove negative windows as negatives very early using only a few feature evaluations, letting other negatives be discarded by the following stages.

recommend using when training with big data

nohup opencv_traincascade -data data -vec positives.vec  -bg bg.txt \
-numPos 1000 -numNeg 600 -numStages 20 -w 50 -h 50 \
-featureType HAAR -mode ALL \
-precalcValBufSize 1024 -precalcIdxBufSize 1024 \
-minHitRate 0.995 -maxFalseAlarmRate 0.5 &

MEANINGS:

-minHitRate is set to 0.995 by default. This means that for this current model, it allows 5 out of 1000 positive samples to get wrongly classified during the training process -maxFalseAlarmRate is set to 0.5 by default. Each stage needs to reach an individual false acceptance rate (good classification of negs)

Gathering more negative images Combine between game scene and gathered images Mostly, 7,500 pos , 3,000 neg

saved !!

Built With

• Dropwizard - The web framework used
• Maven - Dependency Management
• ROME - Used to generate RSS Feeds

Contributing

Please read CONTRIBUTING.md for details on our code of conduct, and the process for submitting pull requests to us.

Versioning

We use SemVer for versioning. For the versions available, see the tags on this repository.

Authors

• Billie Thompson - Initial work - PurpleBooth

See also the list of contributors who participated in this project.

License

This project is licensed under the MIT License - see the LICENSE.md file for details

Acknowledgments

• Hat tip to anyone whose code was used
• Inspiration
• etc

---