Train Cascade Classifier

[scottclowe]

Using shell scripting or C++, use OpenCV to train a Haar and/or LBP cascade classifier to identify the presence of each plankton class.

I think we will need a separate classifier trained for each of the plankton classes, because cascade classifiers are binary/single-class. The output of the classifier will be an XML file which can be read in using the Python wrapper of OpenCV in order to detect the presence of the plankton in the images.

NB: The number of positive and negative samples probably needs to be similar for the classifier to be effective.

http://docs.opencv.org/trunk/doc/user_guide/ug_traincascade.html http://docs.opencv.org/trunk/modules/objdetect/doc/cascade_classification.html https://github.com/Itseez/opencv/blob/master/samples/cpp/dbt_face_detection.cpp

[scottclowe]

Do you want to do this @svepe? No Python needed.