This repositorsy contains 3 hog svm classifers. Each of them can be used using the function classifier in classifiers_svm.py
Parameters : input_image,classifier_number
classifier_number
-
Pass 1 for classification between whether the image is a sign or not, Output: 1 = sign, 2 = Not sign
-
Pass 2 for classification between whether the image is one of the signs we need or not, Output: 1 = yes, 2 = no
-
Pass 3 for classification between whether the image is a sign or not, Output: label of sign
In order to train for 61 classes run python train_61_class_svm.py. This will save the model.
Then run python main.py to start the traffic sign detection algorithm.