I'm working with libsvm and I must perform a multiclass pixel-based classification. I want to classify an image which contains Four classes. For training, I have extracted SURF dense features for each class and out it them in Data_Train.xlsx in which first column is the class and the rest is the SURF features. For test, I have extracted SURF dense features of the image ( that contain four classes at the same time). My question is how to classify each pixel of this image and predict its class using Libsvm and then visualize the results.
https://www.mathworks.com/matlabcentral/answers/50731-algorithms-for-imbalanced-multi-class-classification-in-matlab
The maslab version of libsvm is cripy. Is beeter to use the scikitlean one which is in python but wrapp the libsvm.
http://scikit-learn.org/stable/modules/generated/sklearn.svm.SVR.html
efficient and seay to use.
Place your data is in ar array and your ground truth in an other one.
add run predict.
you could also run the cross validation fonction to get some perf
You can use fitcecoc and train a model.
this worked fine for me.
https://in.mathworks.com/help/stats/classificationecoc-class.html
- Badges
- Science topic
- Similar topics
- Psychology
- Mental Processes
- Learning