Feature selection of Gabor filter

Question

I am working with Gabor filter with level set method as in A Level-Set and Gabor-based Active Contour Algorithm for Segmenting Textured Images. As you known, Gabor filter gave many filter output. Some of them is not necessary. Could you suggest to me some method that can choose best filter output that satisfy maximum the difference between object (zebra) and background that are separated by a contour (white circle)as figure. Thank you. Now, I am working in image segmentation, hence, these best filter outputs will give more accuracy

Answer 1

You have a classification problem.
You may use method of feature selection for classification. For specific policies look at scikit-learn feature selection.