Ask Your Question

Revision history [back]

You can convert your image to HSV color space and make segmentation by H component. Your pipeline we be like this: color conversion -> threshold by H value -> morphology operations if needed -> find contours -> filtering contours, i.e. by size.