Abstract

A method of fish category recognition is proposed. Our novelty is in feature extraction technique, which is based on determining the curvature of the contour of fish image. The category recognition of test fish is done by a suitable linear multi-class SVM classifier. To apply our algorithm, we have developed a fish image database containing six different categories of fish. The overall accuracy of our proposed algorithm is about 94%.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call