Robust features extraction from shape signature for fish images classification

Recently, the process of fish species classification has become one most challenging problems addressed by researchers. In this work, a robust scheme to classify images based on feature extraction from shape signatures is proposed. First, image contour fitted using common approaches named radial basis function neural network (RBFNN) fitting obtain centroid. Afterward, prominent features signature are extracted. These representative shapes because they can distinguish characteristics each class as well being relatively scale and rotation changes. Finally, for purpose, RBFNN used again against commonly techniques called support vector machine (SVM). The proposed paradigm been applied standard dataset acquired live video grouped into twenty-three clusters representing specific species. resulting accuracy SVM was 90.41% 98.04%, respectively.