Published Paper Details:

Automatic classification of fish and its analysis is a stimulating task due to the multifaceted textures, appearance, and diversity found in fishes. Automatic fish identification and classifications have enormous probable applications in expanded parks associated with fishery sciences and domestic commercial uses of fish. Traditionally, the approach of fish classification is founded on various morphological, geometrical, location features, and needs the support of taxonomists to identify fish classes. The dataset used in research work is collection of 500 fish images from Fishbase and Fish4Knowledge. This complex collection, makes the classification process challenging due to variance in data acquisition sensors, location, scale, etc. Since, the considered dataset is complex, presented research work presents a novel approach of Artificial Intelligence (AI) methods like Local Binary Patterns (LBP), Self-Organizing Maps (SOM), and Principal Component Analysis (PCA) exploiting computations of prominent morphological features of fins, gills resulting 97% accuracy.