Fish species classification using a collaborative technique of firefly algorithm and neural network

Abstract Monitoring various Fish Species and its distribution of the species obtains a primary significance in receiving insights to marine ecological-system. After this, visual classification those would aid tracing out movement yield patterns trends fish activities, which provides depth knowledge species. Unconstrained under-water images pose highly variations because orientation changes, Light-intensities, similarity shapes. This create greater challenge for Image-processing techniques accurate or classes. Hence, this reason, Underwater Image Enhancement is implemented combination Morphological-operations pre-processing method. The pre-processed image then subjected feature extraction process by using Speed-up Robust Feature algorithm. followed Firefly Algorithm, applied optimization Region interest selection selected-features. For categorization Fish-species, PatternNet technique employed, classifying 10,000 fish-images five categories ( Dascyllus reticulatus , Plectroglyphidodon dickii Chromis chrysura Amphiprion clarkii Chaetodon lunulatus ). Efficiency proposed-framework performed terms Classification accuracy, execution time, precision value, F-measure recall factors with respect comparison also assessed other existing methods. 98% accuracy rate was produced evaluation results proposed framework lesser average computation time 3.64 s upon different tested images. Thus, higher efficiency proved outcomes study.