Detection and Classification of Fish Using Deep Learning Models

Aquaculture, also known as aqua farming, plays a substantial role in India's agricultural industry. India is a major global fish producer, with aquaculture playing a significant role in the country's economy, food security, and jobs. India's varied climatic conditions and abundant water resources, such as rivers, ponds, lakes, and coastal regions, make it a very suitable place for aquaculture. The enhancement of the maritime economy relies on the accurate detection and identification of marine species. Accurate detection of aquatic species thriving in a demanding marine environment can be benefit the local as well as researchers. This research study proposes a framework for a YOLOv8, YOLOv9, ResNet50, Vgg19, DenseNet201 convolutional neural network that is specifically trained on aquatic marine life. This study used a variety of nine distinct marine animals, including fish, Black-sea sprat, Gilt head bream, Horse mackerel, Red-mullet, Red-sea bream, Sea bass, Shrimp, Stripped red mullet and Trout. When detection model is executed, it found that YOLOv9 performs better than YOLOv8. This research work reveals that in terms of classification of fish species it was found that ResNet50 has higher efficiency compared to YOLOv8, Vgg19, DenseNet201.