Abstract

Aquatic life plays an essential role in the food chain and it is very important to monitor life underwater especially in the current times with increasing global warming. Several existing approaches are designed and implemented to aid marine biologists and scientists in keeping track of marine life, specifically fishes, but these approaches either give results with poor accuracy or fail to detect fishes in the dark or murky underwater environment. In this research, we present the YOLO model for fish detection along with fish identification under various environmental scenarios such as rocky shores, coral reefs, and murky or unclear water conditions to aid Marine biologists and conservationists in easily identifying the fishes in any given region.