An Analysis of Training Artificial Intelligence Techniques into Eco Sounder Machine to Identify Fish

The acoustic backscatter coefficient values obtained from Echosounders provides important information about the presence of fishes in the water. There are several developments in underwater acoustic technology such as single beam echo sounder, multi frequency echo sounder, side scan radar. But there are several challenges associated with this technology such as interpretation of echograms generated in these devices are time consuming, there is requirement of technical experts to understand these echograms and detection of fish species is still a challenge etc. The recent advancement in this field is the integration of echo signal with the Artificial Intelligence algorithms. Machine Learning, Deep Learning, Fuzzy Logic are some of the recent advanced algorithms which are used for automatic classification of fishes that aids the fishermen to identify the locations of the fishes. Hence, this review article focusses on the role of these advance algorithms to identify fishes in the sea that can help fishermen in saving their time by precisely locating the fishes. A trained AI program can locate areas in a scene and recognize fish feature patterns. Fish recognition and species categorization on 3D photos have both been successfully accomplished using the Echo sounder object detection framework.