Fully automatic system for fish biomass estimation based on deep neural network

Tianye Zhang, Yuqiao Yang, Yueyue Liu, Chenglei Liu, Ran Zhao, Daoliang Li, Chen Shi

Abstract: Highlights•A fast, precise and fully automatic system was proposed for free-swimming fish biomass estimation based on binocular stereo vision technology.•An improved target detection algorithm has been developed for real-time detecting and extraction of high-quality images of moving fish.•The multiple factor model was utilized to estimate the weight of fish and exhibited greater accuracy than the single factor model.