Research and Practice in Aquaculture Safety Monitoring based on Multi-source Information Fusion

Abstract

Research on aquaculture is an important part of environmental science. Aiming at the problem of aquaculture safety monitoring, this paper proposes a detection and judgment method for theft based on multi-source information fusion. The iterative Bayesian inference is used to integrate multi-source information such as abnormal human posture recognition, abnormal sound recognition and occurrence time to obtain relevant data on whether the aquaculture farm has been stolen. The experimental results show that, compared with traditional methods such as video detection or audio detection, the algorithm in this paper can determine whether the aquaculture farm is in a safe, suspicious or stolen state more quickly and accurately, thus providing an accurate disposal basis for managers. At the same time, the algorithm has better robustness.