Analysis and comparison of fish posture by image processing

According to observation of fish instinct, environment has direct impact on fish posture. Hence, a good fish posture analysis can indicate environment change. This study used image processing technique to recognize fish posture, and analyze the position and angle of the posture, in order to determine the condition of fish. Morphological Processing, Moving Object Detection and Moving Object Tracking approaches were applied, and Puntius tetrazona was chosen as the subject. The hardware included 18 cm³ fish bin and IP camera for real-time fish posture analysis. This study used moving object detection, moving object tracking and relevant algorithms to build a fish posture analysis system that can identify fish behavior instantly.