Automatic Recognition and Counting Method of Deep-Sea Jellyfish Based on Image Multi-Feature Matching

Jellyfish outbreaks have now become a serious ecological disaster, affecting fishing and industrial production in coastal areas and causing huge economic losses. Monitoring jellyfish populations using low light imaging technology is an effective way to provide early warning of jellyfish outbreaks. However, there is no effective method to analyze the bioluminescent images and count the number of jellyfish automatically now. This paper proposes an automatic identification and counting method for deep-sea jellyfish based on multi-feature matching. First, Jellyfish are identified according to the area of the target, the ratio of contour circumference to area, and the ratio of external convex hull area to contour area. Then, in order to count accurately, the targets belong to consecutive frames are matched by the change rate of the target area, the distance of the center of mass and the contact degree of the rectangle outside the contour. The algorithm is verified with real biological image data, and the results show that the method has high accuracy in jellyfish recognition and counting.