Research on underwater object recognition based on YOLOv3

In recent years, object recognition and detection technology, which is a very important research direction in the field of computer vision, is widely used in human life. The technology has been relatively mature for the recognition of objects such as people and objects on land. However, due to some conditions, it is relatively rare in the marine field. The reasons for the analysis mainly include two points: underwater classification and localization are affected by multiple factors such as illumination uniformity, occlusion and water color, and secondly, underwater video acquisition is also relatively difficult. These issues have long been the focus of attention. Therefore, effective classification and recognition of objects in underwater video is of great significance for the intelligentization of marine equipment. This paper mainly locates and classifies the images of seacucumber, scallop, seaurchin. This paper uses two algorithms that are widely used at present to texperiment with underwater image dataset. The experimental results show that the mean Average Precision (mAP) of YOLOv3 algorithm is 6.4% higher than Faster R-CNN, and the recall rate (Recall) is 13.9% higher. Moreover, the detection speed of the YOLOv3 algorithm is 20Fps, which is 12Fps higher than the speed of Faster R-CNN. The detection speed of the YOLOv3 algorithm basically meets the real-time detection requirements in this experiment.

Automated summary

The paper discusses the importance of object recognition and classification in underwater videos, and uses YOLOv3 algorithm and Faster R-CNN algorithm to detect marine objects, with results showing that YOLOv3 algorithm outperforms Faster R-CNN algorithm in both speed and accuracy.