Underwater scallop recognition algorithm using improved YOLOv5
Songsong Li,Chen Li,Ying Yang,Qi Zhang,Yuheng Wang,Zhongyu Guo
DOI: https://doi.org/10.1016/j.aquaeng.2022.102273
IF: 4
2022-06-22
Aquacultural Engineering
Abstract:The identification of scallops is of great significance for the evaluation of scallop resources. At present, the identification of bottom sowing cultured scallops is mainly though manual inspection, which has strong subjectivity and high error rate. This study proposes an underwater scallop recognition algorithm based on improved Yolov5s. First, a new lightweight backbone network model was designed to replace the original Yolov5s backbone network using group convolution and inverse residual block, which improves the detection accuracy and accelerates the detection speed. Thereafter, k-means algorithm was used for clustering analysis of the dataset to reduce the initial prediction layer of the model from three to two, which further reduced the volume of the model. Then, the image enhancement module mainly used the adaptive dark channel algorithm to improve the clarity of the blurred image. Finally, the scallop dataset collected in the laboratory was used for the verification. The experimental results indicate that the accuracy rate, recall rate, F1, and mAP of the proposed algorithm reached 90.8%, 91.6%, 91.2%, and 88.2%, respectively, which are 7.9%, 1.2%, 4.7%, and 1.6% higher than those of the original Yolov5s. Moreover, the volume of the algorithm model in this study was only 36.5Mbytes, approximately 35% less than that required of the original model, and the detection speed of the proposed model was increased by 39%. This algorithm provides a good solution for recognizing bottom sowing cultured scallops.
fisheries,agricultural engineering