文献类型：期刊文献

英文题名：An detection algorithm for golden pomfret based on improved YOLOv5 network

作者：Yu, Guoyan[1,3];Luo, Yingtong[1,2];Deng, Ruoling[1,3]

机构：[1]Guangdong Ocean Univ, Sch Mech Engn, Zhanjiang 524088, Guangdong, Peoples R China;[2]Guangdong Ocean Univ, Guangdong Marine Equipment & Mfg Engn Technol Res, Zhanjiang 524088, Guangdong, Peoples R China;[3]Southern Marine Sci & Engn Guangdong Lab Zhanjiang, Zhanjiang 524088, Guangdong, Peoples R China

年份：2023

卷号：17

期号：5

起止页码：1997

外文期刊名：SIGNAL IMAGE AND VIDEO PROCESSING

收录：SCI-EXPANDED(收录号：WOS:000900795400004)、、EI(收录号：20225113285433)、Scopus(收录号：2-s2.0-85144254111)、WOS

基金：The work was financially supported by the Guangdong Interregional Collaborative Fund (No. 2019B1515120017), Guangdong Special Project of Ocean Economic Development (No.011Z21001), Zhanjiang project of Innovation and Entrepreneurship Team "Pilot Program" (No.2020LHJH003), and Zhanjiang Key Laboratory of Modern Marine Fishery Equipment. (No. 2021A05023), and program for scientific research start-up funds of Guangdong Ocean University (No. 060302062106).

语种：英文

外文关键词：Golden pomfret; Transformer; YOLOv5; Object detection; Deformable convolution

外文摘要：It is of great significance to realize high-precision detection of golden pomfret for intelligent management of fishery farming. Nevertheless, the highly variable size of the objectives and the degree of overlap between objectives make optimization of the algorithm challenge. To solve the problems mentioned above, we propose a golden pomfret detection algorithm that combines the improved transformer and the YOLOv5 framework to surpass not only the canonical transformer, but also the high-performance convolutional modules. The specific methods are designed as follows: (1) On the transformer frame, this paper designs a transformer with a progressively increasing number of cascaded tokens, that aims to improve detection accuracy by adaptively learning grid parameters based on the size of the golden pomfret in each image. To achieve a high-performance result, the large kernel convolution is included between the input image and feature space mapping. (2) Based on YOLOv5, we redesigned the prediction head to address different sizes of golden pomfret detection. Then, we replace the original prediction heads with deformable prediction heads to further improve network performance and training efficiency through fine-grained feature mapping of golden pomfret. In particular, the deformable convolution uses a novel generalized linear interpolation algorithm to reduce detection errors. (3) Considering the robustness of the network, we introduce the bags of useful strategies such as data augmentation and polynomial interpolation. Experimental results in the golden pomfret test set showed that the mAP is better than the original YOLOv5 network by 22.59%. Therefore, our algorithm can effectively detect golden pomfret in complex ocean scenes.