基于改进YOLOX的水库水面漂浮物目标检测算法  

Improved YOLOX-based object detection algorithm for water surface floating objects in reservoirs

作　　者：谭文群[1,2] 曾祥君 包学才[1,2] 梁义 许小华 TAN Wenqun;ZENG Xiangjun;BAO Xuecai;LIANG Yi;XU Xiaohua(Jiangxi Province Key Laboratory of Water Information Cooperative Sensing and Intelligent Processing,Nanchang Institute of Technology,Nanchang 330099,China;School of Information Engineering,Nanchang Institute of Technology,Nanchang 330099,China;Jiangxi Academy of Water Science and Engineering,Nanchang 330029,China)

机构地区：[1]南昌工程学院江西省水信息协同感知与智能处理重点实验室,江西南昌330099 [2]南昌工程学院信息工程学院,江西南昌330099 [3]江西省水利科学院,江西南昌330029

出　　处：《人民长江》2024年第3期249-256,共8页Yangtze River

基　　金：国家自然科学基金项目(61961026);江西省科技厅重大科技研发专项“揭榜挂帅”制项目(20213AAG01012)。

摘　　要：针对目前水库水面小目标漂浮物检测识别精度低的问题,提出基于改进YOLOX的水库水面漂浮物目标检测算法。此算法引入新型dark2模块融入主干网络并拓展主干网络的分支输出结构,提升主干网络对图片的特征提取能力。在此基础上,提出改进特征融合模块(ZL-FPN),用于增强特征图信息融合,提高对水库水面小目标漂浮物的检测精度。结果表明:改进后算法的mAP值比YOLOv4和原YOLOX算法分别提升了29.93%和12.11%,有效提升了水库水面漂浮物检测精度。研究成果可为提升水库智能化管理水平提供有效技术支撑。To address the issue of low accuracy in detecting small floating objects on a reservoir,we proposed a YOLOX-based detection framework for water surface floating object recognition.The proposed detector introduces a novel dark2 module,which was embedded into the backbone as a plug-and-play module,to develop the branch structure and enhance feature extraction and representation for given images.Furthermore,we designed a modified feature aggregation module(ZL-FPN)to facilitate the fusion and interaction of multi-scale features,and the detection accuracy of small floating objects on a reservoir was improved.The results demonstrated that the proposed model obtained 29.93%and 12.11%performance gains compared with YOLOv4 and the original YOLOX.The research findings can provide effective technical support for improving the level of intelligent management of reservoirs.

关 键 词：水面小目标漂浮物 目标检测 YOLOX算法 水库智能化管理 

分 类 号：TV697.2[水利工程—水利水电工程]