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.
作者
谭文群
曾祥君
包学才
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)
出处
《人民长江》
北大核心
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)。
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
small floating objects on water surface
object detection
YOLOX algorithm
intelligent management of reservoirs
