任英杰,李传奇,王薇,葛召华

• 1:山东大学土建与水利学院
• 2:山东省水利综合事业服务中心

摘要(Abstract)：

【目的】为解决水域监控下漂浮物检测效率低、检测模型复杂度高的问题，提出一种基于改进YOLOv3的轻量化漂浮物检测算法。【方法】使用轻量级网络MobileNetv3代替YOLOv3的主干特征提取网络Darknet53以降低模型计算量和参数；构建简化版加权双向特征金字塔网络(Bi-FPN-tiny)以进行多尺度特征的加权融合；利用Focal Loss优化损失函数，加强对于困难样本的学习。为验证所提算法的有效性，建立了PASCAL VOC格式的漂浮物数据集，并进行数据标注和增广。【结果】结果表明：改进后的算法平均精度均值(mAP)达到92.8%,比原算法提高了7.1%;在NVIDIA Quadro P2200显卡下检测速度达到了86 fps/s,高于YOLOv3算法的47 fps/s;模型体积为43.7 MB,仅为初始算法的17.7%。【结论】改进YOLOv3是一种性能优越且轻量化的模型，为在移动端进行实时漂浮物检测提供了新的契机。

关键词(KeyWords)： YOLOv3算法;漂浮物;目标检测;轻量化;特征融合

Abstract:

Keywords:

基金项目(Foundation): 山东省自然科学基金(ZR2021ME030);; 深圳市可持续发展科技专项项目(KCXFZ20201221173407021);; 济南市水务科技项目(JNSWKJ202106)

作者(Author): 任英杰,李传奇,王薇,葛召华

DOI: 10.13928/j.cnki.wrahe.2023.10.015

参考文献(References)：

• [1] 施巍松，孙辉，曹杰，等.边缘计算：万物互联时代新型计算模型[J].计算机研究与发展，2017,54(5):907-924.SHI Weisong,SUN Hui,CAO Jie,et al.Edge computing-an emerging computing model for the internet of everything era [J].Journal of Computer Research and Development,2017,54(5):907-924.
• [2] 张慧，王坤峰，王飞跃.深度学习在目标视觉检测中的应用进展与展望[J].自动学报，2017,43(8):1289-1305.ZHANG Hui,WANG Kunfeng,WANG Feiyue.Advances and perspectives on applications of deep learning in visual object detection [J].Acta Automatica Sinica,2017,43(8):1289-1305.
• [3] 范丽丽，赵宏伟，赵浩宇，等.基于深度卷积神经网络的目标检测研究综述[J].光学精密工程，2020,28(5):1152-1164.FAN Lili,ZHAO Hongwei,ZHAO Haoyu,et al.Survey of target detection based on deep convolutional neural networks [J].Optics and Precision Engineering,2020,28(5):1152-1164.
• [4] 程旭，宋晨，史金钢，等.基于深度学习的通用目标检测研究综述[J].电子学报，2021,49(7):1428-1438.CHENG Xu,SONG Chen,SHI Jingang,et al.A survey of generic object detection methods based on deep learning [J].Acta Electronica Sinica,2021,49(7):1428-1438.
• [5] LIU W,ANGUELOV D,ERHAN D,et al.SSD:Single shot multibox detector [C].Amsterdam:European conference on computer vision,2016.
• [6] REDMON J,DIVVALA S,GIRSHICK R,et al.You only look once:Unified,real-time object detection [C].Las Vegas:IEEE conference on Computer Vision and Pattern Recognition,2016.
• [7] REDMON J,FARHADI A.YOLOv3:An incremental improvement [EB/OL].[2019-02-25].https://arxiv.org/abs/1804.02767v1.
• [8] REN S,HE K,GIRSHICK R,et al.Faster R-CNN:Towards real-time object detection with region proposal networks [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,39(6):1137-1149.
• [9] HE K,GKIOXARI G,DOLLAR P,et al.Mask R-CNN [C].Venice:IEEE International Conference on Computer Vision,2017.
• [10] 曹天正，骆辉煌，李月等.复苏河湖生态环境背景下地下水环境变化研究[J].水利发展研究，2022,22(10):5-10.CAO Tianzheng,LUO Guanghui,LI Yue,et al.Research on Groundwater Environment Change under the Background of Revitalizing the Ecological Environment of Rivers and Lakes[J].Water Resources Development Research,2022,22(10):5-10.
• [11] SINGLA N.Motion detection based on frame difference method[J].International Journal of Information & Computation Technology,2014,4(15):1559-1565.
• [12] HOESER T,KUENZER C.Object detection and image segmentation with deep learning on earth observation data:A review-part i:Evolution and recent trends[J].Remote Sensing,2020,12(10):1667.
• [13] 李国进，姚冬宜，艾矫燕，等.基于改进YOLOv3算法的水面漂浮物检测方法[J].广西大学学报(自然科学版),2021,46(6):1569-1578.LI Guojin,YAO Dongyi,AI Jiaoyan,et al.Floating objects detection based on improved YOLOv3 [J].Journal of Guangxi University(Natural Science Edition),2021,46(6):1569-1578.
• [14] 李国进，姚冬宜，艾矫燕，等.基于改进Faster R-CNN的水面漂浮物识别与定位[J].信阳师范学院学报(自然科学版),2021,34(2):292-299.LI Guojin,YAO Dongyi,AI Jiaoyan,et al.Detection and localization of floating objects via improved Faster R-CNN [J].Journal of Xinyang Normal University (Natural Science Edition),2021,34(2):292-299.
• [15] 刘伟，王源楠，江山，等.基于Mask R-CNN的水面漂浮物识别方法研究[J].人民长江，2021,52(11):226-233.LIU Wwei,WANG Yuannan,JIANG Shan,et al.Research on recognition method of floating objects on water surface based on Mask R-CNN [J].Yangtze River,2021,52(11):226-233.
• [16] 陈娜，蔺志刚，刘瑾程，等.基于智能视频监控系统的河湖四乱巡检技术研究[J] .水利水电技术(中英文),2022,53( S2):455-462.CHEN Na,LIN Zhigang,LIU Jincheng,et al.Research of inspection technology of four illegal action on the rivers and lakes based on intelligent video monitoring system[J].Water Resources and Hydropower Engineering,2022,53(S2):455-462.
• [17] LIN F,HOU T,JIN Q,et al.Improved YOLO Based Detection Algorithm for Floating Debris in Waterway[J].Entropy,2021,23(9):1111.
• [18] ZHANG L,ZHANG Y,ZHANG Z,et al.Real-Time Water Surface Object Detection Based on Improved Faster R-CNN[J].Sensors (Basel).2019,19(16):3523.
• [19] HOWARD A,SANDLER M,CHU G,et al.Searching for MobileNetv3 [C].International Conference on Information,Communication and Networks (ICICN).Seoul:IEEE,2019.
• [20] TAN M,PANG R,LE Q V.Efficientdet:Scalable and efficient object detection[J].Proceedings of the IEEE/CVF conference on computer vision and pattern recognition 2020.10781-10790.
• [21] LIN T,GOYAL P,GIRSHICK R,et al.Focal loss for dense object detection[C].IEEE International Conference on Computer Vision (ICCV).Venice,Italy:IEEE,2017.
• [22] 邹北骥，雷太航，刘姝，等.自然场景车标数据集的构建及其应用[J].国防科技大学学报，2021,43(1):95-102.ZOU Beiji,LEI Taihang,LIU Shu,et al.Construction of vehicle logo dataset in natural scenes and its application [J].Journal of national university of Defense Technology,2021,43(1):95-102.