包学才,刘飞燕,聂菊根,许小华,柯华盛

• 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
• 2:南昌工程学院信息工程学院
• 3:江西省水利科学院

摘要(Abstract)：

【目的】为解决传统图像处理方法鲁棒性差、常用深度学习检测方法无法准确识别大片漂浮物的边界等问题，【方法】提出一种基于改进DeeplabV3+的水面多类型漂浮物识别的语义分割方法，提高水面漂浮的识别能力。对所收集实际水面漂浮物进行分类，采用自制数据集进行对比试验。算法选择xception网络作为主干网络以获得初步漂浮物特征，在加强特征提取网络部分引入注意力机制以强调有效特征信息，在后处理阶段加入全连接条件随机场模型，将单个像素点的局部信息与全局语义信息融合。【结果】对比图像分割性能指标，改进后的算法mPA(Mean Pixel Accuracy)提升了5.73%,mIOU(Mean Intersection Over Union)提升了4.37%。【结论】相比于其他算法模型，改进后的DeeplabV3+算法对漂浮物特征的获取能力更强，同时能获得丰富的细节信息以更精准地识别多类型水面漂浮物的边界与较难分类的漂浮物，在对多个水库场景测试后满足实际水域环境中漂浮物检测的需求。

关键词(KeyWords)： 深度学习;语义分割;特征提取;漂浮物识别;注意力机制;全连接条件随机场;算法模型;影响因素

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目(61961026);; 江西省水利厅科技项目(202223YBKT19);; 江西省科技厅重大科技研发专项“揭榜挂帅”制项目(20213AAG01012)

作者(Author): 包学才,刘飞燕,聂菊根,许小华,柯华盛

DOI: 10.13928/j.cnki.wrahe.2024.04.015

参考文献(References)：

• [1] 程军蕊，徐继荣，郑琦宏，等.宁波市城区河道水环境综合整治效果评价方法及应用[J].长江流域资源与环境，2015,24(6):1060-1066.CHENG Junrui,XU Jirong,ZHENG Qihong,et al.Evaluation method and application of comprehensive improvement effect on river water environment in urban areas of Ningbo City [J].Resources and Environment of the Yangtze River Basin,2015,24 (6):1060-1066.
• [2] 张馨月，高千红，闫金波，等.三峡水库近坝段水面漂浮物对水质的影响[J].湖泊科学，2020,32(3):609-618.ZHANG Xinyue,GAO Qianhong,YAN Jinbo,et al.The impact of floating debris on water quality in the near dam section of the Three Gorges Reservoir [J].Lake Science,2020,32 (3):609-618.
• [3] 谷源泉，董华，翟冬毅.小浪底和西霞院水库漂浮物综合防治措施研究[C]//中国水利学会.2022中国水利学术大会论文集(第二分册).郑州：黄河水利出版社，2022.GU Yuanyuan,DONG Hua,ZHAI Dongyi.Research on comprehensive prevention and control measures for floating objects in xiaolangdi and xixiayuan reservoirs [C]//Chinese Water Conservancy Society.Proceedings of the 2022 China Water Conservancy Academic Conference (Volume 2).Zhengzhou:Yellow River Water Conservancy Press,2022.
• [4] 蔡莹，唐祥甫，蒋文秀.河道漂浮物对工程影响及研究现状[J].长江科学院院报，2013,30(8):84-89.CAI Ying,TANG Xiangfu,JIANG Wenxiu.The impact of river floating debris on engineering and its research status [J].Journal of the Yangtze River Academy of Sciences,2013,30 (8):84-89.
• [5] 陈佳，董世勇.重庆市三峡水库消落区管理现状与优化研究[J].水利发展研究，2022,22(2):77-82.CHEN Jia,DONG Shiyong.Research on the current situation and optimization of management in the Three Gorges Reservoir water consumption area in Chongqing[J].Water Resources Development Research,2022,22(2):77-82.
• [6] 江杰，李刚.河流漂浮物的自动监测方法研究[J].人民黄河，2010,32(11):49-50.JIANG Jie,LI Gang.Research on automatic monitoring methods for floating debris in rivers [J].People′s Yellow River,2010,32 (11):49-50.
• [7] 吴有富，左建军，吴晶.一种基于SOM的水面污染物智能监控系统环保科技[J].2012,18(2):1-4.WU Youfu,ZUO Jianjun,WU Jing.An SOM based intelligent monitoring system for water surface pollutants in environmental protection technology [J] 2012,18 (2):1-4.
• [8] 於双飞，杨怀斌，孔凡国，等.一种水面漂浮目标的视觉检测方法[J].机电工程技术，2019,48(4):145-147.YU Shuangfei,YANG Huaibin,KONG Fanguo,et al.A visual detection method for floating targets on water surface [J].Mechanical and Electrical Engineering Technology,2019,48 (4):145-147.
• [9] 汤伟，刘思洋，高涵，等.基于视觉的水面垃圾清理机器人目标检测算法[J].科学技术与工程，2019,19(3):141-146.TANG Wei,LIU Siyang,GAO Han,et al.Visual based object detection algorithm for surface garbage cleaning robots [J].Science and Technology and Engineering,2019,19 (3):141-146.
• [10] ALTAF A,RAEISI A.Presenting an effective algorithm for tracking of moving object based on support vector machine[J].Indian Journal of Science & Technology,2015,8(17):1-7.
• [11] 曾文静.基于光视觉的无人艇水面目标检测与跟踪研究[D].哈尔滨：哈尔滨工程大学，2013.ZENG Wenjing.Research on surface target detection and tracking of unmanned craft based on light vision [D].Harbin:Harbin Engineering University,2013.
• [12] KATAOKA T,HINATA H,KAKO S.A new technique for detecting colored macro plastic debris on beaches using webcam images and CIELUV[J].Marine Pollution Bulletin,2012,64(9):1829-1836.
• [13] LIU W,ANGUELOV D,et al.Ssd:Single shot multibox detector[C]//Computer Vision-ECCV 2016:14th European Conference.Amsterdam:Springer International Publishing,2016.
• [14] REDMON J,DIVVALA S,GIRSHICK R,et al.You only look once:Unified,real-time object detection[C]//29th IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas:Institute of Electrical and Electronics Engineers,2016.
• [15] REDMON J,FARHADI A.YOLO9000:Better,faster,stronger[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition:CVPR 2017.Hawaii:Institute of Electrical and Electronics Engineers,2017.
• [16] LIN T Y,GOYAL P,GIRSHICK R,et al.Focal loss for denseobject detection[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision(ICCV).Venice:IEEE,2017.
• [17] GIRSHICK R.Fast R-CNN[C]//2015 IEEE International Conference on Computer Vision.Santiago:Institute of Electrical and Electronics Engineers,2015:1440-1448.
• [18] 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 & Machine Intelligence,2017,39(6):1137-1149.
• [19] 雷李义，艾矫燕，彭婧.基于深度学习的水面漂浮物目标检测评估[J].环境与发展，2019,31(6):117-120.LEI Liyi,AI Jiaoyan,PENG Jing.Detection and evaluation of floating objects on water surface based on deep learning [J].Environment and Development,2019,31 (6):117-120.
• [20] ZHANG H,KYAW Z,YU J,et al.PPR-FCN:Weakly supervised visual relation detection via parallel pairwise R-FCN[C]//IEEE International Conference on Computer Vision.Venice:IEEE,2017.
• [21] 唐小敏，舒远仲，刘文样，等.基于SSD深度网络的河道漂浮物检测技术研究[J].计算机技术与发展，2020,30(9):154-158.TANG Xiaomin,SHU Yuanzhong,LIU Wenxiang,et al.Research on river floating object detection technology based on SSD Deep Network [J].Computer Technology and Development,2020,30 (9):154-158.
• [22] HE K,ZHANG X,REN S,et al.Deep residual learning for image recognition[C]//29th IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:Institute of Electrical and Electronics Engineers,2016.
• [23] 李宁，王雨萱，徐守坤，等.基于AlexNet的小样本水面漂浮物识别[J].计算机应用与软件，2019,36(2):245-251.LI Ning,WANG Yuxuan,XU Shoukun,et al.Identification of small sample floating objects on water surface based on AlexNet [J].Computer Applications and Software,2019,36 (2):245-251.
• [24] MARQUES T P,ALBU A B,O′HARA P,et al.Size-invariant detection of marine vessels from visual time series[C]//2021 IEEE Winter Conference on Applications of Computer Vision.Waikoloa:Institute of Electrical and Electronics Engineers,2021:443-453.
• [25] SHARMA V,TIWARI P,KAUR N,et al.Optical nanosensors based on fluorescent carbon dots for the detection of water contaminants:A review[J].Environmental Chemistry Letters,2021,19:3229-3241.
• [26] KHAN H M,YUNZE C.Ship detection in SAR image using YOLOv2[C]//2018 37th Chinese Control Conference.Wuhan:Institute of Electrical and Electronics Engineers,2018.
• [27] 雷雨萌，陈祖煜，于沭，等.基于深度阈值卷积模型的土石料级配智能检测方法研究[J].水利学报，2021,52(3):369-380.LEI Yumeng,CHEN Zuyu,YU Shu,et al.Research on intelligent detection method for soil and rock grading based on deep threshold convolution model [J].Journal of Water Resources,2021,52 (3):369-380.
• [28] CHEN L C,PAPANDREOU G,KOKKINOS I,et al.Semantic image segmentation with deep convolutional nets an-d fully connected CRFs[J].Computer Science,2014(4):357-361.
• [29] CHEN L C,PAPANDREOU G,SCHROFF F,et al.Rethinking atrous convolution for semantic image segmentation[J].2017.DOI:10.48550/arXiv.1706.05587.
• [30] CHEN L C,PAPANDREOU G,KOKKINOS I,et al.DeepLab:Semantic image segmentation with deep convolutional nets,atrous convolution,and fully connected CRFs [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2018,40(4):834-848.
• [31] CHEN L C,ZHU Y K,PAPANDREOU G,et al.Encoderdecoder with atrous separable convolution for semantic image segmentation [M].Cham:Springer,2018,11211:833-851.
• [32] WOO S,PARK J,LEE J Y,et al.CBAM:Convolutional block attention module[C]//Computer vision-ECCV 2018:15th European Conference.Munich:Springer,2018.
• [33] ZHENG S,JAYASUMANA S,ROMERA-PAREDES B,et al.Conditional random fields as recurrent neural networks[C].//2015 IEEE international conference on computer vision:ICCV 2015.Santiago:Institute of Electrical and Electronics Engineers,2015.
• [34] RONNEBERGER O,FISCHER P,BROX T.U-net:Convolutional networks for biomedical image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015:18th International Conference.Munich:Springer International Publishing,2015.
• [35] ZHAO H,SHI J,QI X,et al.Pyramid scene parsing network[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition:CVPR 2017.Hawaii:Institute of Electrical and Electronics Engineers,2017.