基于Faster R-CNN的海面垃圾检测研究

韦波; 张衡; 王斐; 王书献; 杨昱皞; 姚宇青; 戴阳

doi:DOI:10.13205/j.hjgc.202207022

基于Faster R-CNN的海面垃圾检测研究

doi: DOI:10.13205/j.hjgc.202207022

韦波^1,2,
张衡²,
王斐²,
王书献²,
杨昱皞²,
姚宇青^1,2,
戴阳^2, ,

1. 上海海洋大学信息学院, 上海 201306;
2. 中国水产科学研究院东海水产研究所农业农村部远洋与极地渔业创新重点实验室, 上海 200090

基金项目:

上海市2020年度"科技创新行动计划"(20dz1206400)

上海市青年科技英才扬帆计划(19YF1460000)

详细信息

作者简介:
韦波(1996-),男,硕士研究生,主要研究方向为深度学习。lianpigouhou@163.com

通讯作者:
戴阳(1969-),男,副研究员,主要研究方向为海洋低功耗仪器。13162787852@163.com

计量
- 文章访问数: 165
- HTML全文浏览量: 13
- PDF下载量: 4
- 被引次数: 0
出版历程
- 收稿日期: 2021-09-18
- 网络出版日期: 2022-09-02

RESEARCH ON SEA GARBAGE DETECTION BASED ON FASTER R-CNN

1. School of Information, Shanghai Ocean University, Shanghai 201306, China;
2. Key Laboratory of Innovation in Pelagic and Polar Fisheries, Ministry of Agriculture and Rural Affairs, East China Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Shanghai 200090, China

摘要

摘要: 针对当前全球海洋垃圾污染造成严重生态破坏等问题,提出了一种基于改进Faster R-CNN的海面垃圾检测算法以及视频中前后2帧目标是否为同一目标的方法。改进Faster R-CNN算法通过将常用的VGG16特征提取网络替换为ResNet101网络并融入特征金字塔,提高对小目标的检测精度;判断视频前后两帧目标是否为同一目标的方法则通过对比前后2帧目标的面积、重合度以及颜色差异度确定是否为同一目标。在现场拍摄数据上的实验结果表明,与传统Faster R-CNN相比,该改进Faster R-CNN的mAP值提高了4.9%,损失曲线的收敛速度更快,且在实际检测中的检测效果更好;前后两帧是否同一物体的判断方法在九段视频的最高精测判断精度高达100%,平均准确率为93%。该研究方法主要包括以下优点:1)改进Faster R-CNN在海面小目标垃圾检测上具有更高的精度;2)判断视频中前后2帧目标是否为同一目标的算法代码复杂度小,方便根据实际情况更改判断阈值。
- 改进Faster R-CNN /
- 小目标 /
- 特征金字塔 /
- 同一目标判别法
Abstract: Given the serious ecological damage caused by global marine garbage pollution,a sea garbage detection algorithm based on improved fast R-CNN and a method of whether the targets in the front and back frames of the video are the same target were proposed.The improved fast R-CNN algorithm improved the detection accuracy of small targets by replacing the commonly used VGG16 feature extraction network with ResNet101network and integrating it into the feature pyramid;the method to judge whether the two frames before and after the video were the same target,was to determine whether they were the same target by comparing the area,coincidence and color difference of them.The experimental results on the field shooting data showed that compared with the traditional fast R-CNN,the map value of the improved fast R-CNN in this paper was increased by 4.9%,the convergence speed of the loss curve was faster,and the detection effect in the actual detection was better;the proposed method for judging whether the two frames are the same object had a peak precision of 100% and an average accuracy of 93% in nine videos.The proposed method mainly included the following advantages:1) the improved fast R-CNN had higher accuracy in the detection of small target garbage on the sea surface;2) the algorithm code complexity of judging whether the target of the first and second frames in the video is the same target is small,making it convenient to change the judgment threshold according to the actual situation.
- improved Faster R-CNN /
- small target /
- feature pyramid /
- same target discrimination method

HTML全文

参考文献(29)

[1]	VILLARRUBIA-GOMEZ P, CORNELL S E, FABRES J. Marine plastic pollution as a planetary boundary threat:the drifting piece in the sustainability puzzle[J]. Marine Policy, 2018, 96:213-220.
[2]	马萌婉.北极航道水域环境保护的国际法律体系研究[D].上海:上海师范大学,2021.
[3]	陈熙,高翊尧,凌玮.辽东湾河口区海洋垃圾赋存特征及管理对策[J].环境科学研究,2019,32(12):1959-1965.
[4]	张典,俞炜炜,陈彬.厦门湾海洋塑料垃圾对中华白海豚的摄食风险评价[J].中国环境科学,2020,40(4):1809-1818.
[5]	李潇,杨翼,杨璐.欧盟及其成员国海洋塑料垃圾政策及对我国的启示[J].海洋通报,2019,38(1):14-19.
[6]	XUE C,JIA P,ZHANG X Y.Steering control in autonomous vehicles using deep reinforcement learning[J].The Journal of China Universities of Posts and Telecommunications,2018,25(6):58-64,73.
[7]	CAO C,LIU F,TAN H,et al. Deep learning and its applications in biomedicine[J]. Genomics, Proteomics&Bioinformatics,2018,16(1):17-32.
[8]	BARGSHADY G, ZHOU X J, DEO R C, et al. Enhanced deep learning algorithm development to detect pain intensity from facial expression images[J]. Expert Systems with Applications, 2020, 149:113305.
[9]	HAMMAM A A, SOLIMAN M M, HASSANIEN A E. Real-time multiple spatiotemporal action localization and prediction approach using deep learning[J]. Neural Networks, 2020, 128:331-344.
[10]	陈历.深度学习在遥感领域的研究及应用[D].武汉:华中科技大学,2017.
[11]	李德鑫,闫志刚,孙久运.基于无人机视觉的河道漂浮垃圾分类检测技术研究[J/OL].金属矿山:1-11[2021-08-05 ].http://kns.cnki.net/kcms/detail/34.1055.TD.20210608.1117.005.html.
[12]	耿丽婷,阿里甫·库尔班,米娜瓦尔·阿不拉.改进SSD的可回收垃圾检测方法[J/OL].计算机工程与应用:1-9[2021-08-05 ].http://kns.cnki.net/kcms/detail/11.2127.TP.20210721.1701.028.html.
[13]	NIE Z F, DUAN W J, LI X D. Domestic garbage recognition and detection based on Faster R-CNN[C]//Journal of Physics:Conference Series. IOP Publishing, 2021, 1738(1):012089.
[14]	潘美艳,孙俊,杨予昊.基于Faster R-CNN网络的海面目标检测方法[J].现代雷达,2021,43(6):19-26.
[15]	MENG R H, RICE S G, WANG J, et al. A fusion steganographic algorithm based on faster R-CNN[J]. Computers, Materials&Continua, 2018, 55(1):1-16.
[16]	HUNG J, CARPENTER A. Applying faster R-CNN for object detection on malaria images[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2017:56-61.
[17]	LI J X, ZHANG D X, ZHANG J J, et al. Facial expression recognition with faster R-CNN[J]. Procedia Computer Science, 2017, 107:135-140.
[18]	ZHANG L L, LIN L, LIANG X D, et al. Is faster R-CNN doing well for pedestrian detection?[C]//European Conference on Computer Vision. Springer, Cham, 2016:443-457.
[19]	XU Y Z, YU G Z, WANG Y P, et al. Car detection from low-altitude UAV imagery with the faster R-CNN[J]. Journal of Advanced Transportation, 2017, 2017.
[20]	LIN T Y, DOLLÁR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017:2117-2125.
[21]	ZHANG D M,JIN G Q,DAI F,et al. Saliency target detection algorithm based on deep fusion[J].Chinese Journal of Computers,2019,42(9):2076-2086.
[22]	ZIVKOVIC Z, KROSE B. An EM-like algorithm for color-histogram-based object tracking[C]//Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004. IEEE, 2004, 1:I-I.
[23]	王书献,张胜茂,朱文斌.基于深度学习YOLO v5网络模型的金枪鱼延绳钓电子监控系统目标检测应用[J/OL].大连海洋大学学报:1-17[2021-07-01 ].https://doi.org/10.16535/j.cnki.dlhyxb.2020-333.
[24]	曹磊,王强,史润佳.基于改进RPN的Faster-RCNN网络SAR图像车辆目标检测方法[J].东南大学学报(自然科学版),2021,51(1):87-91.
[25]	张远琴,肖德琴,陈焕坤.基于改进Faster R-CNN的水稻稻穗检测方法[J/OL].农业机械学报:1-12[2021-07-01 ].http://kns.cnki.net/kcms/detail/11.1964.S.20210623.0909.002.html.
[26]	REN S Q, HE K M, GIRSHICK R, et al. Faster r-cnn:towards real-time object detection with region proposal networks[J]. Advances in Neural Information Processing Systems, 2015, 28:91-99.
[27]	WANG W Z, WU B, YANG S X, et al. Road damage detection and classification with faster r-cnn[C]//2018 IEEE international conference on big data. IEEE, 2018:5220-5223.
[28]	SUN L,WU T,SUN G C,et al.Object detection research of SAR image using improved faster regionBased convolutional neural network[J].Journal of Geodesy and Geoinformation Science,2020,3(3):18-28.
[29]	ZHAO Z B, ZHEN Z, ZHANG L, et al. Insulator detection method in inspection image based on improved faster R-CNN[J]. Energies, 2019, 12(7):7-15.