基于YOLOv3的船舶实时监测识别

doi:10.3969/j.issn.1006-2475.2020.03.022

摘要/Abstract

摘要： 针对水面环境复杂多变、远处小目标识别准确率低和目前目标检测算法实时性差的问题，分析以Darknet-53为主干网络的YOLOv3框架相较于其他算法的改进特点，提出一种基于YOLOv3的船舶实时监测识别方法，并在训练阶段对难识别样本进行精细训练。该方法增强了系统在不同情况下船舶分类检测与识别的准确率，提高了整个算法的鲁棒性。〖JP3〗实验数据表明，最终在整个数据集上单类平均准确率最高可达到91.82%。本文方法可应用于船舶智能驾驶的辅助支撑系统。

关键词: 船舶识别, YOLOv3, 强化训练, 船舶智能驾驶, 实时监测

Abstract: Ship detection task faces some challenging problems, such as the changeable environment, long-distance small target, poor real-time performance. Compared with other algorithms, the advanced capability of YOLOv3 with backbone network Darknet-53 is analyzed, and a method of real-time ship monitoring and recognition based on YOLOv3 is put forward. Also for some difficult cases, the samples are further trained. So the mean average precision in these difficult cases are improved, and higher robustness is obtained. It is illustrated by the experimental data that the mean average precision of single class is up to 91.82%. The method can work as a support system for ship intelligent driving.

Key words: ship recognition, YOLOv3, further training, ship intelligent driving, real-time monitoring

中图分类号:

TP391.4

屈雯怡. 基于YOLOv3的船舶实时监测识别[J]. 计算机与现代化, doi: 10.3969/j.issn.1006-2475.2020.03.022.

QU Wen-yi. Real-time Ship Monitoring and Recognition Based on YOLOv3[J]. Computer and Modernization, doi: 10.3969/j.issn.1006-2475.2020.03.022.

参考文献

［1］ KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks［C］// Proceedings of the 25th International Conference on Neural Information Processing Systems. 2012:1097-1105.
［2］ SIMONYAN K, ZISSERMAN A. Very Deep Convolutional Networks for Large-scale Image Recognition ［J/OL］.(2014-09-15)［2019-07-01］. http://arxiv.org/abs/1409.1556v2.
［3］ SZEGEDY C, LIU W, JIA Y Q, et al. Going deeper with convolutions［C］// Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. 2015:1-9.
［4］ GIRSHICK R. Fast R-CNN［C］// Proceedings of IEEE International Conference on Computer Vision. 2015:1440-1448.
［5］ REN S Q, HE K M, 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, 2017,39(6):1137-1149.
［6］ CHEN X, XIANG S, LIU C L, et al. Vehicle detection in satellite images by hybrid deep convolutional neural networks［J］. IEEE Geoscience and Remote Sensing Letters， 2014,11(10):1797-1801.
［7］ LIU W, ANGUELOV D, ERHAN D, et al. SSD: Single shot MultiBox detector［C］// European Conference on Computer Vision. 2016:21-37.
［8］ REDMON J, FARHADI A. YOLO9000: Better, faster, stronger［C］// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. 2017:7263-7271.
［9］胡晨阳. 基于数字图像处理的嵌入式山体滑坡监测系统设计与实现［D］. 成都:西南交通大学, 2017.
［10］李扬. 面向图像目标识别和检测的深度神经网络关键技术研究［D］. 北京:北京邮电大学, 2018.
［11］〖JP3〗AREL I， ROSE D C， KARNOWSKI T P. Deep machine learning: A new frontier in artificial intelligence research［J］. Computational Intelligence Magazine, 2010,5(4):13-18.
［12］HUANG G B， ZHU Q Y， SIEW C K. Extreme learning machine: A new learning scheme of feed forward neural networks［C］// Proceedings of IEEE International Conference on Neural Networks. 2004:985-990.
［13］闫河,王鹏,董莺艳,等. 一种CNN与ELM相结合的船舶分类识别方法［J］. 重庆理工大学学报(自然科学), 2019,33(1):53-57.
［14］PRASAD D K, RAJAN D, RACHMAWATI L, et al. Video processing from electrooptical sensors for object detection and tracking in maritime environment: A survey［J］. IEEE Transactions on Intelligent Transportation Systems, 2017,18(8):1993-2016.
［15］BENJDIRA B, KHURSHEED T, KOUBAA A, et al. Car Detection Using Unmanned Aerial Vehicles: Comparison Between Faster R-CNN and YOLOv3［EB/OL］.(2018-12-28)［2019-07-01］. https://arxiv.org/abs/1812.10968.
［16］REDMON J, DIVVALA S K, GIRSHICK R B, et al. You only look once: Unified, real-time object detection［C］// 2016 IEEE Conference on Computer Vision and Pattern Recognition. 2016:779-788.
［17］HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition［C］// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway. 2016:770.
［18］REDMON J, FARHADI A. YOLOv3: An incremental improvement［C］// IEEE Conference on Computer Vision and Pattern Recognition. 2018:89-95.
［19］温捷文,战荫伟,凌伟林,等. 实时目标检测算法YOLO的批再规范化处理［J］. 计算机应用研究, 2018,35(10):3179-3185.
［20］施辉,陈先桥,杨英.改进YOLO v3的安全帽佩戴检测方法［J］. 计算机工程与应用, 2019,55(11):213-220.
［21］LIN T Y， DOLAR P， GIRSHICK R， et al. Feature pyramid networks for object detection ［C］// 2017 IEEE Conference on Computer Vision and Patern Recognition. 2017:936-944.
［22］LIU W Y, WEN Y D, YU Z D, et al. Large-margin softmax loss for convolutional neural networks［C］// Proceedings of the 33rd International Conference on Machine Learning. 2016:507-516.

[1]	国腾飞, 张则言, 付宏财, 王继选, 牛天宝. 融合注意力机制的轻量级红外高压套管识别算法[J]. 计算机与现代化, 2022, 0(01): 70-76.
[2]	李国进, 荣誉. 基于DCN-SERes-YOLOv3的人脸佩戴口罩检测算法[J]. 计算机与现代化, 2021, 0(09): 12-20.
[3]	李龙, 张重阳. 基于改进YOLOv3的车辆尾灯检测方法[J]. 计算机与现代化, 2021, 0(07): 89-94.
[4]	张磊, 郎贤礼, 王乐. 基于图像融合与YOLOv3的铝型材表面缺陷检测[J]. 计算机与现代化, 2020, 0(11): 8-15.
[5]	王春辉, 王全民. 基于改进YOLOv3网络的学生特定行为识别[J]. 计算机与现代化, 2020, 0(07): 90-96.
[6]	秦嘉, 曹雪虹, 焦良葆. 基于深度学习的安全帽佩戴检测与跟踪[J]. 计算机与现代化, 2020, 0(06): 1-.
[7]	陈凯,祖莉,欧屹. 基于YOLOv3与ResNet50的摄影机器人人脸识别跟踪系统[J]. 计算机与现代化, 2020, 0(04): 30-.
[8]	刘力冉1，曹杰2，杨磊1，仇男豪1. 一种改进YOLOv3-Tiny的行车检测算法[J]. 计算机与现代化, 2020, 0(03): 108-.
[9]	莫蓓蓓，吴克河 . 引入Self-Attention的电力作业违规穿戴智能检测技术研究[J]. 计算机与现代化, 2020, 0(02): 115-.
[10]	杨磊1,王少云2,刘力冉1,龚永富1. 一种智能视频监控系统中的行人检测方法[J]. 计算机与现代化, 2019, 0(11): 69-.
[11]	张雪云,牟艳,张九博. 基于微信小程序的电源监测管理系统设计[J]. 计算机与现代化, 2017, 0(12): 98-102+107.
[12]	张文，牟艳，高振兴，刘志丰. WebSocket技术在工业设备数据实时监测中的应用[J]. 计算机与现代化, 2017, 0(11): 111-115.
[13]	龚安，高云，高洪福. 研究生入学考试信息验证核对系统[J]. 计算机与现代化, 2015, 0(1): 105-108.
[14]	张新林;彭灿明. 基于模态匹配的校园网信息实时过滤方法[J]. 计算机与现代化, 2013, 1(11): 91-94.
[15]	王震;陈跃鹏. 物联网环境下的助盲寻物机器人动态路径规划[J]. 计算机与现代化, 2012, 208(12): 23-28.