Multi-Scale Object Detection Model for Autonomous Ship Navigation in Maritime Environment

被引:24
作者
Shao, Zeyuan [1 ]
Lyu, Hongguang [1 ]
Yin, Yong [1 ]
Cheng, Tao [2 ]
Gao, Xiaowei [2 ]
Zhang, Wenjun [1 ]
Jing, Qianfeng [1 ]
Zhao, Yanjie [3 ,4 ]
Zhang, Lunping [3 ,4 ]
机构
[1] Dalian Maritime Univ, Nav Coll, Dalian 116026, Peoples R China
[2] Univ Coll London UCL, Dept Civil Environm & Geomatic Engn, SpaceTimeLab, Gower St, London WC1E, England
[3] China Ship Sci Res Ctr, Wuxi 214082, Peoples R China
[4] Taihu Lab Deepsea Technol Sci, Wuxi 214082, Peoples R China
基金
中国国家自然科学基金; 国家重点研发计划;
关键词
autonomous ships; sea-surface; object detection; computer vision; convolutional neural network (CNN); VarifocalNet; TRACKING; IMAGE;
D O I
10.3390/jmse10111783
中图分类号
U6 [水路运输]; P75 [海洋工程];
学科分类号
0814 ; 081505 ; 0824 ; 082401 ;
摘要
Accurate detection of sea-surface objects is vital for the safe navigation of autonomous ships. With the continuous development of artificial intelligence, electro-optical (EO) sensors such as video cameras are used to supplement marine radar to improve the detection of objects that produce weak radar signals and small sizes. In this study, we propose an enhanced convolutional neural network (CNN) named VarifocalNet * that improves object detection in harsh maritime environments. Specifically, the feature representation and learning ability of the VarifocalNet model are improved by using a deformable convolution module, redesigning the loss function, introducing a soft non-maximum suppression algorithm, and incorporating multi-scale prediction methods. These strategies improve the accuracy and reliability of our CNN-based detection results under complex sea conditions, such as in turbulent waves, sea fog, and water reflection. Experimental results under different maritime conditions show that our method significantly outperforms similar methods (such as SSD, YOLOv3, RetinaNet, Faster R-CNN, Cascade R-CNN) in terms of the detection accuracy and robustness for small objects. The maritime obstacle detection results were obtained under harsh imaging conditions to demonstrate the performance of our network model.
引用
收藏
页数:20
相关论文
共 51 条
[1]   Rich feature hierarchies for accurate object detection and semantic segmentation [J].
Girshick, Ross ;
Donahue, Jeff ;
Darrell, Trevor ;
Malik, Jitendra .
2014 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2014, :580-587
[2]  
Arshad N, 2010, COMM COM INF SC, V123, P121
[3]   SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation [J].
Badrinarayanan, Vijay ;
Kendall, Alex ;
Cipolla, Roberto .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2017, 39 (12) :2481-2495
[4]   Enhancing Automatic Maritime Surveillance Systems With Visual Information [J].
Bloisi, Domenico D. ;
Previtali, Fabio ;
Pennisi, Andrea ;
Nardi, Daniele ;
Fiorini, Michele .
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2017, 18 (04) :824-833
[5]  
Bochkovskiy A., 2020, PREPRINT
[6]  
Bodla N., 2017, ARXIV
[7]   Stereo obstacle detection for unmanned surface vehicles by IMU-assisted semantic segmentation [J].
Bovcon, Borja ;
Mandeljc, Rok ;
Pers, Janez ;
Kristan, Matej .
ROBOTICS AND AUTONOMOUS SYSTEMS, 2018, 104 :1-13
[8]  
Cai Z., 2017, ARXIV
[9]   Comprehensive comparative evaluation of background subtraction algorithms in open sea environments [J].
Chan, Yi-Tung .
COMPUTER VISION AND IMAGE UNDERSTANDING, 2021, 202
[10]   Ship Type Recognition via a Coarse-to-Fine Cascaded Convolution Neural Network [J].
Chen, Xinqiang ;
Yang, Yongsheng ;
Wang, Shengzheng ;
Wu, Huafeng ;
Tang, Jinjun ;
Zhao, Jiansen ;
Wang, Zhihuan .
JOURNAL OF NAVIGATION, 2020, 73 (04) :813-832