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

Масштабно-многоуровневая модель обнаружения объектов для автономной навигации судов в морской среде
Qianfeng Jing, Tao Cheng, Yong Yin, Zeyuan Shao, Hongguang Lyu, Xiaowei Gao, Wenjun Zhang, Yanjie Zhao, Lunping Zhang
2022-11-19

VarifocalNetautonomous ship navigationdeformable convolutionmaritime object detectionmulti-scale prediction
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.
1
An enhanced VarifocalNet CNN is proposed for detecting sea-surface objects in harsh maritime environments.
2
Experiments show significantly higher detection accuracy and robustness for small maritime objects than SSD, YOLOv3, RetinaNet, Faster R-CNN, and Cascade R-CNN.
3
The method demonstrates effective maritime obstacle detection under challenging imaging conditions, supporting autonomous ship navigation.
4
The model combines deformable convolutions, a redesigned loss function, soft non-maximum suppression, and multi-scale prediction.
5
These components improve feature representation, learning capacity, and detection reliability under turbulent waves, sea fog, and water reflections.

sea-surface objects and maritime obstacles detected by autonomous ships using EO sensors in harsh maritime environments

small-object detection accuracy and robustness under complex sea and imaging conditions, including turbulent waves, sea fog, and water reflection

Publication Details
Publication Date
2022-11-19
Journal
Publisher
ISSN
Cited by
41
Access Type
Author Information
Authors
Qianfeng Jing
Tao Cheng
Yong Yin
Zeyuan Shao
Hongguang Lyu
Xiaowei Gao
Wenjun Zhang
Yanjie Zhao
Lunping Zhang
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%