Abstract
Thanks to the excellent feature representation capabilities of neural networks, deep learning-based methods perform far better than traditional methods on target detection tasks such as ship detection. Although various network models have been proposed for SAR ship detection such as DRBox-v1, DRBox-v2, and MSR2N, there are still some problems such as mismatch of feature scale, contradictions between different learning tasks, and unbalanced distribution of positive samples, which have not been mentioned in these studies. In this article, an improved one-stage object detection framework based on RetinaNet and rotatable bounding box (RBox), which is referred as R-RetinaNet, is proposed to solve the above problems. The main improvements of R-RetinaNet as well as the contributions of this article are threefold. First, a scale calibration method is proposed to align the scale distribution of the output backbone feature map with the scale distribution of the targets. Second, a feature fusion network based on task-wise attention feature pyramid network is designed to decouple the feature optimization process of different tasks, which alleviates the conflict between different learning goals. Finally, an adaptive intersection over union (IoU) threshold training method is proposed for RBox-based model to correct the unbalanced distribution of positive samples caused by the fixed IoU threshold on RBox. Experimental results show that our method obtains 13.26%, 9.49%, 8.92%, and 4.55% gains in average precision under an IoU threshold of 0.5 on the public SAR ship detection dataset compared with four state-of-the-art RBox-based methods, respectively.
Highlights
S YNTHETIC aperture radar can work under all-weather and day-and-night conditions and make very high resolution images
TA-feature pyramid network (FPN) only contributed an AP50 improvement around 0.4% to the model with adaptive IoU threshold (AIT), which indicates that a better structure may be needed to decouple the gradient flow of different tasks in the channel dimension
An rotatable bounding box (RBox)-based neural network detection method is proposed for SAR image ship detection
Summary
S YNTHETIC aperture radar can work under all-weather and day-and-night conditions and make very high resolution images. It plays an important role in remote sensing information extraction and is suitable for remote monitoring. How to detect and identify various targets quickly and accurately in SAR images has been the focus of research. Researches on ship detection are vital in many areas [1], such as marine monitoring, maritime management, and military intelligence acquisition. Many investigations that relate to ship detection in SAR imagery have been carried out recently [2], [3]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have