Abstract
Synthetic Aperture Radar (SAR), an active remote sensing imaging radar technology, has certain surface penetration ability and can work all day and in all weather conditions. It is widely applied in ship detection to quickly collect ship information on the ocean surface from SAR images. However, the ship SAR images are often blurred, have large noise interference, and contain more small targets, which pose challenges to popular one-stage detectors, such as the single-shot multi-box detector (SSD). We designed a novel network structure, a combinational fusion SSD (CF-SSD), based on the framework of the original SSD, to solve these problems. It mainly includes three blocks, namely a combinational fusion (CF) block, a global attention module (GAM), and a mixed loss function block, to significantly improve the detection accuracy of SAR images and remote sensing images and maintain a fast inference speed. The CF block equips every feature map with the ability to detect objects of all sizes at different levels and forms a consistent and powerful detection structure to learn more useful information for SAR features. The GAM block produces attention weights and considers the channel attention information of various scale feature information or cross-layer maps so that it can obtain better feature representations from the global perspective. The mixed loss function block can better learn the positions of the truth anchor boxes by considering corner and center coordinates simultaneously. CF-SSD can effectively extract and fuse the features, avoid the loss of small or blurred object information, and precisely locate the object position from SAR images. We conducted experiments on the SAR ship dataset SSDD, and achieved a 90.3% mAP and fast inference speed close to that of the original SSD. We also tested our model on the remote sensing dataset NWPU VHR-10 and the common dataset VOC2007. The experimental results indicate that our proposed model simultaneously achieves excellent detection performance and high efficiency.
Highlights
As an active remote sensing imaging radar technology, Synthetic Aperture Radar (SAR) remote sensing has a certain surface penetration ability and can work all day and in all weather conditions, which makes up for the shortcomings of optical remote sensing and infrared remote sensing
The evaluation metric of the experiments was the mean average precision, as in Equations (20)–(21), which was calculated from precision, p, and recall, r, where p denotes the proportion of positive samples in all the predicted positive ones, and r denotes the proportion of the samples that are both positive and predicted to be positive in all the positive samples
We presented a novel single-shot one-stage perspe
Summary
As an active remote sensing imaging radar technology, Synthetic Aperture Radar (SAR) remote sensing has a certain surface penetration ability and can work all day and in all weather conditions, which makes up for the shortcomings of optical remote sensing and infrared remote sensing. SAR has been widely applied to disaster monitoring, environmental monitoring, resource exploration, mapping, and military fields. In ship detection, this technology can quickly collect ship information on the ocean surface, which is important for marine safety [1]. It is insufficient to deal with the ship’s SAR images by relying on an object detection structure such as the SSD, which is directly designed for ordinary images. Multi-scale features are not fully utilized to generate enough information for blurry or small targets. In this study, according to the characteristics of the SAR image, we designed a novel network structure named the combinational fusion single-shot multi-box detector (CF-SSD), based on the framework of the original SSD. Our main contributions are summarized as follows: First, we designed a new feature fusion module named combinational fusion (CF)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have