Abstract
Deep learning detection methods use in ship detection remains a challenge, owing to the small scale of the objects and interference from complex sea surfaces. In addition, existing ship detection methods rarely verify the robustness of their algorithms on multisensor images. Thus, we propose a new improvement on the “you only look once” version 3 (YOLOv3) framework for ship detection in marine surveillance, based on synthetic aperture radar (SAR) and optical imagery. First, improved choices are obtained for the anchor boxes by using linear scaling based on the k-means++ algorithm. This addresses the difficulty in reflecting the advantages of YOLOv3's multiscale detection, as the anchor boxes of a single detection target type between different detection scales have small differences. Second, we add uncertainty estimators for the positioning of the bounding boxes by introducing a Gaussian parameter for ship detection into the YOLOv3 framework. Finally, four anchor boxes are allocated to each detection scale in the Gaussian-YOLO layer instead of three as in the default YOLOv3 settings, as there are wide disparities in an object's size and direction in remote sensing images with different resolutions. Applying the proposed strategy to "YOLOv3-spp” and "YOLOv3-tiny,” the results are enhanced by 2%-3%. Compared with other models, the improved-YOLOv3 has the highest average precision on both the optical (93.56%) and SAR (95.52%) datasets. The improved-YOLOv3 is robust, even in the context of a mixed dataset of SAR and optical images comprising images from different satellites and with different scales.
Highlights
S HIP detection, in the context of maritime monitoring services, has a large number of potential applications, including those in ocean environment monitoring, illegal fishing prevention, rescue and disaster relief, irregular migration detection, and military reconnaissance
synthetic aperture radar (SAR) ship detection has several limitations: it is susceptible to noise interference; it is vulnerable to strong winds and waves; small objects are difficult to locate and identify; and different ship types are difficult to differentiate [3]
true positives (TP) denoted the number of ships correctly detected as ships, false positive (FP) denoted the number of backgrounds incorrectly detected as ships, and false negative(FN) denoted the number of ships incorrectly detected as backgrounds
Summary
S HIP detection, in the context of maritime monitoring services, has a large number of potential applications, including those in ocean environment monitoring, illegal fishing prevention, rescue and disaster relief, irregular migration detection, and military reconnaissance. Deep learning based on object detection tends to underperform when applied to remote sensing image-based ship detection. A detection algorithm should provide fast detection and support various types of satellite sensors It should serve as suitable for the remote sensing of multiscene, multiscale, and multiresolution images. We trained and evaluated the state-of-the-art detection methods Faster R-CNN, SSD, and YOLOv3, and the improved-YOLOv3 on ship detection benchmark datasets including multi-scene, multiresolution, and multisize optical and SAR images. This article addresses the problem regarding the centralized distribution of anchor boxes for HONG et al.: MULTISCALE SHIP DETECTION FROM SAR AND OPTICAL IMAGERY detecting a single target type in the dataset.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
