Abstract
Automatic ship detection by Unmanned Airborne Vehicles (UAVs) and satellites is one of the fundamental challenges in maritime research due to the variable appearances of ships and complex sea backgrounds. To address this issue, in this paper, a novel multi-level ship detection algorithm is proposed to detect various types of offshore ships more precisely and quickly under all possible imaging variations. Our object detection system consists of two phases. First, in the category-independent region proposal phase, the steerable pyramid for multi-scale analysis is performed to generate a set of saliency maps in which the candidate region pixels are assigned to high salient values. Then, the set of saliency maps is used for constructing the graph-based segmentation, which can produce more accurate candidate regions compared with the threshold segmentation. More importantly, the proposed algorithm can produce a rather smaller set of candidates in comparison with the classical sliding window object detection paradigm or the other region proposal algorithms. Second, in the target identification phase, a rotation-invariant descriptor, which combines the histogram of oriented gradients (HOG) cells and the Fourier basis together, is investigated to distinguish between ships and non-ships. Meanwhile, the main direction of the ship can also be estimated in this phase. The overall algorithm can account for large variations in scale and rotation. Experiments on optical remote sensing (ORS) images demonstrate the effectiveness and robustness of our detection system.
Highlights
With the rapid development of the Earth Observing Satellite Technique, more high-resolution optical remote sensing (ORS) data become available, which enlarges the potential of ORS data for image analysis
Ships are confused with the interferences introduced by heavy clouds, islands, coastlines, ocean waves, and other uncertain sea state conditions, which further increases the difficulty of ship detection
Since ship candidates appear in very different directions, we use a rotation-invariant gradient descriptor based on Fourier analysis combined with linear support vector machine (SVM) classifier to identify ship targets at arbitrary orientations
Summary
With the rapid development of the Earth Observing Satellite Technique, more high-resolution optical remote sensing (ORS) data become available, which enlarges the potential of ORS data for image analysis. Most current existing methods adopt a two-stage detection mechanism, namely, region proposal and ship target identification. Leng et al [2] employed the same segmentation algorithm to distinguish potential target and clutter pixels These methods have a small computational effort and work well on the simple ocean scenes. A practical region proposal algorithm should be robust to the variety of target size Of equal importance, it should detect the targets accurately and suppress false alarms under complex backgrounds. Unlike other threshold-based methods, the graph segmentation algorithm [30] based on multi-scale saliency maps is constructed to overcome the problem of ship scale change and accurately locate candidate regions. In the ship identification stage, the rotation-invariant HOG descriptor [31], using Fourier analysis in polar coordinates, is investigated to distinguish between the targets and the false alarms. The final Section concludes the paper by summarizing our findings
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
