Abstract
Target detection is one of the important applications in the field of remote sensing. The Gaofen-3 (GF-3) Synthetic Aperture Radar (SAR) satellite launched by China is a powerful tool for maritime monitoring. This work aims at detecting ships in GF-3 SAR images using a new land masking strategy, the appropriate model for sea clutter and a neural network as the discrimination scheme. Firstly, the fully convolutional network (FCN) is applied to separate the sea from the land. Then, by analyzing the sea clutter distribution in GF-3 SAR images, we choose the probability distribution model of Constant False Alarm Rate (CFAR) detector from K-distribution, Gamma distribution and Rayleigh distribution based on a tradeoff between the sea clutter modeling accuracy and the computational complexity. Furthermore, in order to better implement CFAR detection, we also use truncated statistic (TS) as a preprocessing scheme and iterative censoring scheme (ICS) for boosting the performance of detector. Finally, we employ a neural network to re-examine the results as the discrimination stage. Experiment results on three GF-3 SAR images verify the effectiveness and efficiency of this approach.
Highlights
As one of the most effective means of Earth observation, synthetic aperture radar (SAR) has gained widespread attention recently
Proposed method for ship detection consists of selection four steps:ofthe land masking steps: the land convolutional network, the analysis and distribution with fully convolutional network, the analysis and selection of distribution model of clutter in model of sea clutter in GF-3 SAR images, the Constant False Alarm Rate (CFAR) detection using truncated statistic andsea iterative detection using truncated statistic and iterative censoring scheme, censoring scheme, and the discrimination stage based on deep convolutional neural network to and the discrimination stage based on deep network to remove false alarms
This paper proposes a method for ship detection in GF-3 SAR images based on sea clutter
Summary
As one of the most effective means of Earth observation, synthetic aperture radar (SAR) has gained widespread attention recently. Ship detection is an important one when it comes to marine surveillance. A mature ship detection system usually consists of three necessary steps: land masking, prescreening, and discrimination [1]. The land masking stage distinguishes sea from land and defines a scope to be detected for subsequent steps. In the process of prescreening, certain approaches are used to search for potential target pixels throughout the whole image, among which. CFAR is one of the most prevalent by virtue of its convenience and reliability, whose core is based on the sea clutter modeling to find bright pixels. It is worth noting that CFAR detector will inevitably introduce some false alarms, so the stage named discrimination is employed for false alarm elimination as a necessary supplement to the CFAR
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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
