Welcome to Journal of University of Chinese Academy of Sciences,Today is

›› 2010, Vol. 27 ›› Issue (1): 36-42.DOI: 10.7523/j.issn.2095-6134.2010.1.005

• Research Articles • Previous Articles     Next Articles

A new ship detection algorithm based on local K-distribution in SAR images

AI Jia-Qiu1,2, QI Xiang-Yang1   

  1. 1. Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China;
    2. Graduate University, Chinese Academy of Sciences, Beijing 100049, China
  • Received:2009-06-24 Revised:2009-09-14 Online:2010-01-15

Abstract:

The new ship detection algorithm based on local K-distribution proposed in this paper uses a target window and a background window, removes the leaked ship pixels in the background window by using special methods, and estimates the remaining pixels in the background window to get the local clutter’s gray mean and variance. Then the local clutter’s gray probability distribution can be modeled to realize CFAR detection. Compared with K-distribution CFAR detector and local K-distribution CFAR detector, the new algorithm can fit the local change of the sea clutter and detect the targets which are too near. The simulation results show the effectiveness of the new algorithm.

Key words: ship detection, K-distribution, local window, clutter extraction in the background window

CLC Number: