摘要
为更好地获取舰船检测的图像信息,通过对雷达图像中舰船目标与相干斑噪声的分布特点进行分析,提出1种基于非下采样变换(Nonsubsampled Contourlet Transform,NSCT)与蚁群优化(Ant Colony Optim ization,ACO)算法相结合的合成孔径雷达(Synthetic Aperture Radar,SAR)图像舰船目标检测方法.该方法通过对SAR图像进行NSCT自适应阈值去噪,再应用ACO进行边缘检测,实现舰船目标的精确检测.仿真结果显示,该方法能够在复杂相干斑噪声背景中有效提取舰船目标的轮廓信息,很好地保持图像纹理和舰船结构,具有理想的抗干扰性能,保证检测结果的准确性.
In order to obtain better image information of ship detection, and analyze distribution characteristics of ship targets and speckle noise in radar image, a ship detection method of Synthetic Aperture Radar(SAR) image is proposed based on Nonsubsampled Contourlet Transform (NSCT) and Ant Colony Optimization(ACO) algorithm. According to the method, NSCT adaptive threshold method is applied to denoise SAR images, SAR images' edge is detected by ACO, and ship targets are detected precisely. Simulating results indicate that ship targets' contour information can be picked up efficiently in the complex speckle noise, images' texture and ship targets' structure can be kept well, there is an ideal antijamming competence, and the detection accuracy is ensured.
出处
《上海海事大学学报》
北大核心
2009年第4期17-21,共5页
Journal of Shanghai Maritime University
基金
上海市教育委员会科研创新项目(08YZ110)
上海海事大学研究生创新资助项目(YC2008031)
关键词
合成孔径雷达
非下采样变换
蚁群优化
边缘检测
舰船检测
synthetic aperture radar
nonsubsampled contourlet transform
ant colony optimization
edge detection
ship detection