摘要
根据SAR图像的基本特点 ,阐述了用DFBR场模型表达SAR图像的基本原理。在此基础上 ,通过提取SAR海洋图像中每个像素的分形值、分形模型拟合误差和方差统计量特征参数 ,结合图像像素的灰度值 ,形成一个对应于每个像素的特征矢量 ,并利用模糊数学的模糊子集概念 ,提出了一种基于特征矢量匹配的舰船目标检测方法。实际数据处理的结果表明 ,该方法具有较高的可靠性和准确性。
The principle of image Description with Discre te Fractional Brownian incremental Random field model(DFBR model) are expounded ac cording to the basic properties of SAR image firstly. Three parameters including fractional dimension value and fractional model fitting error and the variance statistics are estimated for each pixel in the SAR image. By combining these thr ee parameters with the image pixel grayscale value, a feature vector is formed f or each pixel. And then applying the fuzzy subset theory to the feature vectors, an algorithm to detect the ship targets in SAR ocean imagery based on feature v ector matching is developed. The data processing results show that the algorithm is reliable and can efficiently improve the accuracy of detection.
出处
《现代雷达》
CSCD
北大核心
2004年第8期25-29,共5页
Modern Radar
关键词
SAR图像
目标检测
分形特征
SAR imagery, target detection, fractional feat ure
