摘要
为解决传统均值比(ratio of average,ROA)算子检测SAR(synthetic aperture radar,SAR)图像边缘时出现的受噪声影响大和边缘定位精度低等问题,结合平稳小波变换的优点,提出一种平稳小波去噪和改进ROA算法的SAR图像边缘检测方法。首先,利用平稳小波进行去噪预处理,减少相干斑噪声。然后,通过把传统ROA算子的4个检测方向增加为8个,以及利用非极值抑制进行边缘定位,在检测方向和定位精度两个方面改进ROA算法。实验结果表明,该方法的去噪性能和边缘检测效果较好。研究结论对传统ROA算法做了改进,使其更好地适用于SAR图像边缘检测。
In order to solve the problems of great impact of noise and low accuracy of edge location in detecting SAR edge image with the traditional ROA operator, this paper proposed a method of stationary wavelet denoising and SAR image edge detection by utilizing the advantages of stationary wavelet transformation.Firstly, denoising pretreatment was made by adopting stationary wavelet to reduce coherent speckle noise.Then, the ROA algorithm was improved in both testing direction and positioning accuracy by increasing traditional ROA operators from 4 to 8 and by using the non-extremum suppression to locate the edge.The experimental results show that the proposed method has good effect in denoising performance and edge detection and the improved ROA algorithm can be better applied to SAR edge image detection.
出处
《山东科技大学学报（自然科学版）》
CAS
2016年第6期17-23,共7页
Journal of Shandong University of Science and Technology(Natural Science)
基金
海洋公益性行业科研专项经费项目(201305034-1)
关键词
合成孔径雷达
边缘检测
平稳小波变换
均值比算法
相干斑噪声
synthetic aperture radar
edge detection
stationary wavelet transform
ratio of average （ROA）
coher-ent speckle noise
