空军工程大学学报（自然科学版）2017，Vol.18Issue(2)：49-53,5.DOI:10.3969/j.issn.1009-3516.2017.02.009
仿视皮层V1特性的SAR图像边缘检测
A Method of SAR Image Edge Detection by Modeling the Properties of V1 in Cortex
摘要
Abstract
Synthetic Aperture Radar(SAR)image edge detection is a hot issue in the field of image processing.It is difficult to eliminate the noise and internal texture of SAR image by traditional methods.By making use of the properties of the receptive field and the surround suppression mechanism to texture and noise in the homogeneous region in cells with non-classical receptive field in Primary Visual Cortex (V1) in human visual system,this paper proposes a new method of SAR image edge detection by modeling the information processing mechanism of V1 Through the operations of odd and even decomposition,spatial summation and normalization by Gabor filter,the paper simulates the properties of the classical receptive field.A series of computational steps motivated by surround suppression mechanism are added to suppress the texture and noise in homogeneous region and preserve the contour wanted.The simulation results show that the proposed method can suppress most texture in the homogeneous region of the four kinds of SAR images used in the article,can weaken speckle noise to some degree,and edge detection effect is better than that of the canny detector.The method can be applied to the edge detection of SAR images.关键词
SAR图像/边缘检测/初级视皮层/感受野/Gabor滤波器/环绕抑制Key words
SAR image/edge detection/Primary Visual Cortex/receptive field/Gabor filter/surround suppression分类
信息技术与安全科学引用本文复制引用
辛鹏,许悦雷,马时平,邹洪中,张文达,李帅,吕超..仿视皮层V1特性的SAR图像边缘检测[J].空军工程大学学报（自然科学版）,2017,18(2):49-53,5.基金项目
国家自然科学基金(61472441) （61472441）
