基于局部纹理特征的合成孔径雷达变体目标自动识别算法

SAR variant target automatic recognition algorithm based on local texture characteristic

提出了一种针对变体的识别算法,利用变体与原目标局部纹理之间的相似性进行识别。首先,提出了一种基于清晰边缘的合成孔径雷达(Synthetic aperture radar,SAR)图像配准算法;然后使用结合伽柏(Gabor)变换,局部二值模式(Local binary pattern,LBP)和空间区域直方图的纹理特征来描述SAR图像;最后用基于大特征的直方图序列的匹配做识别。基于MSTAR S2的试验结果证明了本算法的有效性。

A Synthetic Aperture Radar（SAR） Automatic Target Recognition（ATR） algorithm for recognizing target variants is developed.This algorithm uses the local texture similarity between the variant and the original target for recognition.First,a SAR image registration algorithm based on clear edges is proposed.Then,the texture characteristic,which is obtained by combining the Gabor transform,LBP and spatial domain histogram,is employed to describe the SAR image.Finally,histogram sequence matching based on the large characteristic is used to perform recognition.The effectiveness of the proposed algorithm is verified by experimental results on MSTAR S2.