基于图像局部纹理特征的SAR目标识别算法  

SAR Target Recognition Algorithm Based on Local Texture Features of Images

作　　者：宋斐[1] SONG Fei(College of Science, Ningxia Medical University,Yinchuan 750004, Chin)

机构地区：[1]宁夏医科大学理学院,宁夏银川750004

出　　处：《中国电子科学研究院学报》2018年第3期291-296,共6页Journal of China Academy of Electronics and Information Technology

基　　金：教育部春晖计划(Z2011053);多模医学图像快速配准的研究

摘　　要：传统基于Gabor滤波器的SAR目标识别方法根据图像全局特征进行目标识别,忽略图像局部纹理特征,容易受到噪声因素的干扰,获取的SAR目标识别结果精确度较低。因此,提出基于图像局部纹理特征的SAR目标识别算法,对SAR图像纹理特征进行提取,提取SAR图像纹理特征时,采用优化的TPLBP特征描述器提取图像局部纹理特征,获取TPLBP局部纹理特征向量;通过基于ELM分类器的SAR目标识别算法,对TPLBP局部纹理特征向量进行SAR目标分类与识别,获取理想的SAR目标识别结果。实验结果表明,所提方法在SAR目标识别方面具有准确率高、误判率低的优势。The traditional SAR target recognition method based on Gabor filter targets recognition based on the global features of the image,ignoring the local texture features of the image,and easily being interfered by noise factors. The SAR target recognition results obtained by this method are of low accuracy.Therefore,a SAR target recognition algorithm based on local texture features of images is proposed,when the texture features of SAR images are extracted and the texture features of SAR images are extracted,the local texture features of the images are extracted by the optimized TPLBP feature descriptor and the local texture feature vectors of the TPLBP are obtained; Through the SAR target recognition algorithm based on ELM classifier,the TPLBP local texture feature vector is classified and identified by SAR target,and the ideal result of SAR target recognition is obtained. Experimental results show that the proposed method has the advantage of high accuracy and low false positives in SAR target recognition.

关 键 词：SAR图像 局部纹理特征 TPLBP特征描述器 特征提取 ELM分类器 目标识别 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]