基于Gabor滤波器和局部纹理特征提取的SAR目标识别算法

王璐; 张帆; 李伟; 谢晓明; 胡伟

doi:10.12000/JR15076

基于Gabor滤波器和局部纹理特征提取的SAR目标识别算法

DOI: 10.12000/JR15076

(北京化工大学信息科学与技术学院北京 100029)

基金项目:

国家自然科学基金(61302164),中央高校基本科研业务费专项资金(YS1404),北京高等学校青年英才计划(YETP0500)

详细信息

作者简介:
张帆(1981-),男,2008年毕业于中科院电子所信号与信息处理专业,获得工学博士学位,同年在中科院电子所从事博士后研究,现为北京化工大学副教授,硕士生导师。主要研究方向为SAR系统模拟、高性能计算、科学可视化等。E-mail:zhangf@mail.buct.edu.cn李伟(1985-),男,2012年毕业于美国密西西比州立大学电气与计算机工程专业,获得工学博士学位,同年至美国加州大学戴维斯分校从事博士后研究,现为北京化工大学教授,博士生导师。主要研究方向为模式识别、高光谱图像分析与应用、数据压缩等。E-mail:liw@mail.buct.edu.cn谢晓明(1969-),男,1998年毕业于华中理工大学信息与通信工程专业,获得工学博士学位,现为北京化工大学副教授,硕士生导师。主要研究方向为移动通信技术、宽带通信技术等。E-mail:xmxie@mail.buct.edu.cn胡伟(1979-),男,2006年毕业于清华大学计算机系,获得工学博士学位,现为北京化工大学副教授,硕士生导师。主要研究方向为真实感图形实时绘制、图像与视频处理等。E-mail:huwei@mail.buct.edu.cn

通讯作者:
张帆zhangf@mail.buct.edu.cn

计量
- 文章访问数: 3301
- HTML全文浏览量: 888
- PDF下载量: 1953
- 被引次数: 0
出版历程
- 收稿日期: 2015-06-17
- 修回日期: 2015-10-16
- 网络出版日期: 2015-12-28

A Method of SAR Target Recognition Based on Gabor Filter and Local Texture Feature Extraction

(College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China)

Funds:

The National Natural Science Foundation of China (61302164), The Fundamental Research Funds for the Central Universities (YS1404), The Beijing Higher Education Young Elite Teacher Project (YETP0500)

摘要

摘要: 该文提出了一种基于Gabor滤波器和Three-Patch Local Binary Patterns(TPLBP)局部纹理特征提取的合成孔径雷达(Synthetic Aperture Rader, SAR)图像目标识别算法。首先, 利用Gabor滤波器对SAR图像在不同方向上进行滤波, 增强SAR图像中目标及其阴影的关键特征；然后, 利用TPLBP算法对Gabor滤波之后的图像进行局部纹理特征提取, 该算法克服了Local Binary Patterns(LBP)算法无法描述大范围领域纹理特征的缺陷, 并且保持了LBP旋转不变的特性, 减少了SAR图像目标方位变化对识别效果的影响；最后利用极限学习机(Extreme Learning Machine, ELM)分类器实现目标识别。该文通过MSTAR数据库中的3类SAR目标识别实验验证了该算法的有效性。
- SAR目标识别 /
- Gabor滤波器 /
- Three-Patch LBP /
- 特征提取
Abstract: This paper presents a novel texture feature extraction method based on a Gabor filter and Three-Patch Local Binary Patterns (TPLBP) for Synthetic Aperture Rader (SAR) target recognition. First, SAR images are processed by a Gabor filter in different directions to enhance the significant features of the targets and their shadows. Then, the effective local texture features based on the Gabor filtered images are extracted by TPLBP. This not only overcomes the shortcoming of Local Binary Patterns (LBP), which cannot describe texture features for large scale neighborhoods, but also maintains the rotation invariant characteristic which alleviates the impact of the direction variations of SAR targets on recognition performance. Finally, we use an Extreme Learning Machine (ELM) classifier and extract the texture features. The experimental results of MSTAR database demonstrate the effectiveness of the proposed method.
- SAR target recognition /
- Gabor filter /
- Three-Patch Local Binary Patterns (TPLBP) /
- Feature extraction

HTML全文

参考文献(15)

[1]	丁军, 刘宏伟, 王英华. 基于非负稀疏表示的SAR图像目标识别方法[J]. 电子与信息学报, 2014, 36(9): 2194-2200. Ding Jun, Liu Hong-wei, and Wang Ying-hua. SAR image target recognition based on non-negative sparse representation[J]. Journal of Electronics Information Technology, 2014, 36(9): 2194-2200.
[2]	梁胜杰, 张志华, 崔立林, 等. 基于主成分分析与核独立成分分析的降维方法[J]. 系统工程与电子技术, 2011, 33(9): 2144-2148. Liang Sheng-jie, Zhang Zhi-hua, Cui Li-lin, et al.. A reduced dimension method based on PCA and KICA[J]. Systems Engineering and Electronics, 2011, 33(9): 2144-2148.
[3]	周家锐, 纪震, 沈琳琳, 等. 基于Gabor小波与Memetic算法的人脸识别方法[J]. 电子学报, 2012, 40(4): 642-646. Zhou Jia-rui, Ji Zhen, Shen Lin-lin, et al.. Face recognition using Gabor wavelets and Memetic algorithm[J]. Acta Electronica Sinica, 2012, 40(4): 642-646.
[4]	高涛, 何明一, 戴玉超, 等. 多级LBP直方图序列特征的人脸识别[J]. 中国图象图形学报, 2009, 14(2): 202-207. Gao Tao, He Ming-yi, Dai Yu-chao, et al.. Face recognition using multi-level histogram sequence local binary pattern[J]. Journal of Image and Graphics, 2009, 14(2): 202-207.
[5]	Patel V, Nasraba N, and Chellappa R. Sparsity-motivated automatic target recognition[J]. Applied Optics, 2011, 50(10): 1425-1433.
[6]	Novak L M, Owirka G J, and Brower W S. Performance of 10-and 20-target MSE classification[J]. IEEE Transactions on Aerospace and Electronic Systems, 2000, 36(4): 1279-1289.
[7]	Li W and Du Q. Gabor-Filtering based nearest regularized subspace for hyperspectral image classification[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(4): 1012-1022.
[8]	Ojala T, Pietikainen M, and Maenpaa T. Multiresolution gray-scale and rotation invariant texture classification with local binary pattern[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(7): 971-987.
[9]	Lior W, Tal H, and Yaniv T. Descriptor based methods in the wild[C]. Real-Life Images Workshop at the European Conference on Computer Vision (ECCV), Marseille, France, 2008: 1-14.
[10]	张文博, 姬红兵. 融合极限学习机[J]. 电子与信息学报, 2013, 35(11): 2728-2732.
[11]	Zhang Wen-bo and Ji Hong-bing. Fusion of extreme learning machines[J]. Journal of Electronics Information Technology, 2013, 35(11): 2728-2732.
[12]	Zhou Y, Peng J, and Chen C. Extreme learning machine with composite kernels for hyperspectral image classification[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 8(6): 2351-2360.
[13]	Moving and Stationary Target Acquisition and Recognition (MSTAR) Public Dataset[OL]. https://www.sdms.afrl. af.mil/datasets/mstar/.
[14]	尹奎英, 金林, 李成, 等. 融合目标轮廓和阴影轮廓的SAR 图像目标识别[J]. 空军工程大学学报(自然科学版), 2011, 12(1): 24-28.
[15]	Yin Kui-ying, Jin Lin, Li Cheng, et al.. An SAR ATR based on fusion of target contour and shadow contour[J]. Journal of Air Force Engineer University (Natural Science Edition), 2011, 12(1): 24-28.