基于统计和纹理特征的SAS图像SVM分割研究_陈强.pdfVIP

第 34 卷 第 6 期 2013 年 6 月 仪 器 仪 表 学 报 Chinese Journal of Scientific Instrument Vol. 34 No. 6 Jun． 2013 收稿日期: 2012-12 Received Date: 2012-12 * 基金项目: 国家自然科学基金( 资助项目 基于统计和纹理特征的 SAS 图像 SVM 分割研究* 陈 强 ， 田 杰 ， 黄海宁 ， 张春华 ( 中国科学院声学所 北京 100190) 摘 要: 合成孔径声呐图像可以有效反映海底的地形、地貌和底质等情况，但是单幅 SAS 图像通常对应一片较大的区域，需要 按照某种性质将不同性质的区域分割开来，以有利于下一步的图像分析以及目标检测和识别。研究发现，不同底质区域的 SAS 图像具有不同的统计和纹理特征，选取灰度直方图的均值、标准差、峰度等统计特性和灰度共生矩阵的能量、相关性、对比度、熵 值等纹理特性用以描述 SAS 图像的不同区域。将选取的特征作为 SVM 的训练特征，进而得到 SVM 分类器，用于 SAS 图像分 割。实验结果表明，SVM 算法可以很好地对 SAS 图像进行区域分割。 关键词: 合成孔径声呐; 统计特征; 纹理特征; 图像分割 中图分类号: TP751． 1 文献标识码: A 国家标准学科分类代码: 510． 4050 Study on SAS image segmentation using SVM based on statistical and texture features Chen Qiang，Tian Jie，Huang Haining，Zhang Chunhua ( Institute of Acoustics，Chinese Academy of Sciences，Beijing 100190，China) Abstract: Synthetic aperture sonar ( SAS) images can effectively describe the topography，geomorphology and substrate of seabed; however，one single SAS image usually corresponds to a larger area; so it is necessary to segment the SAS image in- to different regions according to certain property，which benefits further analyzing the image，and detecting and identifying the target． Study found that SAS images of different substrates have different statistical and texture features; in this paper， the statistical properties，such as the mean，standard deviation and kurtosis of the grey level histogram，as well as the tex- ture features，such as the energy，correlation，contrast and entropy of the grey level co-occurrence matrix are selected and used to describe different regions of the SAS image． These selected features are used as the support vector machine ( SVM) training characteristics and the classifier is obtained for the SAS image segmentation． The experiment results show that the proposed SVM algorithm is a good segmentation method for the region segmentation of SAS image． Keywords: synthetic aperture sonar ( SAS) ; statistical feature; texture feature; image segmentation 1 引 言 合成孔径