Building extraction from polarimetric SAR image based on feature selection and two-level classification

doi:10.7523/j.issn.2095-6134.2018.01.012

Abstract

Abstract: In this work, a building extraction method, based on feature selection and two-level classification, for polarimetric SAR (POLSAR) image is proposed. Firstly, some polarimetric features and texture features are extracted from POLSAR data via refined Lee filter as the initial feature set. Then, by using the random forest as the primary classifier and evaluating features' importance meanwhile, the feature subset is obtained according to the importance rank. Secondary classification is made for the selected feature subset by the support vector machine, and the final result is achieved by combining the primary results and secondary results together using neighborhood voting. The experimental results on AIRSAR system demonstrate that the proposed method effectively improves the extraction accuracy.

Key words: polarimetric SAR image, feature selection, multiple classifiers

CLC Number:

TP753

ZHANG Miaoran, LIU Chang. Building extraction from polarimetric SAR image based on feature selection and two-level classification[J]. , 2018, 35(1): 89-95.

References

[1] Lee J S, Pottier E. Polarimetric radar imaging:from basics to applications[M]. Boca Raton:CRC press, 2009:1-3.
[2] Matsuoka M, Yamazaki F. Use of satellite SAR intensity imagery for detecting building areas damaged due to earthquakes[J]. Earthquake Spectra, 2004, 20(3):975-994.
[3] Henderson F M, Xia Z G. SAR applications in human settlement detection, population estimation and urban land use pattern analysis:a status report[J]. IEEE transactions on geoscience and remote sensing, 1997, 35(1):79-85.
[4] Xiang D, Tang T, Ban Y, et al. Unsupervised polarimetric SAR urban area classification based on model-based decomposition with cross scattering[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2016, 116(6):86-100.
[5] Zhai W, Shen H, Huang C, et al. Fusion of polarimetric and texture information for urban building extraction from fully polarimetric SAR imagery[J]. Remote Sensing Letters, 2016, 7(1):31-40.
[6] 车美琴, 阿里木, 赛买提, 等. 利用旋转不变特征提取全极化SAR影像人工地物[J]. 遥感学报, 2016, 20(2):303-314.
[7] 陈博, 王爽, 焦李成, 等. 利用0-1矩阵分解集成的极化SAR图像分类[J]. 电子与信息学报, 2015, 37(6):1495-1501.
[8] Baghdadi N, Cresson R, El Hajj M, et al. Estimation of soil parameters over bare agriculture areas from C-band polarimetric SAR data using neural networks[J]. Hydrology and Earth System Sciences, 2012, 9(3):2897-2933.
[9] Salehi M, Sahebi M R, Maghsoudi Y. Improving the accuracy of urban land cover classification using Radarsat-2 PolSAR data[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(4):1394-1401.
[10] Du P, Samat A, Waske B, et al. Random forest and rotation forest for fully polarized SAR image classification using polarimetric and spatial features[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2015, 105(7):38-53.
[11] 孙勋, 黄平平, 涂尚坦, 等. 利用多特征融合和集成学习的极化SAR图像分类[J]. 雷达学报, 2016, 5(6):692-700.
[12] Xia Z G, Henderson F M. Understanding the relationships between radar response patterns and the bio-and geophysical parameters of urban areas[J]. IEEE Transactions on Geoscience and Remote Sensing, 1997, 35(1):93-101.
[13] 赵凌君, 秦玉亮, 高贵, 等. 利用GLCM纹理分析的高分辨率SAR图像建筑区检测[J]. 遥感学报, 2009, 13(3):475-490.
[14] Cloude S R, Pottier E. An entropy based classification scheme for land applications of polarimetric SAR[J]. IEEE Transactions on Geoscience and Remote Sensing, 1997, 35(1):68-78.
[15] Yamaguchi Y, Moriyama T, Ishido M, et al. Four-component scattering model for polarimetric SAR image decomposition[J]. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(8):1699-1706.
[16] Haralick R M, Shanmugam K, Dinstein I. Textural features for image classification[J]. Systems Man & Cybernetics IEEE Transactions on, 1973, smc-3(6):610-621.
[17] Zhao L, Zhou X, Kuang G. Building detection from urban SAR image using building characteristics and contextual information[J]. EURASIP Journal on Advances in Signal Processing, 2013(1):1-16.
[18] 孙萍. 极化SAR图像建筑物提取方法研究[D]. 北京:首都师范大学, 2013.
[19] Dekker R J. Texture analysis and classification of ERS SAR images for map updating of urban areas in the Netherlands[J]. IEEE Transactions on Geoscience and Remote Sensing, 2003, 41(9):1950-1958.
[20] Genuer R, Poggi J M, Tuleau-Malot C. Variable selection using random forests[J]. Pattern Recognition Letters, 2010, 31(14):2225-2236.
[21] 陈刚, 王宏琦, 孙显. 基于核函数原型和自适应遗传算法的SVM模型选择方法[J]. 中国科学院大学学报, 2012, 29(1):62-69.
[22] Díaz-Uriarte R, De Andres S A. Gene selection and classification of microarray data using random forest[J]. BMC bioinformatics, 2006, 7(1):1-13.
[23] National Aeronautics and Space Administration (NASA). Airborne synthetic aperture radar (AIRSAR) system. Tokyo, Japan mission dataset[DB/OL].(2000-10-02)[2016-12-30].https://www.asf.alaska.edu/.