基于改进栈式核稀疏深度模型的大规模极化SAR地物分类算法
|
摘要
地物分类是极化SAR应用的一个重要分支。传统的地物分类方法需要提取特征,通过分类器进行分类。在栈式稀疏自编码模型的基础上,提出一种鲁棒的极化SAR地物分类算法。采用基于Morlet小波核的最小二乘支撑向量机代替深度模型中常用的Softmax分类器。通过与栈式稀疏自编码网络相结合,在一定程度上克服了传统极化SAR影像地物分类方法受相干斑噪声影响,且结果过于粗糙的缺点,保证了分类结果中非匀质区域的连贯性和匀质区域的一致性。真实极化SAR数据仿真实验结果表明,该算法可以有效地提高分类精度,降低相干斑噪声的对分类精度的影响。
Terrain classification is an important branch of polarization SAR(PolSAR) applications. Traditional terrain classification methods require the extraction of features and then classification by classifiers. On the basis of the stack sparse auto-encoding model, we proposed a robust PolSAR terrain classification algorithm. We used Morlet wavelet kernel-based least squares support vector machine to replace the commonly used Softmax classifier in the deep learning model. Combining with stack sparse AutoEncoder network, it overcame the shortcomings of traditional PolSAR image classification methods, which was greatly affected by speckle noise and the result was too rough. It guarantees the consistency of inhomogeneous regions in classification results and the coherence of homogeneous regions. The simulation results of real PolSAR data show that the algorithm can effectively improve the classification accuracy and reduce the influence of speckle noise on the classification accuracy.
Terrain classification is an important branch of polarization SAR(PolSAR) applications. Traditional terrain classification methods require the extraction of features and then classification by classifiers. On the basis of the stack sparse auto-encoding model, we proposed a robust PolSAR terrain classification algorithm. We used Morlet wavelet kernel-based least squares support vector machine to replace the commonly used Softmax classifier in the deep learning model. Combining with stack sparse AutoEncoder network, it overcame the shortcomings of traditional PolSAR image classification methods, which was greatly affected by speckle noise and the result was too rough. It guarantees the consistency of inhomogeneous regions in classification results and the coherence of homogeneous regions. The simulation results of real PolSAR data show that the algorithm can effectively improve the classification accuracy and reduce the influence of speckle noise on the classification accuracy.
引文
[1] 宋婉莹,李明,张鹏,等.基于加权合成核与三重Markov场的极化SAR图像分类方法[J].电子学报,2016,44(3):520-526.
[2] 张嘉峰,朱博,张鹏,等.Wishart分布情形下极化SAR图像目标CFAR检测解析方法[J].电子学报,2018,46(2):433-439.
[3] 张腊梅,段宝龙,邹斌.极化SAR图像目标分解方法的研究进展[J].电子与信息学报,2016,38(12):3289-3297.
[4] 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.
[5] 李旭.基于Wishart分类器和图谱聚类的极化SAR图像分类[J].山西师范大学学报(自然科学版),2015,29(1):35-39.
[6] Li C S,Wang W J,Wang P B,et al.Current situation and development trends of spaceborne SAR technology[J].Journal of Electronics & Information Technology,2016,38(1):229-240.
[7] Boerner W M.The development of multi-band equatorial orbiting POLSAR satellite sensors[C]//IEEE International Conference on Aerospace Electronics & Remotes Sensing Technology,Yogyakarta,Indonesia,2014:127-131.
[8] Ballester J D,Lopezj M.Applying the Freeman-Durden decomposition concept to polarimetric SAR interferometry[J].IEEE Transactions on Geoscience and Remote Sensing,2010,48(1):466-479.
[9] Minh N P,Zou B,Cai H,et al.Forest height estimation from mountain forest areas using general model-based decomposition for PolInSAR image[J].Journal of Applied Remote Sensing,2014,8(1):083676.
[10] Guo S,Tian Y,Li Y,et al.Unsupervised classification based on H/alpha decomposition and Wishart classifier for compact polarimetric SAR[C]//2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).IEEE,2015:1614-1617.
[11] Vincent P,Larochelle H,Lajoie I,et al.Stacked Denoising Autoencoders:Learning Useful Representations in a Deep Network with a Local Denoising Criterion[J].Journal of Machine Learning Research,2010,11(12):3371-3408.
[12] Sun M,Zhang D,Ren J,et al.Brushstroke based sparse hybrid convolutional neural networks for author classification of Chinese ink-wash paintings[C]//Proceedings of the 2015 IEEE International Conference on Image Processing (ICIP).IEEE,2015:626-630.
[13] Wang Y,Wang G,Lan Y.PolSAR image classification based on deep convolutional neural network[J].Metallurgical & Mining Industry,2015(8):366-371.
[14] 吕文涛,陈映鹰,张绍明.基于小波支撑向量机的图像降噪滤波算法[J].遥感信息,2008(3):22-25.
[15] Song H,Yang W,Bai Y,et al.Unsupervised classification of polarimetric SAR imagery using large-scale spectral clustering with spatial constraints[J].International Journal of Remote Sensing,2015,36(11):2816-2830.
[16] Jiao L,Bo L,Wang L.Fast Sparse Approximation for Least Squares Support Vector Machine[J].IEEE Transactions on Neural Networks,2007,18(3):685-697.
[17] Xu J,Yuan C Q,Cheng Y,et al.Active deep learning based polarimetric SAR image classification[J].Remote Sensing for Land & Resources,2018,30(1):72-77.
[18] Zhong H,Zhang J,Liu G,et al.Robust Polarimetric SAR Despeckling Based on Nonlocal Means and Distributed Lee Filter[J].IEEE Transactions on Geoscience and Remote Sensing,2014,52(7):4198-4210.
[2] 张嘉峰,朱博,张鹏,等.Wishart分布情形下极化SAR图像目标CFAR检测解析方法[J].电子学报,2018,46(2):433-439.
[3] 张腊梅,段宝龙,邹斌.极化SAR图像目标分解方法的研究进展[J].电子与信息学报,2016,38(12):3289-3297.
[4] 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.
[5] 李旭.基于Wishart分类器和图谱聚类的极化SAR图像分类[J].山西师范大学学报(自然科学版),2015,29(1):35-39.
[6] Li C S,Wang W J,Wang P B,et al.Current situation and development trends of spaceborne SAR technology[J].Journal of Electronics & Information Technology,2016,38(1):229-240.
[7] Boerner W M.The development of multi-band equatorial orbiting POLSAR satellite sensors[C]//IEEE International Conference on Aerospace Electronics & Remotes Sensing Technology,Yogyakarta,Indonesia,2014:127-131.
[8] Ballester J D,Lopezj M.Applying the Freeman-Durden decomposition concept to polarimetric SAR interferometry[J].IEEE Transactions on Geoscience and Remote Sensing,2010,48(1):466-479.
[9] Minh N P,Zou B,Cai H,et al.Forest height estimation from mountain forest areas using general model-based decomposition for PolInSAR image[J].Journal of Applied Remote Sensing,2014,8(1):083676.
[10] Guo S,Tian Y,Li Y,et al.Unsupervised classification based on H/alpha decomposition and Wishart classifier for compact polarimetric SAR[C]//2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).IEEE,2015:1614-1617.
[11] Vincent P,Larochelle H,Lajoie I,et al.Stacked Denoising Autoencoders:Learning Useful Representations in a Deep Network with a Local Denoising Criterion[J].Journal of Machine Learning Research,2010,11(12):3371-3408.
[12] Sun M,Zhang D,Ren J,et al.Brushstroke based sparse hybrid convolutional neural networks for author classification of Chinese ink-wash paintings[C]//Proceedings of the 2015 IEEE International Conference on Image Processing (ICIP).IEEE,2015:626-630.
[13] Wang Y,Wang G,Lan Y.PolSAR image classification based on deep convolutional neural network[J].Metallurgical & Mining Industry,2015(8):366-371.
[14] 吕文涛,陈映鹰,张绍明.基于小波支撑向量机的图像降噪滤波算法[J].遥感信息,2008(3):22-25.
[15] Song H,Yang W,Bai Y,et al.Unsupervised classification of polarimetric SAR imagery using large-scale spectral clustering with spatial constraints[J].International Journal of Remote Sensing,2015,36(11):2816-2830.
[16] Jiao L,Bo L,Wang L.Fast Sparse Approximation for Least Squares Support Vector Machine[J].IEEE Transactions on Neural Networks,2007,18(3):685-697.
[17] Xu J,Yuan C Q,Cheng Y,et al.Active deep learning based polarimetric SAR image classification[J].Remote Sensing for Land & Resources,2018,30(1):72-77.
[18] Zhong H,Zhang J,Liu G,et al.Robust Polarimetric SAR Despeckling Based on Nonlocal Means and Distributed Lee Filter[J].IEEE Transactions on Geoscience and Remote Sensing,2014,52(7):4198-4210.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |