空-谱融合的条件随机场高光谱影像分类方法

doi:10.11947/j.AGCS.2020.20190042

测绘学报 ›› 2020, Vol. 49 ›› Issue (3): 343-354.doi: 10.11947/j.AGCS.2020.20190042

空-谱融合的条件随机场高光谱影像分类方法

魏立飞¹, 余铭¹, 钟燕飞², 袁自然¹, 黄灿¹

1. 湖北大学资源环境学院, 湖北武汉 430062;
2. 武汉大学测绘遥感信息工程国家重点实验室, 湖北武汉 430079

收稿日期:2019-01-25 修回日期:2019-10-24 发布日期:2020-03-24
通讯作者: 余铭 E-mail:ym17317399819@163.com
作者简介:魏立飞(1983-),男,博士,副教授。研究方向为遥感图像处理、农业遥感、生态遥感。E-mail:weilifei2508@163.com
基金资助:
国家重点研发计划课题（2017YFB0504202）；国家自然科学基金优秀青年科学基金（41622107）；湖北省技术创新专项重大项目（2018ABA078）；空间数据挖掘与信息共享教育部重点实验室开放基金（2018LSDMIS05）；农业部农业遥感重点实验室开放基金（20170007）

Hyperspectral image classification method based on space-spectral fusion conditional random field

WEI Lifei¹, YU Ming¹, ZHONG Yanfei², YUAN Ziran¹, HUANG Can¹

1. Faculty of Resources and Environmental Science, Hubei University, Wuhan 430062, China;
2. National Laboratory for Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, China

Received:2019-01-25 Revised:2019-10-24 Published:2020-03-24
Supported by:
The National Key Research and Development Program of China (No. 2017YFB0504202);The National Natural Science Foundation of China (No. 41622107);The Special Projects for Technological Innovation in Hubei (No. 2018ABA078);The Open Fund of Key Laboratory of Ministry of Education for Spatial Data Mining and Information Sharing (No. 2018LSDMIS05);The Open Fund of Key Laboratory of Agricultural Remote Sensing of the Ministry of Agriculture (No. 20170007)

摘要/Abstract

摘要： 高光谱遥感数据具有光谱信息丰富、图谱合一的特点，目前已经广泛地应用在对地观测中。传统的高光谱分类模型大多过分依赖影像光谱信息，没有充分利用空间特征信息，这使得分类精度还有很大的提升空间。条件随机场是一种概率模型，能够较好地融合空间上下文信息，在高光谱影像分类中已经得到越来越多的关注，但大部分条件随机场模型存在超平滑的现象，会导致影像细节丢失。针对该问题，本文提出了一种优化融合影像空-谱信息的高分辨率/高光谱影像分类方法，该方法将影像的纹理信息与原始光谱信息进行融合，利用SVM分类器对其进行预分类，并将各类概率定义为一元势函数，以融合空间特征信息；然后将空间平滑项和局部类别标签成本项加入二元势函数中，以考虑空间背景信息，并保留各类别中的详细信息。最后，通过两组的高分辨率/高光谱影像数据进行试验。结果表明，与SVM算法、传统的条件随机场方法和面向对象的分类方法相比，本文提出的算法在整体分类精度上分别提高了10%、9%和8%以上，同时在保持地物边缘完整性、避免"同谱异物"与"同物异谱"的现象方面有较明显的优势。

关键词: 高光谱遥感影像, 条件随机场, 空-谱融合, 影像分类

Abstract: Hyperspectral remote sensing image has the characteristics of rich spectral information and combining image with spectrum, which has been widely applied in the earth observation. Most of traditional hyperspectral image classification models don't make fully use of spatial feature information, rely too much on the spectral imformation, making the classification accuracy still have a lot of room to improve. Conditional random field (CRF) is a kind of probability mode that can better integrate spatial context information. It plays a more and more important role in hyperspectral image classification. However, most CRF models have the problem of excess smoothness, which will result in the loss of detail information. Aiming at this problem, this paper proposed a hyperspectral image classification method based on space-spectral fusion conditional random field. The proposed method designs suitable potential functions in a pairwise conditional random field model, fusing the spectral and spatial features to consider the spatial feature information and retain the details in each class. The experiments on two sets of hyperspectral image showed that, compared with the traditional methods, the proposed classification method can effectively improve the classification accuracy, protect the edges and shapes of the features, and relieve excessive smoothing, while retaining detailed information.

Key words: hyperspectral remote sensing imagery, conditional random field, space-spectral fusion, image classification

中图分类号:

P237

魏立飞, 余铭, 钟燕飞, 袁自然, 黄灿. 空-谱融合的条件随机场高光谱影像分类方法[J]. 测绘学报, 2020, 49(3): 343-354.

WEI Lifei, YU Ming, ZHONG Yanfei, YUAN Ziran, HUANG Can. Hyperspectral image classification method based on space-spectral fusion conditional random field[J]. Acta Geodaetica et Cartographica Sinica, 2020, 49(3): 343-354.

参考文献

[1] 李二珠. 半监督支持向量机高光谱遥感影像分类[D]. 徐州:中国矿业大学, 2014. LI Erzhu. Hyperspectral remote sensing image classification based on semi-supervised support vector machine[D]. Xuzhou:China University of Mining and Technology, 2014.
[2] 田彦平, 陶超, 邹峥嵘, 等. 主动学习与图的半监督相结合的高光谱影像分类[J]. 测绘学报, 2015, 44(8):919-926. DOI:10.11947/j.AGCS.2015.20140221. TIAN Yanping, TAO Chao, ZOU Zhengrong, et al. Semi-supervised graph-based hyperspectral image classification with active learning[J]. Acta Geodaetica et Cartographica Sinica, 2015, 44(8):919-926. DOI:10.11947/j.AGCS.2015.20140221.
[3] 张春森, 郑艺惟, 黄小兵, 等. 高光谱影像光谱-空间多特征加权概率融合分类[J]. 测绘学报, 2015, 44(8):909-918. DOI:10.11947/j.AGCS.2015.20140544. ZHANG Chunsen, ZHENG Yiwei, HUANG Xiaobing, et al. Hyperspectral image classification based on the weighted probabilistic fusion of multiple spectral-spatial features[J]. Acta Geodaetica et Cartographica Sinica, 2015, 44(8):909-918. DOI:10.11947/j.AGCS.2015.20140544.
[4] HAN Xiaobing, ZHONG Yanfei, ZHANG Liangpei. Spatial-spectral unsupervised convolutional sparse auto-encoder classifier for hyperspectral imagery[J]. Photogrammetric Engineering & Remote Sensing, 2017, 83(3):195-206.
[5] 谷雨, 徐英, 郭宝峰. 融合空谱特征和集成超限学习机的高光谱图像分类[J]. 测绘学报, 2018, 47(9):1238-1249. DOI:10.11947/j.AGCS.2018.20170476. GU Yu, XU Ying, GUO Baofeng. Hyperspectral image classification by combination of spatial-spectral features and ensemble extreme learning machines[J]. Acta Geodaetica et Cartographica Sinica, 2018, 47(9):1238-1249. DOI:10.11947/j.AGCS.2018.20170476.
[6] ZHONG Yanfei, WANG Xinyu, XU Yao, et al. MINI-UAV borne hyperspectral remote sensing:a review[C]//Proceedings of 2017 IEEE International Geoscience and Remote Sensing Symposium. Fort Worth, TX:IEEE, 2017:5908-5911.
[7] 李竺强, 朱瑞飞, 高放, 等. 三维卷积神经网络模型联合条件随机场优化的高光谱遥感影像分类[J]. 光学学报, 2018, 38(8):0828001. LI Zhuqiang, ZHU Ruifei, GAO Fang, et al. Hyperspectral remote sensing image classification based on three-dimensional convolution neural network combined with conditional random field optimization[J]. Acta Optica Sinica, 2018, 38(8):0828001.
[8] 刘大伟, 韩玲, 韩晓勇. 基于深度学习的高分辨率遥感影像分类研究[J]. 光学学报, 2016, 36(4):0428001. LIU Dawei, HAN Ling, HAN Xiaoyong. High spatial resolution remote sensing image classification based on deep learning[J]. Acta Optica Sinica, 2016, 36(4):0428001.
[9] 敖平平, 孟凡纪. 基于主动学习和空间约束的高光谱影像分类[J]. 测绘与空间地理信息, 2018, 41(8):178-182. AO Pingping, MENG Fanji. Hyperspectral image classification based on active learning and contextual[J]. Geomatics & Spatial Information Technology, 2018, 41(8):178-182.
[10] 樊利恒, 吕俊伟, 邓江生. 基于分类器集成的高光谱遥感图像分类方法[J]. 光学学报, 2014, 34(9):0910002. FAN Liheng, LÜ Junwei, DENG Jiangsheng. Classification of hyperspectral remote sensing images based on bands grouping and classification ensembles[J]. Acta Optica Sinica, 2014, 34(9):0910002.
[11] 董安国, 李佳逊, 张蓓, 等. 基于谱聚类和稀疏表示的高光谱图像分类算法[J]. 光学学报, 2017, 37(8):0828005. DONG Anguo, LI Jiaxun, ZHANG Bei, et al. Hyperspectral image classification algorithm based on spectral clustering and sparse representation[J]. Acta Optica Sinica, 2017, 37(8):0828005.
[12] CHEN Yi, NASRABADI N M, TRAN T D. Hyperspectral image classification using dictionary-based sparse representation[J]. IEEE Transactions on Geoscience and Remote Sensing, 2011, 49(10):3973-3985.
[13] FANG Leyuan, LI Shutao, KANG Xudong, et al. Spectral-spatial classification of hyperspectral images with a superpixel-based discriminative sparse model[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(8):4186-4201.
[14] CAMPS-VALLS G, BRUZZONE L. Kernel-based methods for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(6):1351-1362.
[15] 余健. 联合纹理和光谱特征的高光谱图像分类方法[J]. 韩山师范学院学报, 2017, 38(6):18-26. YU Jian. Texture-spectrum joint classification method for hyperspectral remote sensing image[J]. Journal of Hanshan Normal University, 2017, 38(6):18-26.
[16] BLASCHKE T. Object based image analysis for remote sensing[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2010, 65(1):2-16.
[17] LÜ Pengyuan, ZHONG Yanfei, ZHAO Ji, et al. Unsupervised change detection based on hybrid conditional random field model for high spatial resolution remote sensing imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2018, 56(7):4002-4015.
[18] 王少宇, 焦洪赞, 钟燕飞. 条件随机场模型约束下的遥感影像模糊C-均值聚类算法[J]. 测绘学报, 2016, 45(12):1441-1447. DOI:10.11947/j.AGCS.2016.20150621. WANG Shaoyu, JIAO Hongzan, ZHONG Yanfei. A modified FCM classifier constrained by conditional random field model for remote sensing imagery[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(12):1441-1447. DOI:10.11947/j.AGCS.2016.20150621.
[19] ZHAO Ji, ZHONG Yanfei, SHU Hong, et al. High-resolution image classification integrating spectral-spatial-location cues by conditional random fields[J]. IEEE Transactions on Image Processing, 2016, 25(9):4033-4045.
[20] ZHONG Ping, WANG Runsheng. Learning conditional random fields for classification of hyperspectral images[J]. IEEE Transactions on Image Processing, 2010, 19(7):1890-1907.
[21] ZHANG Liangpei, HUANG Xin. Object-oriented subspace analysis for airborne hyperspectral remote sensing imagery[J]. Neurocomputing, 2010, 73(4-6):927-936.
[22] BEGUET B, CHEHATA N, BOUKIR S, et al. Classification of forest structure using very high resolution Pleiades image texture[C]//Proceedings of 2014 IEEE Geoscience and Remote Sensing Symposium. Quebec City, QC:IEEE, 2014:2324-2327.
[23] ZHANG Liangpei, HUANG Xin. Object-oriented subspace analysis for airborne Hyperspectral remote sensing imagery[J]. Neurocomputing, 2010, 73(4-6):927-936.
[24] 胡荣明, 黄小兵, 黄远程. 增强形态学建筑物指数应用于高分辨率遥感影像中建筑物提取[J]. 测绘学报, 2014, 43(5):514-520. DOI:10.13485/j.cnki.11-2089.2014.0084. HU Rongming, HUANG Xiaobing, HUANG Yuancheng. An enhanced morphological building index for building extraction from high-resolution images[J]. Acta Geodaetica et Cartographica Sinica, 2014, 43(5):514-520. DOI:10.13485/j.cnki.11-2089.2014.0084.
[25] 付乾坤, 吴波, 汪小钦, 等. 基于形态学建筑物指数的城市建筑物提取及其高度估算[J]. 遥感技术与应用, 2015, 30(1):148-154. FU Qiankun, WU Bo, WANG Xiaoqin, et al. Building extraction and its height estimation over urban areas based on morphological building index[J]. Remote Sensing Technology and Application, 2015, 30(1):148-154.
[26] JACKSON Q, LANDGREBE D A. Adaptive bayesian contextual classification based on Markov random fields[J]. IEEE Transactions on Geoscience and Remote Sensing, 2002, 40(11):2454-2463.
[27] ZHAO Ji, ZHONG Yanfei, ZHANG Liangpei. Detail-preserving smoothing classifier based on conditional random fields for high spatial resolution remote sensing imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(5):2440-2452.
[28] RICHARDS J A, JIA Xiuping. Remote sensing digital image analysis:an introduction[M]. 4th ed. Berlin:Springer, 2006.

空-谱融合的条件随机场高光谱影像分类方法

Hyperspectral image classification method based on space-spectral fusion conditional random field

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	左溪冰, 刘冰, 余旭初, 张鹏强, 高奎亮, 朱恩泽. 高光谱影像小样本分类的图卷积网络方法[J]. 测绘学报, 2021, 50(10): 1358-1369.
[2]	刘冰, 左溪冰, 谭熊, 余岸竹, 郭文月. 高光谱影像分类的深度少样例学习方法[J]. 测绘学报, 2020, 49(10): 1331-1342.
[3]	赵泉华, 谢凯浪, 王光辉, 李玉. 全卷积网络和条件随机场相结合的全极化SAR土地覆盖分类[J]. 测绘学报, 2020, 49(1): 65-78.
[4]	余岸竹, 刘冰, 邢志鹏, 杨帆, 杨其淼. 面向高光谱影像分类的显著性特征提取方法[J]. 测绘学报, 2019, 48(8): 985-995.
[5]	黄鸿, 石光耀, 段宇乐, 张丽梅. 加权空-谱联合保持嵌入的高光谱遥感影像降维方法[J]. 测绘学报, 2019, 48(8): 1014-1024.
[6]	左宗成, 张文, 张东映. 融合可变形卷积与条件随机场的遥感影像语义分割方法[J]. 测绘学报, 2019, 48(6): 718-726.
[7]	刘冰, 余旭初, 张鹏强, 谭熊. 联合空-谱信息的高光谱影像深度三维卷积网络分类[J]. 测绘学报, 2019, 48(1): 53-63.
[8]	何鄂龙, 王红平, 陈奇, 刘修国. 一种改进的空间上下文点云分类方法[J]. 测绘学报, 2017, 46(3): 362-370.
[9]	兰泽英, 刘洋. 领域知识辅助下基于多尺度与主方向纹理的遥感影像土地利用分类[J]. 测绘学报, 2016, 45(8): 973-982.
[10]	王少宇, 焦洪赞, 钟燕飞. 条件随机场模型约束下的遥感影像模糊C-均值聚类算法[J]. 测绘学报, 2016, 45(12): 1441-1447.
[11]	王俊淑, 江南, 张国明, 李杨, 吕恒. 融合光谱-空间信息的高光谱遥感影像增量分类算法[J]. 测绘学报, 2015, 44(9): 1003-1013.
[12]	田彦平, 陶超, 邹峥嵘, 杨钊霞, 何小飞. 主动学习与图的半监督相结合的高光谱影像分类[J]. 测绘学报, 2015, 44(8): 919-926.
[13]	石茜,杜博,张良培. 一种基于局部判别正切空间排列的高光谱遥感影像降维方法[J]. 测绘学报, 2012, 41(3): 0-416.
[14]	杨红磊,彭军还. 基于马尔科夫随机场与模糊C均值的遥感影像分类[J]. 测绘学报, 2012, 41(2): 213-218.
[15]	陶超,谭毅华,彭碧发,田金文. 一种基于概率潜在语义模型的高分辨率遥感影像分类方法[J]. 测绘学报, 2011, 40(2): 156-162.