A mutual information-Dempster-Shafer based decision ensemble system for land cover classification of hyperspectral data

doi:10.1007/s11707-016-0611-2

Front. Earth Sci.

2017, Vol. 11

Issue (4) : 774-783 https://doi.org/10.1007/s11707-016-0611-2

RESEARCH ARTICLE

A mutual information-Dempster-Shafer based decision ensemble system for land cover classification of hyperspectral data

Parham PAHLAVANI, Behnaz BIGDELI(

)

Center of Excellence in Geomatics Engineering in Disaster Management, School of Surveying and Geospatial Engineering, College of Engineering, University of Tehran, Tehran 1439957131, Iran

Download: PDF(4124 KB) HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks

Abstract

Hyperspectral images contain extremely rich spectral information that offer great potential to discriminate between various land cover classes. However, these images are usually composed of tens or hundreds of spectrally close bands, which result in high redundancy and great amount of computation time in hyperspectral classification. Furthermore, in the presence of mixed coverage pixels, crisp classifiers produced errors, omission and commission. This paper presents a mutual information-Dempster-Shafer system through an ensemble classification approach for classification of hyperspectral data. First, mutual information is applied to split data into a few independent partitions to overcome high dimensionality. Then, a fuzzy maximum likelihood classifies each band subset. Finally, Dempster-Shafer is applied to fuse the results of the fuzzy classifiers. In order to assess the proposed method, a crisp ensemble system based on a support vector machine as the crisp classifier and weighted majority voting as the crisp fusion method are applied on hyperspectral data. Furthermore, a dimension reduction system is utilized to assess the effectiveness of mutual information band splitting of the proposed method. The proposed methodology provides interesting conclusions on the effectiveness and potentiality of mutual information-Dempster-Shafer based classification of hyperspectral data.

Keywords mutual information Dempster-Shafer hyperspectral classification support vector machine

Corresponding Author(s): Behnaz BIGDELI

Just Accepted Date: 23 November 2016 Online First Date: 15 December 2016 Issue Date: 10 November 2017

Cite this article:

Parham PAHLAVANI,Behnaz BIGDELI. A mutual information-Dempster-Shafer based decision ensemble system for land cover classification of hyperspectral data[J]. Front. Earth Sci., 2017, 11(4): 774-783.

URL:

https://academic.hep.com.cn/fesci/EN/10.1007/s11707-016-0611-2
https://academic.hep.com.cn/fesci/EN/Y2017/V11/I4/774

Fig.1 Flowchart of the mutual Information-Dempster-Shafer based decision ensemble system for classification of HSI.

Fig.2 Decision profile in fuzzy decision ensemble systems.

Fig.3 AVIRIS Indian Pine data. (a) original data and (b) ground truth.

Fig.4 ROSIS Pavia University data, (a) original data and (b) ground truth.

Tab.1 AVIRIS Indiana Pine land covers classes and available reference samples

Tab.2 ROSIS Pavia University lands cover classes and available reference samples

Fig.5 Splitting of AVIRIS HSI based on local minima of MI (Bigdeli et al., 2013).

Tab.3 Band numbers in splitting of spectral information of AVIRIS data and classification results by FML and SVM on bands subsets

Tab.4 Overall accuracies and kappa coefficients for proposed DES and dimension reduction methods on AVIRIS data

Fig.6 Classification maps obtained by different classification methods of Table 4 on AVIRIS data. (a) MI-DS, (b) MI-WMV, (c) PCA, and (d) PSO.

Tab.5 Class by class accuracies for MI-based decision ensemble systems and dimension reduction techniques on AVIRIS data

Fig.7 Splitting of Pavia ROSIS HSI based on local minima of MI (Bigdeli et al., 2013).

Tab.6 Band numbers in splitting of spectral information of ROSIS data and classification results by FML and SVM on bands subsets

Tab.7 Overall accuracies and kappa coefficients for proposed DES and dimension reduction methods on Pavia ROSIS data set

Fig.8 Classification maps obtained by different classification methods of Table 4 on ROSIS data. (a) MI-DS, (b) MI-WMV, (c) PCA, and (d) PSO.

Tab.8 Class by class accuracies for MI-based decision ensemble systems and dimension reduction techniques on ROSIS data

1	Akbari D, Homayuoni S, Saffari A, Mehrshad N (2016). Mapping urban land cover based on spatial-spectral classification of hyperspectral remote- sensing data. Int J Remote Sens, 37(2): 440–454 https://doi.org/10.1080/01431161.2015.1129561
2	Benediktsson J A , Chanussot J , Fauvel M (2007). Multiple classifier systems in remote sensing: from basics to recent developments. In: Haindl M, Kittler J, Roli F, eds. Multiple Classifier Systems. Heidelberg: Springer, 501–512
3	Bigdeli B, Samadzadegan F, Reinartz P (2013). A multiple SVM system for classification of hyperspectral remote sensing data. Journal of the Indian Society of Remote Sensing, 41(4): 763–776 https://doi.org/10.1007/s12524-013-0286-z
4	Borasca B, Bruzzone L, Carlin L , Zusi M (2006). A fuzzy-input fuzzy-output SVM technique for classification of hyperspectral remote sensing images. Signal Processing Symposium, 2006. NORSIG 2006. Proceedings of the 7th Nordic, 2–5
5	Breve F, Ponti M, Mascarenhas N (2007). Multilayer Perceptron Classifier Combination for Identification of Materials on Noisy Soil Science Multispectral Images. Computer Graphics and Image Processing, 2007. SIBGRAPI 2007. XX Brazilian Symposium on. 239–244
6	Camps-Valls G, Bruzzone L (2005). Kernel-based methods for hyperspectral image classification. IEEE Trans Geosci Rem Sens, 43(6): 1351–1362 https://doi.org/10.1109/TGRS.2005.846154
7	Ceamanos X, Waske B, Benediktsson J , Chanussot J , Fauvel M , Sveinsson J (2010). A classifier ensemble based on fusion of support vector machines for classifying hyperspectral data. International Journal of Image and Data Fusion, 1(4): 293–307 https://doi.org/10.1080/19479832.2010.485935
8	Chen C F (1999). Fuzzy training data for fuzzy supervised classification of remotely sensed images. Asian Conference Remote Sensing (ACRS 1999)
9	Chen H, Varshney P K, Arora M K (2003). Performance of mutual information similarity measure for registration of multitemporal remote sensing images. IEEE Trans Geosci Rem Sens, 41(11): 2445–2454 https://doi.org/10.1109/TGRS.2003.817664
10	Del Frate F, Pacifici F, Schiavon G , Solimini C (2007). Use of neural networks for automatic classification from high-resolution images. IEEE Trans Geosci Rem Sens, 45(4): 800–809 https://doi.org/10.1109/TGRS.2007.892009
11	Di W, Pan Q, He L , Cheng Y (2008). Anomaly detection in hyperspectral imagery by fuzzy integral fusion of band-subsets. Photogramm Eng Remote Sensing, 74(2): 201–213 https://doi.org/10.14358/PERS.74.2.201
12	Goel P K, Prasher S O, Patel R M, Landry J A, Bonnell R B, Viau A A (2003). Classification of hyperspectral data by decision trees and artificial neural networks to identify weed stress and nitrogen status of corn. Comput Electron Agric, 39(2): 67–93 https://doi.org/10.1016/S0168-1699(03)00020-6
13	Ham J, Yangchi Chen, Crawford M M , Ghosh J (2005). Investigation of the random forest framework for classification of hyperspectral data. IEEE Trans Geosci Rem Sens, 43(3): 492–501 https://doi.org/10.1109/TGRS.2004.842481
14	Imani M, Ghasemian H (2015). Feature reduction of hyperspectral images: discriminant analysis and the first principal component. Journal of AI and Data Mining, 3(1): 1–9 https://doi.org/10.5829/idosi.JAIDM.2015.03.01.01
15	Jia X (2002). Simplified maximum likelihood classification for HS data in cluster space. In: IEEE Int. Geoscience and Remote Sensing Symp. (IGARSS ‘02), Toronto, Ontario, Canada, 5: 2578–2580
16	Kasiri bidhendi S , Shirazi A , Fotoohi N , Ebadzadeh M (2007). Material Classification of hyperspectral images using unsupervised fuzzy clustering methods. Third International IEEE Conference on Signal-Image Technologies and Internet-Based System, SITIS '07. 619 –623
17	Kuncheva L (2004). Combining Pattern Classifiers methods and algorithms. Hoboken, New jersey. Canada: John Wiley&sons, INC.
18	Li J, Huang X, Gamba P , Bioucas-Dias J M B , Zhang L , Benediktsson J A , Plaza A (2015). Multiple feature learning for hyperspectral image classification. IEEE Trans Geosci Rem Sens, 53(3): 1592–1606 https://doi.org/10.1109/TGRS.2014.2345739
19	Li S, Wu H, Wan D , Zhu J (2011). An effective feature selection method for hyperspectral image classification based on genetic algorithm and support vector machine. Knowl Base Syst, 24(1): 40–48 https://doi.org/10.1016/j.knosys.2010.07.003
20	Martínez-Usó A , Pla F, Sotoca J M, Garcia-Sevilla P (2006). Clustering based multispectral band selection using mutual information, Proceedings of the 18th International Conference on Pattern Recognition (ICPR), 2: 760–763
21	Müller K L , Mika S, Rätsch G, Tsuda K , Schölkopf B (2001). An introduction to kernel-based learning algorithms. IEEE Trans Neural Netw, 12(2): 181–201 https://doi.org/10.1109/72.914517
22	Pepe M, Boschetti L, Brivio P A , Rampini A (2010). Comparing the performance of fuzzy and crisp classifiers on remotely sensed images: a case of snow classification. Int J Remote Sens, 31(23): 6189–6203 https://doi.org/10.1080/01431160903401395
23	Shafer G (1976). A Mathematical Theory of Evidence. Princeton University Press
24	Shen J, Wang C, Wang R , Huang F , Fan C, Xu L (2015). A band selection method for hyperspectral image classification based on improved particle swarm optimization, International Journal of Signal Processing. Image Processing and Pattern Recognition, 8(4): 325–338 https://doi.org/10.14257/ijsip.2015.8.4.28
25	Su H, Yang H, Du Q , Sheng Y (2011). Semi-supervised band clustering for dimensionality reduction of hyperspectral imagery. IEEE Geosci Remote Sens Lett, 8(6): 1135–1139 https://doi.org/10.1109/LGRS.2011.2158185
26	Vaiphasa C (2003). Innovative genetic algorithm for hyperspectral image classification. In: Proc. Int. Conf. Map Asia.
27	Vapnik V N (1998). Statistical Learning Theory.New York: Wiley & Sons, Inc.
28	Yu S, De Backer S, Scheunders P (2002). Genetic feature selection combined with composite fuzzy nearest neighbor classifiers for hyperspectral satellite imagery. Pattern Recognit Lett, 23(1‒3): 183–190 https://doi.org/10.1016/S0167-8655(01)00118-0
29	Zhang C, Qiu F (2012). Hyperspectral image classification using an unsupervised neuro-fuzzy system. J Appl Remote Sens, 6(1): 1–15

[1]	Bartłomiej SZYPUŁA, Małgorzata WIECZOREK. Geomorphometric relief classification with the k-median method in the Silesian Upland, southern Poland[J]. Front. Earth Sci., 2020, 14(1): 152-170.
[2]	Hongchun ZHU, Yuexue XU, Yu CHENG, Haiying LIU, Yipeng ZHAO. Landform classification based on optimal texture feature extraction from DEM data in Shandong Hilly Area, China[J]. Front. Earth Sci., 2019, 13(3): 641-655.
[3]	Kevin C. Tse, Hon-Chim Chiu, Man-Yin Tsang, Yiliang Li, Edmund Y. Lam. An unsupervised learning approach to study synchroneity of past events in the South China Sea[J]. Front. Earth Sci., 2019, 13(3): 628-640.
[4]	Jinwu TANG, Chunyan HU, Xingying YOU, Yunping YANG, Xiaofeng ZHANG, Jinyun DENG, Meng CHEN. The barrier river reach identification and classification in the Middle Yangtze River[J]. Front. Earth Sci., 2019, 13(3): 596-613.
[5]	Xue DAI, Rongrong WAN, Guishan YANG, Xiaolong WANG, Ligang XU, Yanyan LI, Bing LI. Impact of seasonal water-level fluctuations on autumn vegetation in Poyang Lake wetland, China[J]. Front. Earth Sci., 2019, 13(2): 398-409.
[6]	Xiaochen WANG, Yun SHAO, Fengli ZHANG, Wei TIAN. Comparison of C- and L-band simulated compact polarized SAR in oil spill detection[J]. Front. Earth Sci., 2019, 13(2): 351-360.
[7]	Kun JIA, Jingcan LIU, Yixuan TU, Qiangzi LI, Zhiwei SUN, Xiangqin WEI, Yunjun YAO, Xiaotong ZHANG. Land use and land cover classification using Chinese GF-2 multispectral data in a region of the North China Plain[J]. Front. Earth Sci., 2019, 13(2): 327-335.
[8]	Yongfeng WANG, Zhaohui XUE, Jun CHEN, Guangzhou CHEN. Spatio-temporal analysis of phenology in Yangtze River Delta based on MODIS NDVI time series from 2001 to 2015[J]. Front. Earth Sci., 2019, 13(1): 92-110.
[9]	Xiaoping LIU, Shuli CHEN, Li ZHUO, Jun LI, Kangning HUANG. Multi-sensor image registration by combining local self-similarity matching and mutual information[J]. Front. Earth Sci., 2018, 12(4): 779-790.
[10]	Xiaosong HAN, Jiayi PAN, Adam T. DEVLIN. Remote sensing study of wetlands in the Pearl River Delta during 1995---2015 with the support vector machine method[J]. Front. Earth Sci., 2018, 12(3): 521-531.
[11]	Xiaochen WANG, Yun SHAO, Wei TIAN, Kun LI. On the classification of mixed floating pollutants on the Yellow Sea of China by using a quad-polarized SAR image[J]. Front. Earth Sci., 2018, 12(2): 373-380.
[12]	Hongjun SU, Shufang TIAN, Yue CAI, Yehua SHENG, Chen CHEN, Maryam NAJAFIAN. Optimized extreme learning machine for urban land cover classification using hyperspectral imagery[J]. Front. Earth Sci., 2017, 11(4): 765-773.
[13]	Yue LIU,Qiuming CHENG,Qinglin XIA,Xinqing WANG. The use of evidential belief functions for mineral potential mapping in the Nanling belt, South China[J]. Front. Earth Sci., 2015, 9(2): 342-354.
[14]	Hongjun SU,Yehua SHENG,Peijun DU,Chen CHEN,Kui LIU. Hyperspectral image classification based on volumetric texture and dimensionality reduction[J]. Front. Earth Sci., 2015, 9(2): 225-236.
[15]	Jinpeng LI, Shikui DONG, Mingchun PENG, Xiaoyan LI, Shiliang LIU. Vegetation distribution pattern in the dam areas along middle-low reach of Lancang-Mekong River in Yunnan Province, China[J]. Front Earth Sci, 2012, 6(3): 283-290.

Viewed

Full text

Abstract

Cited

Shared

Discussed