Current Science

Open Access Open Access  Restricted Access Subscription Access

Analysis of Worldview-2 Band Importance in Tree Species Classification based on Recursive Feature Elimination


Affiliations
1 School of Land and Tourism, Luoyang Normal University, Luoyang, Henan Province, 471934, China
2 College of forestry, Inner Mongolia Agricultural University, Huhhot, Inner Mongolia, 010019, China
 

In tree species classifications, different spectral bands feature different importance, and the manner of determining the importance of one band is a problem that needs to be solved. In this study, eight bands of the WorldView-2 fusion data were used as information sources, and a recursive feature elimination based on maximum likelihood (MLC-RFE) was used to sort the importance of these bands. According to the results, the importance of the eight bands was sorted as follows (from important to unimportant): nearinfrared 2 > red edge > yellow > red > near-infrared 1 > coastal blue > green > blue. The poorest band combination yielded the lowest overall accuracy (OA) and Kappa coefficient (40.9153%; 0.3080), whereas the optimal band combination presented the highest OA and Kappa coefficient (74.5479%; 0.7029), indicating the large difference in accuracies between the optimal and poorest band combinations. Therefore, selecting important bands bears significance in tree species classifications. The MLC-RFE method significantly solved the band selection problem. Thus, this method should be extended to more complex feature selections.

Keywords

Bands Importance, Maximum Likelihood, Recursive Feature Elimination, Tree Classification, Worldview- 2.
User
Notifications
Font Size

  • Fassnacht, F. E. et al., Review of studies on tree species classification from remotely sensed data. Remote Sens. Environ., 2016, 186(214), 64-87.

  • Ferreira, M. P. et al., Mapping tree species in tropical seasonal semi-deciduous forests with hyperspectral and multispectral data. Remote Sens. Environ., 2016, 179, 66-78.

  • Li, D., Ke, Y., Gong, H. and Li, X., Object-based urban tree species classification using bi-temporal worldview-2 and worldview3 images. Remote Sens-(Basel), 2015, 7(12), 16917-16937.

  • Heumann, B. W., An object-based classification of mangroves using a hybrid decision tree-support vector machine approach. Remote Sens-Basel., 2011, 3(11), 2440-2460.

  • Immitzer, M., Atzberger, C. and Koukal, T., Tree species classification with random forest using very high spatial resolution 8-band WorldView-2 satellite data. Remote Sens-(Basel), 2012, 4(9), 2661-2693.

  • Immitzer, M., Atzberger, C. and Koukal, T., Suitability of WorldView-2 data for tree species classification with special emphasis on the four new spectral bands. Photogramm. Fernerkun. Geoinf., 2012, 5, 573-588.

  • Pu, R. and Landry, S., A comparative analysis of high spatial resolution IKONOS and WorldView-2 imagery for mapping urban tree species. Remote Sens. Environ., 2012, 124(9), 516-533.

  • Cho, M. A. et al., Mapping tree species composition in South African savannas using an integrated airborne spectral and lidar system. Remote Sens. Environ., 2012, 125(10), 214-226.

  • Peerbhay, K. Y., Mutanga, O. and Ismail, R., Investigating the capability of few strategically placed worldview-2 multispectral bands to discriminate forest species in Kwazulu-Natal, South Africa. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 2013, 7(1), 307-316.

  • Deng, S. et al., Interpretation of forest resources at the individual tree level at purple mountain, nanjing city, china, using worldview-2 imagery by combining GPS, RS and GIS technologies. Remote Sens-(Basel), 2013, 6(1), 87-110.

  • Ghosh, A. and Joshi, P. K., A comparison of selected classification algorithms for mapping bamboo patches in lower gangetic plains using very high resolution worldview 2 imagery. Int. J. Appl. Earth Obs. Geoinf., 2014, 26(1), 298-311.

  • Guyon, I., Weston, J., Barnhill, S. and Vapnik, V., Gene selection for cancer classification using support vector machines. Mach. Learn., 2002, 46(1-3), 389-422.

  • Camps-Valls, G., Gomez-Chova, L., Calpe-Maravilla, J. and Martin-Guerrero, J. D., Robust support vector method for hyperspectral data classification and knowledge discovery. IEEE Trans. Geosci. Remote Sens., 2004, 42(7), 1530-1542.

  • Bazi, Y. and Melgani, F., Toward an optimal SVM classification system for hyperspectral remote sensing images. IEEE Trans. Geosci. Remote Sens., 2006, 44(11), 3374-3385.

  • Zhang, R. and Ma, J., A feature selection algorithm for hyperspectual data with SVM-RFE. Geomatics Inf. Sci. Wuhan Univ., 2009, 34(7), 834-837.

  • Zhang, H., Huang, B. and Yu, L., Kernel function in SVM-RFE based hyperspectral data band selection. Remote Sens. Technol. Appl., 2013, 28(5), 747-752 (in Chinese).

  • Liu, H. P., An, H. J., Wang, B. and Zhang, Q. L., Tree species classification using WorldView-2 images based on recursive texture feature elimination. J. Beijing For. Univ., 2015, 37(8), 53-59 (in Chinese).

  • Chavez, P. S., Berlin, G. L. and Sowers, L. B., Statistical method for selecting landsat MSS ratios 147. J. Appl. Photogr. Eng., 1982, 8(1), 23-30.

  • Qaid, A. M. and Basavarajappa, H. T., Application of optimum index factor technique to landsat7 data for geological mapping of north east of Hajjah, Yemen. Am.-Eurasian J. Sci. Res., 2008, 3(1), 84-91.

  • Malhi, A. and Gao, R. X., PCA-based feature selection scheme for machine defect classification. IEEE Trans. Instrum. Meas., 2004, 53(6), 1517-1525.

  • Kumaran, N. and Bhavani, R., PCA-based feature selection for MRI image retrieval system using texture features. Artif. Intell. Evol. Algorithms Eng. Syst., Springer India, 2015, 324, 109-117.


Abstract Views: 176

PDF Views: 70




  • Analysis of Worldview-2 Band Importance in Tree Species Classification based on Recursive Feature Elimination

Abstract Views: 176  |  PDF Views: 70

Authors

Huaipeng Liu
School of Land and Tourism, Luoyang Normal University, Luoyang, Henan Province, 471934, China
Huijun An
College of forestry, Inner Mongolia Agricultural University, Huhhot, Inner Mongolia, 010019, China
Yongxin Zhang
School of Land and Tourism, Luoyang Normal University, Luoyang, Henan Province, 471934, China

Abstract


In tree species classifications, different spectral bands feature different importance, and the manner of determining the importance of one band is a problem that needs to be solved. In this study, eight bands of the WorldView-2 fusion data were used as information sources, and a recursive feature elimination based on maximum likelihood (MLC-RFE) was used to sort the importance of these bands. According to the results, the importance of the eight bands was sorted as follows (from important to unimportant): nearinfrared 2 > red edge > yellow > red > near-infrared 1 > coastal blue > green > blue. The poorest band combination yielded the lowest overall accuracy (OA) and Kappa coefficient (40.9153%; 0.3080), whereas the optimal band combination presented the highest OA and Kappa coefficient (74.5479%; 0.7029), indicating the large difference in accuracies between the optimal and poorest band combinations. Therefore, selecting important bands bears significance in tree species classifications. The MLC-RFE method significantly solved the band selection problem. Thus, this method should be extended to more complex feature selections.

Keywords


Bands Importance, Maximum Likelihood, Recursive Feature Elimination, Tree Classification, Worldview- 2.

References





DOI: https://doi.org/10.18520/cs%2Fv115%2Fi7%2F1366-1374