Extracting Land Use Information of TM Image in Wind-Water Erosion Interlaced Region Based on Texture

DENG Kun,CHANG Qing-rui,YU Lin,WANG De-cai

DOI: https://doi.org/10.13207/j.cnki.jnwafu.2009.01.040

2009-01-01

Abstract:【Objective】 A method for deriving land use information of TM image in the wind-water erosion interlaced region was discussed.【Method】 This paper employed principal component analysis combined with Gray Level Co-occurrence Matrix texture analysis to abstract the land use information from the TM image in the wind-water erosion interlaced region(Yulin).The classification indexes were selected according to mean and the land use information was abstracted by hierarchical classification.【Result】 The results showed that there were significant differences among parameters of texture feature of the different land utilization.In the desertification,the mean and variance were comparatively high.The gray level was asymmetry,the contrasts between neighboring pixels were significant;while in water and irrigable land,the mean,variance was a little bit low,the grey level was symmetry;the mean,variance and gray level in the forest and dry land was between the above(desertification and water and irrigable land).【Conclusion】 Compared with the traditional classification method,the classification based on the texture analysis could acquire higher classifying precision.