Fusion Of Spatial Autocorrelation And Spectral Data For Remote Sensing Image Classification
2016 2ND INTERNATIONAL CONFERENCE ON ADVANCED TECHNOLOGIES FOR SIGNAL AND IMAGE PROCESSING (ATSIP)(2016)
摘要
The pixel-wise classification of remotely sensed images is a very challenging task due to spectral confusion on the one hand and missing, uncertain and imprecise data on the other hand. This paper focuses on the combination of two important image features that are the spectral and spatial information using Dempster-Shafer theory for improving image classification. Our approach extracts the spectral information by the gray levels histograms, while the spatial information is deduced by neighbourhood correlation of pixels using a geostatistical method which is local Moran's index. We modelled this feature as a new source derived from the principal image. The obtained results show the effectiveness of our fusion strategy and the interesting impact of introducing the spatial autocorrelation in the pixel-wise classification process.
更多查看译文
关键词
Pixel-wise classification,Dempster-Shafer theory,Fusion,Local Moran's index,Spatial autocorrelation,Spectral information
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络