Gabor Wavelet Based Feature Extraction and Fusion for Hyperspectral and Lidar Remote Sensing Data

Abstract

In recent years, it has been found that the fusion processing of remote sensing data produced by multiple sensors is often effective for material classification. Specifically, the joint use of hyperspectral image (HSI) and Light Detection And Ranging (LiDAR) data for classification has been an active topic of research in remote sensing field. Since hyperspectral and LiDAR data provide complementary information (spectral reflectance, and vertical structure, respectively), one promising and challenging approach is to fuse these data in the information extraction procedure. In this paper, we propose an efficient feature extraction and fusion method based on Gabor wavelet, leading to a fusion of the spectral, spatial and elevation data. The core idea of the proposed fusion approach is stacking elevation and intensity data of LiDAR as additional channels to spectral bands. Our strategies are based on the Gabor feature stack structure, which are natural and effective. The features extracted by Gabor wavelets have proved to be discriminant features when considered for thematic classification in remote sensing applications especially when dealing with hyperspectral images due to their ability to extract joint spatial and spectrum information from HSI. Experimental results on the real hyperspectral image data have shown the better discriminative power of our approach for classification.