LiDAR aided hyperspectral images segmentation using schroedinger eigenmaps.

The segmentation of hyperspectral images (HSI) based on spatial and spectral information is being used in many fields from target detection to classification. In this paper, a new Schroedinger Eigenmaps (SE) algorithm is proposed to combine both elevation information obtained from LiDAR data and the spatial-spectral information. The potential matrix of SE algorithm is obtained by integrating the proximity between spatial components and the proximity between LiDAR data. The segmentation results are examined with weighting of this potential matrix. In addition, the segmentation results of the proposed method compares with the results of the Spatial-Spectral Schroedinger Eigenmaps (SSSE) algorithm and the results of the Normalized Cut algorithm.