Comparison of synthetic images generated from LiDAR intensity and passive hyperspectral imagery

Pulsed Light Detection And Ranging (LiDAR) intensity has commonly been used as a measure of relative reflectance of materials to aid in both point classification and object identification. However, as LiDAR systems use a single light wavelength, the intensity has had little value for advanced material classification. With the advent of multispectral LiDAR systems, it may be possible to use the LiDAR intensity in multiple spectral bands to assist in automated target recognition. Towards this end, we present a comparison between LiDAR intensity images and passive reflectance from a hyperspectral imaging system in the same spectral bands. Although qualitatively the LiDAR intensity and hyperspectral imagery show good agreement, a quantitative analysis shows there are significant deviations between their respective reflectance measurements, particularly for complex features such as trees.