Date of this Version
Image processing techniques have been widely used in plant science for plant phenotyping studies. These fast algorithms are desired to process massive image data. In this thesis, we analyze RGB digital images taken from different view angles of plants and propose an efficient ad-hoc algorithm to identify structures of plants by 3D volume reconstruction techniques. We study hyperspectral images of plants and extend our scope to other images from different scientific disciplines. Obtaining the spectral and spatial information simul- taneously is a challenging task due to the high dimensionality of hyperspectral images. We first develop a real-time interactive tool for exploring the hyperspectral images as a hy- perspectral data cube. We discover a strong correlation between information entropy and hyperspectral images with respect to the wavelength under which the hyperspectral images are taken. We design an information metric based transfer function allowing users to study the hyperspectral data cube by interactive volume rendering techniques. In this manner, the transfer function changes dynamically with the regions of interest selected by users and both the spatial and spectral information can be preserved. We show the usefulness of our approach in different scientific disciplines including plant science, physics and remote sensing. In addition, our transfer function also works for the traditional volumetric data and our method provides a new interactive way of volume rendering.
Adviser: Hongfeng Yu