Chapter 1: Visualization and Data Analysis

Title: Real-time fusion and exploration of multispectral and hyperspectral data within 3D immersive virtual environment

Author(s): Wei Deng, Lik-Kwan Shark

Address: ADSIP Research Centre, University of Central Lancashire, Preston, PR1 2HE, UK | ADSIP Research Centre, University of Central Lancashire, Preston, PR1 2HE, UK

Reference: Atlantic Europe Conference on Remote Imaging and Spectroscopy pp. 27 - 34

Abstract/Summary: Currently multispectral and hyperspectral data are fused in 2D and then mapped on the corresponding 3D digital terrain for data exploration. This paper presents a new methodology and a unique software, which have been developed to enable users to carry out rapid and interactive fusion of large spectral data sets in real-time and to perform immersive data exploration within a 3D virtual environment. By harnessing fully the power of the latest PC graphics hardware and virtual reality technology, the presented methodology and software are shown to offer (a) Instantaneous spectral data exploration: with all bands of spectral data loaded into video memory and implemented by Cg shader code running directly on GPU (Graphics Processing Unit) in PC graphics hardware, users are able to switch from rendering one band of spectral data as terrain texture to another band of spectral data in real time; (b) Rapid spectral data fusion: wavelet based fusion methods are implemented by Cg shader code, thereby enabling users to change fusion parameters interactively and to view the fused data on 3D terrain in real time; (c) 3D virtual environment: GIS data (e.g. buildings, roads, vegetation) from a wide ranging file formats can be imported into the software and to be superimposed on the spectral image, thereby providing users the experience of viewing and flying through the reconstructed 3D virtual environment; and (d) 3D immersive environment: by using CAVE (Cave Automatic Virtual Environment), users are provided with a 3D immersive environment for spatial exploration of multispectral or hyperspectral image data. In the paper, the potential of the presented methodology and software for data interpretation are demonstrated using image data of a large area acquired via a Compact Airborne Spectrographic Imager (CASI).

Order a copy of this article