Hyperspectral Image Processing

Organizers: Lei Tong (Beijing University of Technology), Jun Zhou (Griffith University, Australia), Xiao Bai (Beihang University, China), and Xiuping Jia (University of New South Wales, Australia)

Motivation

Comparing with traditional grayscale or color imaging that capture one or three image channels, hyperspectral imaging captures images with tens or hundreds of light wavelength indexed bands. It provides detailed description of spectral and spatial distribution of light reflected from the object surface which is closely related to the material property of target of study. Thanks to this capability, hyperspectral imaging has been widely used in remote sensing, environment, mining, military, and medical imaging.
In recent years, because of the dropping cost of hardware components, hyperspectral imaging has been more and more used in close range computer vision research and applications. While this has greatly expanded the sensing capability of cameras to capture scenes or phenomena that are beyond human vision, effective processing, analysing and understanding of hyperspectral images are still challenging, with many unsolved problems. In particular, hyperspectral images are three dimensions in nature and normally in very larger size. As a consequence, though the state-of-the-art image processing techniques have achieved great success on traditional grayscale and colour images, they cannot be directly applied to hyperspectral images. On the other hand, this also brings new opportunities to the research community, as there are strong needs to develop effective and efficient methods for a variety of image processing and computer vision tasks.
The goal of this special session is to provide a forum for researchers and practitioners in the broad image processing and computer community to present their novel and original hyperspectral imaging research. The topics to be covered in this special session include, but not limit to:

1. Cost effective methods for hyperspectral image and video generation
2. Denoising and registration of hyperspectral and multispectral images.
3. Feature extraction from hyperspectral imaging
4. Image indexing and hashing methods
5. Dimensionality reduction techniques for hyperspectral imaging
6. Object detection and recognition
7. Scene analysis and understanding
8. Novel hyperspectral computer vision benchmark datasets and their evaluation
9. Novel applications of hyperspectral computer vision

<<Back