Share Email Print
cover

Proceedings Paper

Transform coding compression of hyperspectral image
Author(s): Chunsheng Wang; Anlong Jiao; Jie Li
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

The High Resolution Imaging Spectrometer (HIRIS) produces tremendous amounts of raw data. Transmission of these data to an earth receiving station is constrained by downlink communication bandwidth. To alleviate limitation, the data must be compressed. In this paper, we present a compression algorithm for hyperspectral image by using the 3D discrete cosine transform coding (3-D DCT). The 3D DCT includes a 1D DCT on spectral direction followed by 2D DCT on the DCT coefficient images produced by the 1D DCT. The main effect of the 1D DCT is to remove the data correlation on the spectral direction, and produces the spectral band corresponded DCT coefficient images. The resulting spectrally decorrelated DCT coefficient images are then compressed by the JPEG algorithm. The compression ratio is controlled by the quantization of JPEG. In the technology of Transform Coding, the quantization error of the transform coefficients is an important factor affecting the error of compression system. According to the statistical characteristics of the 1D DCT coefficients, we design a combined quantization. In utilizing JPEG to compress DCT coefficient images we apply different quantizations to different spectral bands for better compression.

Paper Details

Date Published: 18 August 1998
PDF: 10 pages
Proc. SPIE 3505, Imaging System Technology for Remote Sensing, (18 August 1998); doi: 10.1117/12.317824
Show Author Affiliations
Chunsheng Wang, Jilin Univ. of Technology (China)
Anlong Jiao, Jilin Univ. of Technology (China)
Jie Li, Jilin Univ. of Technology (China)


Published in SPIE Proceedings Vol. 3505:
Imaging System Technology for Remote Sensing
Mingzhi Wei; Xinjian Yi; Jianzhong Han; Fiodor F. Sizov, Editor(s)

© SPIE. Terms of Use
Back to Top