论文信息 - Compression of hyper-spectral images based on quadtree partitioning

Compression of hyper-spectral images based on quadtree partitioning

The paper analyzes the characteristic features of hyper-spectral image and presents a compression of hyper-spectral images based on quadtree partitioning. Quadtree partition is used to get the mean image of the whole image and the significant correlation of image can be decorrelated by subtract the mean image from original image. The difference image is compressed by DCT and encoded with arithmetic code. Experiment show the algorithm is simple and easy to use in real-time image compressing.

[1] I. Aoi,et al. A Prediction Model of Ambient Noise Level Near Roads Using the Building Coverage Rate Estimated from Landsat TM Data , 1986 .

[2] Khalid Sayood,et al. Introduction to data compression (2nd ed.) , 2000 .

[3] Huang Lian-qing,et al. Feature analysis of hyperspectral images , 2001 .

[4] Khalid Sayood,et al. Introduction to Data Compression , 1996 .

[5] Ni Lin. Compression of Remote Sensing Images Based on Adaptive Quadtree Partitioning , 2002 .

[6] X. Ding. Image Bitplane Coding Based on Combination of Subband Quadtrees , 2006 .

[7] Chen Hui. Method based on quarter-tree decomposition of image compression , 2008 .