A PCA based fast vector quantization coding method for spectral imagery

In this paper, a PCA (Principal Component Analysis) based fast vector quantization coding method for spectral imagery has been proposed. Firstly, PCA is used to concentrate effective information of image into a few PCs (Principal Components). Then a PC selected method based on the eigenvalues is used to extract the PCs that contain the main information. Furthermore, an improved fast vector quantization (VQ) algorithm is exploited to encode the selected PCs. Experimental results show that, under the same coding conditions, the proposed method can achieve 13–16 dB higher reconstruction quality than JPEG2000 image coding standard.

[1]  Shen-En Qian,et al.  Hyperspectral data compression using a fast vector quantization algorithm , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[2]  Guillermo Sapiro,et al.  The LOCO-I lossless image compression algorithm: principles and standardization into JPEG-LS , 2000, IEEE Trans. Image Process..

[3]  Luciano Alparone,et al.  Crisp and Fuzzy Adaptive Spectral Predictions for Lossless and Near-Lossless Compression of Hyperspectral Imagery , 2007, IEEE Geoscience and Remote Sensing Letters.

[4]  Qian Du,et al.  Anomaly-Based JPEG2000 Compression of Hyperspectral Imagery , 2008, IEEE Geoscience and Remote Sensing Letters.

[5]  Qian Du,et al.  Hyperspectral Image Compression Using JPEG2000 and Principal Component Analysis , 2007, IEEE Geoscience and Remote Sensing Letters.

[6]  Shen-En Qian,et al.  Near lossless data compression onboard a hyperspectral satellite , 2006, IEEE Transactions on Aerospace and Electronic Systems.

[7]  Robert M. Gray,et al.  An Improvement of the Minimum Distortion Encoding Algorithm for Vector Quantization , 1985, IEEE Trans. Commun..

[8]  Robert M. Gray,et al.  An Algorithm for Vector Quantizer Design , 1980, IEEE Trans. Commun..