Title :
Optimized quadtree for Karhunen-Loeve transform in multispectral image coding
Author_Institution :
Sarnoff Corp, Princeton, NJ, USA
fDate :
4/1/1999 12:00:00 AM
Abstract :
A new multispectral image compression technique based on the Karhunen-Loeve transform (KLT) and the discrete cosine transform (DCT) is proposed. The quadtree for determining the transform block size and the quantizer for encoding the transform coefficients are jointly optimized in a rate-distortion sense. The problem is solved by a Lagrange multiplier approach. After a quadtree is determined by this approach, a one-dimensional (1-D) KLT is applied to the spectral axis for each block before the DCT is applied on the spatial domain. The eigenvectors of the autocovariance matrix, the quantization scale, and the quantized transform coefficients for each block are the output of the encoder. The overhead information required in this scheme is the bits for the quadtree, KLT, and quantizer representation
Keywords :
Karhunen-Loeve transforms; covariance matrices; data compression; discrete cosine transforms; eigenvalues and eigenfunctions; image coding; optimisation; quadtrees; rate distortion theory; transform coding; 1-D KLT; DCT; Karhunen-Loeve transform; Lagrange multiplier; autocovariance matrix; discrete cosine transform; eigenvectors; multispectral image coding; multispectral image compression technique; optimized quadtree; quantization scale; quantized transform coefficients; quantizer representation; rate-distortion optimization; spatial domain; spectral axis; transform block size; transform coefficients; Covariance matrix; Discrete cosine transforms; Discrete transforms; Image coding; Image storage; Karhunen-Loeve transforms; Multispectral imaging; Nonhomogeneous media; Predictive coding; Vector quantization;
Journal_Title :
Image Processing, IEEE Transactions on
