Abstract: This paper proposes a new multispectral image data compression algorithm (3 Dimensional Embedded Block Coding with Optimized Truncation, 3D EBCOT), which can efficiently reduce spatial and spectral redundancies by applying the 3D Discrete Wavelet Transform (3D DWT), and get the embedded code stream by the three-dimensional Embedded Block Coding with Optimized Truncation(3D EBCOT) algorithm in wavelet transform domain. The novelty of the proposed coding strategy consists of two main parts: first, an unsymmetrical 3 dimensional DWT is used to reduce the redundancies in spatial and spectral domains. Then, after taking a 3D DWT, the coefficients are encoded using the 3D EBCOT algorithm which is expanded from JPEG2000. Numerical experiments on a six-band thematic mapper images show that the reconstructed images after using the proposed algorithm exhibit a better quality and higher compression ratio than the conventional JPEG2000 and KLT-3D SPIHT algorithms.

Key words: multispectral images, image compression, EBCOT, 3D DWT, JPEG2000