Hyperspectral Image Lossy-to-Lossless Compression Using 3D SPEZBC Algorithm Based on KLT and Wavelet Transform

In this paper, we propose a hyperspectral image lossy-tolossless compression using three-dimensional Embedded ZeroBlock Coding (3D EZBC) algorithm based on Karhunen-Loève transform (KLT) and wavelet transform (WT). Furthermore, an improved Hao’s matrix factorization method for integer KLT is also presented, which can reduce not only the computational complexity but also the memory requirements. Experimental results indicate that our coder can achieve better lossy-to-lossless compression performance in comparison with several state-of-the-art coding algorithms. Index Terms — image compression, hyperspectral image, 3D integer wavelet transforms, zeroblock coding