Decomposed Soft Compression for Remote Sensing Image

Abstract

Modern imaging modalities from airborne or space platforms generate a large amount of remote sensing images, which places a burden on storage and transmission. These remote sensing images contain lots of ground targets which cannot tolerate fidelity loss in many fields. In this paper, a novel lossless compression method named Decomposed Soft Compression is proposed which takes advantage of rich details in images. The proposed method exploits image redundancy through integer wavelet transform. Then the decomposed images are encoded with shape-based encoder. Transformation on image layers can increase the sparsity and improve the compression ratio. Ex-periments on large-scale remote sensing image datasets show that the proposed method achieves up to 17.1 % improvement in compression ratio compared with JPEG 2000.