A Distortion Controllable Image Compression Technique

摘要

In order to compromise the contradiction between the high quality image and low bandwidth in data transmission and storage of satellite remote sensing images, a new high fidelity image compression technique, named DIstortion COntrollable image compression technique(DICO) is presented. By controlling the maximal gray error between the reconstructed image and the original image, the required image distortion degree can be reached. Contrast to the DWT based SPIHT, the proposed technique has several outstanding characteristics: First, given a threshold T, the image distortion and the PSNR of the reconstructed image can be predicted in advance, which can guarantee the required reconstructed image quality and avoid overmuch loss of original image's information. Second, there is no floating-point calculation and transformation in the compression process, so the compression speed is much faster than that of SPIHT. Finally, the PSNR of the reconstructed images of DICO is better than that of SPIHT, especially for the remote sensing images.