Abstract: There exist a large amount of redundant information in the smooth areas of a star map. Based on this characteristic, a star map compression algorithm is proposed relying on target object. This algorithm divides a star map into two sets of which, one is the smooth background and the other the complicated target object. The former areas can be coded with less data quantity when the latter areas are being coded, wavelet decomposition is applied to remove the relativity among pixels. Then the coefficients can be coded efficiently throngh energy clustering. Experimental results showed that the proposed algorithm achieved good visual quality and the peak signal-to-noise ratio (PSNR) is excellent.