In this paper the authors propose a reversible data hiding method which uses a novel algorithm based on histogram sifting. The distortion that occurs due to embedding can be completely removed during the data extraction. The proposed method uses double embedding to increase the reversibility. Location map information is embedded into the image itself, so that the reversibility can be easily achieved. The lower bound of PSNR for the proposed method is 51.73dB for 20000 bits. Moreover, the proposed method is quite simple and efficient. Experimental results show that the new method performs better than existing methods.