In this paper, a new steganographic method is proposed using region-level sparse representation for high density reversible data hiding. Reversible data hiding (RDH) is widely used in the field of secured communication. Using reversible data hiding, large amount of private messages can be embedded within one image and a stego image can be generated. In this work, the huge similarity of adjacent pixels of in sub image regions is efficiently utilized for reversible data hiding. In order to understand the relationship between pixels in a neighborhood, region-level sparse representation is used. The encryption and decryption processes used in the proposed steganographic technique apply auto keying and manual keying methods. Thus, the confidentiality of the data is increased. Sparse coding technique is used for the patch level linear representation of pixels considered over complete dictionary. © 2016, International Journal Of Pharmacy and Technology. All rights reserved.