Abstract—This paper proposes an improved LSB(least Significant bit) based Steganography technique for images imparting better information security .It presents an embedding algorithm for hiding encrypted messages in nonadjacent and random pixel locations in edges and smooth areas of images. It first encrypts the secret message, and detects edges in the cover-image using improved edge detection filter. Message bits are then, embedded in the least significant byte of randomly selected edge area pixels and 1-3-4 LSBs of red, green, blue components respectively across randomly selected pixels across smooth area of image. It ensures that the eavesdroppers will not have any suspicion that message bits are hidden in the image and standard steganography detection methods can not estimate the length of the secret message correctly. The Proposed approach is better in PSNR value and Capacity as shown experimentally than existing techniques.

Index Terms—Data hiding, image steganography, LSB insertion, edge detection.

Mamta Juneja is with the University Institute of Engineering and Technology, Panjab University, Chandigarh, India (e-mail: er_mamta@yahoo.com).
Parvinder S. Sandhu is with the Rayat and Bahara Institute of Engineering and Bio-Technology, Mohali, India.

Cite:Mamta Juneja and Parvinder S. Sandhu, "An Improved LSB Based Steganography Technique for RGB Color Images," International Journal of Computer and Communication Engineering vol. 2, no. 4, pp. 513-517, 2013.