Analysis on Combination of Watermarking Algorithm: Modified Least Significant Bit Algorithm with Least Significant Bit+1

• Published in: Journal of physics. Conference series 2019-06, Vol.1235 (1), p.12081
• Main Authors: Lubis, R R A, Hardi, S M, Zarlis, M, Jaya, I, Tarigan, J T
• Format: Article
• Language: English
• Subjects: Algorithms; Digital broadcasting; Digital imaging; Digital media; Insertion; Parameters; Physics; Pixels; Watermarking
• ISSN: 1742-6588; 1742-6596
• DOI: 10.1088/1742-6596/1235/1/012081

To protect someone's copyright on digital media such as digital image, watermarking technique can be used, which is a technique to insert copyright marks into digital media. The algorithms that commonly used in the watermark insertion is the Least Significant Bit (LSB). But there is deficiency in LSB algorithm. LSB algorithm requires one byte image pixel to insert one bit watermark. The more byte pixels are inserted, the higher the difference between watermarked image with the original image. Embedded watermark can also easily be known because it is definitely placed on the LSB bits of the pixel image. From these problems, in this research we analyze the combination of two watermarking algorithms: Modified Least Significant Bit (MLSB) with Least Significant Bit (LSB+1). MLSB have the advantage in compressing the watermark to be inserted from eight bit/character into five bit/character. LSB+1 algorithm will cover the deficiency of MLSB algorithm, which the location of the watermark bit is easy to be discovered. Added in this research, the watermark bit had inserted in random image pixel. The Combined algorithm tested by using five image that have different sizes and dimensions, which each of the image inserted by text watermark that have different length. The purpose is to find the value of these three comparison parameters: Mean Squared Error value (MSE), the size of the result image and processing time. It concluded that for the three parameters, the combined algorithm of MLSB-LSB+1 have done better than the LSB+1algorithm but slightly worse than the MLSB algorithm. This happens due to MLSBLSB+1 besides inheriting the advantages of each algorithm, also inherited deficiency of the LSB+1 algorithm which is because of the insertion is in the one bit before the least significant bit of the image pixel, the value of the pixel is greater for two point(in decimal).