On Using LSB Matching Function for Data Hiding in Pixels

The proposedmethod in this paper is based on the LSB matching function ofMielikainen for Steganography. However, there is a significant change with respect to Mielikainen's method in the use of the LSB matching function to compute secret bits. Melikainen extracts secret bits in pixel pairs: the first one is the LSB of the first pixel in a pixel pair and the other is computed by a LSB matching function applied to both pixels. The proposed method does not make any partition, but makes the sequential image processing: the current secret bit is always computed by a LSB matching function where the LSB matching function is modified using only an XOR operation. To make the data extraction compatible with data embedding, the proposed method modifies sparse pixels by adding/subtracting their value to/from one. In the experimental results, the values of embedding efficiency increase while the involved bits increase. Taking test images in the experiments, the numbers of modified pixels in the proposed method are always lower than those in Mielikainen's method.