Abstract
The concept, proposed here, may prove to be an innovative derivation in the field of frequency domain steganography for grayscale images. Sub image blocks, each of size 2x2 in row major order are taken from the carrier image and are passed through the process of Discrete Cosine Transform (DCT) to obtain the corresponding frequency components. In the process of embedding a single bit of secret message/image is inserted into the real part of the frequency component of 2nd, 3rd & 4th carrier image bytes of each block. The level of data obscurity scales up due to the application of a self devised hash function. It generates pseudorandom positions which control embedding and extraction of secret bits in and from the frequency components. In addition an adjustment is made on each embedded frequency component to optimally reduce the noise effect due to embedding. The results after experimenting with the proposed technique prove statistically improved performance compared to other exiting techniques.KeywordsImage AuthenticationSteganographyDCTIDCTMSEPSNRIFSSIM
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
