Secure Imperceptible Digital Color Video Steganography using Diagonal DCT, Pattern Recognition and DLSB Embedding

Abstract

As digital data is transferred and exchanged through the internet, numerous threats are imposed. Many hackers try to infiltrate transferred data unless security is considered during transmission. The process of video steganography is an engineering term which implies implanting video sequence in a cover sequence. The paper proposes a new technique of embedding video sequence data called "secret video" in cover video data. The technique provides a robust method to conceal secret information in cover video using signal transform. A discrete cosine transform (DCT) is applied on each secret video frame to localize the energy of the frame and scan the frame in the transform domain to recognize similar patterns. After converting the patterns to binary bits using fixed block codes, the blocks are embedded in the cover frame using dynamic least significant bit algorithm. This technique has provided a robust and secure embedding technique without apprising the hacker of a certain video data embedded. Many performance parameters are employed to measure the quality of stego video and reconstructed secret video.