Towards Lower Bounds on Embedding Distortion in Information Hiding Younhee Kim, Zoran Duric, and Dana Richards George Mason University, Fairfax VA 22030, USA, {ykim9, zduric, richards}@cs.gmu.edu Abstract. We propose two efficient information hiding algorithms in the least significant bits of JPEG coefficients of images. Our algorithms embed information by modifying JPEG coefficients in such a way that the introduced distortion is minimized. We derive the expected value of the additional error due to distortion as a function of the message length and the probability distribution of the JPEG quantization errors of cover images. We have implemented our methods in Java and performed the extensive experiments with them. The experiments have shown that our theoretical predictions agree closely with the actual introduced distor- tions. 1 Introduction 1.1 Motivation Surveillance video data are often used as evidence of traffic accident or crime. The surveillance system widely uses closed-circuit television (CCTV), which is exemplified by the small camera at ATM machines or parking lots. The CCTV system records scenes in analog film. Due to the high cost of film maintenance, the security industry seeks a way to replace it with digital system and store- in-files instead of films. However, there is one issue in using the digitally stored video as evidence: authentication. Because of the ease of undetectable alteration, it is essential to ensure that the video has not been tampered with after it was archived. Federal Rules of Evidence (FRE) states that the original of a recording is required to prove the content of the recording (FRE 1002) [1]. There is a requirement for authentication of digital image as evidence. An authentication system should detect any tampering in a marked image. It may be desirable for some applications to provide an indication of how much alteration has occurred and where the alteration has occurred. Another requirement is that the message extracting process should not require the original image. There are two technical approaches which provide authentication of digital video data: cryptographic approach [1, 2] and information hiding approach. Cryptographic authentication, creates a digital digest of the original image and encrypts it with the signer’s key, creating a digital signature. This digital signature [25] can be decrypted only by a key that is correspondent to the signer’s key. The digital signature can prove data integrity: if the image is exactly the