Mahalingam Ramkumar

A classifier design for detecting image manipulations

In this paper we present a framework for digital image forensics. Based on the assumptions that some processing operations must be done on the image before it is doctored and an expected measurable distortion after processing an image, we design classifiers that discriminates between original and processed images. We propose a novel way of measuring the distortion between two images, one being the original and the other processed. The measurements are used as features in classifier design. Using these classifiers we test whether a suspicious part of a given image has been processed with a particular method or not. Experimental results show that with a high accuracy we are able to tell if some part of an image has undergone a particular or a combination of processing methods.

An efficient key predistribution scheme for ad hoc network security

We introduce hashed random preloaded subsets (HARPS), a highly scalable key predistribution (KPD) scheme employing only symmetric cryptographic primitives. HARPS is ideally suited for resource constrained nodes that need to operate for extended periods without active involvement of a trusted authority (TA), as is usually the case for nodes forming ad hoc networks (AHNs). HARPS, a probabilistic KPD scheme, is a generalization of two other probabilistic KPDs. The first, random preloaded subsets (RPSs), is based on random intersection of keys preloaded in nodes. The second, proposed by Leighton and Micali (LM) is a scheme employing repeated applications of a cryptographic hash function. We investigate many desired properties of HARPS like scalability, computational and storage efficiency, flexibility in deployment modes, renewability, ease of extension to multicast scenarios, ability to cater for broadcast authentication, broadcast encryption, etc., to support its candidacy as an enabler for ad hoc network security. We analyze and compare the performance of the three schemes and show that HARPS has significant advantages over other KPDs, and in particular, over RPS and LM.

Capacity estimates for data hiding in compressed images

In this paper, we present an information-theoretic approach to obtain an estimate of the number of bits that can be hidden in still images, or, the capacity of the data-hiding channel.We show how the addition of the message signal or signature in a suitable transform domain rather than the spatial domain can significantly increase the channel capacity. Most of the state-of-the-art schemes developed thus far for data-hiding have embedded bits in some transform domain, as it has always been implicitly understood that a decomposition would help. Though most methods reported in the literature use DCT or wavelet decomposition for data embedding, the choice of the transform is not obvious.We compare the achievable data-hiding capacities for different decompositions like DCT, DFT, Hadamard, and subband transforms and show that the magnitude DFT decomposition performs best among the ones compared.

Information theoretic bounds for data hiding in compressed images

We present an information-theoretic approach to obtain an estimate of the number of bits that can be hidden in still images, or, the capacity of the data-hiding channel. We show how addition of the message signal in a suitable transform domain rather than the spatial domain can significantly increase the channel capacity. We compare the capacities achievable with different decompositions like DCT, DFT, Hadamard, and subband transforms.

Pre-loaded key based multicast and broadcast authentication in mobile ad-hoc networks

The nature of mobile ad hoc networks (MANET), demands stringent requirements on primitives that could be used to secure such networks. Mobility imposes restrictions on memory and processor requirements due to limited battery life. The ad hoc nature warrants schemes that could operate for extended periods without referring to a trusted authority (TA). Additionally, any enabling scheme for security should be able to scale well. We introduce a novel key management scheme, RPS - random preloaded subset key distribution - which satisfies all the above requirements. More specifically, RPS is an n-secure r-conference key predistribution scheme. While most of the previously reported key predistribution schemes also meet all the stringent requirements, RPS has many inherent advantages. In this paper we investigate the applicability of RPS in securing MANETs.

MP3 resistant oblivious steganography

Robustness to compression is a basic requirement for any data hiding scheme. We concentrate on MP3 resistant oblivious data hiding. First the MP3 compression algorithm is briefly introduced. We then propose three effective data hiding schemes, where the message is embedded in amplitude, DFT phase domain, and noisy components, respectively. All these schemes demonstrate robustness to MP3 compression.

A robust data hiding scheme for images using DFT

We present a data hiding scheme for still images, in which only the magnitude of the DFT coefficients are altered to embed the hidden information bits. We begin by identifying various components of a typical data hiding scheme, like the decomposition used, signature design, self-noise suppression and signaling. The final choice of components for the proposed data hiding scheme are well tailored by utilizing theoretical reasonings and experimental observations. Performance results show good robustness of the data hiding scheme for both JPEG and SPIHT image compression.

A robust protocol for proving ownership of multimedia content

We explore the problem of proving ownership or origin of multimedia content like image/video or audio signals through watermarking. The need for watermarking arises out of the insufficiency of present copyright laws for claiming ownership of digital content. Watermarking schemes however, are threatened by counterfeit attacks, which primarily use the freedom available in choice of signature or choice of the watermarking method. A restrictive protocol for watermarking could go a long way in rendering counterfeit attacks extremely difficult. We suggest a comprehensive protocol as an extension of the one suggested by Craver et al. , that makes it possible for the true owner to claim ownership unambiguously, while making it extremely difficult for a pirate to do so.

A new method for detection of watermarks in geometrically distorted images

The goal of this study is to investigate the performance of the watermarking scheme proposed by Ramkumar and Akansu (see IEEE Trans. on Multimedia, 1999) for watermarked images distorted by geometrical attacks such as StirMark. Although the geometric distortions are not visually noticeable, the watermark cannot be detected due to the resulting high mean square error between the original and attacked images. In order to extract the watermark from distorted images, the effects of these attacks must be minimized. This work focuses on the recovery of an attacked image using a reliable distortion estimation. The performance test is studied for several images attacked by StirMark. The proposed algorithm is shown to reduce false alarm probability in detection of the watermark from about 0.005 for (for StirMarked images) to less than 10/sup -50/.

Robust scheme for oblivious detection of watermarks/data hiding in still images

We propose a novel, robust scheme for data hiding/oblivious detection of watermarks in still images. While the low- frequency image coefficients are robust, they cannot be used effectively for oblivious detection methods, when correlative processing is employed for detection. However, in the proposed non-linear detection method, the robust low- frequency bands can be used effectively. Thus the proposed method turns out to be more robust than methods employing linear addition and correlative extraction of the signature. We report the results obtained for 7 test images in terms of probability of error in detection of the watermark/hidden bits.