Jun Tian

Reversible data embedding using a difference expansion

Reversible data embedding has drawn lots of interest recently. Being reversible, the original digital content can be completely restored. We present a novel reversible data-embedding method for digital images. We explore the redundancy in digital images to achieve very high embedding capacity, and keep the distortion low.

Wavelet-based reversible watermarking for authentication

In the digital information age, digital content (audio, image, and video) can be easily copied, manipulated, and distributed. Copyright protection and content authentication of digital content has become an urgent problem to content owners and distributors. Digital watermarking has provided a valuable solution to this problem. Based on its application scenario, most digital watermarking methods can be divided into two categories: robust watermarking and fragile watermarking. As a special subset of fragile watermark, reversible watermark (which is also called lossless watermark, invertible watermark, erasable watermark) enables the recovery of the original, unwatermarked content after the watermarked content has been detected to be authentic. Such reversibility to get back unwatermarked content is highly desired in sensitive imagery, such as military data and medical data. In this paper we present a reversible watermarking method based on an integer wavelet transform. We look into the binary representation of each wavelet coefficient and embed an extra bit to expandable wavelet coefficient. The location map of all expanded coefficients will be coded by JBIG2 compression and these coefficient values will be losslessly compressed by arithmetic coding. Besides these two compressed bit streams, an SHA-256 hash of the original image will also be embedded for authentication purpose.

A lossy image codec based on index coding

Summary form only. We propose a new lossy image codec based on index coding. Both Shapiros embedded zerotree wavelet algorithm, and Said and Pearlmans codetree algorithm use spatial orientation tree structures to implicitly locate the significant wavelet transform coefficients. A direct approach to find the positions of these significant coefficients is presented. The new algorithm combines the discrete wavelet transform, differential coding, variable-length coding of integers, ordered bit plane transmission, and adaptive arithmetic coding. The encoding can be stopped at any point, which allows a target rate or distortion metric to be met exactly. The bits in the bit stream are generated in the order of importance, yielding a fully embedded code to successively approximate the original image source; thus it is well suited for progressive image transmission. The decoder can also terminate the decoding at any point, and produce a lower bit rate reconstruction image. Our algorithm is very simple in its form (which will make the encoding and decoding very fast), requires no training of any bind or prior knowledge of image sources, and has a clear geometric structure. The image coding results of it are quite competitive with almost all previous reported image compression algorithms on standard test images.

Embedded Image Coding Using Wavelet Difference Reduction

We present an embedded image coding method, which basically consists of three steps, Discrete Wavelet Transform, Differential Coding, and Binary Reduction. Both J. Shapiro’s embedded zerotree wavelet algorithm, and A. Said and W. A. Pearlman’s codetree algorithm use spatial orientation tree structures to implicitly locate the significant wavelet transform coefficients. Here a direct approach to find the positions of these significant coefficients is presented. The encoding can be stopped at any point, which allows a target rate or distortion metric to be met exactly. The bits in the bit stream are generated in the order of importance, yielding a fully embedded code to successively approximate the original image source; thus it’s well suited for progressive image transmission. The decoder can also terminate the decoding at any point, and produce a lower (bit) rate reconstruction image. Our algorithm is very simple in its form (which will make the encoding and decoding very fast), requires no training of any kind or prior knowledge of image sources, and has a clear geometric structure. The image coding results of it are quite competitive with almost all previous reported image compression algorithms on standard test images.

High capacity reversible data embedding and content authentication

We present a high capacity reversible data embedding algorithm. It serves the purpose of both self authentication and reversible data embedding. As the algorithm is reversible, the original digital content (before data embedding) can be completely restored after authentication. We employ two techniques, difference expansion and generalized least significant bit embedding, to achieve very high embedding capacity, while keeping the distortion (the quality degradation on the digital content after data embedding) low. A noticeable difference between our method and others is that we do not need to compress original values of the embedding area. We explore the redundancy in the digital content to achieve reversibility. The paper considers grayscale images only, but our method can be applied to color images and to digital audio and video as well.

A new class of biorthogonal wavelet systems for image transform coding

We construct general biorthogonal Coifman wavelet systems, a new class of compactly supported biorthogonal wavelet systems with vanishing moments equally distributed for a scaling function and wavelet pair. A time-domain design method is employed and closed-form expressions for the impulse responses and the frequency responses of the corresponding dual filters are derived. The resulting filter coefficients are all dyadic fractions, which is an attractive feature in the realization of multiplication-free discrete wavelet transform. Even-ordered systems in this family are symmetric, which correspond to linear-phase dual filters. In particular, three filterbanks (FBs) in this family are systematically verified to have competitive compression potential to the 9-7 tap biorthogonal wavelet FB by Cohen et al., which is currently the most widely used one in the field of wavelet transform coding. In addition, the proposed FBs have much smaller computational complexity in terms of floating-point operations required in transformation, and therefore indicate a better tradeoff between compression performance and computational complexity.

Image data processing in the compressed wavelet domain

A new embedded image coding method, the wavelet-difference-reduction algorithm, was introduced by Jun Tian et al. (see Proceedings of the IEEE Data Compression Conference, Snowbird, Utah, 1996). It combines the discrete wavelet transform, differential coding, binary reduction, order bit plane transmission, and adaptive arithmetic coding. It is a simple and efficient way to successively approximate the image source and is very suitable for progressive image transmission. In addition, this algorithm provides a clear geometric structure in the compressed wavelet domain, which enables us to process the image data in compressed form. We discuss three aspects of this problem, namely denoising, speckle reduction, and zooming. These outline some ideas of how to facilitate image data processing in the compressed wavelet domain.

A Compressive Sensing Based Secure Watermark Detection and Privacy Preserving Storage Framework

Privacy is a critical issue when the data owners outsource data storage or processing to a third party computing service, such as the cloud. In this paper, we identify a cloud computing application scenario that requires simultaneously performing secure watermark detection and privacy preserving multimedia data storage. We then propose a compressive sensing (CS)-based framework using secure multiparty computation (MPC) protocols to address such a requirement. In our framework, the multimedia data and secret watermark pattern are presented to the cloud for secure watermark detection in a CS domain to protect the privacy. During CS transformation, the privacy of the CS matrix and the watermark pattern is protected by the MPC protocols under the semi-honest security model. We derive the expected watermark detection performance in the CS domain, given the target image, watermark pattern, and the size of the CS matrix (but without the CS matrix itself). The correctness of the derived performance has been validated by our experiments. Our theoretical analysis and experimental results show that secure watermark detection in the CS domain is feasible. Our framework can also be extended to other collaborative secure signal processing and data-mining applications in the cloud.Privacy is a critical issue when the data owners outsource data storage or processing to a third party computing service, such as the cloud. In this paper, we identify a cloud computing application scenario that requires simultaneously performing secure watermark detection and privacy preserving multimedia data storage. We then propose a compressive sensing (CS)-based framework using secure multiparty computation (MPC) protocols to address such a requirement. In our framework, the multimedia data and secret watermark pattern are presented to the cloud for secure watermark detection in a CS domain to protect the privacy. During CS transformation, the privacy of the CS matrix and the watermark pattern is protected by the MPC protocols under the semi-honest security model. We derive the expected watermark detection performance in the CS domain, given the target image, watermark pattern, and the size of the CS matrix (but without the CS matrix itself). The correctness of the derived performance has been validated by our experiments. Our theoretical analysis and experimental results show that secure watermark detection in the CS domain is feasible. Our framework can also be extended to other collaborative secure signal processing and data-mining applications in the cloud.

Wavelet-based image compression and content authentication

In the digital information age, digital content (audio, image, and video) can be easily copied, manipulated, and distributed. Copyright protection and content authentication of digital content has become an urgent problem to content owners and distributors. Digital watermarking has provided a valid solution to this problem. Based on its application scenario, most digital watermarking methods can be divided into two categories: robust watermarking and fragile watermarking. Here, we will concentrate on fragile watermarking of digital images, which is for image content authentication. Our fragile watermarking method is heavily based on the new image compression standard JPEG 2000. We choose a compressed bit stream from JPEG 2000 as the hash of an image, and embed the hash back to the image. The exceptional compression performance of JPEG 2000 solves the tradeoff between small hash size and high hash confidence level. In the authentication stage, the embedded compressed bit stream will be extracted. Then it will be compared with the compressed bit stream of the image to be authenticated. The authentication decision comes from the comparison result. Besides content authentication, we will also show how to employ this watermarking method for hiding one image into another.

Biorthogonal Wavelet Space: Parametrization and Factorization

In this paper we study the algebraic and geometric structure of the space of compactly supported biorthogonal wavelets. We prove that any biorthogonal wavelet matrix pair (which consists of the scaling filters and wavelet filters) can be factored as the product of primitive para-unitary matrices, a pseudo identity matrix pair, an invertible matrix, and the canonical Haar matrix. Compared with the factorization results of orthogonal wavelets, it now becomes apparent that the difference between orthogonal and biorthogonal wavelets lies in the pseudo identity matrix pair and the invertible matrix, which in the orthogonal setting will be the identity matrix and a unitary matrix. Thus by setting the pseudo identity matrix pair to be the identity matrix and using the Schmidt orthogonalization method on the invertible matrix, it is very straightforward to convert a biorthogonal wavelet pair into an orthogonal wavelet.