Image Authentication Under Geometric Attacks Via Structure Matching

Image authentication techniques have recently gained great attention due to its importance for a large number of multimedia applications. Digital images are increasingly transmitted over non-secure channels such as the Internet. Therefore, military, medical and quality control images must be protected against attempts to manipulate them; such manipulations could tamper the decisions based on these images. To protect the authenticity of multimedia images, several approaches have been proposed. These approaches include conventional cryptography, fragile and semi-fragile watermarking and digital signatures that are based on the image content. The aim of this paper is to present a survey and a comparison of emerging techniques for image authentication. Methods are classified according to the service they provide, that is strict or selective authentication, tamper detection, localization and reconstruction capabilities and robustness against different desired image processing operations. Furthermore, we introduce the concept of image content and discuss the most important requirements for an effective image authentication system design. Different algorithms are described and we focus on their comparison according to the properties cited above.

show abstract

“…Monga and Vats tested the same algorithm with another wavelet function [94]. The results were very similar.…”

Section: Content-based Characteristics Extractionsupporting

confidence: 55%

Methods for image authentication: a survey

Haouzia

Noumeir

2007

Multimed Tools Appl

173

View full text Add to dashboard Cite

show abstract

“…Hence the total size of the hash h is L = 920 bits (< 1kB). Figure 3 depicts the ROC curves for 5 state-of-the art methods based on Fourier-Mellin invariants (FMI) [2], radial basis projections(RASH) [3], wavelets [5], SVD [4], structure matching (Feature) [1] compared to the proposed method. The hash component h 1 was used in this comparison.…”

Section: Experiments and Analysismentioning

confidence: 99%

“…Existing image hashing methods (that primarily address the issue of robustness) can be categorized as belonging to (1) exhaustive search based [1] and (2) robust representation based approach [2,3,4,5,7]. In an exhaustive search based approach, the noise N is modeled by some fixed distortion model (e.g., affine transform) and the hash h carries some alignment information about the original.…”

Section: Introductionmentioning

confidence: 99%

Robust Hash for Detecting and Localizing Image Tampering

Roy

Sun

2007

2007 IEEE International Conference on Image Processing

View full text Add to dashboard Cite

An image hash should be (1) robust to allowable operations and (2) sensitive to illegal manipulations (like image tampering) and distinct queries. Although existing methods try to address the first issue, the second issue has not been adequately looked into, primarily the issue of localizing tampering. This is primarily because of the difficulty in meeting two contradictory requirements. First, the hash should be small and second, to detect localized tampering, the amount of information in the hash about the original should be as large as possible. The desynchronization of the query with the original further aggravates the problem. Hence a tradeoff between these factors needs to be found. This paper presents an image hashing approach that is both robust and sensitive to not only detect but also localize tampering using a small signature (< 1kB). To our knowledge this is the first hashing method that can localize image tampering using a small signature that is not embedded into the image, like in watermarking.Index Terms-Locality preserving hashing, edge histogram, local region descriptors.

show abstract

“…First, the system employs the decoded random projections to perform image registration. In this example, the top-left corner of the cropped image is at θ = [13,18]. The coarse search returns the estimateθ C = [16,16], which is further refined toθ R = [13,18] when working at full resolution.…”

Section: An Illustrative Examplementioning

confidence: 99%

“…In this example, the top-left corner of the cropped image is at θ = [13,18]. The coarse search returns the estimateθ C = [16,16], which is further refined toθ R = [13,18] when working at full resolution. Figure 12 illustrates the error surface explored during the refinement step, where each point represents the value of the MSE between y θ andỹ for a given value of the parameter θ.…”

Section: An Illustrative Examplementioning

confidence: 99%

Hash-Based Identification of Sparse Image Tampering

Tagliasacchi

Valenzise

Tubaro

2009

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

In the last decade, the increased possibility to produce, edit and disseminate multimedia contents has not been adequately balanced by similar advances in protecting these contents from unauthorized diffusion of forged copies. When the goal is to detect whether or not a digital content has been tampered with in order to alter its semantics, the use of multimedia hashes turns out to be an effective solution to offer proof of legitimacy and to eventually identify the introduced tampering.We propose an image hashing algorithm based on compressive sensing principles, which solves both the authentication and the tampering identification problems. The original content producer generates a hash using a small bit budget by quantizing a limited number of random projections of the authentic image. The content user receives the (possibly altered) image, and uses the hash to estimate the mean square error distortion between the original and the received image. In addition, if the introduced tampering is sparse in some orthonormal basis or redundant dictionary, an approximation is given in the pixel domain. We emphasize that the hash is universal, e.g. the same hash signature can be used to detect and identify different types of tampering. At the cost of additional complexity at the decoder, the proposed algorithm is robust to moderate content-preserving transformations including cropping, scaling and rotation. In addition, in order to keep the size of the hash small, hash encoding/decoding takes advantage of distributed source codes.

show abstract

Image Authentication Under Geometric Attacks Via Structure Matching

Cited by 36 publications

References 23 publications

Methods for image authentication: a survey

Methods for image authentication: a survey

Robust Hash for Detecting and Localizing Image Tampering

Hash-Based Identification of Sparse Image Tampering

Contact Info

Product

Resources

About