Robust Perceptual Hashing Research Articles

Robust perceptual fingerprint image hashing: a comparative study

This paper presents a robust perceptual hashing scheme for biometric template protection where the input fingerprint image is mapped into a sequence of Boolean values. Our aim is to develop a method that relies on the use of four functions namely SIFT, Harris, DWT and SVD. After extracting the minutiae, the scale-invariant feature transform (SIFT) is applied in order to extract the robust features against geometric attacks. The resulting vector is then filtered using Harris criterion to maintain only the stable key-points. Next, the fingerprint template is produced by image binarisation and decomposed into blocks. The hash code is finally obtained by concatenating the singular values computed on the approximation coefficients of each image block. Similarity between hash codes is evaluated by the normalised Hamming distance (HD). Comparative analysis to three similar methods indicates that the proposed hashing scheme shows better performances in terms of discriminative capability as well as robustness against acceptable image manipulations, such as JPEG compression, gamma correction, speckle noise, Gaussian blur, shearing and slight rotation.

Method of star image identification based on robust perceptual hash feature

Traditional star image identification methods use adjacent stellar angular distances to form identification features. The missed detection of high-magnitude stars due to noise interference affects their identification accuracy. We propose a star image identification method based on perceptual hash features. The multiple Gaussian distribution superposition constructed by stars within a certain range around the center star forms the unique standard identification image of that star, in which the brightness distribution is determined by stellar magnitudes. The perceptual hash feature of the standard identification image is then used as a standard identification feature. The observation feature is extracted from the identification image as-constructed by the star points extracted from the star image. Finally, the observation feature is matched with the standard feature to identify the star image. Simulation results indicate that the proposed method can efficiently identify star images and effectively prevent missed stars from influencing the matching results.

A Visual Model-Based Perceptual Image Hash for Content Authentication

Perceptual image hash has been widely investigated in an attempt to solve the problems of image content authentication and content-based image retrieval. In this paper, we combine statistical analysis methods and visual perception theory to develop a real perceptual image hash method for content authentication. To achieve real perceptual robustness and perceptual sensitivity, the proposed method uses Watson's visual model to extract visually sensitive features that play an important role in the process of humans perceiving image content. We then generate robust perceptual hash code by combining image-block-based features and key-point-based features. The proposed method achieves a tradeoff between perceptual robustness to tolerate content-preserving manipulations and a wide range of geometric distortions and perceptual sensitivity to detect malicious tampering. Furthermore, it has the functionality to detect compromised image regions. Compared with state-of-the-art schemes, the proposed method obtains a better comprehensive performance in content-based image tampering detection and localization.

Zero-watermarking Algorithm for Medical Volume Data Based on Difference Hashing

In order to protect the copyright of medical volume data, a new zerowatermarking algorithm for medical volume data is presented based on Legendre chaotic neural network and difference hashing in three-dimensional discrete cosine transform domain. It organically combines the Legendre chaotic neural network, three-dimensional discrete cosine transform and difference hashing, and becomes a kind of robust zero-watermarking algorithm. Firstly, a new kind of Legendre chaotic neural network is used to generate chaotic sequences, which causes the original watermarking image scrambling. Secondly, it uses three-dimensional discrete cosine transform to the original medical volume data, and the perception of the low frequency coefficient invariance in the three-dimensional discrete cosine transform domain is utilized to extract the first 4*5*4 coefficient in order to form characteristic matrix (16*5). Then, the difference hashing algorithm is used to extract a robust perceptual hashing value which is a binary sequence, with the length being 64-bit. Finally, the hashing value serves as the image features to construct the robust zero-watermarking. The results show that the algorithm can resist the attack, with good robustness and high security.

A Robust Image Perceptual Hashing Scheme Based on Improved Normalization Algorithm

An improved image normalization algorithm is proposed firstly. Then a geometric robust image perceptual hashing scheme is proposed based on image normalization and discrete cosine transform. The original image is preprocessed using our improved image normalization method. And the selected DCT coefficients as image feature are encrypted for security. The geometric robust image hash is achieved by quantizing encrypted DCT coefficients and coding. The experimental results show that the algorithm can resist against common global affine transformations such as rotation, scaling, translation and their combinations.

抗几何失真的视频Hashing 算法研究

In this paper, we propose a robust perceptual hashing algorithm by using the shape of video luminance histogram. The underlying robustness principles are based on three main aspects: 1) since the histogram in shape does not depend on the exact position of a pixel, the algorithm is resistant to geometric deformations; 2) the hash is extracted from the spatial Gaussian-filtering low-frequency component against common video processing operations such as noise permutation, low-pass filtering and lossy compression; 3) temporal Gaussian-filtering operation is designed so that the hash is robust to temporal desynchronization operations, such as frame rate change and dropping. As a result, the hash function is robust to common geometric distortions and video processing operations. Experimental results show that the proposed hashing strategy provides satisfactory robustness and uniqueness.

Perceptual video hashing robust against geometric distortions

In this paper, we propose a robust perceptual hashing algorithm by using video luminance histogram in shape. The underlying robustness principles are based on three main aspects: 1) Since the histogram is independent of position of a pixel, the algorithm is resistant to geometric deformations; 2) the hash is extracted from the spatial Gaussian-filtering low-frequency component for those common video processing operations such as noise corruption, low-pass filtering, lossy compression, etc.; 3) a temporal Gaussian-filtering operation is designed so that the hash is resistant to temporal desynchronization operations, such as frame rate change and dropping. As a result, the hash function is robust to common geometric distortions and video processing operations. Experimental results show that the proposed hashing strategy can provide satisfactory robustness and uniqueness.