Beijing Chen

• Nanjing University of Information Science and Technology

It has become a research hotspot to detect whether a video is natural or DeepFake. However, almost all the existing works focus on detecting the inconsistency in either spatial or temporal. In this paper, a dual-branch (spatial branch and temporal branch) neural network is proposed to detect the inconsistency in both spatial and temporal for DeepFa...

Fake face detection is in dilemma with the rapid development of face manipulation technology. One way to improve the effectiveness of detector is to make full use of intra and inter frame information. In this paper, a novel Xception-LSTM algorithm is proposed by using our new spatiotemporal attention mechanism and convolutional long short-term memo...

With the development of face image synthesis and generation technology based on generative adversarial networks (GANs), it has become a research hotspot to determine whether a given face image is natural or generated. However, the generalization capability of the existing algorithms is still to be improved. Therefore, this paper proposes a general...

In recent years, generative adversarial networks (GANs) have been widely used to generate realistic fake face images, which can easily deceive human beings. To detect these images, some methods have been proposed. However, their detection performance will be degraded greatly when the testing samples are post-processed. In this paper, some experimen...

It has become a research hotspot to detect whether a face is natural or GAN-generated. However, all the existing works focus on whole GAN-generated faces. So, an improved Xception model is proposed for locally GAN-generated face detection. To the best of our knowledge, our work is the first one to address this issue. Some improvements over Xception...

Fully convolutional networks (FCNs) have been efficiently applied in splicing localization. However, the existing FCN-based methods still have three drawbacks: (a) their performance in detecting image details is unsatisfactory; (b) deep FCNs are difficult to train; (c) results of multiple FCNs are merged using fixed parameters to weigh their contri...

Artificial intelligence has been widely studied on solving intelligent surveillance analysis and security problems in recent years. Although many multimedia security approaches have been proposed by using deep learning network model, there are still some challenges on their performances which deserve in-depth research. On the one hand, high computa...

In this paper, we improve the parallel deep neural network (DNN) scheme BusterNet for image copy-move forgery localization with source/target region distinguishment. BusterNet is based on two branches, i.e., Simi-Det and Mani-Det, and suffers from two main drawbacks: (a) it should ensure that both branches correctly locate regions; (b) the Simi-Det...

Deep learning-based watermarking methods have achieved a better performance in capacity and invisibility than some traditional methods. However, their robustness against JPEG lossy compression attack is still to be improved. To enhance the robustness and construct an end-to-end method, it is urgent to simulate the JPEG lossy compression by a neural...

Artificial intelligence has been widely studied on solving intelligent surveillance analysis and security problems in recent years. Although many multimedia security approaches have been proposed by using deep learning network model, there are still some challenges on their performances which deserve in-depth research. On one hand, high computation...

Artificial intelligence has been widely studied on solving intelligent surveillance analysis and security problems in recent years. Although many multimedia security approaches have been proposed by using deep learning network model, there are still some challenges on their performances which deserve in-depth research. On one hand, high computation...

Artificial intelligence has been widely studied on solving intelligent surveillance analysis and security problems in recent years. Although many multimedia security approaches have been proposed by using deep learning network model, there are still some challenges on their performances which deserve in-depth research. On one hand, high computation...

Discrete Tchebyshev moments (DTMs), as a kind of typical discrete orthogonal moments, have been widely used in image analysis. However, the order of DTMs is restricted to an integer, and the fractional versions of DTMs have not been investigated. A novel framework for deriving fractional order DTMs (FrDTMs) by the eigen-decomposition of kernel matr...

With the development of internet technology, the video data has been widely used in multimedia devices, such as video surveillance, webcast, and so on. Lots of visual processing algorithms are developed to handle the corresponding visual task, but the challenging problems still exist. In this paper, we propose a weighted multiple instances based de...

Secure two-party computation evaluates a function among two distributed parties without revealing the parties’ inputs except for the function’s outputs. Secure two-party computation can be applied into various fields like cloud computing, which is a composition of distribute computing, parallel computing and utility computing etc. Rational secure t...

Recently, fully convolutional network (FCN) has been successfully used to locate spliced regions in synthesized images. However, all the existing FCN-based algorithms use real-valued FCN to process each channel separately. As a consequence, they fail to capture the inherent correlation between color channels and the integrity of three channels. So,...

In this paper, fractional quaternion cosine transforms (FrQCT) is proposed to generalize the conventional fractional cosine transforms (FrCT) to quaternion signal processing in a holistic manner. Firstly, the new transform FrQCT is defined and the proof of its inverse transform is presented. An efficient discrete implementation method of FrQCT is t...

In this paper, a novel spatial domain color image watermarking technique is proposed to rapidly and effectively protect the copyright of color image. Firstly, the direct current (DC) coefficient of 2D-DFT obtained in the spatial domain is discussed, and the relationship between the change of each pixel in the spatial domain and the change of the DC...

Recently, deep learning has achieved great success in visual tracking. The goal of this paper is to review the state-of-the-art tracking methods based on deep learning. First, we categorize the existing deep learning based trackers into three classes according to network structure, network function and network training. For each categorize, we anal...

Liu and Pun proposed a method based on fully convolutional network (FCN) and conditional random field (CRF) to locate spliced regions in synthesized images from different source images. However, their work has two drawbacks: (a) FCN often smooths detailed structures and ignores small objects; (b) CRF is employed as a standalone post-processing step...

In this paper, fractional Zernike moments (FrZMs) for complex signals are generalized to fractional quaternion Zernike moments (FrQZMs) for quaternion signal processing in a holistic manner by the quaternion algebra. We first present the definition of FrQZMs and an efficient implementation algorithm for speeding up the computation of FrQZMs through...

Support vector regression (SVR) is an important machine learning algorithm, although some successful applications have been achieved. The algorithm for complex system is still worth studying. Multiple output intuitionistic fuzzy least squares support vector regression (IFLS-SVR) is improved by using the intuitionistic fuzzy to solve the problem of...

In this study, by using the quaternion algebra, multiple-parameter fractional quaternion Fourier transform (MPFrQFT) is proposed to generalise the conventional multiple-parameter fractional Fourier transform (MPFrFT) to quaternion signal processing in a holistic manner. First, the new transform MPFrQFT and its inverse transform are defined. An effi...

As for the fractional transforms, to date only fractional Fourier transform (FrFT) has been applicable to color images in a holistic manner by using the quaternion algebra, yet, discrete fractional random transform (DFRNT) with the useful intrinsic randomness is still to be explored. This paper first defines quaternion DFRNT (QDFRNT) which generali...

This paper proposes a new algorithm of blind color image watermarking based on LU decomposition. Because of the first column the second row element and the first column the third row element of the lower triangular matrix, which obtained by LU decomposition, have higher similarity, and the color image watermark can be embedded into these elements b...

Three-dimensional reconstruction from a single input image is a very difficult issue, especially for the texture images. Moreover, the unknown lighting parameters also make this problem more complex. In this paper, an improved genetic algorithm has been proposed to reconstruct the 3D shape from a single texture image with similar appearances. The p...

This paper proposes a new blind watermarking algorithm, which embedding the binary watermark into the blue component of a RGB image in the spatial domain, to resolve the problem of protecting copyright. For embedding watermark, the generation principle and distribution features of direct current (DC) coefficient are used to directly modify the pixe...

With the rapid development of computer technology, the copyright protection of color image is becoming a more and more urgent problem. In this paper, an improved color image blind watermarking scheme based on Schur decomposition is proposed, which is different from using the binary or gray-level image as watermark. By analyzing the 4 × 4 unitary ma...

Content-aware image resizing is currently widely used because it maintains the original appearance of important objects to the greatest extent when the aspect ratio of an image changes during resizing. Content-aware image resizing techniques, such as seam carving, are also used for image forgery. A new Benford’s law-based algorithm for detecting co...

The quaternion representation (QR) used in current quaternion-based color image processing creates redundancy when representing a color image of three components by a quaternion matrix having four components. In this paper, both RGB and depth (RGB-D) information are considered to improve QR for efficiently representing RGB-D images. The improved QR...

Depth-image-based-rendering (DIBR) is a popular technique for view synthesis. The rendering process mainly introduces artifacts around edges, which leads to degraded quality. This letter proposes a DIBR-synthesized image quality metric by measuring the Statistics of both Edge Intensity and Orientation (SEIO). The Canny operator is first used to det...

It is a challenging work to design a blind color image watermarking scheme for protecting copyright, which is different from the existing schemes used binary image or grayscale image as watermark and is also different from other non-blind watermarking schemes. In this paper, we analyze the feature of the upper Hessenberg matrix, and propose a blind...

While the existing quaternion principal component analysis (QPCA) is a linear tool developed mainly for processing linear quaternion signals, the quaternion representation (QR) used in QPCA creates redundancy when representing a color image signal of three components by a quaternion matrix having four components. In this paper, the kernel technique...

Image segmentation is a crucial step in image processing, especially for medical images. However, the existence of partial volume effect, noise and other artifacts makes this problem much more complex. Fuzzy c-means (FCM), as an effective tool to deal with partial volume effect, cannot deal with noise and other artifacts. In this paper, one modifie...

Fuzzy C-means(FCM) has been adopted to perform image segmentation due to its simplicity and efficiency. Nevertheless it is sensitive to noise and other image artifacts because of not considering spatial information. Up to now, a series of improved FCM algorithms have been proposed, including fuzzy local information C-means clustering algorithm(FLIC...

In this paper, a novel blind color image watermarking based on Contourlet transform and Hessenberg decomposition is proposed to protect digital copyright of color image. Firstly, each color channel of the host image is transformed by Contourlet transform and its low frequency sub-band is divided into 4 × 4 non-overlap coefficient block. Secondly, t...

Most of the existing color image watermarking schemes use binary or grayscale image as watermark, and many of them belong to non-blind watermarking methods. In this paper, an improved color image watermarking algorithm based on QR decomposition is proposed to embed color watermark image into color host image. For embedding watermark, the 24-bits co...

Cloud computing is based on utility and consumption of computer resources. To solve the security issues in cloud computing, secure computing protocols are often used. Recently, rational parties as a new kind of parties are proposed, who wish to maximize their utilities in secure computing protocols. The utility definitions in most previous rational...

Face recognition encounters the problem that multiple samples of the same object may be very different owing to the deformation of appearances. To synthesizing reasonable virtual samples is a good way to solve it. In this paper, we introduce the idea of image-block-stretching to generate virtual images for deformable faces. It allows the neighbored...

Image enhancement is a popular technique, which is widely used to improve the visual quality of images. While image enhancement has been extensively investigated, the relevant quality assessment of enhanced images remains an open problem, which may hinder further development of enhancement techniques. In this paper, a no-reference quality metric fo...

For detecting the image copies of a given original image generated by arbitrary rotation, the existing image copy detection methods can not simultaneously achieve desirable performances in the aspects of both accuracy and efficiency. To address this challenge, a novel effective and efficient image copy detection method is proposed based on two glob...

Photometric stereo (PMS) can reconstruct shape and albedo of an object by using multiple images captured under varied illumination directions. However, PMS may fail if light intensity is varied across different images captured under different unknown lighting directions. This paper presents a method that can estimate shapes and albedo of inhomogene...

Image retouching is fundamental in photography, which is widely used to improve the perceptual quality of a low-quality image. Traditional image quality metrics are designed for degraded images, so they are limited in evaluating the quality of retouched images. This letter presents a RETouched Image QUality Evaluation (RETIQUE) algorithm by measuri...

The representation based classification method (RBCM) has attracted much attention in the last decade. RBCM exploits the linear combination of training samples to represent the test sample, which is then classified according to the minimum reconstruction residual. Recently, an interesting concept, Inverse Representation (IR), is proposed. It is the...

Computational load remains a major concern when processing signals by means of sliding transforms. In this paper, we present an efficient algorithm for the fast computation of one-dimensional and two-dimensional sliding discrete Tchebichef moments. To do so, we first establish the relationships that exist between the Tchebichef moments of two neigh...

Compression and encryption are widely used in network traffic in order to improve efficiency and security of some systems. We propose a scheme to concatenate both functions and run them in a paralle pipelined fashion, demonstrating both a hardware and a software implementation. With minor modifications to the hardware accelerators, latency can be r...

In this paper, we propose a blind color image watermarking scheme based on quaternion discrete Fourier transform (QDFT) and on an improved uniform log-polar mapping (IULPM). The proposed watermarking scheme embeds dual watermarks: one is a meaningful binary image watermark and the other is a bipolar watermark. The former is embedded in the real par...

The development of wind power has a higher requirement for the accurate prediction of wind. In this paper, a trustworthy and practical approach, Multidimensional Support Vector Regression (MSVR) with Data-Dependent Kernel(DDK), is proposed. In the prediction model, we applied the longitudinal component and lateral component of the wind speed, chang...

In this study, a new quaternion filter for removal of Gaussian noise in colour images is presented. It is based on the quaternion representation of colour images and the optimisation of a tight bound of the quaternion mean-square error between the restored colour image and the original one, together with the essential idea of the non-local means fi...

Among the few existing color watermarking schemes, some use quaternion discrete Fourier transform (QDFT). By modulating at least one component of QDFT coefficients, they spread the watermark over two or three of the RGB color channels. However, these schemes do not fully utilize the four-dimensional (4-D) QDFT frequency domain and some also suffer...

In this paper, the quaternion Bessel-Fourier moments are introduced. The significance of phase information in quaternion Bessel-Fourier moments is investigated and an accurate estimation method for rotation angle is described. Furthermore, a new set of invariant descriptors based on the magnitude and the phase information of quaternion Bessel-Fouri...

In this paper, by using the quaternion algebra, the conventional complex-type moments (CTMs) for gray-scale images are generalized to color images as quaternion-type moments (QTMs) in a holistic manner. We first provide a general formula of QTMs from which we derive a set of quaternion-valued QTM invariants (QTMIs) to image rotation, scale and tran...

Based on fixed point theory, this paper proposes a simple but efficient method for image integrity authentication, which is different from Digital Signature and Fragile Watermarking. By this method, any given image can be transformed into a fixed point of a well-chosen function, which can be constructed with periodic functions. The authentication c...

In order to process color images holistically, a quaternion weighted mean filter for Poisson noise removal in color images is presented. First, a color image is represented by a pure quaternion matrix by using the quaternion representation of color images. After that, the quaternion mean square error (QMSE) between the restored color image and the...

In order to make better use of color information for color face recognition, a new algorithm based on quaternion representation has been proposed. First of all, quaternion pseudo-Zernike moments (QPZMs) are defined to generalize the conventional pseudo-Zernike moments (PZMs) for grayscale images to the case of color images using the quaternion repr...

Based on the sparse representation-based classification (SRC) algorithm, taking the advantages of Tchebichef moments in image representation and noise robustness, a face recognition algorithm combining the sparse representation and Tchebichef moments is proposed to recognize the face images with and without additive noise. The mathematical derivati...

Moments and moment invariants have become a powerful tool in pattern recognition and image analysis. Conventional methods to deal with color images are based on RGB decomposition or graying, which may lose some significant color information. In this paper, by using the algebra of quaternions, we introduce the quaternion Zernike moments (QZMs) to de...

The aim of this paper is to resolve the completeness and independence of an invariant set derived by the traditional method, a systematic method for deriving a complete set of pseudo-Zernike moment similarity invariants (translation, scale and rotation) is described. First, the relationship between pseudo-Zernike moments of the original image and t...

The derivation of moment invariants has been extensively investigated in the past decades. In this paper, we construct a set of invariants derived from Zernike moments which is simultaneously invariant to similarity transformation and to convolution with circularly symmetric point spread function (PSF). Two main contributions are provided: the theo...

To resolve the dependence problem of traditional invariant features in pattern recognition, a new method to derive a complete set of Zernike moments similarity invariants is presented. Firstly, based on the connection between Zernike moments and radial moments, the relationship between Zernike moments of the original image and those of the images h...

Discrete orthogonal moments have been recently introduced in the field of image analysis. It was shown that they have better image representation capability than the continuous orthogonal moments. One problem concerning the use of moments as feature descriptors is the high computational cost, which may limit their application to the problems where...

Moments and moment invariants are useful tool in pattern recognition and image analysis. Conventional methods to deal with color images are based on RGB decomposition or graying. In this paper, by using the theory of quaternions, we introduce a set of quaternion Zernike moments (QZMs) for color images in a holistic manner. It is shown that the QZMs...

Illumination change and expression variation are two main factors affecting the performance of some existing face recognition algorithms. Edge feature based methods are robust to illumination change and are easy to implement. But they donpsilat work well for the images with expression variation. In this paper, a novel face recognition method based...

In this paper, a novel face recognition method based on binary face edges is presented to deal with the illumination problem. The Binary Face Edge Map (BFEM) is extracted using the Locally Adaptive Threshold (LAT) algorithm. Based on BEFM, a new image similarity metric is proposed. Experimental results show that face recognition rates of 76.32% and...