Beijing Chen

Color Image Analysis by Quaternion-Type Moments

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 translation transformations by eliminating the influence of transformation parameters. An efficient computation algorithm is also proposed so as to reduce computational complexity. The performance of the proposed QTMs and QTMIs are evaluated considering several application frameworks ranging from color image reconstruction, face recognition to image registration. We show they achieve better performance than CTMs and CTM invariants (CTMIs). We also discuss the choice of the unit pure quaternion influence with the help of experiments.

Quaternion Bessel-Fourier moments and their invariant descriptors for object reconstruction and recognition

(i-j-k)/sqrt{3}

Color image classification via quaternion principal component analysis network

(i-j-k)/3 appears to be an optimal choice.

Removing Gaussian noise for colour images by quaternion representation and optimisation of weights in non-local means filter

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 from a watermark energy loss directly at the embedding stage. In this paper, we first establish the links that exist between the DFT of the three RGB color channels and the components of QDFT coefficients while considering a general unit pure quaternion. Then, for different unit pure quaternions i, j, k or their linear combinations, we discuss the symmetry constraints one should follow when modifying QDFT coefficients in order to overcome the previous drawbacks. We also provide a general watermarking framework to illustrate the overall performance gain in terms of imperceptibility, capacity and robustness we can achieve compared to other QDFT based algorithms, i.e. when fully considering the 4-D QDFT domain. From this framework we derive three schemes, depending on whether i, j or k is used. Provided theoretical analysis and experimental results show that these algorithms offer better performance in terms of capacity and robustness to most common attacks, including JPEG compression, noise, cropping and filtering and so on, than other QDFT based algorithms for the same watermarked image quality.

Fast Computation of Sliding Discrete Tchebichef Moments and Its Application in Duplicated Regions Detection

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-Fourier moments is derived. Experimental results show that quaternion Bessel-Fourier moments lead to better performance for color image reconstruction than the other quaternion orthogonal moments such as quaternion Zernike moments, quaternion pseudo-Zernike moments and quaternion orthogonal Fourier-Mellin moments. In addition, the angles estimated by the proposed moments are more accurate than those obtained by using other quaternion orthogonal moments. The proposed invariant descriptors show also better robustness to geometric and photometric transformations. Quaternion Bessel-Fourier moments for color image reconstruction and recognition.An accurate estimation method for rotation angle throughout quaternion moments.The invariant descriptor using the moment magnitudes and phase coefficients.

Color image zero-watermarking based on SVD and visual cryptography in DWT domain

Graphical abstractDisplay Omitted A robust blind color image watermarking scheme is proposed.An improved uniform log-polar mapping method is used to resist geometric attacks.A dual watermark is embedded into quaternion discrete Fourier transform domain.The holistic embedding manner rather than block-based achieves better robustness. 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 part of mid-frequency QDFT coefficients using quantization index modulation. The latter is used to resynchronize the watermark after the watermarked image has been attacked, making the scheme resistant to geometric attacks. In particular, the IULPM allows for greater accuracy when estimating the rotation angle of a geometric attack. At the same time, the watermark embedding employs the image holistically, rather than in a block pattern. Experimental results demonstrate that the proposed scheme achieves better performance of robustness against both common signal operations and geometric attacks compared to other existing schemes.

Effective and Efficient Image Copy Detection with Resistance to Arbitrary Rotation

The principal component analysis network (PCANet), which is one of the recently proposed deep learning architectures, achieves the state-of-the-art classification accuracy in various datasets and reveals a simple baseline for deep learning networks. However, the performance of PCANet may be degraded when dealing with color images due to the fact that the architecture of PCANet cannot properly utilize the spatial relationship between each color channel in three dimensional color image. In this paper, a quaternion principal component analysis network (QPCANet), which extends PCANet by using quaternion theory, is proposed for color image classification. Compared to PCANet, the proposed QPCANet takes into account the spatial distribution information of RGB channels in color images and ensures larger amount of intra-class invariance by using quaternion domain representation for color images. Experiments conducted on different color image datasets such as UC Merced Land Use, Georgia Tech face, CURet and Caltech-101 have revealed that the proposed QPCANet generally achieves higher classification accuracy than PCANet in color image classification task. The experimental results also verify that QPCANet has much better rotation invariance than PCANet when color image dataset contains lots of rotation information and demonstrate even a simple one-layer QPCANet may obtain satisfactory accuracy when compared with two-layer PCANet.

DIBR-Synthesized Image Quality Assessment via Statistics of Edge Intensity and Orientation

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 filter. The optimal weights are obtained by using the method of Lagrange multipliers. The authors quaternion optimal weights non-local means filter is given by the weighted means of the observed quaternion representation using the optimal weights. Experiments on commonly used images are provided to illustrate the efficiency of the proposed filter.