基于四元数小波变换的清晰度评价

针对当前各种图像清晰度评价方法在清晰度判别过程中单调性和区分度不够以及适用范围较小的问题,提出了一种基于四元数小波变换（QWT）幅值与相位的图像清晰度评价方法。该算法通过四元数小波变换将图像从空间域变换到频率域,对得到的四元数小波变换系数进一步计算之后获得低频子带与高频子带的幅值与相位信息,求得低频子带幅值各方向的梯度之后与对应方向的相位相乘求和,最终得到两个清晰度指标值。采用该算法与多种现有算法对不同内容的图像、不同程度模糊的图像以及含有不同程度噪声的图像进行清晰度评价实验：相对于现有算法,所提算法在对上述多种图像的清晰度评价中都保持着很好的单调性与区分度。实验结果表明,所提算法不但克服了现有算法在单调性与区分度上的不足,而且所提清晰度评价指标可以应用在图像处理中。     

Image Morphing in Frequency Domain

Image morphing is often used in film and television industry to create synthetic visual effects by smooth transformation of one object into another. Based upon spatial representation of images, several image morphing techniques have been proposed. Simple spatial techniques, for example cross-dissolve, suffer from lack of smooth transformation while better quality techniques, like mesh warping or field warping, have significant computational complexity. In this paper we present a simple but good quality image morphing technique based upon frequency domain representation of images. Transformation from a source image to a target image takes place by mixing low frequencies of the source image and high frequencies of the target image in varying proportions. The proposed technique has been applied to a wide variety of images. The resulting sequence of images are better in visual quality and faster in execution time.     

Contrast enhancement for cephalometric images using wavelet-based modified adaptive histogram equalization

Cephalometric images usually have low contrast. The existing techniques for automatic cephalometric analysis usually use histogram equalization for image enhancement. This technique has the advantage of being fully automatic and nonlinear. However, it suffers from spikes, excessive enhancement, and lack of brightness preservation. The proposed technique is an adaptive histogram equalization technique that uses wavelet based gradient histograms. This paper compares its performance with two traditional techniques, three histogram modification based techniques, and two wavelet based techniques. Forty digital and scanned cephalograms are used to conduct tests. In addition to visual histograms and intensity profiles, the proposed method is compared in terms of eight quantitative measures. The various measures are applied to analyze the results in terms of contrast enhancement (EME, CNR), brightness preservation (AMBE), edge conservation and enhancement (H, TEN), preservation of image structures and non-addition distortion (MSSIM, SVD-M). The proposed method gives good contrast enhancement, with better brightness preservation without losing edge information and with the minimum addition of distortions to the enhanced cephalometric images.     

基于全局和局部特征的无参考夜间图像质量评价

针对夜间图像光线暗、特征不易提取的问题,提出一种基于全局和局部特征的无参考夜间图像质量评价方法.首先,利用等高线原理将图像分为亮区域和暗区域2部分,将亮区域占整幅图的比例作为特征1;其次,提取夜间图像的全局亮度信息并将其作为特征2;再次,结合微分算子法求得图像的边缘图作为特征3;然后,将夜间图像从RGB转换到HSI,提取色度、饱和度和亮度分量并将其分别作为特征4、特征5和特征6;最后,结合上述特征通过BP神经网络建立评价模型来评价夜间图像的质量.在公开数据库上的测试结果表明,所提方法与主观分数具有更好的一致性,并且优于现有的图像质量评价方法.     

Expert system for low frequency adaptive image watermarking: Using psychological experiments on human image perception

This paper proposes a new system for low frequency adaptive image watermarking based on the statistical data from psychological experiments on human image perception. The new approach can lead to a reduction of degrading the subjective image quality that often occurs when watermark is embedded into low frequency area. In order to reduce the degrading of image quality, the new approach determines the strength of watermark according to local image characteristics such as brightness and contrast. By conducting a behavioral experiment on human image fidelity based on the psycho-visual image association technique, we were able to infer the relationship between the watermark strength and the different levels of image brightness and contrast information. The exact watermark is extracted according to edge characteristics by adopting a so-called edge mask that exploits the coefficients of subbands in the subsampled discrete wavelet transform images. Thus, our new approach does not require original images for watermark. We also show the new approach is practically validated against some standard images.     

A human perception inspired quality metric for image fusion based on regional information

Comparative evaluation of fused images is a critical step to evaluate the relative performance of different image fusion algorithms. Human visual inspection is often used to assess the quality of fused images. In this paper, we propose some variants of a new image quality metric based on the human vision system (HVS). The proposed measures evaluate the quality of a fused image by comparing its visual differences with the source images and require no knowledge of the ground truth. First, the images are divided into different local regions. These regional images are then transformed to the frequency domain. Second, the difference between the local regional images in frequency domain is weighted with a human contrast sensitivity function (CSF). The quality of a local regional image is obtained by computing the MSE of the weighted difference images obtained from the fused regional image and source regional images. Finally, the quality of a fused image is the weighted summation of the local regional images quality measures. Our experimental results show that these metrics are consistent with perceptually obtained results.     

Moment-preserving piecewise linear approximations of signals andimages

Approximation techniques are an important aspect of digital signal and image processing. Many lossy signal compression procedures such as the Fourier transform and discrete cosine transform are based on the idea that a signal can be represented by a small number of transformed coefficients which are an approximation of the original. Existing approximation techniques approach this problem in either a time/spatial domain or transform domain, but not both. This paper briefly reviews various existing approximation techniques. Subsequently, we present a new strategy to obtain an approximation fˆ(x) of f(x) in such a way that it is reasonably close to the original function in the domain of the variable x, and exactly preserves some properties of the transformed domain. In this particular case, the properties of the transformed values that are preserved are geometric moments of the original function. The proposed technique has been applied to single-variable functions, two-dimensional planar curves, and two-dimensional images, and the results obtained are demonstrative     

基于2D sine Logistic混沌映射的医学图像频域加密算法

针对现有医学图像加密算法在加密效率和安全性上的不足,提出一种基于2D sine logistic混沌映射的医学图像小波域加密算法。算法首先利用整数小波变换将医学图像从空域转换为频域,充分打破像素间的相关性;其次,利用2D sine logistic混沌映射生成混沌序列,选取三级小波分级的低频系数LL3进行扩散和置乱加密,提高加密效率;并且将二级小波分解的中高频系数HL2和LH2进行扩散加密,解决加密图像中存在的明显轮廓问题;最后将加密后的小波系数进行小波逆变换得到加密图像。实验仿真结果表明,算法具有高安全性和加密效率,与现有空域方法相比,加密时间约为1/40;与现有频域方法相比,在保证加密效率情况下具有更好的加密图像隐蔽性。     

An effective local contrast enhancement technique by blending of local statistics and Genetic Algorithm

Nowadays, Image enhancement finds enormous image processing applications, which are related to practical situations, Contrast enhancement is one among the different image enhancement techniques that intends to improve the image visibility. Though several works for local contrast enhancement are available in the literature, the effectiveness remains an issue and the enhancement performance needs to be improved. In this paper, a local contrast enhancement technique is proposed for both gray scale images and RGB color images. The proposed technique is comprised of two stages of enhancement, namely, local statistics-based image enhancement and Genetic Algorithm based local contrast enhancement. The former stage is a pre-enhancement stage and the later is the major stage of enhancement. In the former stage, the image is processed in window basis and the local statistics of the image is obtained. Based on the local statistics, the image is enhanced. In the later stage, the window based operation is performed over the preenhanced image and the local contrast is enhanced. The Genetic Algorithm aids in searching of an optimal contrast factor, which plays vital role in the contrast enhancement. The technique is evaluated with both gray scale images as well as RGB color images and performance is compared with the existing contrast enhancement techniques.     

基于Retinex理论的多曝光图像融合算法

多曝光图像融合技术是将一组场景相同但曝光程度不同的图像序列直接融合成为一幅含有更多场景细节信息的高质量图像。针对现有算法局部对比度差和色彩失真的问题，结合Retinex理论模型提出了一种新的多曝光图像融合算法。首先，基于Retinex理论模型，利用光照估计算法将曝光序列图像分为入射光分量序列和反射光分量序列，然后分别采用不同的融合方法对这两组序列进行处理。对于入射光分量，要保证场景的全局亮度的变化特性并且削弱过曝光和欠曝光区域的影响；而对于反射光分量，要采用适度曝光的评价参数来更好地保留场景的色彩及细节信息。分别从主观和客观两方面对所提算法进行了分析。实验结果表明，同传统基于图像域合成的算法相比，该算法在结构相似度（SSIM）上平均提升了1.7%，另外在图像色彩和局部细节上的处理效果更好。     

Tile selection method based on error minimization for photomosaic image creation

Photomosaic images are composite images composed of many small images called tiles. In its overall visual effect, a photomosaic image is similar to the target image, and photomosaics are also called “montage art”. Noisy blocks and the loss of local information are the major obstacles in most methods or programs that create photomosaic images. To solve these problems and generate a photomosaic image in this study, we propose a tile selection method based on errorminimization. A photomosaic image can be generated by partitioning the target image in a rectangular pattern, selecting appropriate tile images, and then adding them with a weight coefficient. Based on the principles of montage art, the quality of the generated photomosaic image can be evaluated by both global and local error. Under the proposed framework, via an error function analysis, the results show that selecting a tile image using a global minimum distance minimizes both the global error and the local error simultaneously. Moreover, the weight coefficient of the image superposition can be used to adjust the ratio of the global and local errors. Finally, to verify the proposed method, we built a new photomosaic creation dataset during this study. The experimental results show that the proposed method achieves a lowmean absolute error and that the generated photomosaic images have a more artistic effect than do the existing approaches.     

A two-dimensional discrete fractional Fourier transform-based pansharpening scheme

In this paper, a new approach for fusion of multi-spectral (MS) and panchromatic (Pan) images based on 2D-discrete fractional Fourier transform (2D-DFRFT) is proposed. The proposed technique is closer in approach to the other filtering-based pansharpening schemes existing in the literature. In the proposed method histogram equalized Pan image is transformed using the 2D-DFRFT and further used to generate the pansharpened image using appropriate pansharpening rule. The angle parameters associated with the 2D-DFRFT provide additional degrees of freedom which are optimized by single-objective particle swarm optimization (PSO) algorithm for finding better pansharpening results. Simulation results of the proposed technique carried out in MATLAB are presented for IKONOS and GeoEye-1 satellite images and compared with existing fusion methods in terms of both visual perception and objective metrics such as Q-index (Q4), Spectral Angle Mapper (SAM), relative dimensionless global error (ERGAS) and quality with-no reference (QNR). It is observed that the proposed pansharpening scheme provides improved spectral and spatial quality as compared with the existing schemes. The effects of aliasing and mis-registration errors on the proposed method are also investigated and compared with existing pansharpening methods. It is seen that the proposed method is robust against aliasing and mis-registration errors.     

结合频域信息与对抗网络的虚假图像检测

随着深度学习方法的发展, 深度造假(Deepfake)技术越发成熟。大量近似真实自然的图像涌入人们的生活, 在满足个人娱乐兴趣的同时, Deepfake技术的滥用对个人隐私、经济市场乃至国家安全构成了潜在威胁。因此,针对虚假图像的检测方法亟待研究。现有的虚假图像检测技术大多存在准确率低、泛化性差、鲁棒性不足的问题, 因此, 本文从Deepfake技术的图像生成机制出发, 对生成的虚假图像存在缺陷进行分析, 并提出了一种基于生成对抗网络的虚假图像检测模型。该模型利用离散傅里叶变换方法将图像从图像域转换到频域, 并将U-Net结构和谱归一化引入鉴别器; 利用生成对抗网络优异的特征学习和提取能力, 实现了虚假图像的模式分类。此外, 一种新颖的复合损失函数被提出, 以增强模型检测性能。提出的方法分别在7个单独数据集和1个混合数据集上进行实验验证, 并采用3种实验指标进行模型性能分析。本文方法在单独数据集上最高可达到100%准确率, 最低准确率也可达88.53%; 模型检测召回率, 精确率和F1分数平均分别可达98.17%, 98.25%, 98.19%。此外, 无论是在混合数据集, 还是在模型未知的跨数据集上, 提出方法都能获得良好的模型检测性能。即使在图像压缩的情况下, 本文方法仍然具有较强的鲁棒性。实验与理论结果表明, 与现有先进的虚假图像检测方法相比, 本文方法是一种有效且具有良好泛化性和鲁棒性的虚假图像检测方法。     

结合内外先验知识的低照度图像增强与去噪算法

目的 现有大多数低照度图像增强算法会放大噪声,且用于极低照度图像时会出现亮度提升不足、色彩失真等问题。为此,提出一种基于Retinex(retina cortex)的增强与去噪方法。方法 为了增强极低照度图像,首先利用暗通道先验原理估计场景的全局光照,若光照低于0.5,对图像进行初始光照校正;其次,提出一种Retinex顺序分解模型,使低照度图像中的噪声均体现在反射分量中,基于分解结果,利用Gamma校正求取增强后的噪声图像;最后,提出一种基于内外双重互补先验约束的去噪机制,利用非局部自相似性原理为反射分量构建内部先验约束,基于深度学习,为增强后的噪声图像构建外部先验约束,使内外约束相互制约。结果 将本文算法与6种算法比较,在140幅普通低照度图像和162幅极低照度图像上（有正常曝光参考图像）进行主观视觉和客观指标评价比较,结果显示本文方法在亮度提升、色彩保真及去噪方面均有明显优势,对于普通低照度图像,BTMQI(blind tone-mapped quality index)和NIQE(natural image quality evaluator)指标均取得次优值,对于极低照度图像...     

New image matching technique based on hyper-vectorisation of grey level sliced binary image

Image matching is an important area of research in the field of artificial intelligence, machine vision and visual navigation. A new image matching scheme in which grey scale images are quantised to form sub-band binary images is presented. The information in the binary images is then signaturised and the signatures are sorted as per significance. These sorted signatures are then normalised to transform the represented image pictorial features in the form of a hyper-dimensional vector cluster. For the image matching, the two clusters from both the images are compared in the transformed domain. This comparison yields efficient results directly in the image spatial domain avoiding the need of image inverse transformation for the interpretation of results. As compared with the conventional techniques, this comparison avoids the wide range of square error calculations all over the image. It also directly guides the solution in an iterative fashion to converge towards the true match point. The process of signaturisation is based on image local features and is moulded in a way to support the scale and rotation-invariant template matching as well. A four-dimensional solution population scheme has also been presented with an associated matching confidence factor. This factor helps in terminating the iterations when the essential matching conditions have been achieved. The proposed scheme gives robust and fast results for normal, scaled and rotated templates. Speed comparison with older techniques shows the computational viability of this new technique and its much lesser dependence on image size. The method also shows noise immunity at 30 dB additive white Gaussian noise and impulsive noise.     

面向GAN生成图像的被动取证及反取证技术综述

生成对抗网络（generative adversarial network,GAN）快速发展,并在图像生成和图像编辑技术等多个方面取得成功应用。然而,若将上述技术用于伪造身份或制作虚假新闻,则会造成严重的安全隐患。多媒体取证领域的研究者面向GAN生成图像已提出了多种被动取证与反取证方法,但现阶段缺乏相关系统性综述。针对上述问题,本文首先阐述本领域的研究背景和研究意义,然后分析自然图像采集与GAN图像生成过程的区别。根据上述理论基础,详细介绍了现有GAN生成图像的被动取证技术,包括：GAN生成图像检测算法,GAN模型溯源算法和其他相关取证问题。此外,针对不同应用场景介绍基于GAN的反取证技术。最后,通过实验分析当前GAN生成图像被动取证技术所面临的挑战。本文根据对现有技术从理论和实验两方面的分析得到以下结论：现阶段,GAN生成图像的被动取证技术已在空间域和频率域形成了不同技术路线,较好地解决了简单场景下的相关取证问题。针对常见取证痕迹,基于GAN的反取证技术已能够进行有效隐藏。然而,该领域研究仍存在诸多局限：1）取证与反取证技术的可解释性不足;2）取证技术鲁棒性和泛化性较弱;3）反取证技术缺乏多特征域协同的抗分析能力等。上述问题和挑战还需要研究人员继续深入探索。     

基于DCT和SVD的图像哈希水印算法

为使频域水印技术更好地应用于数字图像的版权保护,提出一种基于离散余弦变换和奇异值分解相结合的图像哈希水印算法。利用DWT变换提取载体图像的低频系数矩阵构造水印;对载体图像进行分块DCT变换,提取每个子块的低频系数;对低频系数所组成的矩阵进行SVD变换,在对角阵上嵌入水印;对频域系数进行逆变换得到含水印图。将已有算法和当前所提出的算法进行对比,实验结果表明,所提水印算法具有良好的不可感知性和鲁棒性。     

Texture Repairing by Unified Low Rank Optimization

In this paper, we show how to harness both low-rank and sparse structures in regular or near-regular textures for image completion. Our method is based on a unified formulation for both random and contiguous corruption. In addition to the low rank property of texture, the algorithm also uses the sparse assumption of the natural image: because the natural image is piecewise smooth, it is sparse in certain transformed domain (such as Fourier or wavelet transform). We combine low-rank and sparsity properties of the texture image together in the proposed algorithm. Our algorithm based on convex optimization can automatically and correctly repair the global structure of a corrupted texture, even without precise information about the regions to be completed. This algorithm integrates texture rectification and repairing into one optimization problem. Through extensive simulations, we show our method can complete and repair textures corrupted by errors with both random and contiguous supports better than existing low-rank matrix recovery methods. Our method demonstrates significant advantage over local patch based texture synthesis techniques in dealing with large corruption, non-uniform texture, and large perspective deformation.     

Local Image Reconstruction and Subpixel Restoration Algorithms

This paper introduces a new class of reconstruction algorithms that are fundamentally different from traditional approaches. We deviate from the standard practice that treats images as point samples. In this work, image values are treated as area samples generated by nonoverlapping integrators. This is consistent with the image formation process, particularly for CCD and CID cameras. We show that superior results are obtained by formulating reconstruction as a two-stage process: image restoration followed by application of the point spread function (PSF) of the imaging sensor. By coupling the PSF to the reconstruction process, we satisfy a more intuitive fidelity measure of accuracy that is based on the physical limitations of the sensor. Efficient local techniques for image restoration are derived to invert the effects of the PSF and estimate the underlying image that passed through the sensor. The reconstruction algorithms derived herein are local methods that compare favorably to cubic convolution, a well-known local technique, and they even rival global algorithms such as interpolating cubic splines. Evaluations are made by comparing their passband and stopband performances in the frequency domain, as well as by direct inspection of the resulting images in the spatial domain. A secondary advantage of the algorithms derived with this approach is that they satisfy an imaging-consistency property. This means that they exactly reconstruct the image for some function in the given class of functions. Their error can be shown to be at most twice that of the "optimal" algorithm for a wide range of optimality constraints.