基于多梯度的区域自适应去马赛克算法*

针对Bayer格式的彩色滤波阵列,鉴于传统去马赛克算法存在图像纹理和边缘模糊的现象,提出了一种基于多梯度的区域自适应去马赛克算法.该算法利用区域水平和垂直方向的梯度组算子来判断区域平滑程度和插值方向,从而有效地恢复绿色分量,同时根据色度差恒定原理来还原另外两个丢失的颜色信息.实验结果显示,相对于传统插值算法提高了图像的峰值信噪比,锐化了图像的纹理和边缘,改善了图像的视觉质量.     

基于生成对抗网络的CFA图像去马赛克算法

在单传感器数码相机图像采集系统的彩色滤波阵列中,每个像素仅捕获单一颜色分量,并且在彩色图像重构过程中图像边缘等高频区域的伪影现象尤为明显。提出一种基于生成对抗网络的图像去马赛克算法,通过设计生成对抗网络的生成器、鉴别器和网络损失函数增强学习图像高频信息的能力,其中使用的生成器为具有残差稠密块和远程跳跃连接的深层残差稠密网络,鉴别器由一系列堆叠的卷积单元构成,并且结合对抗性损失、像素损失以及特征感知损失改进网络损失函数,提升网络整体性能。数值实验结果表明,与传统去马赛克算法相比,该算法能更有效减少图像边缘的伪影现象并恢复图像高频信息,生成逼真的重建图像。     

自适应非局部3维全变分彩色图像去噪

目的 许多彩色图像去噪算法未充分利用图像局部和非局部的相似性信息,并且忽略了真实噪声在彩色图像不同区域内分布的差异,对不同图像块和不同颜色通道都进行同等处理,导致去噪图像中同时出现过平滑和欠平滑现象。针对这些问题,本文提出一种自适应非局部3维全变分去噪算法。方法 利用一个非局部3维全变分正则项获取彩色图像块内和块间的相似性信息,同时在优化模型的保真项内嵌入一个自适应权重矩阵,该权重矩阵可以根据每次迭代得到的中间去噪结果的剩余噪声来调整算法在每个图像块、每个颜色分量以及每次迭代中的去噪强度。结果 通过不同的高斯噪声添加方式得到两个彩色噪声图像数据集。将本文算法与其他6个基于全变分的算法进行比较,采用峰值信噪比(peak signal-to-noise ratio,PSNR)和结构相似性(structural similarity, SSIM)作为客观评价指标。相比于对比算法,本文算法在两个噪声图像数据集上的平均PSNR和SSIM分别提高了0.161.76 dB和0.12%6.13%,并获得了更好的图像视觉效果。结论 本文去噪算法不仅更好地兼顾了去噪与保边功能,而且提升了稳定性和鲁棒性,显示了在实际图像去噪中的应用潜力。     

基于总体最小二乘法的图像降噪

来自图像传感器的数字图像会受到各种噪声的干扰,其中主要包括加性噪声、乘性噪声和混合噪声。乘性噪声随信号幅度改变而改变,没有理想的去除方法。为此,运用基于总体最小二乘法的图像估计降噪方法,研究图像块尺寸选取对降噪性能的影响,分析成像系统中去马赛克环节影响噪声传播的内在规律,并通过比较实验给出总体最小二乘法降噪的性能优势。     

基于谱相关性的数字图像真伪鉴别

如何区分逼真的计算机生成图像和真实的自然图像,是数字图像取证领域的一个重要研究方向。提出了一种基于谱间相关性的图像真伪鉴别算法。在单CCD数码成像过程中,每个像素只采集单一颜色值,缺失的颜色值通过颜色滤波阵列插值获得,而基于颜色滤波阵列插值的去马赛克方法会引起彩色图像三个颜色分量之间较高的谱间相关性。算法利用小波变换和标准互相关系数提取颜色组件谱间相关性,作为区分和识别特征。通过在标准图像库上的实验测试,表明所提取特征有效捕获了二类图像的差别,并具有较高的检测率。     

一种基于多通道联合估计的非局部均值彩色图像去噪方法

提出一种基于多通道联合估计的非局部均值彩色图像去噪方法,包括彩色通道联合去噪和彩色通道融合去噪两个步骤：在彩色通道联合去噪步骤,采用经典的彩色图像非局部均值去噪算法对噪声彩色图像去噪,得到预去噪图像作为彩色通道融合去噪步骤的输入;在彩色通道融合去噪步骤,采用广义多通道非局部均值去噪算法对预去噪图像再次去噪,去噪过程应用预去噪图像三通道高频成分的相似性。实验结果表明,与其他经典彩色图像去噪方法相比,本文方法在主观和客观上均具有竞争性。     

Bayer阵列图像去马赛克算法综述

Bayer阵列图像去马赛克技术是对稀疏采样的Bayer阵列图像进行RGB信息重建,图像重建质量是成像设备评价的重要因素之一,同时也对其他计算机视觉任务(如图像分割、人脸识别)产生影响。随着深度学习方法的快速发展,图像去马赛克领域提出了多种高性能算法。为了便于研究者更全面了解图像去马赛克算法的原理和研究进展,本文对该领域的经典算法和深度学习算法进行综述。首先对Bayer采样阵列原理和图像去马赛克技术进行概述。然后将现有方法分为传统方法和基于深度学习方法两类进行总结,同时根据去马赛克任务是否具有独立性,将深度学习方法分为独立去马赛克任务和联合去马赛克任务两类,分析不同方法的原理和优缺点,重点阐述基于深度学习的去马赛克方法的网络结构和重建机理,介绍去马赛克领域常用的公共数据集和性能评价指标,并对图像去马赛克相关实验进行分析对比。最后,围绕网络深度、运算效率和实用性等方面分析了现阶段图像去马赛克技术面临的挑战及未来发展方向。目前,基于深度学习的图像去马赛克方法已成为主流发展方向,但仍然存在计算成本较高、实际应用性不强等问题。因此,如何开发出重建精度高、处理时间短以及实用性强的图像去马赛克方法,是该领域未来重要的研究方向。     

基于数学形态学的彩色噪声图像边缘检测算法*

针对已有的数学形态学边缘检测算法对彩色噪声图像检测到的彩色边缘信息不够完整、清晰,提出了一种基于HSI色彩空间的多尺度多结构元的数学形态学边缘检测算法,采用以尺度和结构两个单位元素进行横向和纵向的拓展,以面的形式对彩色噪声图像进行全面的边缘检测。基于这种理念分别对H和S两个携带颜色信息的分量进行边缘检测,最后将两分量的边缘信息通过加权合成得到彩色图像的彩色边缘。实验证明,该算法的去噪效果明显,得到的彩色边缘轮廓清晰、细节丰富,对彩色边缘的提取具有可行性和有效性。     

基于联合滤波的聚类稀疏表示图像去噪算法

针对非局部均值去噪算法中噪声对结构聚类影响的问题,提出了一种基于联合滤波预处理的聚类稀疏表示图像去噪算法。利用维纳滤波和巴特沃斯滤波联合滤波处理提取含噪图像中的高频分量,同时减小了噪声对聚类的影响;利用非局部均值去噪的思想将高频图像块进行聚类,每一类图像块单独进行字典学习,增强字典的自适应性;利用多循环字典更新的K-SVD算法进行类内字典学习,增强字典的描述能力。实验结果表明,与传统的K-SVD算法相比,该算法能有效保留图像的结构信息,并且提升了图像的去噪效果。     

基于PCA的图像小波去噪方法

目前使用的各种小波去噪方法基本上都是建立在对噪声方差精确估计的基础上，而对噪声方差的精确估计是很困难的．提出了一种采用主分量分析（PCA）提取小波系数的主要特征，通过对小波域中噪声能量的估计来实现去噪的新方法．首先利用PCA对小波高频子带进行局部特征提取；然后以主分量对小波系数进行重建的平均能量作为局部噪声能量的估计；将原小波系数的能量减去噪声能量，就得到去噪后的小波系数；最后用小波逆变换对剔除噪声分量后的小波系数进行恢复得到去噪后的图像．本文算法无需对噪声方差进行估计，因而更具实用价值．本文算法与“软阈值”、“硬阈值”去噪方法相比，峰值信噪比（PNNR）提高了2～8dB．实验证实了本文算法良好的去噪性能。     

Deep Joint Design of Color Filter Arrays and Demosaicing

We present a convolutional neural network architecture for performing joint design of color filter array (CFA) patterns and demosaicing. Our generic model allows the training of CFAs of arbitrary sizes, optimizing each color filter over the entire RGB color space. The patterns and algorithms produced by our method provide high‐quality color reconstructions. We demonstrate the effectiveness of our approach by showing that its results achieve higher PSNR than the ones obtained with state‐of‐the‐art techniques on all standard demosaicing datasets, both for noise‐free and noisy scenarios. Our method can also be used to obtain demosaicing strategies for pre‐defined CFAs, such as the Bayer pattern, for which our results also surpass even the demosaicing algorithms specifically designed for such a pattern.     

彩色图像矢量滤波技术综述

彩色图像滤波是彩色图像处理的最基本的研究领域之一。彩色图像滤波技术可以分成标量滤波法和矢量滤波法两大类。其中,标量滤波法只是早期的滤波方法。大量的研究表明,矢量滤波法比标量滤波法更加有效,因为它更能保护彩色图像的光谱特性。为使人们对彩色图像矢量滤波技术及其应用有个系统的了解,该文首先全面地总结了彩色图像矢量滤波的基本理论和方法,并跟踪该领域的最新进展,同时分析介绍了彩色图像矢量滤波技术的一些典型应用;然后对彩色图像矢量滤波技术进行了分类,并对每种类型的滤波算法中经典和目前最常用的算法做了详细的介绍和阐述;接着结合笔者对该领域的研究,提出了一些新的研究方法;最后,对于一些有代表性、经常使用的矢量滤波算法,以冲击噪声为例,给出了其视觉上的滤波效果和客观的评估数据。     

Automatic estimation and removal of noise from a single image

Image denoising algorithms often assume an additive white Gaussian noise (AWGN) process that is independent of the actual RGB values. Such approaches are not fully automatic and cannot effectively remove color noise produced by todays CCD digital camera. In this paper, we propose a unified framework for two tasks: automatic estimation and removal of color noise from a single image using piecewise smooth image models. We introduce the noise level function (NLF), which is a continuous function describing the noise level as a function of image brightness. We then estimate an upper bound of the real noise level function by fitting a lower envelope to the standard deviations of per-segment image variances. For denoising, the chrominance of color noise is significantly removed by projecting pixel values onto a line fit to the RGB values in each segment. Then, a Gaussian conditional random field (GCRF) is constructed to obtain the underlying clean image from the noisy input. Extensive experiments are conducted to test the proposed algorithm, which is shown to outperform state-of-the-art denoising algorithms.     

RETRACTED ARTICLE: Dense stereovision using mono-CCD color cameras

Most stereo algorithms are based only on an analysis of the luminance information. However, with advances in camera technology, in addition to the fact that color information can robustly improve matching, color stereovision is receiving more and more attention. Color stereovision setups are usually based on single-sensor cameras which provide color filter array (CFA) images. In these images, a single color component is sampled at each pixel rather than the three required components red, green, and blue (RGB). We show that standard demosaicing techniques, which are used to interpolate the missing components, are not well adapted when the resulting color pixels are matched in order to estimate image disparities. In order to avoid this problem while exploiting color information, we propose a new matching system designed for dense stereovision based on pairs of CFA images.     

Markov random field processing for color demosaicing

In digital color imaging, color filter arrays (CFAs) are obtained from single-chip cameras in the form of sampled spectral components (red, green and blue) in an interleaved fashion. Color demosaicing is the process of interpolating these CFAs into dense pixel maps for each spectral component. There are different interpolation techniques for color demosaicing operations. These approaches have their limitations regarding the improvement of the quality of the images. Particularly, they perform poorly in recovering the edges of the images. In this work we have applied Markov random field (MRF) processing over these roughly interpolated images obtained by the existing techniques to improve the quality of the reconstruction. We have observed that the processing improves the image quality in many cases. Particularly, the edges of the reconstructed images are significantly enhanced using MRF processing.     

Edge-preserving color image denoising through tensor voting

This paper presents a new method for edge-preserving color image denoising based on the tensor voting framework, a robust perceptual grouping technique used to extract salient information from noisy data. The tensor voting framework is adapted to encode color information through tensors in order to propagate them in a neighborhood by using a specific voting process. This voting process is specifically designed for edge-preserving color image denoising by taking into account perceptual color differences, region uniformity and edginess according to a set of intuitive perceptual criteria. Perceptual color differences are estimated by means of an optimized version of the CIEDE2000 formula, while uniformity and edginess are estimated by means of saliency maps obtained from the tensor voting process. Measurements of removed noise, edge preservation and undesirable introduced artifacts, additionally to visual inspection, show that the proposed method has a better performance than the state-of-the-art image denoising algorithms for images contaminated with CCD camera noise.     

Evaluation of visual saliency analysis algorithms in noisy images

Most existing visual saliency analysis algorithms assume that the input image is clean and does not have any disturbances. However, this situation is not always the case. In this paper, we provide an extensive evaluation of visual saliency analysis algorithms in noisy images. We analyze the noise immunity of saliency analysis algorithms by evaluating the performances of the algorithms in noisy images with increasing noise scales and by studying the effects of applying different denoising methods before performing saliency analysis. We use 10 state-of-the-art saliency analysis algorithms and 7 typical image denoising methods on 4 eye fixation datasets and 2 salient object detection datasets. Our experiments show that the performances of saliency analysis algorithms decrease with increasing image noise scales in general. An exception is that the nonlinear features (NF) integrated algorithm shows good noise immunity. We also find that image denoising methods can greatly improve the noise immunity of the algorithms. Our results show that the combination of NF and Median denoising method works best on eye fixation datasets and the combination of saliency optimization (SO) and color block-matching and 3D filtering (C-BM3D) method works best on salient object detection datasets. The combination of SO and Average denoising method works best for applications wherein time efficiency is a major concern for both types of datasets.     

I3D: a new dataset for testing denoising and demosaicing algorithms

Multimedia Tools and Applications - In this paper we present a dataset of images to test the performance of image processing algorithms, in particular demosaicing and denoising methods. Despite the...