基于深层特征学习的无参考立体图像质量评价

提出了一种基于深层特征学习的无参考(NR)立体图 像质量评价方 法。与传统人工提取图像特征不同,采用卷积神经网络(CNN)自动提取图像特征,评价过程 分为训练和 测试两阶段。在训练阶段,将图像分块训练CNN网络,利用CNN提取图像块特征,并结合不同 的整合方式 得到图像的全局特征,通过支持向量回归(SVR)建立主观质量与全局特征的回归模型;在测 试阶段,由已训练的CNN网 络和回归模型,得到左右图像和独眼图的质量。最后,根据人眼双目视觉特性融合左图像、 右图像和独眼 图的质量,得到立体图像质量。本文方法在LIVE-I和LIVE-II数据库上的Spearman等级系 数(SROCC)分别达 到了0.94,评价结果准确,与人眼的主 观感受一致。     

基于视觉显著性的立体图像质量评价方法

人眼的视觉显著性能够影响图像质量评价,结合平面显著性和中央偏移因子,提出基于视觉显著性的立体图像质量评价算法(VS-SSIM);同时考虑中央偏移因子的影响,提出基于中央偏移的结构相似度算法(CB-SSIM)。实验结果表明,中央偏移因子和视觉显著性均能提高立体图像质量评价算法;相比CB-SSIM,VS-SSIM算法与主观评价结果的相关性更高。     

基于显著性目标分类的无参考模糊图像质量评价

近年来有大量关于无参考模糊图像质量评价的研究,但是目前很多方法都忽略了图像内容对评价结果的影响.针对纯背景的无显著性目标图像和含背景的显著性目标图像的模糊评价方式是不同的,基于人眼注意力机制,前者侧重于图像的整体模糊,而后者更侧重于图像的局部细节模糊.整体模糊指的是图像整体内容的锐度信息,局部细节模糊指的是图像不同位置的局部锐度信息,二者可以将视觉显著性和图像内容更好地结合起来.针对上述问题,提出了一种基于显著性目标分类的无参考模糊图像质量评价方法.首先提出了一种基于显著性检测的目标分类算法,对待评价图像进行显著性目标分类,然后根据分类结果提取其局部模糊特征和全局模糊特征,最后对这两个特征进行融合得到最终的质量评估分数.实验结果表明,该算法不仅在BLUR数据库上取得最优的评价效果,同时在LIVE、CSIQ和TID2013数据库上也有较好的结果,具有很好的鲁棒性.此外,本文算法在各数据库中也表现出了优异的统计性能.     

基于视觉感知特性的红外图像质量评价方法

根据红外成像特点,设计了一种基于视觉感知特性的红外图像质量评价算法。该算法结合人眼视觉和红外图像的结构信息对图像的失真程度进行描述,通过提取图像的边缘特征、对比度特征,然后利用视觉显著模型对特征进行差异融合,从而实现对红外失真图像的质量评测。实现结果表明,本文方法可对失真红外图像进行有效评价,与传统方法相较,此评价指标与人眼主观感知更一致。     

基于双目特征联合的无参考立体图像质量评价

通过模拟人类视觉系统(HVS)的双目视觉行为,提 出一种基于双目特征联合的无参考立 体图像质量评价(NR-SIQA)方法。首先分析立体视觉感知中的双目联合行为,提出 可应用于立体图像质量预 测的双目联合模型;然后采用学习和统计分析的方法,分别提取局部和全局特征并联合作 为感知特征; 最后采用机器学习算法,建立特征和质量的关系模型,并结合基于特征的双目联合模型预测 立体图像质量。实验结果表明,本文方法在对称立体图像库上的Pearson线性相关系数(PLCC)和Spearman等级系数(SRCC)高于0.93,在非对称库上高于0.87,优 于现有评价方法。     

基于血管分割的视网膜图像无参考质量评价方法

为了研究不同失真类型和不同失真程度对血管分割 的影 响,本文将图像的失真类型和失真程度量化为图像血管分割精确度,由于现有公开库中包含 血管分割标签 的图像中均为低失真甚至无失真图像,因此本文构建了一个视网膜失真图像数据库,共包含 2种失真类型, 每种失真类型的图像均有8个等级的失真程度,共552幅视网膜失真图像,并将每幅失真图 像对应的血管 分割精确度作为该图像的标签。此外,本文提出了一种基于血管分割方法的视网膜图像无参 考质量评价方 法,通过提取视网膜图像的像素值统计特征、图像纹理特征以及血管形状特征得到最终视网 膜图像的质量。 在提出的数据库上测试结果显示,皮尔逊线性相关系数值高于0.96, 斯皮尔曼等级相关系数值高于0.95。 与现有评价方法相比,该方法优于传统的无参考评价方法,更能够客观的反映不同失真图像 对血管分割这一应用的影响。     

一种无参考运动模糊图像的质量评价方法

针对高动态环境下视觉导航系统的可靠性受到运动模糊制约,而现有的结构相似度算法（SSIM）对模糊图像评价不敏感的问题,提出了一种基于改进结构相似度的无参考运动模糊图像质量评价算法。首先,将原始运动模糊图像经过同运动方向的模糊算子生成再模糊图像;之后,将原始模糊图像与再模糊图像进行8×8分块,并在相应子块中找到边缘信息丰富的子块;然后,计算原始模糊图像和再模糊图像中对应边缘信息丰富子块的结构相似度;最后,通过边缘置信度对子块加权得到整幅图像的模糊评价指标。实验结果表明,此评价方法结果与主观评价结果具有较高的一致性,能够准确地对运动模糊图像进行评价。由于本方法能有效地优化边缘子块的权值,与传统的平均结构相似度（MSSIM）评价方法相比,具有更高的灵敏性。     

基于显著性的无参考模糊图像质量评价

为了有效得评价模糊图像的质量分数,提出了一种基于显著性的无参考模糊图像质量评价方法.该方法首先通过改进的的显著值将图像分为显著块和非显著块,并舍弃非显著块;然后,根据模糊检测概率计算块的整体模糊,此外为了在评价过程中引入图像的结构细节信息,将标准差作为块的细节模糊;最后,对整体模糊和细节模糊进行几何融合得到块的质量分数,从而将人眼视觉信息与图像结构细节信息相融合,并利用改进的显著值对其所在的块进行加权,最终得到模糊图像质量评价方法.试验结果显示在LIVE、TID2013、CSIQ三大数据库上,该方法的评价效果都具有较好的准确性与预测性.该算法利用的图像信息相对比较全面,主要适用于纹理信息比较丰富的模糊图像.     

基于联合字典的无参考真实失真图像的质量评价

针对真实失真图像提出一种基于联合字典的无参考 (NR)图像质量评价(IQA)方法,分为训练和测试两个阶 段。在训练阶段,首先对真实失真图像提取美学特征和自然场景统计特征,然后对图像特征 和标签进行联 合字典学习,训练得到特征字典和质量字典。在测试阶段,根据特征字典和质量字典计算真 实失真图像的 质量值。在LIVE Challenge数据库上的实验结果表明,本文方法的评价结果与主观评价结果 有较好的相关 性,符合人类视觉系统的感知,相比较传统的无参考方法,具有更好的优越性。     

Blind quality assessment of tone-mapped images using multi-exposure sequences

The tone mapping operator (TMO) enables high dynamic range (HDR) images to be presented on low dynamic range (LDR) consumer electronic devices. However, the results obtained by this method are not always ideal due to the reduced number of bits. In comparison, the multi-exposure image fusion (MEF) bypasses the intermediate HDR image composition and directly produces an image presented on standard devices. Inspired by this, this paper proposes a quality assessment method for tone-mapped image (TMI) based on generating multi-exposure sequences. Specifically, the method uses a generative adversarial network (GAN) to generate a set of sequences with different exposure levels based on the TMIs. Then a two-branch convolutional neural network (CNN) is used to extract features from the tone-mapped images and the multi-exposure reference sequences, respectively. Finally, the transformer is used to mine the intrinsic connections between TMIs and multi-exposure sequences and learn the mapping relationships from feature space to quality space. We conducted extensive experiments on the ESPL-LIVE HDR database. The applicability and effectiveness of the proposed method are verified by comparing and analyzing relevant features and model configurations with existing mainstream evaluation algorithms.     

Blind quality assessment for multiply distorted stereoscopic images towards IoT-based 3D capture systems

Empowered by 5G mobile communication networks, multimedia processing has been considered as a very promising application of Internet-of-Things (IoT). Stereoscopic image quality assessment (SIQA), as an important part of 3D capture system, can be embedded in the cloud or fog servers to automatically monitor the perceptual quality of the collected stereoscopic images. In this paper, a novel blind image quality assessment method towards IoT-based 3D capture systems is developed for multiply-distorted stereoscopic images (MDSIs), in which five complementary channels, including left view, right view, cyclopean map, summation map and difference map, are jointly considered in dictionary learning for characterizing the monocular receptive field (MRF) and binocular receptive field (BRF) properties of the visual cortex in response to MDSIs. Additionally, the high order statistics scheme is adopted by utilizing the statistical differences between the codebook and images to ensure the stable and robust quality prediction performance for MDSIs. The proposed method shows competitive prediction performances on four benchmark databases compared with the existing SIQA metrics.     

Simple quality assessment for binary images

Usually image assessment methods could be classified into two categories： subjective assessments and objective ones. The latter are judged by the correlation coefficient with subjective quality measurement MOS （Mean Opinion Score）. This paper presents an objective quality assessment algorithm special for binary images. In the algorithm, noise energy is measured by Euclidean distance between noises and signals and the structural effects caused by noise are described by Euler number change. The assessment on image quality is calculated quantitatively in terms of PSNR （Peak Signal to Noise Ratio）. Our experiments show that the results of the algorithm are highly correlative with subjective MOS and the algorithm is more simple and computational saving than traditional objective assessment methods.     

Blind image quality assessment in the contourlet domain

No-reference/blind image quality assessment (NR-IQA/BIQA) algorithms play an important role in image evaluation, as they can assess the quality of an image automatically, only using the distorted image whose quality is being assessed. Among the existing NR-IQA/BIQA methods, natural scene statistic (NSS) models which can be expressed in different bandpass domains show good consistency with human subjective judgments of quality.In this paper, we create new ‘quality-aware’ features: the energy differences of the sub-band coefficients across scales via contourlet transform, and propose a new NR-IQA/BIQA model that operates on natural scene statistics in the contourlet domain. Prior to applying the contourlet transform, we apply two preprocessing steps that help to create more information-dense, low-entropy representations. Specifically, we transform the picture into the CIELAB color space and gradient magnitude map. Then, a number of ‘quality-aware’ features are discovered in the contourlet transform domain: the energy of the sub-band coefficients within scales, and the energy differences between scales, as well as measurements of the statistical relationships of pixels across scales. A detailed analysis is conducted to show how different distortions affect the statistical characteristics of these features, and then features are fed to a support vector regression (SVR) model which learns to predict image quality. Experimental results show that the proposed method has high linearity against human subjective perception, and outperforms the state-of-the-art NR-IQA models.     

POCS-based restoration of space-varying blurred images

We propose a new method for space-varying image restoration using the method of projection onto convex sets (POCS). The formulation allows the use of a different blurring function at each pixel of the image in a computationally efficient manner. We illustrate the performance of the proposed approach by comparing the new results with those of the ROMKF method on simulated images. We also present results on a real-life image with unknown space-varying out-of-focus blur.     

Modeling for edge detection problems in blurred noisy images

The aim of this paper is to provide a theoretical set up and a mathematical model for the problem of image reconstruction. The original image belongs to a family of two-dimensional (2-D) possibly discontinuous functions, but is blurred by a Gaussian point spread function introduced by the measurement device. In addition, the blurred image is corrupted by an additive noise. We propose a preprocessing of data which enhances the contribution of the signal discontinuous component over that one of the regular part, while damping down the effect of noise. In particular we suggest to convolute data with a kernel defined as the second order derivative of a Gaussian spread function. Finally, the image reconstruction is embedded in an optimal problem framework. Now convexity and compactness properties for the admissible set play a fundamental role. We provide an instance of a class of admissible sets which is relevant from an application point of view while featuring the desired properties.     

A fuzzy operator for the enhancement of blurred and noisy images

Rule-based fuzzy operators are a novel class of operators specifically designed in order to apply the principles of approximate reasoning to digital image processing. This paper shows how a fuzzy operator that is able to perform detail sharpening but is insensitive to noise can be designed. The results obtainable by the proposed technique in the enhancement of a real image are presented.