Discrete quality assessment in IPTV content distribution networks

Maintaining the quality of videos in resource-intensive IPTV services is challenging due to the nature of packet-based content distribution networks (CDN). Network impairments are unpredictable and highly detrimental to the quality of video content. Quality of the end user experience (QoE) has become a critical service differentiator. An efficient real-time quality assessment service in distribution networks is the foundation of service quality monitoring and management. The perceptual impact of individual impairments varies significantly and is influenced by complex impact factors. Without differentiating the impact of quality violation events to the user experience, existing assessment methodologies based on network QoS such as packet loss rate cannot provide adequate supports for the IPTV service assessment. A discrete perceptual impact evaluation quality assessment (DEQA) framework is introduced in this paper. The proposed framework enables a real-time, non-intrusive assessment service by efficiently recognising and assessing individual quality violation events in the IPTV distribution network. The discrete perceptual impacts to a media session are aggregated for the overall user level quality evaluation. With its deployment scheme the DEQA framework also facilitates efficient network diagnosis and QoE management. To realise the key assessment function of the framework and investigate the proposed advanced packet inspection mechanism, we also introduce the dedicated evaluation testbed—the LA2 system. A subjective experiment with data analysis is also presented to demonstrate the development of perceptual impact assessment functions using analytical inference, the tools of the LA2 system, subjective user tests and statistical modelling.     

基于结构相似度的自适应图像质量评价

考虑到在结构相似度(SSIM,structural similarity)模型中,亮度、对比度和结构度3个评价因子对不同失真类型图像质量评价(QA)的贡献程度不同,本文提出了根据图像失真类型分析的自适应SSIM(ASSIM)的IQA方法。首先,分析失真图像和参考图像的小波子带能量、傅里叶功率谱和幅度谱的数据特点,据此判定图像失真类型,包括高斯白噪声(WN)、JPEG压缩(JPEM)、高斯模糊(Gblur)及类JP2K4类失真;接着,通过优化算法确定SSIM在评价不同失真类型图像时最佳的评价因子权重;最后,将图像的失真类型判别和评价因子的调整相结合,实现对图像的自适应评价。实验结果表明,由于失真类型的判断和评价因子权值的优化,ASSIM对各类失真图像的评价效果都要优于SSIM,特别是对Gblur失真的图像进行评价时,Pearson系数(CC)值提高了0.05,Spearman等级相关系数(SROCC)值的提高超过0.039。     

No-reference image quality assessment using structural activity

Presuming that human visual perception is highly sensitive to the structural information in a scene, we propose the concept of structural activity (SA) together with a model of SA indicator in a new framework for no-reference (NR) image quality assessment (QA) in this study. The proposed framework estimates image quality based on the quantification of the SA information of different visual significance. We propose some alternative implementations of SA indicator in this paper as examples to demonstrate the effectiveness of the SA-motivated framework. Comprehensive testing demonstrates that the model of SA indicator exhibits satisfactory performance in comparison with subjective quality scores as well as representative full-reference (FR) image quality measures.     

MLSIM: A Multi-Level Similarity index for image quality assessment

Image quality assessment (IQA) is of great importance to numerous image processing applications, and various methods have been proposed for it. In this paper, a Multi-Level Similarity (MLSIM) index for full reference IQA is proposed. The proposed metric is based on the fact that human visual system (HVS) distinguishes the quality of an image mainly according to the details given by low-level gradient information. In the proposed metric, the Prewitt operator is first utilized to get gradient information of both reference and distorted images, then the gradient information of reference image is segmented into three levels (3LSIM) or two levels (2LSIM), and the gradient information of distorted image is segmented by the corresponding regions of reference image, therefore we get multi-level information of these two images. Riesz transform is utilized to get corresponding features of different levels and the corresponding 1st-order and 2nd-order coefficients are combined together by regional mutual information (RMI) and weighted to obtain a single quality score. Experimental results demonstrate that the proposed metric is highly consistent with human subjective evaluations and achieves good performance.     

Image quality assessment using a SVD-based structural projection

The development of objective image quality assessment (IQA) metrics aligned with human perception is of fundamental importance to numerous image-processing applications. Recently, human visual system (HVS)-based engineering algorithms have received widespread attention for their low computational complexity and good performance. In this paper, we propose a new IQA model by incorporating these available engineering principles. A local singular value decomposition (SVD) is first utilised as a structural projection tool to select local image distortion features, and then, both perceptual spatial pooling and neural networks (NN) are employed to combine feature vectors to predict a single perceptual quality score. Extensive experiments and cross-validations conducted with three publicly available IQA databases demonstrate the accuracy, consistency, robustness, and stability of the proposed approach compared to state-of-the-art IQA methods, such as Visual Information Fidelity (VIF), Visual Signal to Noise Ratio (VSNR), and Structural Similarity Index (SSIM).     

Robust image retrieval by cascading a deep quality assessment network

The performance of computer vision algorithms can severely degrade in the presence of a variety of distortions. While image enhancement algorithms have evolved to optimize image quality as measured according to human visual perception, their relevance in maximizing the success of computer vision algorithms operating on the enhanced image has been much less investigated. We consider the problem of image enhancement to combat Gaussian noise and low resolution with respect to the specific application of image retrieval from a dataset. We define the notion of image quality as determined by the success of image retrieval and design a deep convolutional neural network (CNN) to predict this quality. This network is then cascaded with a deep CNN designed for image denoising or super resolution, allowing for optimization of the enhancement CNN to maximize retrieval performance. This framework allows us to couple enhancement to the retrieval problem. We also consider the problem of adapting image features for robust retrieval performance in the presence of distortions. We show through experiments on distorted images of the Oxford and Paris buildings datasets that our algorithms yield improved mean average precision when compared to using enhancement methods that are oblivious to the task of image retrieval. ¹     

A new image quality assessment method to detect and measure strength of blocking artifacts

Block based transform coding is one of the most popular techniques for image and video compression. However it suffers from several visual quality degradation factors, most notably from blocking artifacts. The subjective picture quality degradation caused by blocking artifacts, in general, does not agree well with the popular objective quality measure such as PSNR.A new image quality assessment method that detects and measures strength of blocking artifacts for block based transform coded images is proposed. In order to characterize the blocking artifacts, we utilize two observations: if blocking artifacts occur on the block boundary, the pixel value changes abruptly across the boundary and the same pixel values usually span along the entire length of the boundary. The proposed method operates only on a single block boundary to detect blocking artifacts. When a boundary is classified as having blocking artifacts, corresponding blocking artifact strength is also computed. Average values of those blocking artifact strengths are converted into a single number representing the subjective image quality. Experiments on various JPEG compressed images with various bit rates demonstrated that the proposed blocking artifacts measuring value matches well with the subjective image quality judged by human observers.     

极端学习机在立体图像质量客观评价中的应用

基于传统神经网络训练速度慢、易陷入局部极小值和泛化性能低等问题,提出采用极端学习机(ELM,extreme learning machine)对立体图像质量进行了客观评价。ELM是单隐层前馈神经网络(SLFNs)的泛化,输入权重可以随机赋值并通过解析获得输出权值。与传统神经网络算法相比,ELM算法具有参数选择简单、学习速度快及泛化性能好等优点。实验结果表明,以sigmoid为激励函数,对241幅不同等级的立体图像测试样本进行测试,其正确等级分类率达到93.85%。研究了不同激励函数条件下不同隐藏层节点数对极端学习机网络性能的影响,且将ELM和传统BP及支持向量机(SVM)在立体图像质量评价中的性能进行了分析比较。     

No-reference quality assessment for DCT-based compressed image

A blind/no-reference (NR) method is proposed in this paper for image quality assessment (IQA) of the images compressed in discrete cosine transform (DCT) domain. When an image is measured by structural similarity (SSIM), two variances, i.e. mean intensity and variance of the image, are used as features. However, the parameters of original copies are actually unavailable in NR applications; hence SSIM is not widely applicable. To extend SSIM in general cases, we apply Gaussian model to fit quantization noise in spatial domain, and directly estimate noise distribution from the compressed version. Benefit from this rearrangement, the revised SSIM does not require original image as the reference. Heavy compression always results in some zero-value DCT coefficients, which need to be compensated for more accurate parameter estimate. By studying the quantization process, a machine-learning based algorithm is proposed to estimate quantization noise taking image content into consideration. Compared with state-of-the-art algorithms, the proposed IQA is more heuristic and efficient. With some experimental results, we verify that the proposed algorithm (provided no reference image) achieves comparable efficacy to some full reference (FR) methods (provided the reference image), such as SSIM.     

Stitched image quality assessment based on local measurement errors and global statistical properties

Image stitching is developed to generate wide-field images or panoramic images for virtual reality applications. However, the quality assessment of stitched images with respect to various stitching algorithms has been less studied. Effective stitched image quality assessment (SIQA) is advantageous to evaluate the performance of various stitching methods and optimize the design of stitching methods. In this paper, we propose a novel SIQA method by exploiting local measurement errors and global statistical properties for feature extraction. Comprehensive image attributes including ghosting, misalignment, structural distortion, geometric error, chromatic aberrations and blur are considered either locally or globally. The extracted local and global features are aggregated into an overall quality via regression. Experimental results on two benchmark databases demonstrate the superiority of the proposed metric over both the state-of-the-art quality models designed for natural images and stitched images.     

Blind image blur assessment by using valid reblur range and histogram shape difference

The presence of blur artifact is an annoyance to image viewers, and affects the perceived quality of the image. Telecommunication service providers and imaging product manufacturers are interested in this quality feedback for their process and product improvement. However, human-based quality feedback is tedious, expensive and has to be done in compliance with the standards for subjective evaluation such as the ITU-R BT. 500 standard. Thus, automatic assessment of images is proposed to overcome the difficulties in human-based evaluation. The automatic assessment is basically an objective estimation to predict the blur severity of an image. In this paper, a new model for blind estimation is proposed by using reblur algorithm to create reblur image and measure valid reblur range. Shape difference of local histograms is measured between the reblur and test images to produce the blur score. The proposed model is performed in the spatial domain without the need of data conversion or training. Experiment results show that the proposed model is highly correlated to human perception of blurriness, and performs better than other state-of-the-art blur metrics in the spatial domain.