BIAQ论文一览

浅浅的花香味﹌ 2022-08-20 14:11 137阅读 0赞

1.  STRUCTURE-PRESERVING IMAGE QUALITY ASSESSMENT  
    保留结构的图像质量评价  
    Abstract： developing an MSE-like metric  
    introducing structure-preserving kernelization into a MSE-like formulation.  
    在类标准差评价标准的假想中引入保留的核心化结构，发展出FR和NR两种评价方法
2.  No-Reference Quality Assessment of Contrast-Distorted Images Based on Natural Scene Statistics  
    基于NSS的图像失真对比的图像质量评价  
    Abstract：build NSS models based on moment and entropy features  
    evaluate images’ unnaturalness characterized by the degree of deviation from the NSS models  
    SVR is employed to predict MOS from multiple NSS features as the input  
    先用大量的数据库图像中的力矩和熵特征建立NSS模型，评估待测图像的不自然特性和NSS模型的相差等级，用SVR预测图像质量
3.  Visual-PSNR measure of image quality 可视的PSNR图像质量评价方法（FR）  
    VPSNR takes into account some features of the HVS , which is a modified PSNR measure  
    引入了一些基于HVS的特征改进PSNR的全参考图像质量评价方法
4.  ☆ Non-distortion-specific no-reference image quality assessment: A survey  
    A survey of the recent no-reference image quality algorithm：  
    first half ：specifically for non-distortion-specific cases，natural scene statistics-based and learning-based  
    second half：their performance and limitations are discussed before current research trends  
    Finally： possible future research directions are proposed  
    一个基于目前已有的通用型无参考图形质量评价方法的研究，将已有方法进行了梳理、分类和比较
5.  Blind Image Quality Assessment Based on Multi-Channel Features Fusion and Label Transfer  
    基于多通道特征融合和label变换的无参考图像质量评价  
    Abstract： a new feature fusion scheme by combining an image’s statistical information frommultiple domains (i.e., DCT, wavelet and spatial domains)and multiple color channels (i.e., Y, Cb, Cr）  
    the predicted image quality is generated from a nonparametric model, which is referred to as the label transfer with an image retrieval procedure.  
    k-nearest-neighbors (KNN)  
    从多种域提取融合结构特征，将待测图像特征进行label变换找到k个相邻的已知质量的图像，加权平均得到待测质量
6.  Image quality assessment: A sparse learning way  
    图像质量评价中一种稀疏学习方法  
    Abstract：include two steps  
    sparse representations of local image patches are computed to simulate the low level characteristic of the HVS and represent the meaningful image structure  
    these local distortion measurements are fused into a single image quality score  
    by using kernel ridge regression (KRR)核心岭回归  
    模仿HVS代表结构的浅层特性提取稀疏特征，通过对比失真和非失真图像的特征来判断待测图像块失真程度，再利用KRR给图像质量赋分
7.  Perceptual Quality Assessment for Multi-Exposure Image Fusion  
    可感知多次曝光融合图像的质量评价  
    Abstract：we first build an MEF database  
    based on the principle of the structural similarity approach  
    a novel measure of patch structural consistency.  
    自己建立了一个MEF图像数据库，提取了结构相似性特征，建立模型得到待测质量
8.  Massive Online Crowdsourced Study of Subjective and Objective Picture Quality\*\*重点内容\*\*  
    主观和客观图片质量的大量的在线众包（把任务分配给大众网络的做法）学习  
    Abstract：the LIVE In the Wild Image Quality Challenge Database  
    Our database consists of over 350,000 opinion scores on  
    1,162 images evaluated by over 8100 unique human observers.  
    这篇是对LIVE的最新图片数据库live challenge database制作过程的介绍
9.  High Dynamic Range Image Compression by Optimizing Tone Mapped Image Quality Index  
    通过优化色调映射的高动态范围图像压缩质量指标  
    Abstract：Tone mapping operators (TMOs) aim to compress high dynamic range (HDR) images to low dynamic range (LDR)  
    色调映射算子旨在把HDR图像压缩恒LDR图像，本文旨在设计一种完全不同的TMO，使高动态图像更高质量的转化为LDR图像在标准显示器上显现
10. A Feature-Enriched Completely Blind Image Quality Evaluator\*\*重点内容\*\*  
    一种特征被强化的完全无参考图像质量评价  
    本文介绍了一种opinion-unaware方法，预先基于大量的优质自然图形建立基于NSS特征的MVG模型，再将待测图像的NSS特征和MVG模型比较得到待测质量
11. 1.  Utilizing Image Scales Towards Totally Training Free Blind Image Quality Assessment\*\*重点内容\*\*  
        the natural images to exhibit redundant information over various scales  
        low pass error、high pass error  
        uses the intrinsic global change of the query image across scales  
        这个方法也是opinion-unaware的，而且不需要训练模型，只是从失真图像本身各尺度信息的改变来评价质量
    2.  The Application of Visual Saliency Models in Objective Image Quality Assessment:A Statistical Evaluation  
        视觉统计模型在客观图像质量评价中的应用：统计评估
    3.  Tasking on Natural Statistics of Infrared Images  
        红外图像自然统计工作
    4.  Image Quality Assessment Using Human Visual DOG Model Fused With Random Forest  
        利用人类视觉高斯差异模型融合随机森林评价图像质量
    5.  Geometrical and Statistical Properties of Vision Models obtained via Maximum Differentiation  
        基于最大变异的视觉模型几何和统计性质的获得
    6.  Objective Quality Assessment for Color-to-Gray Image Conversion  
        彩色转灰度变换图像的客观质量评价  
        we propose a C2G structural similarity (C2G-SSIM) index  
        Inspired bythe philosophy of the structural similarity index  
        真是好灵感啊，作者该去研究哲学或者神学
    7.  No-reference image quality assessment with shearlet transform  
        and deep neural networks  
        基于剪切波变换和深度神经网络的无参考图像质量评价  
        features are extracted by a new multiscale directional transform (shearlet transform) and the sum of subband coefficient amplitudes (SSCA)  
        stacked autoencoders make primary features more discriminative.  
        以shearlet变换后子代系数振幅和为特征，利用堆栈自编码放大差异，建立深度神经网络模型评价图像质量
    8.  Perceptual image quality assessment by independent feature detector  
        利用独立特征探测器评价图像质量
    9.  Objective Quality Assessment for Multiexposure Multifocus Image Fusion  
        多次聚焦多电极对焦融合图像的客观质量评价  
        1) contrast preservation; 2) sharpness; and3) structure preservation  
        create an image fusion database  
        利用对比保存、清晰度、结构保存三方面评价多次曝光、对焦图像的质量
    10. Objective Quality Assessment of Interpolated Natural Images  
        以内插值替换的自然图像客观质量评价  
        adopts a natural scene statistics (NSS) framework  
        deviation of its statistical features from the NSS models trained upon high-quality natural images  
        interpolated natural image distortion (IND) and weighted IND  
        图像内插技术广泛用于将低分辨率图像变成高分辨率图像，本文提出了一种评价内插后图形质量的方法，通过NSS框架来评价待测图像，比较提出的两种失真情况