Image quality assessment based on a degradation model

By Niranjan Damera-Venkata, T. Kite, Wilson S. Geisler, Brian L. Evans, A. Bovik. IEEE Transactions on Image Processing, 2000

923 citations recorded by the source corpus.

Subfields: Atomic and Molecular Physics, and Optics, Cognitive Neuroscience, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/b97f4bcfe9c9b240689855ce83c1412e2db98393

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atomic and Molecular Physics, and Optics

In Cognitive Neuroscience

In Computer Vision and Pattern Recognition

Citations in the corpus

FSIM: A Feature Similarity Index for Image Quality Assessment (2011, methodology): “Such kinds of models emphasize the importance of HVS’ sensitivity to different visual signals, such as the luminance, the contrast, the frequency content, and the interaction between different signal components [2-4].”
41 OBJECTIVE VIDEO QUALITY ASSESSMENT (2003, background): “proposed a metric for quantifying performance of image restoration systems, in which the degradation is modelled as a linear frequency distortion and additive noise injection [46].”
A Comparative Study for Single Image Blind Deblurring (2016, methodology): “For full-reference metrics, we choose several widely used metrics , PSNR and SSIM [40], as well as WSNR [22], MS- SSIM [41], IFC [37], NQM [4], UIQI [39], VIF [36].”
A Statistical Evaluation of Recent Full Reference Image Quality Assessment Algorithms (2006, background): “Works with luminance only. NQM [ 23 ]. Works with luminance only.”
The Design of High-Level Features for Photo Quality Assessment (2006, background): “Initial work on image quality assessment required the original undistorted image to assess the quality of the degraded image [5, 19].”
An information fidelity criterion for image quality assessment using natural scene statistics (2005, background): “Previously, researchers have noted that as the correlation of the noise with the reference signal increases, MSE becomes poorer in predicting perceptual quality [35].”
Single-Image Super-Resolution: A Benchmark (2014, background): “Our experimental results indicate that for SR images four metrics are more effective such as the multi-scale structure similarity index (MSSSIM) [40], information fidelity criterion (IFC) [29], weighted peak signal-to-noise ratio (WPSNR), and noise quality measure (NQM) [4].”
Subjective evaluation of stereoscopic image quality (2013, neutral): “Noise Quality Measure (NQM) [42]”
VSNR: A Wavelet-Based Visual Signal-to-Noise Ratio for Natural Images (2007, background): “In particular, we have shown in [12] and [13] that natural images impose both image-selective and spatial-frequency-selective masking of targets consisting of wavelet distortions.”
Image database TID 2013 : Peculiarities , results and perspectives (2016, methodology): “Spearman rank order correlation PSNR PSNRc MSSIM SSIM VSNR VIFP VIF NQM WSNR PSNRHVSM PSNRHVS PSNRHMA PSNRHA FSIM FSIMc SFF UQI MSDDM SR-SIM DCTUNE CW-SSIM IFC IWPSNR IWSSIM LinLab MADindex XYZ BMMF Kendall rank order correlation Fig.”
DehazeNet: An End-to-End System for Single Image Haze Removal (2016, methodology): “ehazing result. Apart from the widely used mean square error (MSE) and the structural similarity (SSIM) [46] indices, we adopt three additional evaluation matrices, namely noise quality measure (NQM) [47], peak signal-to-noise ratio (PSNR) and weighted peak signal-to-noise ratio (WPSNR) [48]. We deﬁ”
Global-Local Face Upsampling Network (2016, background): “Quantitative Results: In addition to the standard peak signal-to-noise ratio (PSNR) and structural similarity (SSIM), we computed the information fidelity criterion (IFC) [44], weighted PSNR (WPSNR), and noise quality mea- sure (NQM) [45] for quantitative evaluations as suggested in [9]2.”
Foveated video compression with optimal rate control (2001, methodology): “MPEG and H.263-compliant video compression algorithms were demonstrated in [14] and used in the development of an automated algorithm for assessing the quality of foveated video streams using a human visual model [1], [ 15 ], [16].”
New edge-directed interpolation (2001, background): “, jaggy artifacts) are not predicted by the degradation models considered in [20].”
Deep Image Deblurring: A Survey (2022, methodology): “(2005), NQM Damera-Venkata et al. (2000), UIQI Wang and Bovik (2002), VIF Sheikh and Bovik (2006), and LPIPS Zhang et al. (2018). Among these, PSNR and SSIM are the most commonly usedmetrics in image restoration tasks (Gao et al.”
Image Super-Resolution Using Deep Convolutional Networks (2014, methodology): “When we take a look at other evaluation metrics, we observe that SC, to our surprise, gets even lower scores than the bicubic interpolation on IFC and NQM.”