Academic
Publications
Blind Image Quality Assessment: From Natural Scene Statistics to Perceptual Quality

Blind Image Quality Assessment: From Natural Scene Statistics to Perceptual Quality,10.1109/TIP.2011.2147325,IEEE Transactions on Image Processing,Anu

Blind Image Quality Assessment: From Natural Scene Statistics to Perceptual Quality   (Citations: 2)
BibTex | RIS | RefWorks Download
Our approach to blind image quality assessment (IQA) is based on the hypothesis that natural scenes possess certain statistical properties which are altered in the presence of distortion, rendering them un-natural; and that by character- izing this un-naturalness using scene statistics, one can identify the distortion afflicting the image and perform no-reference (NR) IQA. Based on this theory, we propose an (NR)/blind algorithm—the Distortion Identification-based Image Verity and INtegrity Evaluation (DIIVINE) index—that assesses the quality of a distorted image without need for a reference image. DIIVINE is based on a 2-stage framework involving distortion identification followed by distortion-specific quality assessment. DIIVINE is capable of assessing the quality of a distorted image across multiple distortion categories, as against most NR IQA algorithms that are distortion-specific in nature. DIIVINE is based on natural scene statistics which govern the behavior of natural images. In this paper, we detail the principles underlying DIIVINE, the statistical features extracted and their relevance to perception and thoroughly evaluate the algorithm on the popular LIVE IQA database. Further, we compare the performance of DIIVINE against leading full-reference (FR) IQA algorithms and demonstrate that DIIVINE is statistically superior to the often used measure of peak signal-to-noise ratio (PSNR) and statistically equivalent to the popular structural similarity index (SSIM). A software release of DIIVINE has been made available online: http://live.ece.utexas.edu/research/quality/DIIVINE_release.zip for public use and evaluation.
Journal: IEEE Transactions on Image Processing , vol. 20, no. 12, pp. 3350-3364, 2011
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...Our research has lead us to believe that the kurtosis of coefficient distributions in each of the subbands is a good measure of the activity in a frame [18, 19]...
    • ...1 We note that we have experimented with some other measures, including optical flow information, spatial activity as defined in [17], and subband features as defined in [18, 19]...

    Anush Krishna Moorthyet al. H.264 visually lossless compressibility index: Psychophysics and algor...

    • ...More recently, true no-reference (NR) image QA algorithms have been developed that are also distortion-agnostic, by training NSS model-based classifiers on large databases of distorted images and associated human opinion scores [27], [28]...
    • ...The most relevant NSS models for visual QA are the gaussian scale mixture (GSM) model [5], [24], [26], [27] and the generalized gaussian distribution model (GGD) [23], [25]-[28]...
    • ...This observation makes it possible to use machine learning methods to train a classifier, using just the simple parameters (, ) as features, to determine the likeliest distortions afflicting an image [27], [32]...
    • ...Once the likely distortions are identified, the same parameters can be used, possibly with other NSS statistics to capture dependencies over scale or orientations, to accomplish image QA with great efficacy [27]...
    • ...Either approach is completely no-reference and also distortion-agnostic in the sense that the distortions are unknown prior to the process of QA, although they are ostensibly limited to those distortions that the QA algorithm has been trained on. We have very recently developed algorithms of both the two-stage and single-stage variety, named DIIVINE [27] and BLIINDS-II [28], that supply QA results that are quite competitive with even top ...
    • ...Table 2. Performance of four full-reference still image quality assessment indices: PSNR, SS-SSIM [21], MS-SSIM [6], and VIF [5]; and also two no-reference indices: BLIINDS-II [28] and DIIVINE [27]...

    Al Bovik. Perceiving distortions in visual signals

Sort by: