Abstract
The difficulty of no-reference image quality assessment (NR IQA) often lies in the lack of knowledge about the distortion in the image, which makes quality assessment blind and thus inefficient. To tackle such issue, in this article, we propose a novel scheme for precise NR IQA, which includes two successive steps, i.e., distortion identification and targeted quality evaluation. In the first step, we employ the well-known Inception-ResNet-v2 neural network to train a classifier that classifies the possible distortion in the image into the four most common distortion types, i.e., Gaussian white noise (WN), Gaussian blur (GB), jpeg compression (JPEG), and jpeg2000 compression (JP2K). Specifically, the deep neural network is trained on the large-scale Waterloo Exploration database, which ensures the robustness and high performance of distortion classification. In the second step, after determining the distortion type of the image, we then design a specific approach to quantify the image distortion level, which can estimate the image quality specially and more precisely. Extensive experiments performed on LIVE, TID2013, CSIQ, and Waterloo Exploration databases demonstrate that (1) the accuracy of our distortion classification is higher than that of the state-of-the-art distortion classification methods, and (2) the proposed NR IQA method outperforms the state-of-the-art NR IQA methods in quantifying the image quality.
- [1] . 2018. A fast Uyghur text detector for complex background images. IEEE Trans. Multimedia 20, 12 (2018), 3389–3398.Google Scholar
Digital Library
- [2] . 2019. Cross-modality bridging and knowledge transferring for image understanding. IEEE Trans. Multimedia 21, 10 (2019), 2675–2685.Google Scholar
Digital Library
- [3] . 2019. STAT: Spatial-temporal attention mechanism for video captioning. IEEE Trans. Multimedia 22, 1 (2019), 229–241.Google Scholar
- [4] . 2019. Real-time indoor scene reconstruction with RGBD and inertial input. In 2019 IEEE International Conference on Multimedia and Expo (ICME’19). IEEE, 7–12.Google Scholar
Cross Ref
- [5] . 2018. Measuring individual video QoE: A survey, and proposal for future directions using social media. ACM Trans. Multimedia Comput. Commun. Appl. 14, 2s (
May 2018), 30:1–30:24. Google ScholarDigital Library
- [6] . 2006. Defining user perception of distributed multimedia quality. ACM Trans. Multimedia Comput. Commun. Appl. 2, 4 (
Nov. 2006), 241–257. Google ScholarDigital Library
- [7] . 2004. Image quality assessment: From error visibility to structural similarity. IEEE Trans. Image Process. 13, 4 (2004), 600–612. Google Scholar
Digital Library
- [8] . 2018. Reduced-reference image quality assessment in free-energy principle and sparse representation. IEEE Trans. Multimedia 20, 2 (2018), 379–391. Google Scholar
Digital Library
- [9] . 2020. Blind image quality assessment by natural scene statistics and perceptual characteristics. ACM Trans. Multimedia Comput. Comm. Appl. (TOMM) 16, 3 (2020), 1–91.Google Scholar
Digital Library
- [10] . 2017. Quality assessment for real out-of-focus blurred images. J. Vis. Comm. Image Represent. 46 (2017), 70–80. Google Scholar
Digital Library
- [11] . 2012. A fast wavelet-based algorithm for global and local image sharpness estimation. IEEE Signal Process. Lett. 19, 7 (2012), 423–426.Google Scholar
Cross Ref
- [12] . 2013. Image sharpness assessment based on local phase coherence. IEEE Trans. Image Process. 22, 7 (2013), 2798–2810.Google Scholar
Cross Ref
- [13] . 2009. Scale invariance and noise in natural images. In Proceedings of the IEEE International Conference on Computer Vision.2209–2216.Google Scholar
Cross Ref
- [14] . 2013. Single-image noise level estimation for blind denoising. IEEE Trans. Image Process. 22, 12 (
Dec. 2013), 5226–5237. Google ScholarDigital Library
- [15] . 2000. Blind measurement of blocking artifacts in images. In Proceedings 2000 International Conference on Image Processing (Cat. No.00CH37101), Vol. 3. 981–984.Google Scholar
Cross Ref
- [16] . 2004. A locally-adaptive algorithm for measuring blocking artifacts in images and videos. In 2004 IEEE International Symposium on Circuits and Systems (IEEE Cat. No.04CH37512), Vol. 3. III–925.Google Scholar
- [17] . 2019. Blind quality assessment of camera images based on low-level and high-level statistical features. IEEE Trans. Multimedia 21, 1 (2019), 135–146.Google Scholar
Digital Library
- [18] . 2015. Using free energy principle for blind image quality assessment. IEEE Trans. Multimedia 17, 1 (2015), 50–63.Google Scholar
Cross Ref
- [19] . 2010. A two-step framework for constructing blind image quality indices. IEEE Signal Process. Lett. 17, 5 (2010), 513–516.Google Scholar
Cross Ref
- [20] . 2019. Unsupervised blind image quality evaluation via statistical measurements of structure, naturalness and perception. IEEE Trans. Circuits Syst. Video Technol. 30, 4 (2019), 929–943.Google Scholar
- [21] . 2014. VSI: A visual saliency-induced index for perceptual image quality assessment. IEEE Trans. Image Process. 23, 10 (2014), 4270–4281.Google Scholar
Cross Ref
- [22] . 2011. Blind image quality assessment: From natural scene statistics to perceptual quality. IEEE Trans. Image Process. 20, 12 (2011), 3350–3364. Google Scholar
Digital Library
- [23] . 2018. End-to-end blind image quality assessment using deep neural networks. IEEE Trans. Image Process. 27, 3 (2018), 1202–1213.Google Scholar
Cross Ref
- [24] . 2017. Inception-v4, inception-resnet and the impact of residual connections on learning. In 31st AAAI Conference on Artificial Intelligence. Google Scholar
Digital Library
- [25] . 2016. Waterloo exploration database: New challenges for image quality assessment models. IEEE Trans. Image Process. 26, 2 (2016), 1004–1016. Google Scholar
Digital Library
- [26] . 2002. No-reference perceptual quality assessment of JPEG compressed images. In Proceedings International Conference on Image Processing.Google Scholar
Cross Ref
- [27] . 2005. Image blockiness evaluation based on Sobel operator. In IEEE International Conference on Image Processing 2005, Vol. 1. I–389.Google Scholar
Cross Ref
- [28] . 2017. Unified blind quality assessment of compressed natural, graphic, and screen content images. IEEE Trans. Image Process. 26, 11 (
Nov. 2017), 5462–5474.Google ScholarDigital Library
- [29] . 2002. A no-reference perceptual blur metric. In Proceedings of the IEEE International Conference on Image Processing.57–60.Google Scholar
Cross Ref
- [30] . 2009. A no-reference objective image sharpness metric based on the notion of just noticeable blur (JNB). IEEE Trans. Image Process. 18, 4 (2009), 717–728. Google Scholar
Digital Library
- [31] . 2017. No-reference quality metric of contrast-distorted images based on information maximization. IEEE Trans. Cybernetics 47, 12 (
Dec. 2017), 4559–4565.Google ScholarCross Ref
- [32] . 2020. No-reference quality assessment for contrast-distorted images. IEEE Access 8 (2020), 84105–84115.Google Scholar
Cross Ref
- [33] . 2012. No-reference image quality assessment in the spatial domain. IEEE Trans. Image Process. 21, 12 (2012), 4695–4708. Google Scholar
Digital Library
- [34] . 2012. Blind image quality assessment: A natural scene statistics approach in the DCT domain. IEEE Trans. Image Process. 21, 8 (2012), 3339–3352. Google Scholar
Digital Library
- [35] . 2013. Making a completely blind image quality analyzer. IEEE Signal Process. Lett. 20, 3 (2013), 209–212.Google Scholar
Cross Ref
- [36] . 2018. Opinion-unaware blind quality assessment of multiply and singly distorted images via distortion parameter estimation. IEEE Trans. Image Process. 27, 11 (
Nov. 2018), 5433–5448. Google ScholarDigital Library
- [37] . 2014. Convolutional neural networks for no-reference image quality assessment. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’14). 1733–1740. Google Scholar
Digital Library
- [38] . 2018. Deep neural networks for no-reference and full-reference image quality assessment. IEEE Trans. Image Process. 27, 1 (
Jan. 2018), 206–219.Google ScholarCross Ref
- [39] . 2020. End-to-end blind image quality prediction with cascaded deep neural network. IEEE Trans. Image Process. 29 (2020), 7414–7426.Google Scholar
Digital Library
- [40] . 2017. RankIQA: Learning from rankings for no-reference image quality assessment. In 2017 IEEE International Conference on Computer Vision (ICCV’17). 1040–1049.Google Scholar
Cross Ref
- [41] . 2017. Fully deep blind image quality predictor. IEEE J. Sel. Topics Signal Process. 11, 1 (2017), 206–220.
DOI: DOI: http://dx.doi.org/10.1109/JSTSP.2016.2639328Google ScholarCross Ref
- [42] . 2021. Uncertainty-aware blind image quality assessment in the laboratory and wild. IEEE Trans. Image Process. 30 (2021), 3474–3486.
DOI: DOI: http://dx.doi.org/10.1109/TIP.2021.3061932Google ScholarDigital Library
- [43] . 2019. Two stream convolutional networks for blind image quality assessment. IEEE Transactions on Image Processing 28, 5 (2019), 2200–2211. Google Scholar
Digital Library
- [44] . 1973. Textural features for image classification. IEEE Trans. Syst. Man Cybernetics 6 (1973), 610–621.Google Scholar
Cross Ref
- [45] . 1998. Indoor-outdoor image classification. In Proceedings of the 1998 IEEE International Workshop on Content-Based Access of Image and Video Database. IEEE, 42–51. Google Scholar
Digital Library
- [46] . 1990. Invariant image recognition by Zernike moments. IEEE Trans. Pattern Anal. Mach. Intell. 12, 5 (1990), 489–497. Google Scholar
Digital Library
- [47] . 2016. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 770–778.Google Scholar
Cross Ref
- [48] . 2016. Rethinking the inception architecture for computer vision. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2818–2826.Google Scholar
Cross Ref
- [49] . 2013. A dual-model approach to blind quality assessment of noisy images. In 2013 Picture Coding Symposium (PCS’13). 29–32.Google Scholar
- [50] . 2001. Natural image statistics and neural representation. Ann. Rev. Neurosci. 24, 1 (2001), 1193–1216.Google Scholar
Cross Ref
- [51] . 1988. Feature detection in human vision: A phase-dependent energy model. Proc. Roy. Soc. London B Biol. Sci. 235, 1280 (1988), 221–245.Google Scholar
Cross Ref
- [52] . 1986. Mach bands are phase dependent. Nature 324, 6094 (1986), 250–253.Google Scholar
Cross Ref
- [53] . 2015. Quality assessment for out-of-focus blurred images. In Visual Communications and Image Processing (VCIP’15). IEEE, 1–4.Google Scholar
- [54] . 1999. Image features from phase congruency. Videre: Journal of Computer Vision Research 1, 3 (1999), 1–26.Google Scholar
- [55] . 1992. Biorthogonal bases of compactly supported wavelets. Comm. Pure Appl. Math. 45, 5 (1992), 485–560.Google Scholar
Cross Ref
- [56] . 2016. Blind quality assessment of compressed images via pseudo structural similarity. In 2016 IEEE International Conference on Multimedia and Expo (ICME’16). IEEE, 1–6.Google Scholar
Cross Ref
- [57] . 2014. Learning structural regularity for evaluating blocking artifacts in JPEG images. IEEE Signal Process. Lett. 21, 8 (2014), 918–922.Google Scholar
Cross Ref
- [58] . 2014. Learning structural regularity for evaluating blocking artifacts in JPEG images. IEEE Signal Process. Lett. 21, 8 (2014), 918–922.Google Scholar
Cross Ref
- [59] . 1994. Good features to track. In 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. IEEE, 593–600.Google Scholar
- [60] . 2013. A no-reference quality assessment algorithm for JPEG2000-compressed images based on local sharpness. In Image Quality and System Performance X, Vol. 8653. International Society for Optics and Photonics, 865302.Google Scholar
- [61] . 2012. S3: A spectral and spatial measure of local perceived sharpness in natural images. IEEE Trans. Image Process. 21, 3 (2012), 934–945. Google Scholar
Digital Library
- [62] . 2006. A statistical evaluation of recent full reference image quality assessment algorithms. IEEE Trans. Image Process. 15, 11 (2006), 3440–3451. Google Scholar
Digital Library
- [63] . 2010. Most apparent distortion: full-reference image quality assessment and the role of strategy. Journal of Electron. Imaging 19, 1 (2010), 011006.Google Scholar
Cross Ref
- [64] . 2015. Image database TID2013: Peculiarities, results and perspectives. Signal Process. Image Comm. 30 (2015), 57–77. Google Scholar
Digital Library
- [65] . 2000. Final report from the video quality experts group on the validation of objective models of video quality assessment. ITU-T Standards Contribution COM (2000), 9–80.Google Scholar
- [66] . 2015. A highly efficient method for blind image quality assessment. In IEEE International Conference on Image Processing.339–343.Google Scholar
Digital Library
- [67] . 2013. Learning without human scores for blind image quality assessment. In Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition.995–1002. Google Scholar
Digital Library
- [68] . 2015. A feature-enriched completely blind image quality evaluator. IEEE Trans. Image Process. 24, 8 (2015), 2579–2591.Google Scholar
Digital Library
Index Terms
Precise No-Reference Image Quality Evaluation Based on Distortion Identification
Recommendations
No-Reference Image Sharpness Assessment in Autoregressive Parameter Space
In this paper, we propose a new no-reference (NR)/ blind sharpness metric in the autoregressive (AR) parameter space. Our model is established via the analysis of AR model parameters, first calculating the energy- and contrast-differences in the locally ...
Analysis and Design of No-Reference Image Quality Assessment
MMIT '08: Proceedings of the 2008 International Conference on MultiMedia and Information TechnologyNo-Reference (NR) Image Quality Assessment (IQA) seeks to assign quality scores that are consistent with human perception but without an explicit comparison with the reference image. Unfortunately, the field of NR IQA has been largely unexplored. The ...
No-Reference image quality assessment for facial images
ICIC'11: Proceedings of the 7th international conference on Advanced Intelligent Computing Theories and Applications: with aspects of artificial intelligenceImage quality assessment traditionally means the comparison of original image with its distorted version using conventional methods like Mean Square Error (MSE) or Peak Signal to Noise Ratio (PSNR). In case of Blind Quality Evaluation with no prior ...






Comments