A.L. Priorov - Dr.Sc. (Eng.), Associate Professor, P.G. Demidov Yaroslavl State University. E-mail: andcat@yandex.ru I.S. Nenakhov - Post-graduate Student, P.G. Demidov Yaroslavl State University. E-mail: zergoodsound@gmail.com V.V. Khryashchev - Ph.D. (Eng.), Associate Professor, P.G. Demidov Yaroslavl State University. E-mail: vhr@yandex.ru

This paper presents the new algorithm for no-reference blind image quality assessment (NRQ LBP). This algorithm does not need a priori information about possible types of image distortions before assessment. No transformation to another coordinate frame (DCT, wavelet, etc.) is required, distinguishing it from prior no reference quality assessment approaches. NRQ LBP is based on machine learning and uses extremely randomized trees method for mapping quality features with subject quality score (DMOS). Quality features are bins of a histogram of local binary patterns calculated for neighborhood radiuses 1, 2, 3 pixels. Comparative experimental results a given for modern image quality assessment algorithms (PSNR, SSIM, MS-SSIM, LBIQ, LD-TS, GRNN, BRISQUE, NRLBPS). Images from standard LIVE database are used as training and testing datasets. Spearman correlation coefficient, Pearson correlation coefficient and RMSE are used to determine the accuracy of compared algorithms. The proposed approach has the highest correlation with subject quality scores between all tested no reference algorithms and competitive with reference MS-SSIM. NRQ LBP is more accurate for images distorted with JPEG2000 and Gaussian blur. For images distorted with bit errors the proposed algorithm is more accurate than tested no reference algorithms and is accurate as reference algorithms SSIM and MS-SSIM. Moreover, NRQ LBP has very low computational complexity, making it well suited for real time applications.

 

1. Dvorkovich V.P., Dvorkovich A.V. Cifrovye videoinformacionnye sistemy (teorija i praktika). M.: Tekhnosfera. 2012.
2. Bovik A. Handbook of Image and Video Processing. New York: Academic. 2005.
3. Wang Z., Bovik A. Modern Image Quality Assessment. Morgan and Claypool Publishing Company. 2006.
4. Mittal A., Moorthy A.K., Bovik A.C. No-reference image quality assessment in the spatial domain // IEEE Trans. Image Process. 2012. V. 7. № 12. P. 4695-4708.
5. Priorov A.L., KHrjashhev V.V., Nenakhov I.S. Issledovanie metoda neehtalonnojj ocenki kachestva izobrazhenijj v prostranst-vennojj oblasti // Dokl. 3-jj Vseros. konf. «Radioehlektronnye sredstva poluchenija, obrabotki i vizualizacii informa-cii». M. 2015. S. 147-151.
6. Chandler D., Hemami S. VSNR: a wavelet-based visual signal-to-noise ratio for natural images // IEEE Trans. Image Process. 2007. V. 16. № 9. P. 2284-2298.
7. Saad M., Bovik A., Charrier C. Blind image quality assessment: A natural scene statistics approach in the DCT domain // IEEE Trans. Image Process. 2012. V. 21. № 8. P. 3339-3352.
8. Li C., Bovik A.C., Wu X. Blind image quality assessment using a general regression neural network // IEEE Trans. Neural Networks. 2011. V. 22. P. 793-799.
9. Tong H., Li M., Zhang H., Zhang C. Learning no-reference quality metric by examples // Proc. 11th Int. Multimedia Modelling Conf. 2005. P. 247-254.
10. Ye P., Doermann D. No-reference image quality assessment using visual codebook // IEEE International Conference on Image Processing. 2012. V. 21. № 7. P. 3129-3138.
11. Pietikainen M., Hadid A., Zhao G., Ahonen T. Computer vision using local binary patterns. Springer. 2011.
12. Ojala T., Pietikainen M., Maenpaa T. Multiresolution gray-scale and rotation invariant texture classification with local binary pat-terns // IEEE Transactions on Pattern Analysis and Machine Intelligence. 2002. V. 24. № 7. P. 971-987.
13. Khryashchev V., Shmaglit L., Shemyakov A. The application of machine learning techniques to real time audience analysis system // In: Favorskaya M., Jain L.C. (eds.) Computer Vision in Control Systems-2, Intelligent Systems Reference Library, 2015. V. 75. P. 49-69.
14. Nenakhov I.S., KHrjashhev V.V., Pavlov V.A. Ocenka kachestva izobrazhenija lica v zadachakh biometricheskojj identifikacii // Dokl. na 18-jj Mezhdunar. konf «Cifrovaja obrabotka signalov i ee primenenie (DSPA-2016) (Moskva)». 2016. T. 2. S. 954-959.
15. Sheikh H., Sabir M., Bovik A. A statistical evaluation ofrecent full reference image quality assessment algorithms // IEEE Trans. Image Process. 2006. V. 15. № 11. P. 3440-3451.
16. Wang Z., Simoncelli E., Bovik A. Multi-scale structural similarity for image quality assessment // IEEE Asilomar Conf. Signals, Systems, and Computers. 2003. P. 1398-1402.
17. Tang H., Joshi N., Kapoor A. Learning a blind measure of perceptual image quality // International Conference on Computer Vision and Pattern Recognition. 2011. P. 305-312.
18. Gao X., Li X., Tao D. Universal no reference image quality assessment metrics based on local dependency // IEEE The First Asian Conference on Pattern Recognition. 2011. P. 298-302.
19. Zhang M., Xie J., Zhou X., Fujita H. No reference image quality assessment based on local binary pattern statistics // Vis. Commun. Image Process. (VCIP). 2013. P. 1-6.
20. Geurts P., Ernst D., Wehenkel L. Extremely randomized trees // Machine Learning. 2006. V. 36. № 1. P. 3-42.
21. Pedregosa F., Varoquaux G., Gramfort A. et al. Scikit-learn: machine learning in python // Journal of Machine Learning Research. 2011. V. 12. P. 2825-2830.