NSS是一个强大的通用NR-IQA工具。其动机是高质量的自然场景图像服从某种统计特性,而质量下降可能会偏离这些统计特性。
一个典型的基于NSS的NR-IQA测量包括3个关键步骤:特征提取、NSS建模和特征回归。特征可以从空间域或变换域中提取。参数化模型如GGD、多元GGD和非对称GGD(AGGD)被用于NSS建模步骤。最后利用支持向量机(SVM)和支持向量回归(SVR)对NSS模型的参数进行回归得到最终的质量。
在DCT域对NSS进行建模具有计算量小的优点。
随着机器学习研究的兴起,近年来提出了许多基于学习的NR-IQA方法。
基于codebook的传统机器学习方法以后再补充。
基于CNN的方法:
Reference:
[1] Guangtao ZHAI,Xiongkuo MIN.Perceptual image quality assessment: a survey[J/OL].Science China(Information Sciences):1-52[2020-05-12].
[2] Moorthy A K, Bovik A C. A two-step framework for constructing blind image quality indices. IEEE Signal Process Lett, 2010, 17: 513–516
[3] Moorthy A K, Bovik A C. Blind image quality assessment: from natural scene statistics to perceptual quality. IEEE Trans Image Process, 2011, 20: 3350–3364
[4] Tang H, Joshi N, Kapoor A. Learning a blind measure of perceptual image quality. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2011. 305–312
[8] Saad M A, Bovik A C, Charrier C. A DCT statistics-based blind image quality index. IEEE Signal Process Lett, 2010, 17: 583–586
[9] Saad M A, Bovik A C, Charrier C. Blind image quality assessment: a natural scene statistics approach in the DCT domain. IEEE Trans Image Process, 2012, 21: 3339–3352
[177] Mittal A, Soundararajan R, Bovik A C. Making a “completely blind” image quality analyzer. IEEE Signal Process Lett, 2013, 20: 209–212
[223] Kang L, Ye P, Li Y, et al. Convolutional neural networks for no-reference image quality assessment. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2014. 1733–1740
[224] Bosse S, Maniry D, Wiegand T, et al. A deep neural network for image quality assessment. In: Proceedings of IEEE International Conference on Image Processing, 2016. 3773–3777
[225] Kang L, Ye P, Li Y, et al. Simultaneous estimation of image quality and distortion via multi-task convolutional neural networks. In: Proceedings of IEEE International Conference on Image Processing, 2015. 2791–2795
[226] Kim J, Lee S. Fully deep blind image quality predictor. IEEE J Sel Top Signal Process, 2017, 11: 206–220
[227] Gu J, Meng G, Redi J A, et al. Blind image quality assessment via vector regression and object oriented pooling. IEEE Trans Mult, 2018, 20: 1140–1153
[228] Kim J, Nguyen A D, Lee S. Deep CNN-based blind image quality predictor. IEEE Trans Neural Netw Learn Syst, 2019, 30: 11–24
[229] Pan D, Shi P, Hou M, et al. Blind predicting similar quality map for image quality assessment. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2018. 6373–6382
[230] Ma K, Liu W, Zhang K, et al. End-to-end blind image quality assessment using deep neural networks. IEEE Trans Image Process, 2018, 27: 1202–1213
[231] Lin K Y, Wang G. Hallucinated-iqa: no-reference image quality assessment via adversarial learning. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2018. 732–741
[232] Liu X, van de Weijer J, Bagdanov A D. Rankiqa: learning from rankings for no-reference image quality assessment. In: Proceedings of IEEE International Conference on Computer Vision, 2017. 1040–1049
[233] Talebi H, Milanfar P. NIMA: neural image assessment. IEEE Trans Image Process, 2018, 27: 3998–4011
[234] Guan J, Yi S, Zeng X, et al. Visual importance and distortion guided deep image quality assessment framework. IEEE Trans Mult, 2017, 19: 2505–2520