Image Quality Assessment Based on Visual Characteristics

doi:10.19678/j.issn.1000-3428.0064923

Abstract

Abstract:

To solve the problems of conventional image quality assessment algorithms, i.e., single features and low correlation between global and local features, visual characteristics based image quality assessment algorithm are proposed based on human visual characteristics and by combining global and local features of images.The image is segmented into several visual regions based on the response of human vision to brightness and location information. Small regions are segmented based on brightness and location information; subsequently, they are merged using color and location information to generate visual bionic regions.Using the color distribution within the region and the Kullback-Leibler(KL) distance between regions to represent local and global features, respectively, the KL distance of the color distribution in adjacent regions in any space of the image is calculated to construct a binary relationship matrix, namely, the image region color similarity matrix.Subsequently, Singular Value Decomposition(SVD) is performed on the matrix to extract the similarity information of the matrix and construct image quality assessment indicators.The results of testing on three public image databases, i.e., LIVE, CSIQ, and TID2008, show that compared with the results of Peak Signal-to-Noise Ratio(PSNR), Structural Similarity(SSIM), Blind/Referenceless Image Spatial Quality Evaluator(BRISQUE), and Blind Image Quality Index(BIQI) algorithms, the Pearson Linear Correlation Coefficient(PLCC), Spearman Rank Order Correlation Coefficient(SROCC), and Kendall Rank Order Correlation Coefficient (KROCC) of the algorithm show stable and excellent results, and the PLCC, SROCC, and KROCC indicate values of (0.959 0, 0.947 8, 0.865 3) on the LIVE database and(0.940 4, 0.938 8, and 0.801 7) on the CSIQ database.The assessment results yielded by the proposed algorithm are highly consistent with the subjective quality assessment results of the human eye; thus the proposed algorithm can accurately evaluate image quality.

Key words: global features, local features, image quality assessment, Kullback-Leibler(KL) distance, color similarity matrix

摘要：

针对传统图像质量评价算法特征单一、全局和局部特征相关性不高的问题，从人眼视觉特性出发，结合图像全局与局部特征，提出一种基于视觉特性的图像质量评价算法。依据人眼视觉对亮度的响应和位置信息将图像划分为各个视觉区域，根据亮度和位置信息分割小区域，再利用颜色和位置信息对小区域进行合并以生成视觉仿生区域。利用区域内的颜色分布和区域之间的KL距离分别表示局部特征和全局特征，计算图像任意空间近邻区域颜色分布的KL距离以构建二元关系矩阵，即图像区域颜色相似性矩阵。在此基础上，对矩阵进行奇异值分解，提取矩阵的相似性信息，构造图像质量评价指标。在LIVE、CSIQ、TID2008这3个公开图像数据库上进行测试，结果表明，与PSNR、SSIM、BRISQUE、BIQI等算法相比，该算法的Pearson线性相关系数（PLCC）、Spearman等级相关系数（SROCC）、Kendall等级相关系数（KROCC）均取得了稳定且优异的结果，在LIVE和CSIQ数据库上的PLCC、SROCC和KROCC分别达到0.959 0、0.947 8、0.865 3和0.940 4、0.938 8、0.801 7，其评价结果与人眼主观质量评价结果具有高度的一致性，能够较好地评价图像质量。

关键词: 全局特征, 局部特征, 图像质量评价, KL距离, 颜色相似性矩阵

Ziwei HUI, Kun HE, Ben FENG, Yao SU. Image Quality Assessment Based on Visual Characteristics[J]. Computer Engineering, 2023, 49(7): 189-195.

惠子薇, 何坤, 冯犇, 苏曜. 基于视觉特性的图像质量评价[J]. 计算机工程, 2023, 49(7): 189-195.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0064923

https://www.ecice06.com/EN/Y2023/V49/I7/189

Figures/Tables 11

References 30

1	ZHAI G T, MIN X K. Perceptual image quality assessment: a survey. Science China(Information Sciences), 2020, 63 (11): 84- 135.
2	熊超, 沈远彤, 杨迪威. 一种针对高斯模糊图像的无参质量评价方法. 计算机工程与应用, 2013, 49 (15): 192-194, 237 doi: 10.3778/j.issn.1002-8331.1111-0475
	XIONG C, SHEN Y T, YANG D W. No reference image quality assessment method for Gaussian blurred images based on wavelet power spectrum. Computer Engineering and Applications, 2013, 49 (15): 192-194, 237 doi: 10.3778/j.issn.1002-8331.1111-0475
3	曹健. 基于局部特征的图像目标识别技术研究[D]. 北京: 北京理工大学, 2010.
	CAO J. Research on image target recognition technology based on local features[D]. Beijing: Beijing Institute of Technology, 2010. (in Chinese)
4	WANG Q, CHU J, XU L, et al. A new blind image quality framework based on natural color statistic. Neurocomputing, 2016, 173, 1798- 1810. doi: 10.1016/j.neucom.2015.09.057
5	高绍姝, 金伟其, 王岭雪, 等. 基于颜色协调性的典型场景彩色融合图像颜色质量评价. 北京理工大学学报, 2012, 32 (10): 1054- 1060. doi: 10.3969/j.issn.1001-0645.2012.10.012
	GAO S S, JIN W Q, WANG L X, et al. Color-quality assessment for color fusion images of typical scenes based on color harmony. Transactions of Beijing Institute of Technology, 2012, 32 (10): 1054- 1060. doi: 10.3969/j.issn.1001-0645.2012.10.012
6	WANG Z, BOVIK A C, SHEIKH H R, et al. Image quality assessment: from error visibility to structural similarity. IEEE Transactions on Image Processing, 2004, 13 (4): 600- 612. doi: 10.1109/TIP.2003.819861
7	廖宇, 郭黎. 基于梯度相关性分解的无参考图像质量评价方法. 计算机应用, 2013, 33 (3): 691- 694. URL
	LIAO Y, GUO L. New no-reference image quality assessment method based on decomposition of gradient similarity. Journal of Computer Applications, 2013, 33 (3): 691- 694. URL
8	ZHANG Z C, SUN W, MIN X K, et al. A full-reference quality assessment metric for fine-grained compressed images[C]//Proceedings of International Conference on Visual Communications and Image Processing. Washington D. C., USA: IEEE Press, 2022: 1-4.
9	ZHAN Y B, ZHANG R, WU Q. A structural variation classification model for image quality assessment. IEEE Transactions on Multimedia, 2017, 19 (8): 1837- 1847. doi: 10.1109/TMM.2017.2689923
10	WANG X, TIAN B F, LIANG C, et al. Blind image quality assessment for measuring image blur[C]//Proceedings of Congress on Image and Signal Processing. Washington D. C., USA: IEEE Press, 2008: 467-470.
11	MOUSAVI S M H, MOSAVI S M H. A new edge and pixel-based image quality assessment metric for colour and depth images[C]//Proceedings of the 9th Iranian Joint Congress on Fuzzy and Intelligent Systems. Washington D. C., USA: IEEE Press, 2022: 1-11.
12	LIU D B, CHEN Z B, MA H D, et al. No reference block based blur detection[C]//Proceedings of International Workshop on Quality of Multimedia Experience. Washington D. C., USA: IEEE Press, 2009: 75-80.
13	ZHAN Y B, ZHANG R. No-reference JPEG image quality assessment based on blockiness and luminance change. IEEE Signal Processing Letters, 2017, 24 (6): 760- 764. doi: 10.1109/LSP.2017.2688371
14	AGHAPOUR MALEKI S, GHASSEMIAN H. Spatial quality assessment of pansharpened images based on gray level co-occurrence matrix[C]//Proceedings of International Conference on Machine Vision and Image Processing. Washington D. C., USA: IEEE Press, 2022: 1-6.
15	ZHU X, MILANFAR P. Automatic parameter selection for denoising algorithms using a no-reference measure of image content. IEEE Transactions on Image Processing: A Publication of the IEEE Signal Processing Society, 2010, 19 (12): 3116- 3132. doi: 10.1109/TIP.2010.2052820
16	SHEIKH H R, BOVIK A C, CORMACK L. No-reference quality assessment using natural scene statistics: JPEG2000. IEEE Transactions on Image Processing, 2005, 14 (11): 1918- 1927. doi: 10.1109/TIP.2005.854492
17	MOORTHY A K, BOVIK A C. A two-step framework for constructing blind image quality indices. IEEE Signal Processing Letters, 2010, 17 (5): 513- 516. doi: 10.1109/LSP.2010.2043888
18	MITTAL A, MOORTHY A K, BOVIK A C. No-reference image quality assessment in the spatial domain. IEEE Transactions on Image Processing: A Publication of the IEEE Signal Processing Society, 2012, 21 (12): 4695- 4708. doi: 10.1109/TIP.2012.2214050
19	MA K D, LIU W T, ZHANG K, et al. End-to-end blind image quality assessment using deep neural networks. IEEE Transactions on Image Processing, 2018, 27 (3): 1202- 1213. doi: 10.1109/TIP.2017.2774045
20	GAO F, WANG Y, LI P, et al. DeepSim: deep similarity for image quality assessment. Neurocomputing, 2017, 257, 104- 114. doi: 10.1016/j.neucom.2017.01.054
21	JIN C, ZHAO X N, XIONG Q, et al. Blind image quality assessment for multiple distortion image. Circuits, Systems, and Signal Processing, 2022, 41 (10): 5807- 5826. doi: 10.1007/s00034-022-02055-x
22	刘海, 杨环, 潘振宽, 等. 基于生成感知差异的无参考图像质量评价模型. 计算机工程, 2021, 47 (5): 205- 212. URL
	LIU H, YANG H, PAN Z K, et al. No-reference image quality assessment model based on generated perceptual difference. Computer Engineering, 2021, 47 (5): 205- 212. URL
23	史玉华, 张闯, 迟兆鑫. 基于多元特征的立体图像质量评价方法. 计算机工程, 2021, 47 (12): 256- 265. URL
	SHI Y H, ZHANG C, CHI Z X. Method for stereoscopic image quality assessment based on multiple features. Computer Engineering, 2021, 47 (12): 256- 265. URL
24	SHEIKH H R. LIVE image quality assessment database[EB/OL]. [2022-05-05]. http://live.ece.utexas.edu/research/quality.
25	LARSON E C, CHANDLER D M. Categorical subjective image quality CSIQ database[EB/OL]. [2022-05-05]. http://vision.okstate.edu/csiq.
26	PONOMARENKO N, LUKIN V, ZELENSKY A. TID2008—a database for evaluation of full-reference visual quality assessment metrics[EB/OL]. [2022-05-05]. https://ponomarenko.info/papers/mre2009tid.pdf.
27	MITTAL A, SOUNDARARAJAN R, BOVIK A C. Making a "completely blind" image quality analyzer. IEEE Signal Processing Letters, 2013, 20 (3): 209- 212. doi: 10.1109/LSP.2012.2227726
28	YE P, KUMAR J, KANG L, et al. Unsupervised feature learning framework for no-reference image quality assessment[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2012: 1098-1105.
29	HOU W L, GAO X B, TAO D C, et al. Blind image quality assessment via deep learning. IEEE Transactions on Neural Networks and Learning Systems, 2015, 26 (6): 1275- 1286. doi: 10.1109/TNNLS.2014.2336852
30	HE L H, TAO D C, LI X L, et al. Sparse representation for blind image quality assessment[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2012: 1146-1153.

[1]	ZHU Qinquan, WANG Tonghan, JIA Huizhen. Dehazed Image Quality Assessment Method Based on Fog Density Perception Enhancement [J]. Computer Engineering, 2023, 49(6): 208-216,226.
[2]	WANG Kuan, YANG Huan, PAN Zhenkuan, SI Jianwei. Full-Reference Model for Stereoscopic Image Quality Assessment Based on Monocular and Binocular Visual Information [J]. Computer Engineering, 2022, 48(2): 207-214,223.
[3]	LIU Hai, YANG Huan, PAN Zhenkuan, HUANG Baoxiang, HOU Guojia. No-Reference Image Quality Assessment Model Based on Generated Perceptual Difference [J]. Computer Engineering, 2021, 47(5): 205-212.
[4]	SHI Yuhua, ZHANG Chuang, CHI Zhaoxin. Method for Stereoscopic Image Quality Assessment Based on Multiple Features [J]. Computer Engineering, 2021, 47(12): 256-265.
[5]	XIE Rui, SHAO Kun, HUO Xing, MITHUN Md Masud Parvej. An Improved DIQaM_FR/NR Image Quality Assessment Model [J]. Computer Engineering, 2020, 46(8): 258-263,270.
[6]	XU Shaoping,LIN Guanxi,ZENG Xiaoxia,JIANG Yinnan,TANG Yiling. Research and Prospect of Stereoscopic Image Quality-aware Feature Extraction [J]. Computer Engineering, 2018, 44(6): 239-248.
[7]	MA Yun,WANG Xiaodong,FU Xianzu,LOU Daping,ZHANG Lianjun. Non-reference Stereoscopic Image Quality Assessment Based on GA-SVR Model [J]. Computer Engineering, 2017, 43(5): 234-239,247.
[8]	FENG Mingkun,ZHAO Shengmei,SUN Lihui,SHI Xiang. Image Quality Assessment Based on Local Gaussian Weighted Fusion [J]. Computer Engineering, 2016, 42(8): 237-242.
[9]	CHEN Hao,CHEN Jian,YE Qingzhou,CAI Zhiming. Autofocus Algorithm Based on Adjacent Pixel Difference and NRSS [J]. Computer Engineering, 2015, 41(9): 261-265.
[10]	YI Sanli,MIAO Ying,QIAN Jie,GUO Beibei,XIANG Yan. Image Quality Assessment of Edge Structure Similarity Based on Region of Interest [J]. Computer Engineering, 2015, 41(6): 216-220.
[11]	ZHOU Zheng, TIAN Xin, ZHOU Cheng, FENG Wei-dong, LI Tao, XIONG Cheng-yi. Video Image Quality Assessment Method Based on Corner Feature Detection [J]. Computer Engineering, 2013, 39(9): 30-33,44.
[12]	ZHANG Chao, LIANG De-Qun, WANG Xin-Nian. Image Fuzzy Degree Assessment Method Based on Sparse Representation [J]. Computer Engineering, 2013, 39(4): 267-271.
[13]	ZHANG Tao, LIANG De-qun, WANG Xin-nian, LIU Li-juan. Image Blur Degree Assessment Method Based on 3D Structure Tensor [J]. Computer Engineering, 2013, 39(11): 230-234.
[14]	MA Li-Gong, GONG Zi-Beng. Image Quality Assessment Method for Frequency and Direction Sensitive SSIM [J]. Computer Engineering, 2012, 38(5): 19-24.
[15]	LONG Lu-Lu, LI Cong-Li, LUO Jun. Image Quality Assessment Method Based on TV and SSIM [J]. Computer Engineering, 2012, 38(3): 215-217.

Please choose a citation manager

Content to export