All Issue

2022 Vol.10, Issue 1 Preview Page

Research Article

No-Reference Image Quality Assessment Based on Statistical Characterization of Frequency Bands

주파수 대역의 통계적 특성 분석 기반 무 참조 영상 품질 평가 방법

Do-Kyung Shin1, Won-Oh Lee2, Jin-Ho Woo3, Gyu-Tae Park4

신도경1, 이원오2, 우진호3, 박규태4

1,4Chief Research Engineer, Maritime Surface Sonar Systems, LIG Nex1
2,3Research Engineer, Maritime Surface Sonar Systems, LIG Nex1
1,4LIG 넥스원 해양연구소 소나체계개발단 수석연구원
2,3LIG 넥스원 해양연구소 소나체계개발단 선임연구원
31 March 2022. pp. 81-101
PDF Citation
Abstract
디지털 영상은 장비의 물리적인 한계 또는 네트워크의 한정된 대역폭 등으로 인한 정보 손실 및 왜곡이 발생할 수 있다. 영상이 전송되는 과정에서 발생하는 왜곡과 정보의 손실은 영상 품질에 영향을 미치며 다양한 분야에서의 활용을 위해서 왜곡의 정도를 분석하고 개선하기 위한 노력은 매우 중요해 지고 있다. 영상 품질 평가 기술은 크게 참조 영상의 사용 여부에 따라서 전 참조법(FR, Full Reference), 반 참조법(RR, Reduced Reference), 무 참조법(NR, No Reference)로 분류된다. 본 논문에서는 실제로 사람이 영상의 품질을 인지하는 원리를 이용한 시각 집중 맵과 주파수 대역 분포에 대한 통계적 특성 분석 기반으로 접근한 무 참조 영상 품질 평가 방법을 제안한다. 제안한 방법의 결과를 검증하기 위해서 KADID 및 TID2013 데이터 셋과 주관적인 점수 환산 값인 DMOS를 이용하여 결과를 분석하였다. 제안한 방법은 기존의 방법보다 색상의 변화, 밝기의 변화 등의 왜곡에서 품질 평가 정확도가 매우 높은 것을 확인할 수 있었다. 각 왜곡에 대한 SRCC/PLCC의 평균은 0.841, 0.844로 기존 방법보다 0.057~0.114 높은 정확도를 보였다. 특히 기존 방법은 color saturation1, color saturation2, brighten, darken, mean shift, quantization의 왜곡 항목에서 품질 평가 정확도가 매우 낮지만 제안한 방법은 높은 정확도를 도출하였다.
In digital images, information loss and distortion may occur due to physical limitations of equipment or limited bandwidth of networks. Distortion and loss of information that occur in the process of image transmission have a great impact on image quality. In order to utilize image information in various fields, it is very important to research technology to analyze and improve image distortion. Image quality Assessment is classified into FR (Full Reference), RR (Reduced Reference), and NR (No-Reference) depending on whether a reference image is used. In this paper, we propose a no-reference image quality assessment based on a visual saliency map and statistical characteristic analysis of frequency bands using the principle of human perception of image quality. To verify the results of the proposed method, KADID and TID2013 were used as data sets. In addition, the quality evaluation results were analyzed using DMOS, which was subjectively converted to scores. The proposed method was able to confirm higher accuracy in distortion such as color change and brightness change than the existing method. The averages of SRCC/PLCC for each distortion were 0.841 and 0.844, which showed 0.057~0.114 higher accuracy than the existing method. In particular, the existing method showed very low quality evaluation accuracy in the distortion items of color saturation1, color saturation2, brighten, darken, mean shift, and quantization, but the proposed method showed high accuracy.
Keywords
No-reference
IQA
Distortion image
Saliency map
Wavelet transform
Homomorphic filtering
References
  1. 임철수, "빅데이터 서비스 모델 및 사업화 주요 이슈에 대한 연구", 한국차세대컴퓨팅학회 논문지, 제10권 제3호, pp.90-98, 2014.
  2. 고종국, 배유석, 박종열, 박경, "영상 빅데이터 분석기술 동향", 전자통신동향분석, 제29권 제4호, pp.21-29, 2014.
  3. 이상광, 유원영, 서영호, "영상 화질평가 기술동향", 전자통신동향분석, 제27권 제3호, pp.83-91, 2012.
  4. 신도경, 김재경, "특징 정보 공간 도메인 변환 기반의 무 기준 영상 품질 평가를 이용한 왜곡 영상 분류", 융복합지식학회 논문지, 제9권 제2호, pp.49-69, 2021.
  5. C. H. Lee, "Data Quality Assessment Procedure Manual", KDB, pp. 1-204, 2009.
  6. 이대열, 김종호, 정세윤, 조승현, 김휘용, 최진수, "정지영상 및 동영상 인지화질 측정 기술 동향", 전자통신동향분석, 제33권 제3호, pp.11-21, 2018.
  7. H. R. Sheikh, A. C. Bovik, and L. Cormack, "No-Reference Quality Assessment Using Natural Scene Statistics:JPEG2000", IEEE Transactions on Image Processing, Vol. 14, No. 11, pp. 1918-1927, 2005.https://doi.org/10.1109/TIP.2005.854492PMid:16279189
  8. S. A. Golestaneh and K. Kitani, "No-Reference Image Quality Assessment Via Feature Fusion and Multi-Task Learning", Computer Vision and Pattern Recognition, pp. 1-10, 2020.
  9. D. Chen, Y. Wang, and W. Gao, "No-Reference Image Quality Assessment: An Attention Driven Approach", IEEE Transactions on Image Processing, Vol. 29, pp. 6496-6506, 2020.https://doi.org/10.1109/TIP.2020.2990342PMid:32386153
  10. J. J. M. Escobar, O. M. Matamoros, I. L. Reyes, R. T. Padilla, and L. C. Hernandez, "Defining a No-Reference Image Quality Assessment by Means of the Self-Affine Analysis", Multimedia Tools and Applications, pp. 1-16, 2021.https://doi.org/10.1007/s11042-020-10245-5PMid:33500679PMCid:PMC7820528
  11. Y. H. Liu, K. F. Yang, and H. M. Yan, "No-Reference Image Quality Assessment Method Based on Visual Parameters", Journal of Electronic Science and Technology, Vol. 17, No. 2, pp. 171-184, 2019.
  12. A. K. Moorthy and A. C. Bovik, "Blind Image Quality Assessment: From Natural Scene Statistics to Perceptual Quality", IEEE Transactions on Image Processing, Vol. 20, No. 12, pp. 3350-3364, 2011.https://doi.org/10.1109/TIP.2011.2147325PMid:21521667
  13. P. Ye and D. Doermann, "No-Reference Image Quality Assessment using Visual Codebook", IEEE Transactions on Image Processing, Vol. 21, No. 7, pp. 3129-3138, 2012.https://doi.org/10.1109/TIP.2012.2190086PMid:22410336
  14. H. Tang, N. Joshi, and A. Kapoor, "Learning a Blind Measure of Perceptual Image Quality", CVPR (Computer Vision Pattern Recognition), pp. 305-312, 2011.https://doi.org/10.1109/CVPR.2011.5995446
  15. M. Saad, A. C. Bovik, and C. Charrier, "Blind Image Quality Assessment: A Natural Scene Statistics Approach in the DCT Domain", IEEE Transactions on Image Processing, Vol. 21, No. 8, pp. 3339-3352, 2012.https://doi.org/10.1109/TIP.2012.2191563PMid:22453635
  16. A. Mittal, A. K. Moorthy, and A. C. Bovi, "No-Reference Image Quality Assessment in the Spatial Domain", IEEE Transactions on Image Processing, Vol. 21, No. 12, pp. 4695-4708, 2012.https://doi.org/10.1109/TIP.2012.2214050PMid:22910118
  17. A. Mittal, R. Soundararajan, and A. C. Bovik, "Making a 'Completely Blind' Image Quality Analyzer", IEEE Signal Processing Letters, Vol. 20, No. 3, pp. 209-212, 2013.https://doi.org/10.1109/LSP.2012.2227726
  18. Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, "Image Quality Assessment: From Error Visibility to Structural Similarity", IEEE Transactions on Image Processing, Vol. 13, No. 4, pp. 600-612, 2004.https://doi.org/10.1109/TIP.2003.819861PMid:15376593
  19. N. Venkatanath, D. Praneeth, Bh. M. Chandrasekhar, S. S. Channappayya, and S. S. Medasani, "Blind Image Quality Evaluation Using Perception Based Features", In Proceedings of the 21st National Conference on Communications (NCC), pp. 1-6, 2015.https://doi.org/10.1109/NCC.2015.7084843
  20. L. Zhang, L. Zhang, and A. C. Bovik, "A Feature-Enriched Completely Blind Image Quality Evaluator", IEEE Transactions on Image Processing, Vol. 24, No. 8, pp. 2579-2591, 2015.https://doi.org/10.1109/TIP.2015.2426416PMid:25915960
  21. H. Lin, V. Hosu, and D. Saupe, "KADID-10k: A Large-Scale Artificially Distorted IQA Database", 2019 Eleventh International Conference on Quality of Multimedia Experience, pp. 1-3, 2019.https://doi.org/10.1109/QoMEX.2019.8743252
  22. N. Ponomarenko, L. Jin, O. Ieremeiev, V. Lukin, K. Egiazarian, J. Astola, B. Vozel, K. Chehdi, M. Carli, F. Battisti, and C. C. J. Kuo, "Image Database TID2013: Peculiarities Results and Perspectives", Signal Processing: Image Communication, Vol. 30, pp. 57-77, 2015.https://doi.org/10.1016/j.image.2014.10.009
  23. J. Li, "Visual Saliency Based on Scale-Space Analysis in the Frequency Domain", In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 35, pp. 996-1010, 2012.https://doi.org/10.1109/TPAMI.2012.147PMid:22802112
  24. T. Ell, "Quaternion-Fourier Transforms for Analysis of Two-Dimensional Linear Time-Invariant Partial Differential Systems", Proceedings of the 32nd IEEE Conference on Decision and Control, Vol. 2, pp. 1830-1841, 1993.
  25. C. C. Tseng and S. L. Lee, "A Weak-Illumation Image Enhancement Method Using Homomorphic Filter and Image Fusion", 2017 IEEE 6th Global Conference on Consumer Electonics (GCCE), pp. 1-2, 2017.https://doi.org/10.1109/GCCE.2017.8229192
  26. 이진욱, 이현욱, 유철상, "이변량 웨이블릿 분석을 위한 모 웨이블릿 선정", 한국수자원학회 논문집, 제52권 제11호, pp.905-916, 2019.
Information
  • Publisher :The Society of Convergence Knowledge
  • Publisher(Ko) :융복합지식학회
  • Journal Title :The Society of Convergence Knowledge Transactions
  • Journal Title(Ko) :융복합지식학회논문지
  • Volume : 10
  • No :1
  • Pages :81-101
  • DOI :https://doi.org/10.22716/sckt.2022.10.1.009
Journal Informaiton The Society of Convergence Knowledge Transactions The Society of Convergence Knowledge Transactions
Online Submission
  • NRF
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close