한국컴퓨터정보학회논문지 (Journal of the Korea Society of Computer and Information)

  • 제29권11호
  • /
  • Pages.145-152
  • /
  • 2024
  • /
  • 1598-849X(pISSN)
  • /
  • 2383-9945(eISSN)
한국컴퓨터정보학회 (Korean Society of Computer Information)
권호 표지
DOI QR코드

DOI QR Code

Improvement of Non-Local Means Algorithm Using Similarity in Image

  • Jeongwoo Lee (Division of Digital Healthcare, Yonsei University) ;
  • Heeyeon Jo (Division of Digital Healthcare, Yonsei University) ;
  • Jiyun Byun (Software Division, Yonsei University) ;
  • Hongrae Lee (Software Division, Yonsei University)
  • 투고 : 2024.10.29
  • 심사 : 2024.11.21
  • 발행 : 2024.11.29
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

스마트폰의 보급화에 의해 영상은 쉽게 획득할 수 있지만 야간의 조명 조건의 불균형성 그리고 영상 데이터의 전송과 압축 과정에서 열화와 잡음(Noise)이 생성되기도 한다. 이러한 잡음을 최소화하고 영상 화질을 개선하기 위해 Non-Local Means(NLM)은 주변의 픽셀값이 아닌 이미지 내에서 현재 patch와 유사한 patch를 찾아 잡음을 제거한다. 하지만, 유사도를 측정하는 데 있어서 유사한 pactch가 커질수록 유용성이 떨어지는 단점을 가지고 있다. 본 논문에서는 Sum of Absolute Differences 연산을 이용하여 유사도를 계산하고 유사도에 따라 가중치를 적용하는 잡음 제거 방법을 제안한다. 제안한 알고리즘을 사용하여 Salt and Pepper 잡음 이미지에 PSNR 개선이 평균 7.98dB 향상되며 NLM 대비 1.06dB 개선된다. 제안한 알고리즘을 기존 NLM 최적화 논문에 적용 하였을 때 성능 향상을 기대할 수 있다.

With the widespread adoption of smartphones, acquiring images has become easier. However, challenges arise due to uneven lighting conditions at night and the degradation and noise introduced during image transmission and compression. To minimize this noise and improve image quality, Non-Local Means (NLM) techniques are used, which unlike traditional methods, seek out patches within the image that are similar to the current patch to eliminate noise. However, a drawback of NLM is the diminishing utility as the similar patches become larger. This paper proposes a noise reduction method that utilizes the Sum of Absolute Differences to calculate similarity and applies weights accordingly. The proposed algorithm demonstrates an average improvement of 6.911dB in Peak Signal-to-Noise Ratio (PSNR) on Salt and Pepper noise images, showing a 0.713dB improvement over traditional NLM. When the proposed algorithm is applied to existing NLM optimization papers, performance improvements can be expected.

키워드

과제정보

This research was supported by the MISP(Ministry of Science and ICT), Korea, under the National Program for Excellence in SW supervised by the IITP(Institute of Information & communications Technology Planning & Evaluation) (2019-0-01219).

참고문헌

  1. Jihun Jung, Subin Choi, and Seong-Eun Kim, "A Study on the detection of photo of urine test strip using white balancing algorithm," Proceedings of Symposium of the Korean Institute of communications and Information Sciences, pp. 1335-1336, 2022.
  2. Ji-Hyeon Baek, and Nam-Ho Kim, "Noise Removal Method using Entropy in High-Density Noise Environments," Journal of the Korea Institute of Information and Communication Engineering, Vol. 24, No. 10, pp. 1255-1261, 2020.
  3. Sung Jun Kim, and Yong Ju Jung, "Deep Network for Detail Enhancement in Image Denoising," Journal of Korea Multimedia Society, Vol. 22, No. 6, pp. 646-654, 2019.
  4. MoonWon Choo, "Image Contrast Enhancement Based on Genetic Algorithm Using A Variant Histogram Equalization," The Journal of Image and Cultural Contents, Vol. 14, pp. 219-230, 2018. DOI: https://doi.org/10.24174/jicc.2018.06.14.219
  5. Eon Seung Seong, Ji Hye Heo, Seong Hyun Han, Dong Hoon Lim, "Medical image denoising using convolutional dual-decoder autoencoder," Journal of the Korean Data And Information Science Society, Vol. 33, No. 6, pp. 1065-1083, 2022. DOI: http://dx.doi.org/10.7465/jkdi.2022.33.6.1065
  6. Young-Ro Kim, "Spatio-temporal Non-local Means Filter for Noise Reduction," Journal of the Institute of Electronics and Information Engineers, Vol. 59, No. 12, pp. 97-103, 2022. DOI: 10.5573/ieie.2022.59.12.97
  7. A. Buades, B. Coll and J. . -M. Morel, "A non-local algorithm for image denoising," 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05), pp. 60-65, Vol. 2, San Diego, CA, USA, 2005. DOI: 10.1109/CVPR.2005.38.
  8. D. Van De Ville and M. Kocher, "SURE-Based Non-Local Means," IEEE Signal Processing Letters, Vol. 16, No. 11, pp. 973-976, Nov. 2009. DOI: 10.1109/LSP.2009.2027669.
  9. J. Wang, Y. Guo, Y. Ying, Y. Liu and Q. Peng, "Fast Non-Local Algorithm for Image Denoising," 2006 International Conference on Image Processing, pp. 1429-1432, Atlanta, GA, USA, 2006. DOI: 10.1109/ICIP.2006.312698.
  10. Y. Wu, B. Tracey, P. Natarajan and J. P. Noonan, "Probabilistic Non-Local Means," in IEEE Signal Processing Letters, Vol. 20, No. 8, pp. 763-766, Aug. 2013. DOI: 10.1109/LSP.2013.2263135.
  11. J. Orchard, M. Ebrahimi and A. Wong, "Efficient nonlocal-means denoising using the SVD," 2008 15th IEEE International Conference on Image Processing, pp. 1732-1735, San Diego, CA, USA, 2008. DOI: 10.1109/ICIP.2008.4712109.
  12. Antoni Buades, Bartomeu Coll, and Jean-Michel Morel, "Non-Local Means Denoising," Image Processing On Line, Vol. 1 pp. 208-212, 2011. DOI: https://doi.org/10.5201/ipol.2011.bcm_nlm
  13. Jeonghwan Kim, Minjeong Lee, and Jechang Jeong, "A Study to Calculate an Efficient Covariance Matrix of Non-local Means with Principal Components Analysis," Proceedings of the Korean Society of Broadcast Engineers Conference, pp. 205-207, 2015.
  14. Yang-Soo Kim, Jin-Woo Jeong, Yoon-Sik Choe, "Enhanced RGB Video Coding Based on Correlation in the Adjacent Block," The transactions of The Korean Institute of Electrical Engineers, Vol. 58, No. 12, pp. 2538-2541, 2009.
  15. R. Verma and R. Pandey, "Non local means algorithm with adaptive isotropic search window size for image denoising," 2015 Annual IEEE India Conference (INDICON), pp. 1-5, New Delhi, India, 2015. DOI: 10.1109/INDICON.2015.7443193.
  16. Praveen Venkatesh, Non-Local-Means, https://github.com/praveenVnktsh/Non-Local-Means
  17. OpenCV, opencv, https://github.com/opencv/opencv
  18. Metehan Cetinkaya, Python-Median-Filter, https://github.com/MeteHanC/Python-Median-Filter/tree/master