Journal of the Korea Institute of Information and Communication Engineering (한국정보통신학회논문지)

The Korea Institute of Information and Commucation Engineering (한국정보통신학회)
권호 표지
DOI QR코드

DOI QR Code

Salt and Pepper Noise Removal using Histogram

히스토그램을 이용한 Salt and Pepper 잡음 제거

  • Kwon, Se-Ik (Dept. of Control and Instrumentation Eng., Pukyong National University) ;
  • Kim, Nam-Ho (Dept. of Control and Instrumentation Eng., Pukyong National University)
  • Received : 2015.10.27
  • Accepted : 2015.12.04
  • Published : 2016.02.29
Download PDF
⟨ Previous Next ⟩

Abstract

Currently, with the rapid development of the digital age, multimedia-related image devices become popular. However image deterioration is generated by multiple causes during the transmission process, with typical example of salt and pepper noise. When the noise of high density is added, existing methods are deteriorated in the characteristics of removal noise. After judging the noise condition to remove the salt and pepper noise, if the center pixel is the non-noise pixel, it is replaced with the original pixel. On the other hand, if it is the noise pixel, algorithm is suggested by the study, where the histogram of the corrupted image and the median filters are used. And for objective judgment, the proposed algorithm was compared with existing methods and PSNR(peak signal to noise ratio) was used as judgment standard. As the result of the simulation, The proposed algorithm shows a high PSNR of 32.57[dB] for Lena images that had been damaged of a high density salt and pepper noise(P=60%), Compared to the existing CWMF, A-TMF and AWMF there were improvements by 21.67[dB], 18.07[dB], and 20.13[dB], respectively.

현재, 디지털 시대의 급속 발전과 함께 멀티미디어 관련 영상 장치들이 대중화 되고 있다. 그러나 영상 데이터는 전송하는 과정에서 여러 원인으로 열화가 발생하며 주로 salt and pepper 잡음이 대표적이다. 그러나 기존의 방법들은 고밀도 잡음이 첨가된 경우, 잡음 제거 특성이 저하된다. 따라서 본 논문에서는 salt and pepper 잡음을 제거하기 위해 잡음 판단 후, 비잡음인 경우 원 화소로 대치하고, 잡음인 경우 훼손된 영상의 히스토그램과 메디안 필터를 이용하여 처리하는 알고리즘을 제안하였다. 그리고 객관적 판단을 위해 기존의 방법들과 비교하였으며, 판단의 기준으로 PSNR(peak signal to noise ratio)을 사용하였다. 시뮬레이션 결과, 제안한 알고리즘은 salt and pepper 잡음(P=60%)의 고밀도 잡음에 훼손된 Lena 영상에서 32.57[dB]의 높은 PSNR을 보이고 있고, 기존의 CWMF, A-TMF, AWMF에 비해 각각 21.67[dB], 18.07[dB], 20.13[dB] 개선되었다.

Keywords

References

  1. R. C. Gonzalez and R.E. woods, Eds., Digital Image Processing, Prentice Hall, 2007.
  2. K. K. V. Toh, H. Ibrahim, and M. N. Mahyuddin, "Salt-and-pepper noise detection and reduction using fuzzy switching median filter," IEEE trans. Consumer Electron., vol. 54, no. 4, pp. 1956-1961, Nov. 2008. https://doi.org/10.1109/TCE.2008.4711258
  3. Xu Long and Nam-Ho Kim, "A Study on the Spatial Weighted Filter in AWGN Environments," JICCE, vol. 17, no. 3, Mar. 2013.
  4. Se-Ik Kwon and Nam-Ho Kim, "A Study on Modified Spatial Weighted Filter in Mixed Noise Environments," JICCE, vol. 19, no. 1, Jan. 2015.
  5. Oten, Remzi and De Figueiredo, Rlui J P, "Adaptive Alpha-Trimmed Mean Filters Under Deviations From Assumed Noise Model," IEEE Trans, Image Processing, vol. 13, no. 5, pp. 627-639, May 2004. https://doi.org/10.1109/TIP.2003.821115
  6. Xu Long and Nam-Ho Kim, "An Improved Weighted Filter for AWGN Removal," JKIICE, vol. 17, no. 5, pp. 1227-1232, May. 2013.
  7. Jiahui Wang and Jingxing Hong, "a New Selt-Adaptive Weighted Filter for Removing Noise in Infrared images," IEEE Information Engineering and Computer Science, ICIECS International Conference, pp. 1-4, Dec. 2009.

Cited by

  1. 표준편차 및 3차 스플라인 보간법을 이용한 영상 복원 알고리즘에 관한 연구 vol.21, pp.9, 2016, https://doi.org/10.6109/jkiice.2017.21.9.1689