Performance Analysis of Smoothing Techniques in context with Image Processing

  • Tejas G. Patil
  • S K. Kusekar
  • Masresha Adasho
Keywords: Image smoothing; Median filtering;Robust filtering; PSNR; Real time image.

Abstract

Typical flow of an image processing application involve stages like pre-processing, feature extraction and classification or recognition. Image smoothing is one of the pre-processing tasks which balance the effect of noise while capturing the images in specific applications. Smoothing address poor quality of captured image by reducing the noise from it, thereby enhancing accuracy of pattern classification or recognition algorithm. Performance of some frequently used smoothing techniques viz–Median filtering, Gaussian filtering, Robust filtering and Mean filtering (averaging)is analyzed in this work. Sample high quality images are contaminated with predefined levels of Gaussian noise, Speckle noise and Impulse valued noise (salt and pepper noise) manually. Performance of mentioned smoothing techniques is also checked for real time captured images through webcam. Peak Signal to Noise Ratio (PSNR) is used as quality metric to compare the versatility of one method with the other.  Entire experimentation is implemented on an i3 processor machine using Matlab14a. It is found that the Median filtering method outperforms all other three methods to address the effect of noise.

References

[1] Aditya Goyal, Akhilesh Bijalwan, Kuntal Chowdhury,“A Comprehensive Review of Image Smoothing Techniques”International Journal of Advanced Research in Computer Engineering & Technology Volume 1, Issue 4, June 2012.
[2] Param Bagchi, Debotosh Bhattacharjee, Mita Nasipuri, , “A novel approach for nose tip detection using smoothing by weighted median filtering applied to 3-D face images in variant poses”, International conference on Pattern Recognition, Informatics and Medical Engineering (PRIME), 21-23 March 2012, Salem, TN, India.
[3] Hu cheng, Feng Huang, “MRI Image intensity correction with extrapolation and adaptive filtering,” Journal of Magnetic Resonance in Medicines, 55:959-966, 2006.
[4] Jian Yang, Jingfan Fan, Dani Ai, Shoujun Zhou, Songyuan Tang, Yongtian Wang, “ Brain MR image denoising for rician noise using pre smooth non local filter”, Journal of Biomedical Engineering, 2015.
[5] Gu Sui Cheng, Tan Ying, He XinGui, “Laplacian smoothing transform for face recognition”, J. Information sciences, Science China, Vol. 53 No. 12: 2415–2428, Dec. 2010.
[6] Ajay Boyat, Brijendra Kumar Joshi, “A review paper: noise models in digital image processing”, I J of Signal & Image Processing, Vol-6, April2015
Published
2019-04-10
How to Cite
Patil, T. G., Kusekar, S. K., & Adasho, M. (2019). Performance Analysis of Smoothing Techniques in context with Image Processing. Asian Journal For Convergence In Technology (AJCT) ISSN -2350-1146, 4(3). Retrieved from https://asianssr.org/index.php/ajct/article/view/838

Most read articles by the same author(s)

Obs.: This plugin requires at least one statistics/report plugin to be enabled. If your statistics plugins provide more than one metric then please also select a main metric on the admin's site settings page and/or on the journal manager's settings pages.