Loading...

Wavelet image denoising based on improved thresholding neural network and cycle spinning

Sahraeian, M. E ; Sharif University of Technology | 2007

183 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/ICASSP.2007.365975
  3. Publisher: 2007
  4. Abstract:
  5. In this paper we propose a new method for image noise reduction based on wavelet transform. In this method we: introduce an improved version of thresholding neural networks. (TNN) by utilizing a new class of smooth nonlinear thresholding functions as the activation function. Using this approach we will find the best thresholds in the sense of minimum mean square error (MMSE). Then using, TNN with obtained thresholds, we employ a cycle-spinningbased technique to reduce image artifacts. Experimental results indicate that the proposed method outperforms several other established wavelet denoising techniques, in terms of Peak-Signal-to-Noise-Ratio (PSNR) and visual quality. © 2007 IEEE
  6. Keywords:
  7. Activation functions ; Cycle spinning ; Image denoising ; Nonlinear thresholding functions ; Image enhancement ; Mean square error ; Neural networks ; Signal to noise ratio ; Wavelet transforms ; Image analysis
  8. Source: 2007 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '07, Honolulu, HI, 15 April 2007 through 20 April 2007 ; Volume 1 , 2007 , Pages I585-I588 ; 15206149 (ISSN); 1424407281 (ISBN); 9781424407286 (ISBN)
  9. URL: https://ieeexplore.ieee.org/document/4217147