Fine-tuning convolutional neural network based on relaxed Bayesian-optimized support vector machine for random-valued impulse noise removal
Authors:
- 1 January 2023
- journal article
- research article
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 32 (01) , 013006-21
Abstract
Denoising convolutional neural network (DnCNN) has achieved competitive denoising performance for Gaussian noise using residual learning. The same idea can also be applied in the impulse noise (IN) removal by changing the training data. However, as a discriminative learning-based algorithm, DnCNN lacks flexibility in removing spatially varying noise, such as random-valued impulse noise (RVIN). To address this problem, we propose a fine-tuning CNN based on relaxed Bayesian-optimized support vector machine (BO-SVM) for RVIN removal using a relaxed BO-SVM as a detector and a fine-tuning DnCNN as a denoiser. The relaxed BO-SVM is developed to classify all pixels of the test image into two categories: noise and noise-free. A relaxed rule is provided to preserve the information of slightly corrupted pixels, and the system is trained with an optimal feature set, which comprises of newly introduced local ternary pattern with previously used gray-level difference, average background difference, and median value difference. The key hyperparameters of the relaxed SVM are obtained using the Bayesian optimization technique. Then the DnCNN works for image reconstruction through fine-tuning experiments by the use of the corrupted pixels identified by the detector. The experimental results show that the proposed method reports promising results on RVIN removal over other state-of-the-art denoising methods.Keywords
Funding Information
- National Natural Science Foundation of China (62075028)
This publication has 53 references indexed in Scilit:
- Comparative Analysis of K-Means and Fuzzy C-Means AlgorithmsInternational Journal of Advanced Computer Science and Applications, 2013
- Predictive-based adaptive switching median filter for impulse noise removal using neural network-based noise detectorSignal, Image and Video Processing, 2012
- Decision-based filter based on SVM and evidence theory for image noise removalNeural Computing and Applications, 2011
- Robust Video Restoration by Joint Sparse and Low Rank Matrix ApproximationSIAM Journal on Imaging Sciences, 2011
- An efficient approach for the removal of impulse noise from the corrupted image using neural network based impulse detectorImage and Vision Computing, 2010
- Digital Image Processing2008
- Recursive weighted median filters admitting negative weights and their optimizationIEEE Transactions on Signal Processing, 2000
- Progressive switching median filter for the removal of impulse noise from highly corrupted imagesIEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing, 1999
- Center weighted median filters and their applications to image enhancementIEEE Transactions on Circuits and Systems, 1991
- A New Method of Locating the Maximum Point of an Arbitrary Multipeak Curve in the Presence of NoiseJournal of Basic Engineering, 1964