Proceedings of the
The Nineteenth International Conference on Computational Intelligence and Security (CIS 2023)
December 1 – 4, 2023, Haikou, China
Adaptive Regularization Parameter Selection in L1-TV Model for Impulse Noise Removal
1Faculty of Science, Kunming University of Science and Technology, P.R.China.
2College of Engineering, Computing and Cybernetics, Australian National University, Australia.
3Yunnan Key Laboratory of Modern Analytical, Yunan Normal University, P.R.China.
ABSTRACT
In the fields of medical imaging, military target recognition, network security, and image processing, the problem of impulse noise removal arises. The L1-TV model, which combines the L1 data-fitting term and total variation term, is commonly applied to recover degraded images. However, the quality of the recovered image depends on the selection of the regularization parameter. In this paper, we propose an adaptive strategy to select the regularization parameter such that the data-fitting term is bounded by an estimated value. Experimental results show that the proposed method is superior to existing variational image restoration methods.
Keywords: Image denoising, De-blur, Impulse noise, L1-TV model, Regularization parameter.
Download PDF
1Faculty of Science, Kunming University of Science and Technology, P.R.China.
2College of Engineering, Computing and Cybernetics, Australian National University, Australia.
3Yunnan Key Laboratory of Modern Analytical, Yunan Normal University, P.R.China.
ABSTRACT
In the fields of medical imaging, military target recognition, network security, and image processing, the problem of impulse noise removal arises. The L1-TV model, which combines the L1 data-fitting term and total variation term, is commonly applied to recover degraded images. However, the quality of the recovered image depends on the selection of the regularization parameter. In this paper, we propose an adaptive strategy to select the regularization parameter such that the data-fitting term is bounded by an estimated value. Experimental results show that the proposed method is superior to existing variational image restoration methods.
Keywords: Image denoising, De-blur, Impulse noise, L1-TV model, Regularization parameter.
Download PDF