Search papers, labs, and topics across Lattice.
This paper introduces a hyperspectral image denoising framework that balances data fidelity with noise prior modeling by using an adaptive weight tensor to dynamically adjust the influence of fidelity and regularization terms. They model comprehensive noise priors with fewer parameters and combine a pixel-wise model with a representative coefficient total variation regularizer. Experiments on both simulated and real-world datasets show the proposed method achieves state-of-the-art denoising performance with competitive computational efficiency.
Achieve state-of-the-art hyperspectral image denoising by adaptively balancing data fidelity and noise priors, outperforming existing methods that overemphasize image priors.
The core challenge of hyperspectral image denoising is striking the right balance between data fidelity and noise prior modeling. Most existing methods place too much emphasis on the intrinsic priors of the image while overlooking diverse noise assumptions and the dynamic trade-off between fidelity and priors. To address these issues, we propose a denoising framework that integrates noise prior reduction and a spatial-spectral adaptive fidelity term. This framework considers comprehensive noise priors with fewer parameters and introduces an adaptive weight tensor to dynamically balance the fidelity and prior regularization terms. Within this framework, we further develop a fast and robust pixel-wise model combined with the representative coefficient total variation regularizer to accurately remove mixed noise in HSIs. The proposed method not only efficiently handles various types of noise but also accurately captures the spectral low-rank structure and local smoothness of HSIs. An efficient optimization algorithm based on the alternating direction method of multipliers is designed to ensure stable and fast convergence. Extensive experiments on simulated and real-world datasets demonstrate that the proposed model achieves superior denoising performance while maintaining competitive computational efficiency.
