Deep image prior (DIP) is an unsupervised deep learning framework that has been successfully applied to a variety of inverse imaging problems. However, DIP-based methods are inherently prone to overfitting, which leads to performance degradation and necessitates early stopping. In this paper, we propose a method to mitigate overfitting in DIP-based hyperspectral image (HSI) denoising by jointly combining robust data fidelity and explicit sensitivity regularization. The proposed approach employs a Smooth $\ell_1$ data term together with a divergence-based regularization and input optimization during training. Experimental results on real HSIs corrupted by Gaussian, sparse, and stripe noise demonstrate that the proposed method effectively prevents overfitting and achieves superior denoising performance compared to state-of-the-art DIP-based HSI denoising methods.

翻译：深度图像先验（Deep Image Prior, DIP）是一种非监督深度学习框架，已成功应用于多种逆成像问题。然而，基于DIP的方法本质上容易出现过拟合，导致性能退化并需要提前停止训练。本文提出一种方法，通过联合运用鲁棒的数据保真项和显式敏感性正则化，来缓解基于DIP的高光谱图像（Hyperspectral Image, HSI）去噪中的过拟合问题。所提方法在训练过程中采用平滑$\ell_1$数据项、基于散度的正则化以及输入优化。对受高斯噪声、稀疏噪声和条带噪声污染的真实高光谱图像的实验结果表明，所提方法有效防止了过拟合，与当前最先进的基于DIP的高光谱图像去噪方法相比，实现了更优越的去噪性能。