Poisson2Sparse: Self-Supervised Poisson Denoising From a Single Image
Poisson2Sparse: Self-Supervised Poisson Denoising From a Single Image
复制标题
DOI：
10.1007/978-3-031-16452-1_53
复制
发表时间：
2022-06
期刊：
影响因子：
--
通讯作者：
Calvin-Khang Ta;Abhishek Aich;Akash Gupta;A. Roy-Chowdhury
中科院分区：
文献类型：
--
作者：
Calvin-Khang Ta;Abhishek Aich;Akash Gupta;A. Roy-Chowdhury
图像增强方法通常假设噪声是信号无关的，并将退化模型近似为零均值加性高斯模型。然而，这种假设不适用于生物医学成像系统，其中基于传感器的噪声源与信号强度成比例，并且噪声更好地表示为泊松过程。在这项工作中，我们探索了一种基于稀疏性和字典学习的方法，并提出了一种新的自监督学习方法，用于单图像去噪，其中噪声近似为泊松过程，不需要干净的地面实况数据。具体来说，我们近似传统的迭代优化算法的图像去噪与递归神经网络，强制稀疏相对于网络的权重。由于稀疏表示是基于底层图像的，因此能够抑制图像块中的伪分量（噪声），从而通过网络结构为去噪任务引入隐式正则化。两个生物图像数据集上的实验表明，我们的方法优于国家的最先进的方法在PSNR和SSIM。我们的定性结果表明，除了在标准定量指标上具有更高的性能外，我们还能够比其他比较方法恢复更多微妙的细节。我们的代码可在https://github.com/tacalvin/Poisson2Sparse上公开获取。
Image enhancement approaches often assume that the noise is signal independent, and approximate the degradation model as zero-mean additive Gaussian. However, this assumption does not hold for biomedical imaging systems where sensor-based sources of noise are proportional to signal strengths, and the noise is better represented as a Poisson process. In this work, we explore a sparsity and dictionary learning-based approach and present a novel self-supervised learning method for single-image denoising where the noise is approximated as a Poisson process, requiring no clean ground-truth data. Specifically, we approximate traditional iterative optimization algorithms for image denoising with a recurrent neural network that enforces sparsity with respect to the weights of the network. Since the sparse representations are based on the underlying image, it is able to suppress the spurious components (noise) in the image patches, thereby introducing implicit regularization for denoising tasks through the network structure. Experiments on two bio-imaging datasets demonstrate that our method outperforms the state-of-the-art approaches in terms of PSNR and SSIM. Our qualitative results demonstrate that, in addition to higher performance on standard quantitative metrics, we are able to recover much more subtle details than other compared approaches. Our code is made publicly available at https://github.com/tacalvin/Poisson2Sparse.