无干净参考图像超分辨率：在电子显微镜中的应用
No-Clean-Reference Image Super-Resolution: Application to Electron Microscopy
AI总结：
针对大体积脑组织难以获取干净高分辨率电镜图像的问题，提出无干净参考的超分辨率网络EMSR，对比多种训练策略，在9个脑部数据集上验证了方法的有效性与竞争力。
摘要：
无法在大体积脑组织上获取干净的高分辨率（HR）电子显微镜（EM）图像，这一问题阻碍了诸多神经科学研究。为应对这一挑战，我们提出一种基于深度学习的图像超分辨率（SR）方法，可从含噪低分辨率（LR）采集结果中，通过计算重建出大视场（FoV）的干净HR三维电子显微镜（3D-EM）图像。我们的贡献包括：一）研究了采用无干净参考的训练方式，分别使用ℓ₂和ℓ₁损失函数开展实验；二）提出一种名为EMSR的新型网络架构，可在提升LR EM图像分辨率的同时降低固有噪声；三）对比了不同训练策略，包括使用采集到的LR与HR图像对（即受真实伪影污染的无干净参考真实图像对）、合成LR与采集HR的图像对，以及采集LR与去噪HR的图像对。在9个脑部数据集上的实验表明，使用真实图像对训练可生成高质量超分辨率结果，证明了两种损失函数下采用非干净参考训练的可行性。此外，使用去噪参考与含噪参考训练时，在视觉效果和数值指标上均能得到相当的结果。另外，利用由HR图像合成生成的LR图像训练得到的网络，可有效产出令人满意的SR结果，在部分情况下甚至优于真实图像对训练的效果。我们将所提SR网络与多种成熟SR技术进行了定量和定性对比，结果表明该方法在抑制噪声、恢复精细细节方面要么性能更优，要么具备竞争力。
英文摘要：
The inability to acquire clean high-resolution (HR) electron microscopy (EM) images over a large brain tissue volume hampers many neuroscience studies. To address this challenge, we propose a deep-learning-based image super-resolution (SR) approach to computationally reconstruct clean HR 3D-EM with a large field of view (FoV) from noisy low-resolution (LR) acquisition. Our contributions are I) Investigating training with no-clean references for $\ell_2$ and $\ell_1$ loss functions; II) Introducing a novel network architecture, named EMSR, for enhancing the resolution of LR EM images while reducing inherent noise; and, III) Comparing different training strategies including using acquired LR and HR image pairs, i.e., real pairs with no-clean references contaminated with real corruptions, the pairs of synthetic LR and acquired HR, as well as acquired LR and denoised HR pairs. Experiments with nine brain datasets showed that training with real pairs can produce high-quality super-resolved results, demonstrating the feasibility of training with non-clean references for both loss functions. Additionally, comparable results were observed, both visually and numerically, when employing denoised and noisy references for training. Moreover, utilizing the network trained with synthetically generated LR images from HR counterparts proved effective in yielding satisfactory SR results, even in certain cases, outperforming training with real pairs. The proposed SR network was compared quantitatively and qualitatively with several established SR techniques, showcasing either the superiority or competitiveness of the proposed method in mitigating noise while recovering fine details.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
