Image quality is a nebulous concept with different meanings to different people. To quantify image quality a relative difference is typically calculated between a corrupted image and a ground truth image. But what metric should we use for measuring this difference? Ideally, the metric should perform well for both natural and scientific images. The structural similarity index (SSIM) is a good measure for how humans perceive image similarities, but is not sensitive to differences that are scientifically meaningful in microscopy. In electron and super-resolution microscopy, the Fourier Ring Correlation (FRC) is often used, but is little known outside of these fields. Here we show that the FRC can equally well be applied to natural images, e.g. the Google Open Images dataset. We then define a loss function based on the FRC, show that it is analytically differentiable, and use it to train a U-net for denoising of images. This FRC-based loss function allows the network to train faster and achieve similar or better results than when using L1- or L2- based losses. We also investigate the properties and limitations of neural network denoising with the FRC analysis.
翻译:图像质量是一个模糊的概念, 给不同的人带来不同的含义。 为了量化图像质量, 通常在损坏的图像和地面真实图像之间计算出相对的差别。 但是, 我们用什么衡量尺度来衡量这一差异? 理想的是, 衡量尺度应该对自然图像和科学图像都有良好的效果。 结构相似性指数( SSIM) 是衡量人类如何看待图像相似性的良好尺度, 但对于在显微镜中具有科学意义的差异并不敏感。 在电子和超分辨率显微镜中, 经常使用 Fourier Ring Concrelation( FRC), 但在这些领域之外却鲜为人所知。 我们在这里显示, FRC 也可以同样适用于自然图像, 例如 Google Open 图像数据集 。 我们然后根据 FRC 定义一个损失函数, 显示它具有分析性的差别, 并用它来训练一个U- net 来去除图像。 这个基于 FRC 的损失功能使得网络能够比使用 L1- 或 L2- 以 L2 为基础的损失更快地培训相似或取得更好的结果。 我们还调查神经网络的属性和限制 与 FRC 分析 。