Abstract:The phenomenon of continuous out-of-focus imaging often occurs in high-magnification optical microscopy when observing large-scale targets. Lacking of accurate depth-varying point spread functions (DVPSFs) for blurred regions at different depths, it is difficult to locally reconstruct the clear images of these blurred regions using traditional deblurring methods, making it unreasonable to globally observe the optical features of large-scale targets in high-magnification optical microscopy. This paper proposes a global deblurring method for continuous out-of-focus images of large-scale sphere samples. In this study, first we analyze the energy diffusion characteristics of the optical imaging process, integrating the relationship between high-frequency energy parameters, optical range distance, and depth of field, and we propose a three-dimensional continuous energy diffusion model for optical imaging. Next, we propose an adaptive weight depth calculation method for a continuously changing surface based on the depth varying diffusion model by introducing the sample surface curvature variation and light direction. Finally, we propose a universal method for deblurring continuous out-of-focus images of large-scale sphere samples for the purpose of observing the global optical features in high-magnification optical microscopy. Moreover, we use dynamic microspheres of different sizes to verify the effectiveness of our proposed method. The results prove that our proposed method can accurately calculate the depth of the sample surface and the energy diffusion parameters at each depth, and it can achieve the image deblurring of a continuously changing surface and the global deblurring of multiple samples in a wide field of view.

Shape from defocus via diffusion

Shape Reconstruction Within an Optical Diffusion Model.

Global Depth from Defocus with Fixed Camera Parameters

Behavioral Analysis of Anisotropic Diffusion in Image Processing .

Depth reconstruction with Fresnel diffraction

Depth Recovery from Defocus Based on Relative Blur and Edge Enhancement

Depth from Defocus in the Wild.

Shape Reconstruction Based on a New Blurring Model at the Micro/Nanometer Scale

Multi-focus Image Fusion Based on Depth Extraction with Inhomogeneous Diffusion Equation

Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution

Nanoscale Depth Reconstruction from Defocus: Within an Optical Diffraction Model.

Global Deblurring for Continuous Out-of-focus Images Using a Depth-Varying Diffusion Model.

Defocus Fuzzy Image Restoration

A Novel Approach to Recovering Depth from Defocus

Defocus Blur Synthesis and Deblurring via Interpolation and Extrapolation in Latent Space

Novel Algorithm of Recovering Depth from Defocus Image

Nondestructive Image De-Blurring Based on Diffraction Blurring Model

Curved Diffusion: A Generative Model With Optical Geometry Control

A NEW DEPTH RECOVERY ALGORITHM BASED ON DEFOCUS IMAGE

Diffuse3D: Wide-Angle 3D Photography Via Bilateral Diffusion

ShapeFusion: A 3D diffusion model for localized shape editing