Abstract
Confocal microscopy is an established technique with manifold applications that offers the capability to perform nondestructive through-plane imaging. However, depth resolution typically decreases when focusing below the surface of a sample, which limits the applicability. A computational model is introduced that calculates the axial resolution, its decay, and the attenuation coefficient from confocal through-plane scans of translucent layers. The model is benchmarked with different polymers and objectives (air, water, oil) using a confocal Raman microscope. The algorithm requires a single through-plane scan that allows to identify the sample by signal intensity differences. It fits the point spread function of the objective at the top and bottom interface of the specimen to extract the resolution at both interfaces and the attenuation coefficient of the sample. It provides robust outputs on various and even multilayered samples if the signal-to-noise ratio of the input is sufficient and if the layers are planar and homogeneous. The algorithm of the model is provided open-source for MATLAB and Python. Quantifying microscope resolution in through-plane scans can improve image analysis in multiple fields, and this study is a comprehensive proof-of-concept for the presented model. It establishes an accessible tool to quantify the depth resolution of confocal microscopy.