Binary pseudo-random array for calibration of imaging optical microscopes

Imaging systems are crucial in various fields, and extracting accurate information from raw images is essential. However, real-world limitations affect image data quality. Therefore, thoroughly understanding imaging system performance and developing techniques to improve data accuracy remains a major challenge. To develop a technology for quantitative calibration and correction of image processing imperfections that is applicable to any imaging system without hardware changes, we introduce and validate a binary pseudo-random array (BPRA) test artifact. The fabricated BPRA test artifact is designed to create a contrast pattern with binary reflectance or transmission that is suitable for Instrument Transfer Function (ITF) characterization of optical imaging microscopes. We have demonstrated the efficacy of BPRA test artifacts to characterize the resolution limit and the geometrical distortion of a Nikon MM-800/L optical microscope and have used a highly-randomized array (HRA) to measure the ITF of a Zeiss Elyra 7 microscope and reconstruct images of yeast cells. We found that the frequency response of the Zeiss microscope exhibits complex behavior beyond Gaussian approximation, emphasizing the importance of empirical ITF characterization. In this respect, the Power Spectral Density (PSD) of an HRA measurement (after correction for sampling due to the detector pixel size) provides an empirical function for the microscope ITF, even in the absence of an imaging model. If one seeks to understand how different microscope settings impact the ITF, the BPRA provides a comprehensive answer in a single measurement. Thus, proof-of-principle efficacy and the value of the BPRA test artifacts developed for optical microscopy were successfully demonstrated.