Modifications to Image Phase Alignment Super-sampling Produce up to 4.4 times Increased Image Resolution

TL;DR

This paper presents modifications to the ImPASS algorithm that significantly enhance image resolution, achieving up to 4.4 times higher resolution and surpassing previous limits by improving alignment and deconvolution techniques.

Contribution

The study introduces algorithmic modifications that improve image alignment and deconvolution in ImPASS, leading to substantial resolution gains beyond prior capabilities.

Findings

Image resolution improved by up to 4.41 times.

Modified algorithm surpasses diffraction limit by a factor of 2.57.

Further improvements are possible as physical limits are not yet reached.

Abstract

Image Phase Alignment Super-sampling (ImPASS) is a computational method for combining displaced low-resolution images into a single high-resolution image. The general steps include measuring the relative displacements, up-sampling, aligning and combining the images, followed by a blind deconvolution. Previous ImPASS studies have shown that the resulting image resolution can significantly subceed the diffraction limit of the imaging system. Characteristics that potentially limit the processed image resolution include optical parameters, detector noise, image alignment accuracy, or deconvolution parameters. In this report, modifications have been made to the algorithm to improve the image alignment accuracy and deconvolution. Applications of the modified algorithm improved image resolution by a factor up to 1.81. Compared to the original image resolution, the modified ImPASS achieved a resolution improvement factor up to 4.41 while subceeding the diffraction limit by a factor of 2.57. This suggests that limitations imposed by the physical properties of the system have not yet been reached, and further improvement of the algorithm is warranted.