Automated sample tracking and parameter adaption for scanning laser optical tomography

Abstract

Non-destructive, three-dimensional imaging techniques are of great importance in medicine as well as in technical analysis. In this context, it is of particular importance to generate reliable and repeatable results of high quality. This can be aided by automation of manual processes. One of these imaging techniques, the Scanning Laser Optical Tomography, currently requires manual sample alignment by the user to achieve the highest possible image quality. This alignment demands skillful hand-eye coordination as well as experience on the part of the user, and thus often leads to inconsistent imaging results. To overcome this problem, this paper presents a technique for software-based automation of this challenge. The sample is not physically aligned, but digitally detected and tracked during the acquisition. Residual motion artifacts that interfere with tomographic reconstruction are corrected using a second automation algorithm. The combination of the two new algorithms significantly improves the quality of imaging and also increases the reliability and degree of automation of the system, making it accessible to a wide range of users.