An Auto-Focusing Approach for Micro Objects at Different Focal Planes

In this paper, we propose an auto focusing method to find the best plane of focus amongst different depth micro objects. By achieving this, the ultimate goal of the project will be autofocusing on motile micro-organisms to check their quality, e.g., semen motility. Firstly, the auto focusing system cycles through the rough focal range of the micro objects, which represent cell culture by moving in steps of 2 μ m towards best focus, passing it, and completing the scanning process. Simultaneously, at each step, an image of the sample is collected and transformed by use of a Prewitt edge detection operator in order to ascertain focal quality. Secondly, the system moves to the best focus position that is calculated via Prewitt edge detection approach. Thirdly, to begin actively monitoring the focal plane, the system makes minute adjustments to keep the target objects in-focus. This accounts for any changes in the culture, such as cells moving in and out of the focal window or up and down within the semen fluid. The Prewitt operator is tested for focusing performance on multi-microobjects at different focal planes and is compared with other edge detection gradient functions to confirm superiority.