Automatic Focus Algorithms for a Sequence of Microscopic Cytological Images

S. Schüpp and H. Cardot (France)


Autofocus, automated microscopy, medical imaging, image acquisition


This study takes place in a project of cell classification from cellular colored spreading on slides. Segmentation is compute after on acquired images. Then, neuronal networks classify identified cells from 46 characteristics on each cell. Organ type and staining are sensible points for later steps that are segmentation and classification. Another sensible point of the acquisition is the type of preparation. The movements of the slides under microscope depend on the type of preparation. In this paper, we consider only the movements to obtain regular sampling. Autofocus algorithms are of particular importance in scanning microscope systems. The focus may have to be adjusted when the system mechanically moves from field to field. In general, algorithms that determine optimal focus for an image are based upon maximizing or minimizing some focus function. As the total scan time is usually important, algorithms have to be fast. In our paper, we make a review of different criteria for automated focus. After choosing the best criteria, we determine the best strategy of movement for acquiring a sequence of images. We propose a method that solves the problem of objects at different levels of depth.

Important Links:

Go Back