Automated detection of live cells and microspheres in low contrast bright field microscopy

Abstract

Automated robotic manipulation of micro scale objects using optical tweezers requires processing of workspace images to extract real-time information on the positions and orientations of the objects. This information is particularly challenging to extract in workspaces comprising mixture of biological and inanimate objects, such as live cells and microspheres, when the popular imaging modality of low contrast bright field microscopy is used. To address this challenge, we present a robust image processing method that uses a novel combination of well-known processing techniques that have been successfully used in other contexts. Results on time-lapse microscopy images show that our method works well on objects of both regular and irregular shapes, is able to distinguish between object types, and accurately detects object configurations even when they are located in close proximities to each other.