Particle recognition in microfluidic applications using a template matching algorithm

We herein examined the ability of a template matching algorithm to recognize particles with diameters ranging from 1 to 20 mu m in a microfluidic channel. The algorithm consisted of measurements of the distance between the templates and the images captured with a high-speed camera in order to search for the presence of the desired particle. The results obtained indicated that the effects of blur and diffraction rings observed around the particle are important phenomena that limit the recognition of a target. Owing to the effects of diffraction rings, the distance between a template and an image is not exclusively linked to the position of the focus plane; it is also linked to the size of the particle being searched for. By using a set of three templates captured at different Z focuses and an 800x magnification, the template matching algorithm has the ability to recognize beads ranging in diameter from 1.7 to 20 mu m with a resolution between 0.3 and 1 mu m. (C) 2016 The Japan Society of Applied Physics