A tracking framework for laboratory experiments with multiple targets and occlusion

The automatic tracking of components allows a quick and consistent analysis of parameters related to biological experiments. The register of parameters is done by a system that analyzes a sequence of images and computes a number of descriptors characterizing the behavior of each target. Our objective is to create a framework for tracking in biological experiments, with multiple targets that may suffer occlusion. Such framework must be capable of being adapted to other classes of tracking. Here we present the framework, a new method of cells segmentation, and an optimization method for occlusion solving.