Menu

Machine learning for image segmentation, classification and object tracking

We have been developing software for image classification, segmentation and object tracking, based on computer vision and machine learning approaches to data analysis; These algorithms allow one to track individual molecules or cells in large populations over very long periods of time, reconstruct lineages and study cell movement or sub-cellular protein localisation. More details can be found at: http://github.com/quantumjot


Here is an example of automatically segmenting, tracking and classifying data from timelapse microscopy. The image data is shown on the left, and the class scores on the right:

Below is an example of tracking a field of cells from timelapse data: