The Active segmentation ImageJ plugin was developed in the scope of GSOC 2016 - 2018. The plugin provides a general-purpose environment that allows biologists and other domain experts to use transparently state-of-the-art techniques in machine learning to improve their image segmentation results. ImageJ[1] is a public domain Java image processing program extensively used in life and material sciences. The program was designed with an open architecture that provides extensibility via plugins.
Motivation: Particle tracking enables researchers to analyse dynamic structures e.g. neurons, cells or subcellular regions in an elaborate way. There is a number of software platforms (for example Tracemate[2], TrackPy [4]) that provide a set of particle linking algorithms based on combination of several techniques like gaussian template matching, kalman filter to tackle linear motion and nearest neighbor based search.
The project idea: We will use correlation in feature space to improve tracking of objects. The object will be defined on a raw 2D image, after which the feature space will be computed across different frames and potential object displacements will be calculated. The trajectory estimate will be denoised, for example by using a Kalman filter.
The applications of the proposed method are twofold: 1) object tracking in consecutive frames. 2) object tracking in disjoint frame sets by similarity. Application 1 will be focused on using time-lapse light microscopic imaging datasets where position of cells changes with time.
The student will extend the existing Active Segmentation platform, based on ImageJ
Tasks
- Fix existing issues and bugs
- Add-on to the the user interface for trajectory display
- Provide a reference implementation
- Minimal set of deliverables
- Requirement specification - Prepared by the candidate after understanding the functionality.
- System Design - Detailed plan for development of the plugin and test cases.
- Implementation and testing - Details of implementation and testing of the plugin.
Desired skills: experience with ImageJ, signal processing
Mentors: Dimiter Prodanov (dimiterpp@gmail.com), INCF Belgian Node; (backup) Sumit Vohra, ZIB, Berlin, Germany
References:
-
ImageJ: https://imagej.nih.gov/
-
TraceMate: https://www.sciencedirect.com/science/article/pii/S1046202316303346
-
Objective Comparison of Particle Tracking Methods: https://www.nature.com/articles/nmeth.2808
-
Target tracking using gaussian processes:[ https://www.diva-portal.org/smash/get/diva2:1109971/FULLTEXT01.pdf
-
](https://www.diva-portal.org/smash/get/diva2:1109971/FULLTEXT01.pdf)TrackPy: https://soft-matter.github.io/trackpy/v0.3.2/