Publication details
Tool for Generation of Synthetic Image Datasets for Time-Lapse Fluorescence Microscopy
| Authors | |
|---|---|
| Year of publication | 2011 |
| Type | Conference abstract |
| MU Faculty or unit | |
| Citation | |
| Description | In the field of fluorescence microscopy image analysis, motion tracking and segmentation algorithms are indispensable tools for tracking cells and for time-resolved analysis of cell characteristics or events. Although there are many such algorithms in everyday use, most of them are not properly validated. We see the ground truth (GT) information as a very important tool for the verification of image processing algorithms. Unfortunately, it is not readily available. Many algorithms in this field of study are, therefore, validated only against GT consisting of real data manually annotated by a human expert. This annotation is, however, a very difficult, cumbersome, and time consuming task, especially, when single 3D image or even 3D image sequence is considered. We present a novel simulation tool that is capable of generating fully synthetic fluorescence microscopy 3D image data accompanied with GT. The generated data can be either static images or time-lapse sequences. Regarding GT, it is represented by binary image masks, to facilitate evaluation of segmentation algorithms, and by flow fields, to facilitate evaluation of tracking algorithms. Such image data can be then processed by selected algorithms, their results can be compared with GT and the quality of the applied algorithm can be measured. The simulation technique consists of generating shape, structure, and also motion of selected biological objects. It preserves time-lapse continuity of shape and structure in the generated sequences. It is focused on the generation of synthetic objects at various scales ranging from microspheres to tissues. The tool will be publicly available and will replace an older version at http://cbia.fi.muni.cz/simulator web page. This work was supported by the Czech Ministry of Education (Projects MSM0021622419, LC535 and 2B06052). |
| Related projects: |
|