Help in background removal

Dear Team

We are currently working on an experiment to track cells. The attached pipeline works for a few experiments, however it doesn’t work for a few other experiments due to the background. We have attached the images for the set of experiments for which the pipeline doesn’t work. Could you please help us optmise the pipeline for reasonably accurate segmentation of cells.
TrackObjects20131211.cp (8.83 KB)
feature_001.rar (2.76 MB)


Attached is a pipeline which attempts to find the region of interest (the wedge-shaped region), mask out all pixels outside of this region, then perform some illumination correction. To adjust the amount of correction, adjust the smoothing filter size in CorrectIlluminationCaclulate. This pre-processing should put you in a better position to detect the cells more effectively.

2013_12_11.cp (7.35 KB)

Thanks a lot!