Project 2_3: Motor Analysis
Image processing and tracking of all the cells in the given E.coli movie (1.6GB video files) using u-track 2.0; and perform data analysis. For the video: 1 pixel=0.65 $ \mu m$, 1 frame=0.1 second.
First, The video is divided to 999 frame tiff files using ImageJ 1.51j8 (N0.456 frame is broken in origin video). We use the u-track2.2.1 to track the cell by method of Single-Particles and Gaussian Mixture-model Fitting .We can found that some break exist in a complete trajectory in all 3183 tracks.
For a sigle bacteria, we should get the following result.But we find that here are some much noise in our data. Anyway, result suggested that higher speed in the end which is more smooth.(Code: Prj2_3.m)
General, we should get the following result, but we dont't success to calculate and get a wrong result. Because the number of E.Coli is not constant at differet moment, the big problem is how to deal with the merge track and caluculate the mean of distance during difference time. ( Code:Prj2_3_2 )
- http://www.ruanyifeng.com/blog/2012/11/gaussian_blur.html
- http://www.rowland.harvard.edu/labs/bacteria/index.php
- http://www.tautelab.org/research.html
- E.Coli in Motion, p35, Howar C.Berg(2003).