User:TheLarry/Notebook/Larrys Notebook/2009/11/11

{| width="800"
 * style="background-color: #EEE"|[[Image:owwnotebook_icon.png|128px]] Smoothing Data
 * style="background-color: #F2F2F2" align="center"|  |Main project page
 * style="background-color: #F2F2F2" align="center"|  |Main project page


 * colspan="2"|
 * colspan="2"|

Finding Velocities
So here is what I'll do to find velocities for each data set.
 * 1) Remove obviously bad points.
 * 2) *For example if the microtubule goes off the screen or if the tracking lost the microtubule for a frame. Things that I know are not correct data
 * 3) Find Velocity from $$\frac{\sqrt{(x_1-x_2)^2+(y_1-y_2)^2}}{frame_1-frame_2}$$
 * 4) *The denominator is 1 second but i put it in there for completeness
 * 5) Put that data through Koch's Sliding Window Histogram 2009 vi
 * 6) *Use a window size of 2
 * 7) *Record the velocity
 * 8) Do this for length of the microtubule
 * 9) *Except use $$\sqrt{(x_{f,1}-x_{b,1})^2+(y_{f,1}-y_{b,1})^2}$$
 * 10) *Steve Koch 12:59, 11 November 2009 (EST):This is probably going to be a problem for Andy's data, since it's sort of very curvy. Does the magic wang VI spit out the contour length?  That's what you actually want, but you don't want to have to write your own contour integral VI at this point.
 * TheLarry 13:00, 11 November 2009 (EST): You make a good point. But since it isn't that big of a deal yet since this poster won't be looking at length variable I'll think about it when i get this data in.


 * 1) Record this information in possibly an excel file
 * 2) *Also record frame number that the tracking started from and the concentration percentage

I am having a hard time removing the bad points. but i'll get it right
 * Steve Koch 12:57, 11 November 2009 (EST):One reason for the sliding window histogram is to avoid the need for eliminating bad points. As long as they're a small percentage, they shouldn't affect the peak of the histogram too much.  (You can test this by manually adding bad points, and seeing which version (gaussian / boxcar) is better)
 * TheLarry 12:58, 11 November 2009 (EST): I have seen this. You are right I shouldn't spend so much time on removing bad points

Done
It took me a while but i am done. For the record I am never gonna do this again without some kind of automation. This sucked. But anyway, Andy the excel file is under the z drive Z:\Data for Andy\09Nov04 Velocity Data.xlsx


 * }