User:TheLarry/Notebook/Larrys Notebook/2009/08/25

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


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

Fixing my removing bad point problem
haiqing brough to my attention that the removal of my bad points doesn't work if the tracking is particularly shitty. So I need to come up with a better job of doing it. I am tired of pussey footing around it. The derivative has the answer in it. And it is painfully obvious. But that only tells the slope. I need to use that to remove all the bad points from the original and then best fit it. However i want to cheat. Sometimes when the box loses the track it will still rotate the rectangle through the correct amount of radians. So it loses it for one point and rotates correctly however from the wrong starting point. i don't want to lose these points. So what i want to do is keep them and add to them the amount the wrong point dropped. This hopefully will keep the most amount of points I can.

Koch, I know this doesn't make any sense but i'll try to do it, and show you. See what you think about it. I am just hungry now and a bit frustrated from this not working. I should have the morning to work on this and get it right.

OK i think i think i might get this working better. I take out all the bad points from the derivative, and then i integrate that to get the data back. and then best fit a line to that. This might be a better way to handle this. I haven't put it in the .vi yet because working from home is so slow. but i can do it tomorrow morning. i am excited to see if it works better. And the Rockies are in extra innings again. blown save. agh i hate manny ramirez


 * }