How to analyze results with high variance

[andradesilva]

Hi,

I am working on a project and we need to obtain data that results from the readings of a Computer Vision system that a python script exports to a text file. I would like to analyze those results for several numbers of frames. The problem is: the results are very different each time. How can I analyse the results, considering that fact?

Best regards,

 

[IRstuff]

This doesn't sound like an optics problem

Nevertheless, doesn't the variance flag a fundamental problem with your algorithm?

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[sunnysky]

The large variance must be noise so by defining the noise, try to optically crop it or filter out the noise. (chromatic and other tools to reduce the noise on your target.) There may be some sync tools for alignment. Please define or show the discrepancies.