Methods for object detection during sudden illumination change..
3 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
anshuman
am 2 Apr. 2014
Kommentiert: Image Analyst
am 21 Feb. 2017
I am trying to detect objects from video sequences having sudden illumniation changes.I have studied few papers which try to solve this issue, but the math behind it was quite complicated. Could anyone suggest some good resources for solving this problem??
0 Kommentare
Akzeptierte Antwort
Image Analyst
am 2 Apr. 2014
Some simple ways might be to just subtract the mean and divide by the standard deviation. Of course those won't be as good as the papers you read because it's a very simple method.
3 Kommentare
Image Analyst
am 2 Apr. 2014
I don't know why it matters if the mean changes suddenly or not suddenly if you just subtract the mean of each frame from itself. But yes, I told you it's primitive and it you want something better you may have to just knuckle down and understand the math in the papers.
Yes the Computer Vision System Toolbox is more "object oriented" style of programming than the Image Processing Toolbox, and that will take some getting used to. I don't have the CVST toolbox. There are demos and examples provided that you can go through. Anand can also be a helpful resource for you.
Weitere Antworten (1)
Anand
am 2 Apr. 2014
Go through the documentation links on this page:
There are examples at the end of each function reference, which you can learn from. Also, the links at the end of the page pointing to Examples and How To are very instructional.
3 Kommentare
Image Analyst
am 21 Feb. 2017
Yes. There's no guarantee that the illumination will be the same in frames or snapshots taken at different instances in time unless you take special care to control your illumination.
Siehe auch
Kategorien
Mehr zu Detection finden Sie in Help Center und File Exchange
Produkte
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!