hmm, I certainly wouldn't keep my webcam pointed at the sun for hours, but I haven't tested it out. You should use the Hough circles algorithm to detect a circle in the image, then you can quickly find its centre.
If you find that it's picking up circles other than the sun, just get the integral image within a few of the more popular circles and pick the brightest one.
If you have to take snapshots for hours, I'd recommend using Boost time libraries.
I'd also recommend that book mentioned - the one by O'Reilly.
You don't need to know anything about DirectShow.
good luck :)
--- In OpenCV@yahoogroups.com, "pradeeptumati" <ptumati@...> wrote:
> I am currently designing a solar tracker that takes a snapshot of the sun and analyzes its position in the sky (http://www.iaeng.org/publication/WCE2010/WCE2010_pp844-846.pdf). I have never developed webcam/machine vision based applications and so I need some advice on how to approach this issue; I learn things accordingly and get things going from there.
> I intend to put a webcam on a dual axis heliostat. This webcam takes snap shots of the sky at frequent intervals (say once every minute). The application that I intend to develop is supposed to analyze each snapshot, locate the sun and compute its centroid; it essentially identifies the brightest spot in the sky and tries to identify its shape (hopefully it is going to turn out to be a circle). Since this is a basic webcam application, I am assuming that I don't need to study machine vision algorithms etc What topics should I learn to achieve this ? As far as I understand, I should learn how to interface a webcam using directshow and grab bitmaps . How can I analyze the bitmaps ? There should be some algorithms to identify the brightest spot . Can I use opencv for this? Are there any books that I should start off on this topic. I looked at some books in amazon; I just don't know where to start. Any advice is highly appreciated
> BTW: Can webcams withstand the intensity of the sun if I directly point them towards the sun?
Get Emails (get all posts)
Daily Digest (one summary email per day)
Read on the web (read posts on the web only)Or Unsubscribe by mailing OpenCVemail@example.com