Ok I was wrong. The uC and the camera are not going out of sync due to the bug fixes. The uC is fast enough to reach the "sleep for href" well in time.
So now.. why is there a huge difference in thresholds seen in the captured snapshot and the tracking frames, I mean to track an object I need to set RGB values on by trial and error (not easy) since the RGB values seen in the captured frame do not seem to work.
I was under the impression that demosiacing was the culprit, since while tracking we are skipping a few pixels. I thought that the RGB values displayed in the snapshot are after doing some demosiacing. So I wrote a Java application to print the raw RGB values and oh no I am wrong. The raw values match with those shown by AVRcamView!!
Ok there is only one explanation. Since while capture a snapshot, the frame rate is being reduced, perhaps the exposure rate is increased. Yes this must be it. Ok off to find out if I am right this time!!
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment