Question on the color info

This forum is used for discussing hardware, software, and technical details of the AVRcam embedded system.

Question on the color info

Postby kiteswing » Wed Jun 22, 2005 6:07 pm

Hi john,

The OV6620 image sensor seems have more image data information. Can you tell me why you just use 4 bits color data per pixel.

Moreover, If I want the camera can can have higher color quality, and realize the OV6620's other features like sharpness enhancement etc, how can I realize it? Do you have any suggestion on how to do that?

Thanks a lot!
kiteswing
 
Posts: 5
Joined: Tue May 31, 2005 10:40 pm

Postby Guest » Wed Jun 29, 2005 6:05 am

Hi Kiteswing,
The AVRcam only uses the upper four bits of each pixel to simplify the processing that needs to be done. As each pixel is sampled by the mega8, it is immediately used to index into a lookup table to extract some information regarding whether or not the pixel value could potentially be in the user-configured colorMap. Using only four bits of each pixel means that the lookup table can be only 16 bytes long (one lookup table is needed for red, one for green, and one for blue). If the full 8 bits were used, each lookup table would need to be 256 bytes long. On a processor with only 1K of RAM, this would be almost all of the systems RAM. And...early testing showed that colors could be tracked reasonably well if only 4-bits of color information was present, instead of the full 8.

If you want to adjust other parameters of the camera itself (saturation, gain, etc), you can do this by simply building and sending the appropriate command over the serial port. All of the configurable registers of the OV6620 can be configured through this interface. The command to use is the "change register" command, or CR. Check the user's manual for more info, as well as the OV6620 data sheet.
Guest
 


Return to AVRcam Embedded System

Who is online

Users browsing this forum: No registered users and 14 guests

cron