cbrake's profile - activity

What I'm looking for is:

• USB (2 or 3)
• 3.2MP color
• global shutter is probably not required as this is not tracking moving objects
• provide source code to libraries/drivers such that we can make it work with Yocto Linux on x86 (Minnowboard Max)
• 20fps or so (not real critical)

In the past, we've used a Point Grey Firefly MV with libdc1394 on a ARM Linux system, and that worked OK.

We've tried new USB3 cameras from Point Grey, but libdc1394 does not work with it, and I can't get their binary libs to work on our embedded platform with the new camera. The PG libs work fine a Linux PC, but that does not get us there. We also tried a camera from a Chinese vendor, but their software is also only available in binary form, and I can't even get it working at all, even on a PC.

So, we're looking for a camera vendor who can support us with software for a custom Embedded Yocto Linux system (no Ubuntu here). Open source would be preferred, as its my experience that if vendors open source their software, it is much higher quality than those who do not.

Thanks, Cliff

thanks for the thoughts -- template matching is proving to work somewhat, but requires a lot of threshold adjustments from one problem to the next, so is no real practical for real work use.

thanks for the ideas!

need to identify components

Hi,

I would like to use opencv to detect printed circuit board components in a camera image. What technologies/approaches would you advise? Thus far I've tried simple template matching with CV_TM_SQDIFF_NORMED. It works somewhat for small components (like a surface mount capacitor), but for larger components with more detail it does not very well, and overall is very sensitive to threshold settings, etc.

Thanks, Cliff