Hi there,
The method you have described seems to make sense, and using a binary image would certainly give you the area information you require. There may be a few things you'd wish to consider when capturing the initial image - for example using diffuse backlighting would allow you to obtain a silhouette of the object in question making conversion to a binary image much less complicated.
Just to clarify, are you using the IMAQ Vision Assistant to develope your process?
NI-IMAQ shouldn't have any issues reading an image stored on your hard disk. So, in principle, provided your camera's thirdparty software can capture stills, this will give you a somewhat cumbersome method of testing your theory, but a method all the same.
I've had a bit of a trawl to find out more about the iVision camera, but as yet I have not found anything that explicitly states its ability for capturing stills. Have you managed to capture a still image in any way, shape, or form? This should be your first goal before trying to deal with IMAQ.
As I mentioned earlier, the only method by which you will be able to capture images into NI-IMAQ directly from a firewire camera source would be using the NI IEEE-1394 drivers.
If you had already got these drivers, but you are still having troubles acquiring an image then you may want to test the camera feed via Measurement & Automation Explorer (MAX). This is the best first step of debugging any problems.
Let me know how you get on.
Thx,
Rob
National Instruments | Northern California