Video processed in the laptop.

Hi,
I have a project that I want to do for uni. I want to make like an avoider of collision for several cars. Users control cars and once those gets close to one another, then, the system take the priority over the user in order to avoid accident. I did some research, but I still have some questions:

I know arduino can receive video. What I do not know is if it can accurately enough (in the case of my system) process it.
Can I use my computer's processor instead of the arduino in order to have a faster speed of fram per second and a faster processing? Because I am affraid arduino is not enough. The following video inspired me, but it does not show me exactely how the laptop process has been done. iphone controlled Internet PC robot using arduino - YouTube

Can any webcam work with it? (Like included in any laptop)

Is there any other library to install like OpenCV? (I hope we only need arduino)

Can anyone tell me?

Thank you by advance.

These videos show that I can use Skype for receiving video, but not processing it.

I know arduino can receive video.

What makes you think that then?

The following video.

Except if they used another thing than arduino.

Except if they used another thing than arduino.

Precisely.

I have a project that I want to do for uni. I want to make like an avoider of collision for several cars. Users control cars and once those gets close to one another, then, the system take the priority over the user in order to avoid accident.

There are a limited number of things you can do along these lines, but your professer probably didn't mention that people have been working on computer vision for 50 years, and the problem is just beginning to be solved. Not quite as good as what animals do. IOW, it's a non-trivial matter, but still worth a go. Can lead to a long career working on same.

Some of the initial DARPA Challenge vehicles a couple of years ago tried using binocular vision for main guidance, but it proved too difficult to be reliable. Since then, EVERYONE uses laser scanners, and they've had 1000% better success. Computer vision is probably best combined with other types of sensors - eg, sonars.

You might research the Darpa Urban Challenge,
http://www.google.com/custom?q=darpa+urban+challenge&btnG=Search