Automotive manufacturing companies have a long history of investing in machine vision. Two decades ago, attempts to integrate vision systems into automotive parts production and final assembly resulted in decidedly mixed successes, with promises sometimes exceeding the ability of integrators to deliver fully functional systems.

In the automotive industry, hemming presses have been replaced by roller technology where a flexible hemming head with two rollers is mounted to the end of a robot arm to join independently stamped inner panels such as hoods, decks, and fender reinforcements to the outer panels that make up a car's body.

Using geometric-based pattern recognition, an automated system sorts numerous wheel types and sizes according to their model numbers.

Because most camera vendors currently offer products based on imaging sensors from well-known companies, it has become increasingly difficult to differentiate between cameras based on sensor technology alone. Recognizing this, camera vendors are now looking to customize their product offerings by forging partnerships with less well known CMOS design houses.

By classifying features such as edges, color, and shape of images as radial basis functions, neural network systems can be trained to classify parts based on numerous feature vectors.

One of the perks of being a journalist in the trade press is that, once in a while, I am able to enjoy an excellent dinner at a first-class restaurant. During the March Automate 2011 show in Chicago, I was lucky enough to be seated at such a dinner next to one Randall Hinton, a solutions engineer from Edmund Optics, who, like myself, had earned a degree in physics.