Machine vision has migrated from being customized for a single application and being used by large OEMs to being deployable by the masses. And becoming deployable by a wider audience means that machine vision systems have to be user-friendly.
Steve Geraghty, director of ipd (Billerica, MA), says his company's products run the gamut from a simple interface for the end customer to powerful and flexible products for the integrator. "We've seen our company move from frame-grabber type products to more fully integrated solutions."
Michael Schreiber, director of applied engineering at DVT Corp. (Duluth, GA) says, "The clear trend is toward more intuitive and user-friendly interfaces, combined with more powerful and robust algorithms. DVT's introduction of Intellect software is an example of the trend toward increased functionality and ease-of-use. Users now will find that soft sensors are grouped by tasks rather than by algorithm types. For example, if you need to conduct measuring, you would look in the Measurement menu: If you need to count items, look in the Counting menu.
"The trend also is toward more "point-and-click" or "plug-and-play" vision applications. Examples of these are DVT's development of vertical market applications that address specific inspection requirements while minimizing the amount of setup required. These single-purpose solutions provide out-of-the-box simplicity with step-by-step wizards that guide the user through setup, and require little or no experience with vision systems."
Mark Sippel, principal product marketing manager for In-Sight Vision Sensors at Cognex Corp. (Natick, MA) agrees that machine vision has become more user-friendly, but he warns against over-promoting the simplicity factor. "You have to be able to correctly image and have the right light source in place. If you can't get it imaged right, it doesn't matter how easy the integration capability is."
Now that machine vision has become more user-friendly, there are other avenues to explore.
"Resolution is huge and increasing exponentially," Sippel says. The problem is that once resolution increases, speed decreases. He says that gigabyte Ethernet holds promise for the future of machine vision through increased communication speed.
Generally images cannot be removed from the camera and trying to transmit out of the box proves to be too slow, Sippel says, if not impossible, because of the amount of information contained in the image file. With gigabyte Ethernet capabilities, not only will the data such as pass/fail be available, the image can be attached to the file as well, expanding the technology's capabilities.
The future of machine vision allows for the improvement in the quality of products without the need to spend hundreds of thousands of dollars while also providing intangible benefits such as saving a company's reputation. "Having a bad product in the marketplace is a death wish," Geraghty says. Highlighted below are some machine-vision products that can save companies from an untimely demise.