Using an Ethernet connection to tie vision/imaging system nodes together produces several performance advantages. For example, major subsystems, such as camera, image processor, and host computer, have individual Internet Protocol addresses so that intrasystem communications can use standard network protocols and software. Moreover, Ethernet communications allow seamless system integration into a factory-floor-wide network, enabling users to distribute and manage multiple vision sensors. Cognex Corp. (Natick, MA) calls it the Vision Area Network.
Rather than integrate proprietary cameras, frame grabbers, and software into stand-alone platforms, some companies are focusing on less-expensive systems that can be programmed and deployed more easily and rapidly. In this month's Product Focus, Andy Wilson explains how some firms are offering machine-vision sensors that act as Ethernet nodes that can share information with other networked locations on a factory floor and allow reconfiguration from a remote site (see p. 47).
When motorcycle-manufacturer Harley Davidson ordered a test system to check the timing-chain installation on its new engine, it expected just one pass-fail inspection. The resulting test cell, however, included an automated six-axis robotic arm, lighting, a camera subsystem, a PC, custom software, and Ethernet networking. This integrated vision and imaging system, says contributing editor Charles Masi, is capable of making 30 sequences of orientation, image acquisition, image analysis, pass/fail decisions, and network reporting with a minute (see p. 39).
In forestry, university researchers are exploring optical signatures and computed-tomography (CT) scans of tree logs to increase wood-production yields. Multispectral imaging is being used to identify surface defects and CT scanning to examine interior defects. As reported by contributing editor Joe Hallett, researchers are implementing imaging, analyzing, and modeling color scans of sliced logs to determine their suitability for small, clear wooden parts (see p. 33).
To prevent contamination, pharmaceutical companies are inspecting the film coating added to the inside surface of elastomeric stoppers or covers in medicinal containers with an automated vision system. This improves product quality, reduces labor costs, and increases production yields. As described by contributing editor Larry Curran, the vision system looks for several defects and determines product acceptance or rejection depending on established database criteria (see p. 25).
George Kotelly,Editor in Chief[email protected]
