Automation vision

The semiconductor industry has become the largest user of automated vision systems. A silicon wafer that will become hundreds of microchips starts as a finely machined disc about 7.9 in (200 mm) in diameter. Before the disc is split into individual chips, the wafer undergoes dozens of steps—some of which are indiscernible by the human eye. To ensure the wafer maintains that sequence, sorting systems using optical character recognition (OCR) identify each wafer, sort it in a clean room environment and report the results to a central network. 

For example, a company which experienced a customer complaint involving a large glass fragment in an aseptically filled powder vial introduced procedural preventive measures but concluded that the issue required automated vision inspection equipment. Once the corrective equipment was identified, a validation master plan detailed the key qualification elements for hardware, software, defect detection system, infeed/outfeed links, but also the specification requirements of the component and component quality, e.g ... 

In order to place fluid accurately and consistently within the desired tolerance, it is often necessary to use automated vision systems to align the parts to the dispensing robot. This can add flexibility to a system and reduce the complexity of hard tooling which requires pins or mechanical alignment devices. 

Tighter size and space tolerances frequently make an automated vision system necessary. 

Because so much of test and inspection is now concerned with vision systems, whether human or automated, this section will use automated vision as an extended example through which to discuss issues of automation in general. A typical example has already been presented, that of rose grading, and will be used to illustrate concepts as they arise. 

Any recent conference on automated vision from an automation or applied optics viewpoint will provide copious current examples of AVIS applications in a vruiety of industries. Table 4 lists a nmnber of recent applications of automated vision to show the variety of products and industries covered by the pace of this form of airtomation. Unfortunately, few techniced papers contain the depth of performance evaluation required to assess system reliability or its components. The Steinmetz and... 

Tsai, R.Y. A versatile camera calibration technique for high-accuracy 3D machine vision meterology using off-the-shell tv cameras and lenses. IEEE. 1.Robotics Automation, Vol. RA-3(4),August 1988, pp. 323-344. 

Benson and Ponton (1993) and Ponton (1996) have speculated on the ultimate results of continuing efforts for process minimization. They envision a twenty-first century chemical industry totally revolutionized by technological innovation, automation, and miniaturization. Small, distributed manufacturing facilities would produce materials on demand, at the location where they are needed. Raw materials would be nonhazardous, and the manufacturing processes would be waste free and inherently safe. While their vision of future technology is speculative, we are beginning to see progress in this direction. 

In another report, aspects for automating preparative chemistry are described [130]. A comprehensive description of the Ugi reaction is given in [132] and the vision of a micro multi-component reaction as automated parallel micro-channel synthesis is sketched. An interesting point is to convert aldehydes, chiral primary amines, carboxylic adds and isocyanates into corresponding a-amino acids and peptides (U-4CR). 

Sandak, B., Nussinov, R., and Woleson, H.J. An automated computer vision and robotics-based technique for 3D flexible biomolecular docking and matching. Comput. Appl. Biol. Sci. 1995, 11, 87-99. 

Machine vision will be an important tool in the operation of unattended, fully automated renewable energy processes. Vision is the response of the eye and brain to light. Machine vision is the artificial response of a device to spectral radiation. Machine vision also extends the human range of the visible spectrum into the IR, UV, and x-ray regions. Machine vision can be used to make decisions faster or more accurately and precisely than a human can. Machine vision may combine online defect monitors with shade monitoring, which will be an important tool in optimizing the positioning of solar collectors (Table 3.128). 

An Automated Computer Vision and Robotics-Based Technique For 3-D Flexible Biomolecular Docking and Matching. 

Gkoutos, G. V., H. Rzepa, R. M. Clark, O. Adjei, and H. Johal. 2003. Chemical machine vision automated extraction of chemical metadata from raster images. J. Chem. Inf. Comput. Sci. 43(5) 1342-1355. 

Where is shown after the system investment, add 181,000 for an automated inspection system comprising an X-ray metal detection machine, and 3 machine vision units to verify closure in place and label and bar code are correct and in place. 

---