Vision system checks the identity of pistons

Engineers at Olmec-UK (Barton-Upon-Humber, UK) have developed an automatic vision inspection system for an automotive engine manufacturer that can read the alphanumeric product identifiers cast into pistons.

The pistons must be checked immediately prior to engine testing to ensure that the correct ones are used. Failure to do so leads to extreme and costly damage when the engine is tested.

Automatic reading of the alphanumeric identifiers posed a huge challenge for the vision system because the characters were never intended for machine reading.

Although the characters stand raised from the background material, there is very little contrast between the two. This is made even worse because the material is very reflective. In addition, pistons are produced from a variety of different moulds, so there is significant variation in the quality of the characters from batch to batch.

The engines to be inspected arrive at the vision inspection station developed by Olmec-UK on a conveyor. The conveyor is stopped and four cameras, each equipped with their own ring lights, image each of the pistons in the engine simultaneously. System software then processes the images taking into account the font, quality, scale, skew and slant of the characters.

If any piston fails the inspection, the process is stopped to allow an operator to make an additional visual assessment.

Recent articles on automotive technology that you might also find of interest.

1. System sorts nuts for the automotive industry

Engineers at Avinashilingam University (Coimbatore, India) have developed a prototype machine vision system to sort nuts used in the automotive industry.

2. Terahertz sensor to measure paint quality on automobiles

Researchers at the Department of Chemical Engineering at Cambridge University (Cambridge, UK) have been awarded a grant of over £240,000 to develop a terahertz sensor that can be used to measure the quality of paint coatings on automobiles.

3. Smart headlight could improve visibility for drivers

Researchers at Carnegie Mellon's School of Computer Science (Pittsburgh, PA, USA) have developed a novel vision system that is capable improving the vision of drivers at night.

Vision Systems Design magazine and e-newsletter subscriptions are free to qualified professionals. To subscribe, please complete the form here.

-- Dave Wilson, Senior Editor, Vision Systems Design