Skip to main content
European Commission logo print header

Neural Network-Based Vision and Signal System for Industrial Quality Control

Exploitable results

NEUROQUACS is an automated, advanced quality inspection system which uses machine vision and other sensors. Manual inspection is an inefficient, and often ineffective, method of industrial quality control. Not only is it resource intensive, but it also allows certain defects to go undetected, and limits the introduction of inspection standards. Integrating neural networks in an automated inspection system can significantly increase the uniformity and reliability of quality control, and enhance the working environment. An automatic and advanced inspection system using machine vision and other sensors has been developed by the consortium involved in the project 'NEUROQUACS'. By utilizing neural networks, the system has the ability to learn by example, like a human. The modular nature of both the hardware and software allow cost-optimized sizing of the application system. Parallel processors provide a flexible hardware configuration that can be scaled up or down to suit the required processing power. The software development environment, available on SUN workstations, enables the integration of image and signal processing modules, and uses a comprehensive software library. Three applications have already been demonstrated successfully. In wood processing, the system is being used to improve quality control by classifying prepared timber into quality classes according to the occurrence of knots and splits. In another application, a sensor is being used to classify the resistance of potentiometers; and an acoustic sensor application is being used to detect knocking in car engines.

Searching for OpenAIRE data...

There was an error trying to search data from OpenAIRE

No results available