Low cost, user friendly embedded machine vision system implementation for high-speed industrial manufacture

Worcester, F. ORCID: 0000-0003-4616-6842, Breedon, P. ORCID: 0000-0002-1006-0942, Iaquinta, K., Anderson, M., Brooks, S. and Sprinks, J. ORCID: 0000-0003-3781-5484, 2019. Low cost, user friendly embedded machine vision system implementation for high-speed industrial manufacture. In: Everything Connected Conference: Industrial Systems in the Digital Age, Nottingham, June 2019.

Preview

Text 14174_Sprinks.pdf - Published version Download (860kB) | Preview

Abstract

The future of manufacturing environments is becoming increasingly intelligent. With Industry 4.0, and lean production strategies, many manufacturers are considering investing in intelligent, real time, and non-destructive machine vision (MV) systems for product manufacture fault analysis [2]. This technology is improving rapidly in both the industrial and consumer sectors [3]. This work presents the development of a low cost, user friendly embedded machine vision system for use in industrial snack manufacturing. Previous attempts at using a high cost ~£10000 system failed, due to inadequate knowledge for successful integration. The OpenMV cam M7 was procured as a low cost ~£100, simpler alternative. The system achieved 88% ±1% sensitivity and precision in detecting products on a live production line. Therefore this is a viable prototyping option for machine vision novices, enabling informed decisions thereafter.

Item Type:	Conference contribution
Creators:	Worcester, F., Breedon, P., Iaquinta, K., Anderson, M., Brooks, S. and Sprinks, J.
Date:	June 2019
Divisions:	Schools > School of Architecture, Design and the Built Environment
Record created by:	Linda Sullivan
Date Added:	28 Jun 2019 13:28
Last Modified:	28 Jun 2019 13:28
URI:	https://irep.ntu.ac.uk/id/eprint/36959

Actions (login required)

Edit View

Views

Downloads