Industry is placing ever-greater emphasis on automation, particularly with the increasing deployment of Industry 4.0 technology. The latest developments in machine vision systems open up opportunities for the manufacturing sector to raise quality and throughput simultaneously. Furthermore, new applications are starting to appear in other industries, e.g. object detection and recognition in automotive systems or facial-based security in consumer electronics. In order to keep latency levels down, some industrial machine vision implementations must be situated at the network edge.
The following article investigates the role AI has to play in machine vision systems, and outlines the various hardware options available for edge-based machine vision applications.