A Faster R-CNN Implementation of Presence Inspection for Parts on Industrial Produce

Abstract

Presence inspection has been conducted as a matter of course during acceptance/shipment and on production lines. Due to increased factory automation (FA) in recent years, vision systems and image processing technologies have been actively introduced. This paper describes the advantages of using image processing for presence inspection, and the basic principles and practical applications of presence inspection using image processing. This paper demonstrates the use of Faster R-CNN, TensorFlow object detection API, and Computer Vision for the detection of the presence or absence of an industrial part. This solution would present itself as a cost-effective method for the quality assessment phase of a production line in a factory.