Real-Time Object Classification on an Enamel Paint Coating Conveyor Line Using Mask R-CNN

Abstract

The rising demand to efficiently acquire live production data has added more significance to automated monitoring and reporting within the industrial manufacturing sector. Real-time parts screening requiring repetitive human intervention for data input may not be a feasible solution to meet the demands of modern industrial automation. The objective of this study is to automatically classify and report on manufactured metal sheet parts. The metal components are mechanically suspended on an enamel paint-coating conveyor line in a household appliance manufacturing plant. At any given instant, the parts may not be in the exact coordinates within the desired area of interest and the classes of objects vary based on changing production requirements. To mitigate these challenges, this study proposes the use of a trained Mask R-CNN model to detect the objects and their associated class. Images are acquired in real-time using a video camera located next to the enamel coating line which are subsequently processed using the object detection algorithm for automated entry into the plant management information system. The highest achieved average precision obtained from the model was 98.27% with an overall accuracy of 98.24% using the proposed framework. The results surpassed the acceptable standard for the average precision of 97.5% as set by the plant production quality engineers.