Exploring an FPGA-Based Edge Computing Solution for Smart Manufacturing Monitoring: A Case Study on Droplet Recognition

This study presents a novel droplet recognition method based on Field Programmable Gate Array (FPGA). Unlike traditional droplet image recognition techniques, this approach offers enhanced flexibility, cost-effectiveness, and a reduced design cycle while maintaining high reliability. The experimental device is capable of effectively detecting target droplets and efficiently calculating their size and position. The droplet image is preprocessed using the median filtering algorithm, which eliminates some bright spots in the image. The image processing procedures, including image acquisition, preprocessing, segmentation, and droplet feature extraction, are all performed on an FPGA device. Its adaptability to the Internet of Things (IoT) paves the way for constructing a distributed edge computing monitoring network in smart manufacturing. This paper delves into the intricacies of developing such an edge computing unit and explores the potential of a smart monitoring network integrating multiple units for advanced manufacturing. We aim to catalyze further research and applications in this domain.