Scalable Object Detection, Tracking and Pattern Recognition Model Using Edge Computing

Abstract

The Internet of Things (IoT) devices such as sensors and video cameras have small memories and less computational power. The video analytics of traditional approaches for detection, tracking and pattern recognition of moving objects used only in the cloud. This approach suffered from high latency and more network bandwidth to transfer data into the cloud. We address this problem by using edge computing devices between IoT devices and the cloud. We propose a new framework for scalable object detection, tracking and pattern recognition of moving objects that relies on dimensionality reduction with edge computing architecture. We also propose a scalable object detection and tracking method based on You Only Look Once (YOLO) method. The experiment demonstrates that our proposed method will save network bandwidth and processing time. The performance of object detection and tracking model is greater than 96%. This shows that our method has greater performance than existing models.