Optimize Real-World Computer Vision Processing Performance Through The QUIC Transport Protocol And Edge Cloud Computing Model

The advancement of technologies and their practical use in computer vision has led to the emergence of edge computing as a viable solution for alleviating the burden on centralized servers. However, there is a trade-off in performance as edge computing relies on the processing capabilities of edge hardware, which may have limited processing power. Addressing and improving this issue is essential to enhance the effectiveness of computer vision when deployed on hardware devices. This article will present a new approach to address this issue by incorporating the optimization of the data transmission process into edge cloud computing to improve and address the hardware limitation problem. This will be achieved by utilizing the advanced high-speed transmission protocol known as Quick UDP Internet Connections (QUIC). Edge cloud computing is a solution that brings the robust processing abilities of cloud computing services closer to users, aiming to enhance transmission speeds while retaining the powerful processing capabilities of traditional cloud computing. This reduction in transfer times significantly improves the performance of real-world computer vision applications, leading to significant changes in the field of computer vision, which requires substantial real-time processing capabilities. The combination and utilization of the high-speed transmission of QUIC, a protocol developed from User Datagram Protocol (UDP), is set to deliver outstanding transmission performance for applications Practical implementation of computer vision relies on real-time processing capabilities that are not achievable through edge computing. The system under consideration has been assessed and has yielded positive results based on three critical factors, including real-time data transmission speed, the real-time processing capacity of machine vision models, and the ability to offload processing traffic to a central server. With its demanding resource requirements, top-notch data input, and the potential to alleviate data center workload, experiments indicate the robust potential for real-world deployment of this approach.