A Deep Learning Image Recognition Method Based on Edge Cloud Computing

Authors

  • Rui Wei School of Intelligent Science and Information Engineering, Xi’ an Peihua University, Xi’ an 710125, Shaanxi, China

Abstract

Due to the continuous development of computer technology, digital image technology has penetrated all fields of production and everyday life. However, although the transmission and storage facets of this technology are quite mature, image recognition has always undergone constant improvement both in domestic and foreign research. Edge cloud computing and deep learning can both be used to effectively improve the data processing capacity. To ascertain the effectiveness of these methods, a study was conducted using the deep learning method of edge cloud computing for license plate recognition. This article examines the current image recognition methods that perform well, focusing mainly on the recognition of license plates, and exploring the ways in which the existing traffic monitoring and recognition system can be upgraded and improved by using the deep learning method of edge cloud computing. The principles related to deep learning in edge cloud computing were studied in detail, and on this basis, a Hadoop cloud platform was built and applied to the traffic monitoring and identification system. In addition, the algorithm and simulation of license plate recognition are introduced and analyzed. The research results show that license plate recognition based on the deep learning method of edge cloud computing method is feasible, and the overall traffic monitoring recognition system is faster and more accurate. Compared with the traditional monitoring recognition system, the recognition accuracy increases by 15% and the recognition speed is improved by around 7.5%.

Keywords: Edge Cloud Computing, Deep Learning, License Plate Recognition, Monitoring and Recognition System

Cite As

R. Wei, "A Deep Learning Image Recognition Method Based on Edge Cloud Computing", Engineering Intelligent Systems,
vol. 31 no. 1, pp. 5-12, 2022.


Published

2023-01-01