Design and Implementation of Natural Scene Text Recognition System Based on Deep Learning

Abstract

With the popularization and rapid development of information technology such as modern multimedia and mobile Internet, people have more and more ways to obtain natural landscape images. How to collect and extract necessary information from rich images becomes more and more important. With the current popularization and development of technologies such as artificial intelligence and deep machine learning, the application requirements for effectively obtaining text resources are also expanding rapidly. This paper aims to study the design and implementation of a natural scene text recognition system based on deep learning. Based on the analysis of system functional requirements, system non-functional requirements and text detection methods, each functional module of the system is designed in detail, and finally in SVT. The performance of the algorithm in this article is analyzed on the public data set, and the test results are compared with the existing algorithms. The test results show that the method and recognition results used in this paper are significantly higher than other methods, and the text recognition results have reached a very high accuracy.