Research on Extraction and Translation of English Public Signs in Tourist Attractions Based on Machine Vision

Abstract

There are many English public signs in tourist attractions. Using digital imaging technology and computer technology to extract and translate these English public language is helpful for tourists to understand or warn the surrounding environment. This study constructs the recognition system based on machine vision technology and convolutional neural network (CNN) to extract, recognize and translate English public signs. The results show that the average recognition accuracy of the system is 98%; the average accuracy of translation is 96.5%. The above results show that the recognition system can effectively extract and translate English public signs, which is helpful for tourists to understand the information conveyed by scenic spots.