Application of improved image restoration algorithm and depth generation in English intelligent translation teaching system

Abstract

Abstract The difficulty of English book translation lies in the need to recognize English characters in a complex background and perform accurate classification and translation. Based on the deep generative model, this paper constructs an intelligent translation system for English books based on machine learning and proposes an improved image restoration algorithm based on LSGAN on the basis of the image restoration algorithm based on DCGAN. Moreover, this paper optimizes the structure of the LSGAN network to a certain extent, improves the stability of the network, combines context loss and perceptual loss to find the most suitable generated image for filling. In addition, based on the VAE and GAN network structure, this paper proposes an improved image restoration algorithm based on VAE-GAN to improve the integrity of English characters and perform intelligent recognition. Finally, this paper analyzes through the experimental form, and after setting the identification object, the model constructed in this paper is used for translation, the translation accuracy and classification accuracy rate are calculated, and the performance analysis is performed after the statistical data. From the research results, it can be seen that the artificial intelligence model constructed in this paper has certain practical effects.