End-to-end text recognition method, model training method and device

Abstract

The invention discloses an end-to-end text recognition method, a model training method and device. The model training method comprises the steps of constructing an initial end-to-end text recognition model, wherein the initial end-to-end text recognition model comprises an initial text detection module and an initial text recognition module; obtaining a training sample data set; processing training samples in the training sample data set by using a sample generation algorithm to generate an amplified training sample data set so as to increase the number of the training samples used for training the initial text recognition module; and training the initial end-to-end text recognition model by using the training sample data set and the amplified training sample data set to obtain an end-to-end text recognition model. According to the technical scheme, a large number of training samples used for training the text recognition module are generated through the sample generation algorithm, so that the problems of over-fitting of the text detection module and under-fitting of the text recognition module are effectively solved, and the recognition precision of the end-to-end text recognition model is improved.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, and more specifically, to an end-to-end text recognition method, a training method and a device for an end-to-end text recognition model. Background technique [0002] The end-to-end text recognition method generally integrates the text detection module and the text recognition module into a network model, where the text detection module is used to detect the position of the text, and the text recognition module is used to recognize the content of the text. End-to-end text recognition has a wide range of applications in areas such as autonomous driving, machine translation, and commodity retrieval. However, for different application fields, the text recognition module and text detection module in the model need to be trained to achieve better recognition accuracy. The ratio of the number of samples required for training the text recognition module to the number of samples...

AI Technical Summary

Problems solved by technology

However, the ratio of the required number of samples in the current end-to-end text recognition model training method is less than 10, which does not meet the required amount of training samples

Therefore, in these methods, the insufficient number of samples for training the text recognition module leads to the overfitting of the text detection module and the underfitting of the text recognition module, which greatly limits the accuracy of end-to-end text recognition.

Images