Character recognition system and method based on combination of neural network and attention mechanism

Abstract

The invention claims to protect a character recognition system and method based on the combination of a neural network and an attention mechanism, the system comprising: a convolution neural network feature extraction module, which is used for spatial feature of character image; The spatial features extracted by the convolution neural network are input to the bi-directional long-short memory network module, and the bi-directional long-short memory network can extract the sequence features of characters. The extracted feature vectors are semantically encoded, and then the attention weights of feature vectors are assigned through the attention mechanism, so that the attention is focused on the feature vectors with higher weights. In the decoding part of the model, the features extracted fromattention and the prediction information of the previous time are used as the inputs of the nested long-short memory network. The purpose of using the long-short memory network is to keep the temporal characteristics of the eigenvectors and make the attention points of the model constantly change with time. In the decoding part, the features extracted from attention and the prediction informationof the previous time are used as the inputs of the nested long-short memory network. The invention can more accurately detect the text area in the natural scene, and has good detection effect on thesmall target text and the text with small tilt angle.

Description

technical field [0001] The invention belongs to character and image recognition in natural scenes, and relates to a correlation algorithm combining a convolutional neural network, a long-term short-term memory network and an attention mechanism. Background technique [0002] Natural scenes are the living environment we live in. Natural scene images contain a variety of visual information, such as text, cars, landscapes, organisms, and architectural landscapes. These elemental information constitute the main content of natural scenes. Element. [0003] Digital recognition in natural scenes belongs to the category of text recognition in natural scenes. The research on text recognition in natural scenes began in the 1990s, but it is still an unsolved problem until now. Generally speaking, the text recognition task in natural scenes consists of two parts: text region detection and text recognition. Text recognition is based on detection, and the detected text box is used as re...

AI Technical Summary

Problems solved by technology

However, due to many factors such as the position, deformation, and illumination of the characters in the natural scene, and the background of the characters in the natural scene is also quite complicated, there are many technical difficulties that need to be overcome in the recognition

[0004] At present, many research methods are mostly based on top-down algorithm models. Jaderberg et al. designed an end-to-end recognition method based on convolutional neural network and structured output, but the length of the text needs to be fixed. The text recognition effect is not good. Shi et al. proposed an end-to-end recognition method based on "convolutional neural network + cyclic neural network + sequence classification", but this method is not effective for complex text image recognition.

Images