A Random Sample Generation Method for Complicated Character Recognition

Abstract

The invention relates to the field of image recognition, in particular to a random sample generation method for complex character recognition. In complex text recognition, by analyzing the reasons for the complexity of the text, a large number of samples containing the noise model of the image to be recognized and the distorted feature model generated by a random sample generator are used on the basis of standard characters similar to the characters to be recognized. The training samples automatically generated by the random sample generator contain various complex noises and distortions, which can meet the needs of various complex text recognition; inputting the above random samples as training samples into the deep neural network can solve the problem of training the deep neural network. The problem of requiring a lot of manual labeling when recognizing text makes the automatic recognition of complex text images easier and significantly saves related labor costs.

Description

technical field [0001] The invention relates to the field of image recognition, in particular to a random sample generation method for complex character recognition. Background technique [0002] Image recognition is of great significance in the field of intelligent recognition. With the advancement of technology and the development of society, the demand for automatic recognition of text in pictures is also increasing rapidly. Traditional Optical Character Recognition (OCR) systems are often used to identify documents scanned using optical devices, such as digitized ancient books, business cards, invoices, forms, etc. Usually this type of scanned document has a relatively high resolution and contrast, and the printed fonts are generally relatively single and regular, making it easier to extract a single text for recognition. Therefore, the core of this type of document recognition is to eliminate noise. There are many ways to eliminate noise: for example, use Gaussian for ...

AI Technical Summary

Problems solved by technology

By analyzing the reasons for the complexity of the text, a large number of training samples containing various noise and distortion features that can be used by the deep neural network are automatically generated, which solves the problem of requiring a large amount of human labeling when using the deep neural network to recognize text in the prior art , significantly saving labor costs; improving the efficiency of identification

Images