OCR training sample generation method, device and system

Abstract

The invention discloses an OCR training sample generation method, device and system, and relates to the field of computer vision. The method comprises the following steps: a character contour extraction step: extracting all character contours based on an original image, determining an erasure area mask by combining an erasure area coordinate, and obtaining a repair area mask; an image repairing and filling step: performing image repairing and filling according to the mask of the repairing area and the pixel information around the repairing area to obtain a background template after the characters are erased; and a random text generation step: generating a random text in each generation area so as to obtain a new sample picture and a labeling information file corresponding to the new sample picture. According to the method, a character contour extraction algorithm, an image restoration technology and other technologies are combined, background information of an original picture is fully utilized, a high-quality training picture is generated, meanwhile, an annotation file (including character content and position information) corresponding to the picture is generated, the burdensome and labor-consuming annotation work is omitted, and the method can be directly used for OCR model training.

Description

technical field [0001] The present invention relates to the field of computer vision, in particular to a method, device and system for generating OCR training samples. Background technique [0002] OCR (Optical Character Recognition, Optical Character Recognition) tasks widely exist in life and business scenarios. The current best method for this task is to use deep learning technology for text positioning and recognition. However, in real scenarios, there is often a shortage of training samples for a certain format, and deep learning relies on huge training samples. Therefore, a method for automatically generating OCR training samples emerges at the historic moment. [0003] The current methods for generating OCR samples are roughly divided into two categories: [0004] The first category is template-based methods. That is, input a clean template picture, and then write random text in the specified position of the template to generate an OCR sample of the template style...

AI Technical Summary

Problems solved by technology

This method has two main defects: one is that we cannot obtain the template image in most cases; the other is that the template image is often a clean and undistorted ideal image, which is quite different from the real image mixed with various noises. Sample authenticity is poor

This type of method has four main defects: The first defect is that the generalization ability of the model is poor. Since the OCR image features and text features of different formats are very different, for image formats or text styles that have not appeared in the model training process, The model often performs poorly; the second defect is that it is difficult to obtain training data. Due to the limited generalization ability of the model, if you want to train an image editing model for a certain format, you first need a large amount of training data similar to the format, and the current scene is originally It is because of the lack of data that image editing technology is needed; the third defect is that the samples for training image editing models need pixel-level annotation, which is very expensive and time-consuming; the fourth defect is that the model reasoning speed is relatively slow, and the operation often A deep learning reasoning framework is required, which greatly increases engineering complexity

Images