Table structure identification method and device

Abstract

The embodiment of the invention provides a table structure identification method and device, and the method comprises the steps: inputting a target image with a table structure into a pre-trained neural network model, and obtaining a table line segment in the target image; obtaining a table structure according to a table line segment in the target image, wherein the neural network model is an improved L-CNN model, and the improved L-CNN model is trained by taking a sample image with a table structure as a sample and taking a table line segment in the sample image as a sample label, the bent table line segment in the sample image is represented by a broken line formed by a plurality of continuous straight line segments, and the proportion of the line feature map loss in the improved L-CNN model to the calculation loss is greater than the point feature map loss. According to the embodiment of the invention, the improved L-CNN model can be well applied to the field of table recognition, and compared with the prior art, the recognition efficiency and accuracy are significantly improved.

Description

technical field [0001] The present invention relates to the technical field of image processing, and more specifically, to a table structure recognition method and device. Background technique [0002] Tables are widely used as an effective data organization and presentation method, and have also become common page objects in various documents, such as scientific journals, reports, and financial statements. Usually people need to manually process table documents, extract and collect the required information, and with the explosive growth of the number of documents, this will consume huge labor costs. [0003] Due to the development of Optical Character Recognition (OCR) technology, people began to use deep learning technology to extract effective text information in pictures, and achieved certain results. As a special form of information carrier, the table is different from simple text OCR. It requires strict row-column correspondence, and extracts the text content in each ...

AI Technical Summary

Problems solved by technology

Usually people need to manually process table documents, extract and collect the required information, and with the explosive growth of the number of documents, this will consume huge labor costs

Images