Pdf table structure identification method based on image identification

Abstract

The invention discloses a pdf table structure identification method based on image identification. According to the invention, the method includes converting a pdf document into an image; and for eachinput image, identifying the position of a table, intercepting a table area, identifying a text blob block in the table area, finding an adjacent blob for each blob, predicting the relationship between the blob and each adjacent blob, and finally obtaining the structure of the table through the relationship. According to the method, the image features are removed, the features of the edges of theadjacent blocks are increased, the search range of the text blocks in the table is narrowed by using the blob field, and the convergence rate and the recognition accuracy are greatly increased. The problem of character omission is eliminated through detection and post-processing of the text blob.

Description

technical field [0001] The invention relates to image recognition technology, in particular to an image recognition-based pdf table structure recognition method. Background technique [0002] In the application scenarios of big data and artificial intelligence, it is necessary to collect, process, and analyze a large amount of information, structure the data, and discover the laws in the data to guide production. Information exists in various and unstructured ways. A large amount of information exists in tables, and tables may exist in pdfs, web pages, and images. For tables in pdf, currently existing table parsing methods generally include parsing tables by reading xml information of pdf (such as xpdf tool), converting pdf to xml, html, word and other formats and then parsing (such as pdf-docx tool ), converting the pdf into an image and then performing structural recognition. The first two methods cannot be accurately analyzed due to the information loss of the pdf file i...

AI Technical Summary

Problems solved by technology

For tables in pdf, currently existing table parsing methods generally include parsing tables by reading xml information of pdf (such as xpdf tool), converting pdf to xml, html, word and other formats and then parsing (such as pdf-docx tool ), convert the pdf to an image and then perform structure recognition. The first two methods cannot be accurately analyzed due to the information loss of the pdf file itself. The third method mainly relies on image recognition algorithms. The existing methods are not suitable for complex forms. can accurately identify

Images