PDF文档表格信息的识别与提取

为编辑PDF文档中的表格信息，设计一种基于Python平台的，包含文件选取与文件转换两大功能模块的信息提取软件。该软件利用Python内置库，针对PDF中不同表格的结构设计算法，识别表格内的文字信息与表格结构,将得到的表格结构还原至Word与Excel文档中,文字信息同样复原至对应单元格内。实验表明:开发的软件完整快速地提取了PDF中的表格信息，并将其转化为易于编辑的Excel和Word文档，达到了预期目的；其转换速度与收费软件WPS相当，转换速度明显快于其他免费转换软件；识别精确度与迅捷PDF转换器、Smallpdf等相比有所提高。

Recognition and Extraction of Table Information from PDF Documents

An information extraction software to edit the table information in PDF documents was designed on the Python platform,which contains modules of file selection and file conversion.The software uses the Python built in library to design algorithms for structures of different tables in PDF,recognizes the text information and table structure in the table,and restores the obtained table structure to Word documents or Excel documents and the text information to the corresponding cells.Experiments show that the software developed extracts table information from PDF documents completely and quickly,and converts it into easy to edit Excel and Word documents,achieving the expected results.It has a conversion speed equivalent to that of paid software WPS and significantly faster than those of other free conversion software,and its recognition accuracy is better than Xunjiepdf,Smallpdf and the likes.