Design of Document Secret Level Consistency Check Software Based on OCR
Shuhao Qi, Qiao Sun, Chen Cheng, Zhenjie Wang, Ziwen Gao, Juntao Liu · 2024
The secret level of secret documents should be correctly labeled, especially when they are printed or copied from the company’s internal network to mobile media. At this point, it is necessary to perform a consistency check on the file name of the document, the document cover, and the content of the document to ensure the consistency of the secret levels among the three. Traditional consistency checks require manual labor and can be time-consuming when there are many documents. This article uses Python language and OCR to design a document secret level consistency check software. The software determines the secret level of the document cover by extracting the cover image from a doc/docx or PDF file and recognizing text in the upper left corner. The content secret level of the document is also determined through character matching. The test shows that the software can replace manual verification of secret level consistency, and the processing speed meets the requirements.