Tools for scanning documents and extracting texts from them are nothing new. Optical Character Recognition (OCR) technology can be found in software for almost any platform, including web platforms.
The interesting thing about this is when options that refresh the proposal with something new and affordable burst into the middle. This is the case of Docsumo.
Enterprise-grade scan tool, available free
Docsumo is a startup from Bombay, India, which for two years has been offering business solutions for the digitization of administrative tasks, implementing document processing automation systems, to facilitate their management and analysis.
As things usually work under this model, originally Docsumo’s features were offered as paid services. Although that remains, in the last time some tools were released for free use.
One of them is Extract Tables, which allows you to capture tables and recognize key-value binomials from scanned images or PDF files. For more accurate results, you can train the algorithm, providing some samples to guide it. Likewise, after scanning and before exporting, you can manually verify the data and correct it if necessary.
If you need to export the data in JSON format, you can configure rules so that the exported document is valid to work with a particular API. Also, you can export the scanned table in XLSX format, to edit it in a spreadsheet processor. Two other options are in plain text (TXT) format and a copy of the original image.
Although it is free to access, this tool limits its offer to 20 documents per day.
This utility, released for free recently, comes as a complement to other tools of a similar category that are also offered free, such as a OCR tool high precision web and a PDF to Excel converter, with less paraphernalia than the aforementioned function.