This repo shows how to extract information from forms saved as images. For example,

and get structured data from it:

# Data Output
{'cargo': 'AGENTE FISCAL 2014-08-01',
 'ciudad': '‘Quito',
 'civil': '‘SOLTERO',
 'desde': '2014-08-01',
 'gestion': 'INICIO DE GESTION     x PERIODICA             FIN DE GESTION          ',
 'hasta': '',
 'institucion': 'FISCAIA GENERAL DEL ESTADO, AGENTE ‘4. INFORMACION PATRIMONIAL',
 'prov': 'PICHINCHA,'}

I show examples using two OCR agents Tesseract and Google Cloud Vision.

• Tesseract is free. To install follow the instructions in its own repo.
• Google Cloud Vision is free for the first 1k images, then it runs on a pay-per-use base. For instructions on the installation follow this link.