Start date:
01/01/2022
End date:
30/09/2022
Artificial Intelligence system for document digitization
PIADOCS aims to design and develop an advanced document digitization system based on artificial intelligence models, neural networks and advanced data compression techniques. The final solution will allow the precise extraction of common working documents in the management of an organization, providing interoperability capabilities with third-party systems and a mechanism for measuring the quality of information capture.
Main activities
The objective of the project will be addressed by the CIC research team by carrying out the following activities:
- Training of an algorithm based on AI techniques and neural networks for OCR recognition, covering all the typologies of documents common in the management of an organization.
- Development of a system based on AI algorithms to compress the documents that are inserted into the system.
- Adjustment of the algorithm by using machine learning techniques in a wide variety of document types.
- Training to identify types of documents in scenarios in which an entry of information must be discriminated against before being processed.
- Design and prototyping of interoperability elements that allow connecting the different components that are part of the process.
- Development of a user interface capable of offering all the necessary actions for the personalized training process.
Funding
This project has received funding from SODERCAN (Cantabria Government), R&D on ICT projects, call 2021.
