NanScan is library for developing tools that allow image and document processing, as well as extracting key information. It doesn't aim to be a generic image processing library but one that eases extracting information from text based enterprise documents. Such as invoices, orders or contracts.

Although the tool can be easily integrated with other applications, for example it integrates very well with the Koo platform, NanScan is not an independent application but it always requires a project to integrate and adapt it to your needs.

If you have to spend a lot of money in document information processing in your business, we can probably help you. Contact us and tell us what you need.

If you want to use NanScan to integrate it with other applications, but want to do it yourself, NaN offers training services so you can start using it efficiently in a few days.

Some of the features of NanScan:

• It's free software available under GPL license.
• Public source code repository at GitHub.
• As it's developed with Python programming language, it allows easy integration of new functionalities.
• Provides a programming interface to access scanner device under Windows and Linux.
• Implements several text comparison algorithms.
• Can scan the document using several open source OCR and new ones can be easily added.
• Recognizes standard barcodes and new DataMatrix.
• Includes a deskew algorithm, for fixing rotation usually introduced by scanner devices.
• Includes a template management system, very useful for document classification and key data extraction.
• Contains some graphic elements, such as PyQt widgets for scanning images or a visual template designer (Planta).
• Most costly processes (scanning and recognition) can be executed in background (multi-threaded).
• Includes an extension for Koo application so users can scan documents from within the application.
• Integrates an OpenERP server module for queue document processing, classify them using templates designed with Planta application, extract key information from them, attach the image to the desired document, and trigger an action, such as creating a new product or an invoice. Everything done automatically.
• Recently we've started the development of a generic module for detection and recognition of invoices. This system will allow extraction of key information from documents without the need of previously designed templates.
• Currently the application works from images, but it will be easily extended so it can be feeded with OpenOffice.org, Word or PDF documents.



Planta template desinger