Intelligent Document Processing
Extract data from semi-structured and unstructured documents and automate your processes with the GdPicture.NET Intelligent Document Processing set of tools.
Add Key-Value Pair extraction, smart redaction, and table extraction capabilities to your applications thanks to OCR and artificial intelligence technologies.
IDP technologies for unstructured documents
What are unstructured documents?
Any document that does not have a pre-defined data model or is not organized in a pre-defined manner has unstructured data, which represents about 90% of all documents generated.
These documents are:
What about PDF?
The purpose of Intelligent Document Processing (IDP) is to extract data from unstructured and semi-structured documents, including PDFs, but also images, emails, and more using OCR and artificial intelligence technologies.
The GdPicture.NET IDP technologies
The GdPicture.NET Intelligent Document Processing tools rely on various technologies, including heuristics, mathematics, and Artificial Intelligence capabilities while making the best use of resources available.
Document Layout Analysis (DLA)
Document Layout Analysis is the identification and categorization of regions on a document.
It implies a geometric analysis of tables, pictures, equations, and barcodes and a logical layout analysis (paragraphs, lines, words, characters) of the document.
Optical Character Recognition (OCR)
For Intelligent Document Processing purposes, a traditional/standard OCR is not enough, especially in everything that is not typed text on a perfectly white background. So, for documents with:
Traditional OCR won’t work well.
This also means that solutions built on this system are also hard to scale because they will require a lot of verification.
The GdPicture.NET IDP tools use its own OCR engine combined with AI technologies like machine learning and deep learning, to mitigate the traditional OCR limitations.
Textual Content Key-Value Association (KVP)
Key-Value Pairs are two related data items, a key, and a value. The key defines the data and is fixed, and the value is variable and describes the key.
Natural Language Processing (NLP)
NLP is an AI technology that enables machines to understand human speech in text or voice form to communicate with humans in their own natural language.
NLP is essential for extracting data from unstructured documents, as it is, with deep learning, the technology that will make sense of the information extracted.
Named-Entity Recognition (NER)
NER is a form of Natural Language Processing (NLP), a subfield of artificial intelligence.
It is a sub-task of information extraction that tries to locate and classify named entities in unstructured text into predefined categories such as a person’s name, ID number, address, organization, etc. This technology is used for key-value pair extraction and smart redaction in unstructured/semi-structured documents.
Our Intelligent Document Processing technologies
KVP – Key-Value Pair Extraction
ID documents like passports, visas, and other ID cards contain a Machine Readable Zone (MRZ) which makes them readable by machines. The GdPicture.NET MRZ recognition engine allows you to create applications to extract and decode MRZ characters on all types of documents.
Smart Redaction
The GdPicture.NET OMR engine helps to detect the content of a checkbox, fill-in-area, multiple-choice examination form, or any area where highlighting is required to indicate a certain choice.
It also provides an anchoring mechanism (also known as template recognition) to specify the area that needs to be processed.
Table Extraction
The GdPicture.NET MRC engine is producing spectacular results by automatically adjusting the tradeoff between quality and compression rate to provide top quality PDF MRC documents at the lower possible size.
It uses very elaborated adaptative document learning algorithms permitting to identify and classify any form of any nature very quickly.