Key Value Pair Extraction SDK

Bring Intelligent Document understanding and processing features to your unstructured and semi-structured documents with a key value pair extraction sdk.

The engine is able to understand there is information with labels and values in a document, extract them and qualify the value, in an instant.

Trusted since 15+ years, by 3000+ and Fortune500 companies, and 10K+ developers.

Free Download
Schedule a Demo

Key Value Pair Extraction and Intelligent Document Processing

Key Value Pair

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.

Example:
Date (key) : 06/14/22 (value)

Depending on the type of document, the key-value pair fields are different. For instance, the ones on an invoice will be different than the ones on a survey or a government form.
It’s easy to get key-value pairs from structured documents like Excel files because the values are all named.

Key-value pair fields for invoices can be:

  • Invoice Number
  • Date
  • Total Amounts
  • Taxes

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.
For these documents, you will need a key value pair extraction engine to retrieve the information.

The GdPicture Key Value Pair Extraction Engine

The KVP extract engine is fully part of the GdPicture.NET OCR engine and like the other OCR technologies (MICR, MRZ, OMR, contextual OCR, and more), it benefits from a hybrid approach that includes heuristics, mathematics, and ML capabilities.
We also use an adaptive layout understanding and the same underlying elements techniques as NLP technologies.

The GdPicture.NET key value pair extraction engine automatically adapts to the document and searches for the right approach, making the best use of resources available.

This approach allows us to have excellent results on the usual weaknesses of traditional OCR and pure Machine Learning engines, especially with:

  • Text recognition in documents with lots of noise thanks to adaptive despeckling,
  • Dotted lines: ML engines often fail on recognizing them,
  • Touching & broken characters, thanks to character segmentation,
  • Text on colored background: we’re using thresholding and image segmentation to make the image easier to analyze by converting an image from color or grayscale into a binary image,
  • Underlined text,
  • Skewed text,
  • Text in graphics and tables.

The GdPicture.NET KVP extraction engine also provides two additional fields besides Key and Value: Type and Accuracy.

  • The Type data provides the nature of the content. For example: phone number, IBAN, name, credit card number, etc.
  • The Accuracy data is a confidence level. This confidence level is computed by taking into account various parameters like OCR results at character and word levels, type of key, position on the page, and more.
Screenshot of an IBAN showing KVP boxes and results
GdPicture.NET KVP extractor results on an IBAN

Capabilities and Benefits

Document management capabilities

  • enhanced document indexing
  • automatic labeling
  • automatic removal of sensitive information (helps with redaction)
  • invoice processing

Global benefits

  • Less errors
  • Saves time… and money
  • Easier for compliance

Areas of Use of the Key Value Pair Extraction

Banking & Finance

Insurance

Healthcare

Government

HR

Examples of usage

GetKeyValuePairCount Method (GdPictureOCR)

How to use

Download and install GdPicture.NET package from here.

You will find a compiled demo application in
[Install directory]\Samples\Bin\

You will find C# and VB.NET demo applications including source code in
[Install directory]\Samples\WinForm\

You will find other code snippets within the online reference guide found here

GdPicture.NET Guides

FAQs

What is Key-Value Pair (KVP) Extraction?

Key-Value Pair (KVP) Extraction is a technique used in Intelligent Document Processing (IDP) to automatically identify and extract structured data from unstructured documents. A key represents a label (e.g., “Invoice Number”), while the value is the corresponding data (e.g., “INV-12345”). This process is widely used in invoice processing, form recognition, and document automation.

How does Key-Value Pair Extraction work?

KVP extraction typically involves a combination of Optical Character Recognition (OCR) and Natural Language Processing (NLP). The process includes:

Data Structuring – Storing extracted pairs in a structured format for further processing or automation.

Preprocessing – Enhancing document quality for better recognition.

Text Extraction – Using OCR to convert scanned images or PDFs into machine-readable text.

Key-Value Detection – Identifying pairs using predefined rules, machine learning models, or AI-based pattern recognition.

What types of documents benefit from Key-Value Pair Extraction?

KVP extraction is particularly useful for:

Medical & Insurance Documents – Extracting patient IDs, claim numbers, and diagnosis codes.

Invoices & Receipts – Extracting vendor details, total amounts, and payment terms.

Forms & Surveys – Recognizing handwritten or digital form fields.

Contracts & Legal Documents – Identifying client names, agreement dates, and key terms.

What are the challenges of Key-Value Pair Extraction?

Some common challenges include:

Multilingual & Special Characters – Processing documents in different languages or with special symbols adds complexity.

Variability in Document Layouts – Different formats require adaptive models.

Low-Quality Scans – Poorly scanned or handwritten documents can impact accuracy.

Context Understanding – AI models must correctly associate keys with values, even when not explicitly labeled.

How can I implement Key-Value Pair Extraction in my workflow?

You can integrate KVP extraction into your document processing pipeline using AI-powered SDKs or APIs. Solutions like GdPicture’s Intelligent Document Processing offer machine-learning-based key-value extraction that can be embedded into business applications, RPA workflows, and enterprise automation systems.

Our Intelligent Document Processing technologies

Intelligent Document Processing

Intelligent Document Processing

Smart Redaction

Smart Redaction

Table Extraction

Table Extraction

Try GdPicture.NET Now!

60-day free trial

Download GdPicture.NET SDK
Contact sales