Azure Document Intelligence client library for .NET - version 1.0.0

Artikel
12/27/2024

Note: on July 2023, the Azure Cognitive Services Form Recognizer service was renamed to Azure AI Document Intelligence. Any mentions of Form Recognizer or Document Intelligence in documentation refer to the same Azure service.

Azure AI Document Intelligence is a cloud service that uses machine learning to analyze text and structured data from your documents. It includes the following main features:

Layout - Extract text, selection marks, table structures, styles, and paragraphs, along with their bounding region coordinates from documents.
Read - Read information about textual elements, such as page words and lines in addition to text language information.
Prebuilt - Analyze data from certain types of common documents using prebuilt models. Supported documents include receipts, invoices, business cards, identity documents, US W2 tax forms, and more.
Custom analysis - Build custom document models to analyze text, field values, selection marks, table structures, styles, and paragraphs from documents. Custom models are built with your own data, so they're tailored to your documents.
Custom classification - Build custom classifier models that combine layout and language features to accurately detect and identify documents you process within your application.

Source code | Package (NuGet) | API reference documentation | Product documentation | Samples

Getting started

Install the package

Install the Azure Document Intelligence client library for .NET with NuGet:

dotnet add package Azure.AI.DocumentIntelligence

Note: This version of the client library defaults to the 2024-11-30 version of the service.

Prerequisites

An Azure subscription.
A Document Intelligence or an Azure AI services resource to use this package.

Create a Document Intelligence or an Azure AI services resource

Document Intelligence supports both multi-service and single-service access. Create an Azure AI services resource if you plan to access multiple Azure AI services under a single endpoint and key. For Document Intelligence access only, create a Document Intelligence resource. Please note that you will need a single-service resource if you intend to use Microsoft Entra ID authentication.

You can create either resource using:

Option 1: Azure Portal.
Option 2: Azure CLI.

Below is an example of how you can create a Document Intelligence resource using the CLI:

# Create a new resource group to hold the Document Intelligence resource
# If using an existing resource group, skip this step
az group create --name <your-resource-name> --location <location>

# Create the Document Intelligence resource
az cognitiveservices account create \
    --name <resource-name> \
    --resource-group <resource-group-name> \
    --kind FormRecognizer \
    --sku <sku> \
    --location <location> \
    --yes

For more information about creating the resource or how to get the location and sku information see here.

Authenticate the client

In order to interact with the Document Intelligence service, you'll need to create an instance of the DocumentIntelligenceClient class. An endpoint and a credential are necessary to instantiate the client object.

Get the endpoint

You can find the endpoint for your Document Intelligence resource using the Azure Portal or the Azure CLI:

# Get the endpoint for the Document Intelligence resource
az cognitiveservices account show --name "<resource-name>" --resource-group "<resource-group-name>" --query "properties.endpoint"

Either a regional endpoint or a custom subdomain can be used for authentication. They are formatted as follows:

Regional endpoint: https://<region>.api.cognitive.microsoft.com/
Custom subdomain: https://<resource-name>.cognitiveservices.azure.com/

A regional endpoint is the same for every resource in a region. A complete list of supported regional endpoints can be consulted here. Please note that regional endpoints do not support Microsoft Entra ID authentication.

A custom subdomain, on the other hand, is a name that is unique to the Document Intelligence resource. They can only be used by single-service resources.

Create DocumentIntelligenceClient with Microsoft Entra ID

You can authenticate with Microsoft Entra ID by using the Azure Identity library. Note that regional endpoints do not support identity-based authentication. Create a custom subdomain for your resource in order to use this type of authentication.

To use the DefaultAzureCredential provider shown below, or other credential providers provided with the Azure SDK, please install the Azure.Identity package:

dotnet add package Azure.Identity

You will also need to register a new Microsoft Entra application and grant access to Document Intelligence by assigning the "Cognitive Services Data Reader" role to your service principal.

string endpoint = "<endpoint>";
var credential = new DefaultAzureCredential();
var client = new DocumentIntelligenceClient(new Uri(endpoint), credential);

Create DocumentIntelligenceClient with AzureKeyCredential

It is strongly recommended to use Microsoft Entra ID as your default authentication approach. On the other hand, using an AzureKeyCredential can be helpful on getting-started scenarios since it can be set up fastly.

Get the API Key

The API key can be found in the Azure Portal or by running the following Azure CLI command:

az cognitiveservices account keys list --name "<resource-name>" --resource-group "<resource-group-name>"

Create the client

Once you have the value for the API key, create an AzureKeyCredential. With the endpoint and key credential, you can create the DocumentIntelligenceClient:

string endpoint = "<endpoint>";
string apiKey = "<apiKey>";
var client = new DocumentIntelligenceClient(new Uri(endpoint), new AzureKeyCredential(apiKey));

Key concepts

DocumentIntelligenceClient

DocumentIntelligenceClient provides operations for:

Analyzing input documents using prebuilt and custom models through the AnalyzeDocument API.
Detecting and identifying custom input documents with the ClassifyDocument API.

Sample code snippets are provided to illustrate using a DocumentIntelligenceClient here.

More information about analyzing documents, including supported features, locales, and document types can be found in the service documentation.

DocumentIntelligenceAdministrationClient

DocumentIntelligenceAdministrationClient provides operations for:

Building custom models to analyze specific fields you specify by labeling your custom documents.
Compose a model from a collection of existing models.
Managing models created in your account.
Copying a custom model from one Document Intelligence resource to another.
Getting or listing operations created within the last 24 hours.
Building and managing document classification models to accurately detect and identify documents you process within your application.

See examples for Build a Custom Model, Manage Models, and Build a Document Classifier.

Please note that models and classifiers can also be built using a graphical user interface such as the Document Intelligence Studio.

Thread safety

We guarantee that all client instance methods are thread-safe and independent of each other (guideline). This ensures that the recommendation of reusing client instances is always safe, even across threads.

Additional concepts

Examples

The following section provides several code snippets illustrating common patterns used in the Document Intelligence .NET API. Most of the snippets below make use of asynchronous service calls, but keep in mind that the Azure.AI.DocumentIntelligence package supports both synchronous and asynchronous APIs.

Extract Layout
Use Prebuilt Models
Build a Custom Model
Manage Models
Build a Document Classifier
Classify a Document

Extract Layout

Extract text, selection marks, table structures, styles, and paragraphs, along with their bounding region coordinates from documents.

Uri uriSource = new Uri("<uriSource>");
Operation<AnalyzeResult> operation = await client.AnalyzeDocumentAsync(WaitUntil.Completed, "prebuilt-layout", uriSource);
AnalyzeResult result = operation.Value;

foreach (DocumentPage page in result.Pages)
{
    Console.WriteLine($"Document Page {page.PageNumber} has {page.Lines.Count} line(s), {page.Words.Count} word(s)," +
        $" and {page.SelectionMarks.Count} selection mark(s).");

    for (int i = 0; i < page.Lines.Count; i++)
    {
        DocumentLine line = page.Lines[i];

        Console.WriteLine($"  Line {i}:");
        Console.WriteLine($"    Content: '{line.Content}'");

        Console.Write("    Bounding polygon, with points ordered clockwise:");
        for (int j = 0; j < line.Polygon.Count; j += 2)
        {
            Console.Write($" ({line.Polygon[j]}, {line.Polygon[j + 1]})");
        }

        Console.WriteLine();
    }

    for (int i = 0; i < page.SelectionMarks.Count; i++)
    {
        DocumentSelectionMark selectionMark = page.SelectionMarks[i];

        Console.WriteLine($"  Selection Mark {i} is {selectionMark.State}.");
        Console.WriteLine($"    State: {selectionMark.State}");

        Console.Write("    Bounding polygon, with points ordered clockwise:");
        for (int j = 0; j < selectionMark.Polygon.Count; j++)
        {
            Console.Write($" ({selectionMark.Polygon[j]}, {selectionMark.Polygon[j + 1]})");
        }

        Console.WriteLine();
    }
}

for (int i = 0; i < result.Paragraphs.Count; i++)
{
    DocumentParagraph paragraph = result.Paragraphs[i];

    Console.WriteLine($"Paragraph {i}:");
    Console.WriteLine($"  Content: {paragraph.Content}");

    if (paragraph.Role != null)
    {
        Console.WriteLine($"  Role: {paragraph.Role}");
    }
}

foreach (DocumentStyle style in result.Styles)
{
    // Check the style and style confidence to see if text is handwritten.
    // Note that value '0.8' is used as an example.

    bool isHandwritten = style.IsHandwritten.HasValue && style.IsHandwritten == true;

    if (isHandwritten && style.Confidence > 0.8)
    {
        Console.WriteLine($"Handwritten content found:");

        foreach (DocumentSpan span in style.Spans)
        {
            var handwrittenOptions = result.Content.Substring(span.Offset, span.Length);
            Console.WriteLine($"  {handwrittenOptions}");
        }
    }
}

for (int i = 0; i < result.Tables.Count; i++)
{
    DocumentTable table = result.Tables[i];

    Console.WriteLine($"Table {i} has {table.RowCount} rows and {table.ColumnCount} columns.");

    foreach (DocumentTableCell cell in table.Cells)
    {
        Console.WriteLine($"  Cell ({cell.RowIndex}, {cell.ColumnIndex}) is a '{cell.Kind}' with content: {cell.Content}");
    }
}