Azure Cognitive Service Form Recognizer

Azure Form Recognizer is a cloud-based Azure Applied AI Service that enables you to build intelligent document processing solutions. Massive amounts of data, spanning a wide variety of data types, are stored in forms and documents. Form Recognizer enables you to effectively manage the velocity at which data is collected and processed and is key to improved operations, informed data-driven decisions, and enlightened innovation.

Use cases

Document processing models

Azure Form Recognizer supports a wide variety of models that enable you to add intelligent document processing to your apps and flows. You can use a prebuilt document analysis or domain specific model or train a custom model tailored to your specific business needs and use cases. Form Recognizer can be used with the REST API or Python, C#, Java, and JavaScript SDKs.

Models include:

• Document analysis models (read OCR, layout analysis, etc.)
• Prebuilt models (health insurance card, W-2, invoices, etc.)
• Custom models

Form Recognizer Studio

Form Recognizer Studio is an online tool for visually exploring, understanding, and integrating features from the Form Recognizer service into your applications. Use the Form Recognizer Studio quickstart to get started analyzing documents with pretrained models. Build custom template models and reference the models in your applications using the Python SDK and other quickstarts.

Use Form Recognizer SDKs

The Form Recognizer SDK is a set of libraries and tools that enable you to easily integrate Form Recognizer models and capabilities into your applications. Form Recognizer SDK is available across platforms in C#/.NET, Java, JavaScript, and Python programming languages.

Security

Identity-based authentication

We configured authentication using a managed identity and RBAC instead of using the traditional API key. This is a more secure method for your application to use the Form Recognizer API since we aren’t exposing your API key. A managed identity is automatically configured with your application’s runtime, and the Form Recognizer bundle informs the application which RBAC roles are needed to access the service. Those roles are automatically applied when the connection between Form Recognizer and your application is made in Massdriver.

Secure data in transit (for scanning)

All Cognitive Services endpoints, including the Form Recognizer API URLs, use HTTPS URLs for encrypting data during transit. The client operating system needs to support Transport Layer Security (TLS) 1.2 for calling the endpoints. For more information, see Azure Cognitive Services security.

Encrypts input data for processing

The incoming data is processed in the same region where the Cognitive Services Azure resource was created. When you submit your documents to a Form Recognizer operation, it starts the process of analyzing the document to extract all text and identify structure and key values in a document. Your data and results are then temporarily encrypted and stored in Azure Storage.

Data stored by Form Recognizer

For all analysis: To facilitate asynchronous analysis and checking the completion status and returning the extracted results to the customer upon completion, the data and extracted results are stored temporarily in Azure Storage in the same region. All customers in the same region share the temporary storage. The customer’s data is logically isolated from other customers with their Azure subscription and API credentials.

For customer trained models: The Custom model feature allows customers to build custom models from training data stored in customer’s Azure blob storage locations. The interim outputs after analysis and labeling are stored in the same location. The trained custom models are stored in Azure storage in the same region and logically isolated with their Azure subscription and API credentials.

Deletes data: For all features, the input data and results are deleted within 24 hours and not used for any other purpose. For customer trained models, the customers can delete their models and associated metadata at any time by using the API.

Trade-offs

• VNet integration is not supported