Reading data from images

Suppose you are given thousands of images and asked to transfer the text on the images to a computer database. The scanned images have text organized in different formats and contain multiple languages. What are some ways you could complete the project in a reasonable time frame and make sure the data is entered with a high degree of accuracy?

Companies around the world are tackling similar scenarios every day. Without AI services, it would be challenging to complete the project, especially if it were to change in scale.

Using AI services, we can treat this project as a computer vision scenario and apply Optical Character Recognition (OCR). OCR allows you to extract text from images, such as photos of street signs and products, as well as from documentsinvoices, bills, financial reports, articles, and more.

To build an automated AI solution, you need to train machine learning models to cover many use cases. Azure's Computer Vision service is a Cognitive Service that gives access to advanced algorithms for processing images and returns data to secure storage.

The Computer Vision service offers two APIs that you can use to read text.

OCR API:
Use this API to read small to medium volumes of text from images.
The API can read text in multiple languages.
Results are returned immediately from a single function call.

Read API:
Use this API to read small to large volumes of text from images and PDF documents.
This API uses a newer model than the OCR API, resulting in greater accuracy.
The Read API can read printed text in multiple languages, and handwritten text in English.

The initial function call returns an asynchronous operation ID, which must be used in a subsequent call to retrieve the results.
The computer vision service offers the OCR API and Read API.

You can access both technologies via the REST API or a client library. In the next few units, we'll show you how to call the REST API and return a JSON response. Then for the exercise, you'll use a client library to return objects that abstract the JSON response.