Extract text from images or PDFs by using the Computer Vision service

Azure Computer Vision is a powerful service that allows you to extract text from images or PDFs with ease. Whether you need to process scanned documents, extract information from receipts, or digitize handwritten notes, the Computer Vision service can assist you in automating these tasks. In this article, we will explore how to use the Computer Vision service to extract text from images or PDFs in a Designing and Implementing a Microsoft Azure AI Solution exam scenario.

Prerequisites

To get started, you will need an Azure subscription and the Azure Computer Vision service provisioned in your Azure portal. If you haven’t set up the service yet, you can follow the documentation to create a Computer Vision resource.

Once you have the required resources set up, you can use the Computer Vision API to perform Optical Character Recognition (OCR) on images or PDF files. OCR is the process of converting images of text into machine-readable text.

Steps to Extract Text from Images

To use the Computer Vision service, you need an API key and an endpoint URL. You can find these details in the Azure portal after provisioning the Computer Vision resource. Make sure to keep your API key secure.

Let’s begin by extracting text from an image. Assume that you have a PNG image named “sample_image.png” stored locally on your machine. Here’s how you can extract the text from the image using the Computer Vision service:


const apiKey = "YOUR_API_KEY";
const endpoint = "YOUR_ENDPOINT_URL";

In the above code, we first load the required SDKs and create an instance of the Computer Vision client using your API key and endpoint. Then, we use the `readFileSync` function from the Node.js `fs` module to read the image file. Finally, we make an asynchronous call to the `recognizePrintedTextInStream` method to perform OCR on the image. The result is then processed to extract the recognized text.

Steps to Extract Text from PDFs

Now, let’s explore how to extract text from a PDF. Assume that you have a PDF file named “sample_pdf.pdf” stored locally on your machine. Here’s an example of how you can extract the text from the PDF using the Computer Vision service:


// Instantiate the Computer Vision client and read the PDF file
const extractTextFromPDF = async () => {
const pdfPath = "sample_pdf.pdf";
const result = await client.recognizePrintedTextInStream(fs.readFileSync(pdfPath));

extractTextFromPDF();


In this code snippet, we again instantiate the Computer Vision client and read the PDF file using `readFileSync`. Then, we call the `recognizePrintedTextInStream` method, passing in the PDF file. The resulting text is extracted and displayed.

Using the above examples, you can easily extract text from images or PDFs by leveraging the Computer Vision service. This functionality can be extremely useful in automating data extraction and digitizing processes. Make sure to refer to the Microsoft documentation for more details and explore the various additional features offered by the Computer Vision service to enhance text extraction capabilities in your Azure AI solution.