HomeiOS DevelopmentCapturing Textual content inside Picture Utilizing Dwell Textual content API and SwiftUI

Capturing Textual content inside Picture Utilizing Dwell Textual content API and SwiftUI

Final 12 months, iOS 15 got here with a really helpful characteristic often called Dwell Textual content. You could have heard of the time period OCR (brief for Optical Character Recognition), which is the method for changing a picture of textual content right into a machine-readable textual content format. That is what Dwell Textual content is about.

Dwell Textual content is built-into the digicam app and Photographs app. In case you haven’t tried out this characteristic, merely open the Digicam app. If you level the machine’s digicam at a picture of textual content, you will see that a Dwell Textual content button on the lower-right nook. By tapping the button, iOS mechanically captures the textual content for you. You may then copy and paste it into different functions (e.g. Notes).

This can be a very highly effective and handy options for many customers. As a developer, wouldn’t it’s nice if you happen to can incorporate this Dwell Textual content characteristic in your personal app? In iOS 16, Apple launched the Dwell Textual content API for builders to energy their apps with Dwell Textual content. On this tutorial, let’s see the way to use the Dwell Textual content API with SwiftUI.

Allow Dwell Textual content Utilizing DataScannerViewController

Within the WWDC session about Capturing Machine-readable Codes and Textual content with VisionKit, Apple’s engineer confirmed the next diagram:


Textual content recognization just isn’t a brand new characteristic on iOS 16. On older model of iOS, you should utilize APIs from the AVFoundation and Imaginative and prescient framework to detect and acknowledge textual content. Nevertheless, the implementation is sort of difficult, particularly for individuals who are new to iOS growth.

Within the subsequent launch of iOS, all of the above is simplified to a brand new class known as DataScannerViewController in VisionKit. By utilizing this view controller, your app can mechanically show a digicam UI with the Dwell Textual content functionality.

To make use of the category, you first import the VisionKit framework after which verify if the machine helps the info scanner characteristic:

The Dwell Textual content API solely helps units launched in 2018 or newer with Neural engine. On prime of that, you additionally must verify the provision to see if the consumer approves the usage of knowledge scanner:

As soon as each checks come again with constructive outcomes, you’re prepared to begin the scanning. Right here is the pattern code to launch a digicam with Dwell Textual content:

All you want is create an occasion of DataScannerViewController and specify the acknowledged knowledge varieties. For textual content recognition, you cross .textual content() as the info kind. As soon as the occasion is prepared, you possibly can current it and begin the scanning course of by calling the startScanning() methodology.

Working with DataScannerViewController in SwiftUI

The DataScannerViewController class now solely helps UIKit. For SwiftUI, it wants a bit of labor. We now have to undertake the UIViewControllerRepresentable protocol to make use of the category in SwiftUI initiatives. On this case, I create a brand new struct named DataScanner like this:

This struct accepts two binding variables: one for triggering the info scanning and the opposite is a binding for storing the scanned textual content.

To efficiently undertake the UIViewControllerRepresentable protocol, we have to implement the next strategies:

Within the makeUIViewController methodology, we return an occasion of DataScannerViewController. For updateUIViewController, we begin (or cease) the scanning relying on the worth of startScanning.

To seize the scanned textual content, we’ve to undertake the next methodology of DataScannerViewControllerDelegate:

The tactic is named when the consumer faucets the detected textual content, so we are going to implement it like this:

We verify the acknowledged merchandise and retailer the scanned textual content if any textual content is acknowledged. Lastly, insert this line of code within the makeUIViewController methodology to configure the delegate:

This controller is now prepared to be used in SwiftUI views.

Capturing Textual content Utilizing DataScanner

For instance, we are going to construct a textual content scanner app with a quite simple consumer interface. When the app is launched, it mechanically shows the digicam view for reside textual content. When textual content is detected, customers can faucet the textual content to seize it. The scanned textual content is displayed within the decrease a part of the display screen.


Assuming you’ve created a regular SwiftUI mission, open ContentView.swift and the VisionKit framework.

Subsequent, declare a few state variables to manage the operation of the info scanner and the scanned textual content.

For the physique half, let’s replace the code like this:

We begin the info scanner when the app launches. However earlier than that, we name DataScannerViewController.isSupported and DataScannerViewController.isAvailable to make sure Dwell Textual content is supported on the machine.

The demo app is nearly prepared to check. Since Dwell Textual content requires Digicam entry, please bear in mind to go to the mission configuration. Add the important thing Privateness – Digicam Utilization Description within the Data.plist file and specify the explanation why your app must entry the machine’s digicam.


After the adjustments, you possibly can deploy the app to an actual iOS machine and take a look at the Dwell Textual content operate.

Aside from English, Dwell Textual content additionally helps French, Italian, German, Spanish, Chinese language, Portuguese, Japanese, and Korean.


For the supply code of this demo mission, you possibly can test it out on GitHub.

Observe: We’re updating our Mastering SwiftUI e-book for iOS 16. If you wish to begin studying SwiftUI, try the e-book right here. You’ll obtain a free replace later this 12 months.



Please enter your comment!
Please enter your name here

- Advertisment -
Google search engine

Most Popular

Recent Comments