Apple Vision framework – Text extraction from image - swift

Question

I am using Vision framework for iOS 11 to detect text on image.
The texts are getting detected successfully, but how we can get the detected text?

Answer 1

Recognizing text in an image
VNRecognizeTextRequest works starting from iOS 13.0 and macOS 10.15 and higher.
In Apple Vision you can easily extract text from image using VNRecognizeTextRequest class, allowing you to make an image analysis request that finds and recognizes text in an image.
Here's a SwiftUI solution showing you how to do it (tested in Xcode 13.4, iOS 15.5):
import SwiftUI
import Vision
struct ContentView: View {
var body: some View {
ZStack {
Color.black.ignoresSafeArea()
Image("imageText").scaleEffect(0.5)
SomeText()
}
}
}
The logic is the following:
struct SomeText: UIViewRepresentable {
let label = UITextView(frame: .zero)
func makeUIView(context: Context) -> UITextView {
label.backgroundColor = .clear
label.textColor = .systemYellow
label.textAlignment = .center
label.font = .boldSystemFont(ofSize: 25)
return label
}
func updateUIView(_ uiView: UITextView, context: Context) {
let path = Bundle.main.path(forResource: "imageText", ofType: "png")
let url = URL(fileURLWithPath: path!)
let requestHandler = VNImageRequestHandler(url: url, options: [:])
let request = VNRecognizeTextRequest { (request, _) in
guard let obs = request.results as? [VNRecognizedTextObservation]
else { return }
for observation in obs {
let topCan: [VNRecognizedText] = observation.topCandidates(1)
if let recognizedText: VNRecognizedText = topCan.first {
label.text = recognizedText.string
}
}
} // non-realtime asynchronous but accurate text recognition
request.recognitionLevel = VNRequestTextRecognitionLevel.accurate
// nearly realtime but not-accurate text recognition
request.recognitionLevel = VNRequestTextRecognitionLevel.fast
try? requestHandler.perform([request])
}
}
If you wanna know a list of supported languages for recognition, read this post please.

Answer 2

Not exactly a dupe but similar to: Converting a Vision VNTextObservation to a String
You need to either use CoreML or another library to perform OCR (SwiftOCR, etc.)

Answer 3

This will return a overlay image with rectangle box on detected text
Here is the full xcode project
https://github.com/cyruslok/iOS11-Vision-Framework-Demo
Hope it is helpful
// Text Detect
func textDetect(dectect_image:UIImage, display_image_view:UIImageView)->UIImage{
let handler:VNImageRequestHandler = VNImageRequestHandler.init(cgImage: (dectect_image.cgImage)!)
var result_img:UIImage = UIImage.init();
let request:VNDetectTextRectanglesRequest = VNDetectTextRectanglesRequest.init(completionHandler: { (request, error) in
if( (error) != nil){
print("Got Error In Run Text Dectect Request");
}else{
result_img = self.drawRectangleForTextDectect(image: dectect_image,results: request.results as! Array<VNTextObservation>)
}
})
request.reportCharacterBoxes = true
do {
try handler.perform([request])
return result_img;
} catch {
return result_img;
}
}
func drawRectangleForTextDectect(image: UIImage, results:Array<VNTextObservation>) -> UIImage {
let renderer = UIGraphicsImageRenderer(size: image.size)
var t:CGAffineTransform = CGAffineTransform.identity;
t = t.scaledBy( x: image.size.width, y: -image.size.height);
t = t.translatedBy(x: 0, y: -1 );
let img = renderer.image { ctx in
for item in results {
let TextObservation:VNTextObservation = item
ctx.cgContext.setFillColor(UIColor.clear.cgColor)
ctx.cgContext.setStrokeColor(UIColor.green.cgColor)
ctx.cgContext.setLineWidth(1)
ctx.cgContext.addRect(item.boundingBox.applying(t))
ctx.cgContext.drawPath(using: .fillStroke)
for item_2 in TextObservation.characterBoxes!{
let RectangleObservation:VNRectangleObservation = item_2
ctx.cgContext.setFillColor(UIColor.clear.cgColor)
ctx.cgContext.setStrokeColor(UIColor.red.cgColor)
ctx.cgContext.setLineWidth(1)
ctx.cgContext.addRect(RectangleObservation.boundingBox.applying(t))
ctx.cgContext.drawPath(using: .fillStroke)
}
}
}
return img
}