Converting a Vision VNTextObservation to a String - ios

I'm looking through the Apple's Vision API documentation and I see a couple of classes that relate to text detection in UIImages:
1) class VNDetectTextRectanglesRequest
2) class VNTextObservation
It looks like they can detect characters, but I don't see a means to do anything with the characters. Once you've got characters detected, how would you go about turning them into something that can be interpreted by NSLinguisticTagger?
Here's a post that is a brief overview of Vision.
Thank you for reading.

This is how to do it ...
//
// ViewController.swift
//
import UIKit
import Vision
import CoreML
class ViewController: UIViewController {
//HOLDS OUR INPUT
var inputImage:CIImage?
//RESULT FROM OVERALL RECOGNITION
var recognizedWords:[String] = [String]()
//RESULT FROM RECOGNITION
var recognizedRegion:String = String()
//OCR-REQUEST
lazy var ocrRequest: VNCoreMLRequest = {
do {
//THIS MODEL IS TRAINED BY ME FOR FONT "Inconsolata" (Numbers 0...9 and UpperCase Characters A..Z)
let model = try VNCoreMLModel(for:OCR().model)
return VNCoreMLRequest(model: model, completionHandler: self.handleClassification)
} catch {
fatalError("cannot load model")
}
}()
//OCR-HANDLER
func handleClassification(request: VNRequest, error: Error?)
{
guard let observations = request.results as? [VNClassificationObservation]
else {fatalError("unexpected result") }
guard let best = observations.first
else { fatalError("cant get best result")}
self.recognizedRegion = self.recognizedRegion.appending(best.identifier)
}
//TEXT-DETECTION-REQUEST
lazy var textDetectionRequest: VNDetectTextRectanglesRequest = {
return VNDetectTextRectanglesRequest(completionHandler: self.handleDetection)
}()
//TEXT-DETECTION-HANDLER
func handleDetection(request:VNRequest, error: Error?)
{
guard let observations = request.results as? [VNTextObservation]
else {fatalError("unexpected result") }
// EMPTY THE RESULTS
self.recognizedWords = [String]()
//NEEDED BECAUSE OF DIFFERENT SCALES
let transform = CGAffineTransform.identity.scaledBy(x: (self.inputImage?.extent.size.width)!, y: (self.inputImage?.extent.size.height)!)
//A REGION IS LIKE A "WORD"
for region:VNTextObservation in observations
{
guard let boxesIn = region.characterBoxes else {
continue
}
//EMPTY THE RESULT FOR REGION
self.recognizedRegion = ""
//A "BOX" IS THE POSITION IN THE ORIGINAL IMAGE (SCALED FROM 0... 1.0)
for box in boxesIn
{
//SCALE THE BOUNDING BOX TO PIXELS
let realBoundingBox = box.boundingBox.applying(transform)
//TO BE SURE
guard (inputImage?.extent.contains(realBoundingBox))!
else { print("invalid detected rectangle"); return}
//SCALE THE POINTS TO PIXELS
let topleft = box.topLeft.applying(transform)
let topright = box.topRight.applying(transform)
let bottomleft = box.bottomLeft.applying(transform)
let bottomright = box.bottomRight.applying(transform)
//LET'S CROP AND RECTIFY
let charImage = inputImage?
.cropped(to: realBoundingBox)
.applyingFilter("CIPerspectiveCorrection", parameters: [
"inputTopLeft" : CIVector(cgPoint: topleft),
"inputTopRight" : CIVector(cgPoint: topright),
"inputBottomLeft" : CIVector(cgPoint: bottomleft),
"inputBottomRight" : CIVector(cgPoint: bottomright)
])
//PREPARE THE HANDLER
let handler = VNImageRequestHandler(ciImage: charImage!, options: [:])
//SOME OPTIONS (TO PLAY WITH..)
self.ocrRequest.imageCropAndScaleOption = VNImageCropAndScaleOption.scaleFill
//FEED THE CHAR-IMAGE TO OUR OCR-REQUEST - NO NEED TO SCALE IT - VISION WILL DO IT FOR US !!
do {
try handler.perform([self.ocrRequest])
} catch { print("Error")}
}
//APPEND RECOGNIZED CHARS FOR THAT REGION
self.recognizedWords.append(recognizedRegion)
}
//THATS WHAT WE WANT - PRINT WORDS TO CONSOLE
DispatchQueue.main.async {
self.PrintWords(words: self.recognizedWords)
}
}
func PrintWords(words:[String])
{
// VOILA'
print(recognizedWords)
}
func doOCR(ciImage:CIImage)
{
//PREPARE THE HANDLER
let handler = VNImageRequestHandler(ciImage: ciImage, options:[:])
//WE NEED A BOX FOR EACH DETECTED CHARACTER
self.textDetectionRequest.reportCharacterBoxes = true
self.textDetectionRequest.preferBackgroundProcessing = false
//FEED IT TO THE QUEUE FOR TEXT-DETECTION
DispatchQueue.global(qos: .userInteractive).async {
do {
try handler.perform([self.textDetectionRequest])
} catch {
print ("Error")
}
}
}
override func viewDidLoad() {
super.viewDidLoad()
// Do any additional setup after loading the view, typically from a nib.
//LETS LOAD AN IMAGE FROM RESOURCE
let loadedImage:UIImage = UIImage(named: "Sample1.png")! //TRY Sample2, Sample3 too
//WE NEED A CIIMAGE - NOT NEEDED TO SCALE
inputImage = CIImage(image:loadedImage)!
//LET'S DO IT
self.doOCR(ciImage: inputImage!)
}
override func didReceiveMemoryWarning() {
super.didReceiveMemoryWarning()
// Dispose of any resources that can be recreated.
}
}
You'll find the complete project here included is the trained model !

SwiftOCR
I just got SwiftOCR to work with small sets of text.
https://github.com/garnele007/SwiftOCR
uses
https://github.com/Swift-AI/Swift-AI
which uses NeuralNet-MNIST model for text recognition.
TODO : VNTextObservation > SwiftOCR
Will post example of it using VNTextObservation once I have it one connected to the other.
OpenCV + Tesseract OCR
I tried to use OpenCV + Tesseract but got compile errors then found SwiftOCR.
SEE ALSO : Google Vision iOS
Note Google Vision Text Recognition - Android sdk has text detection but also has iOS cocoapod. So keep an eye on it as should add text recognition to the iOS eventually.
https://developers.google.com/vision/text-overview
//Correction: just tried it but only Android version of the sdk supports text detection.
https://developers.google.com/vision/text-overview
If you subscribe to releases:
https://libraries.io/cocoapods/GoogleMobileVision
Click SUBSCRIBE TO RELEASES
you can see when TextDetection is added to the iOS part of the Cocoapod

Apple finally updated Vision to do OCR. Open a playground and dump a couple of test images in the Resources folder. In my case, I called them "demoDocument.jpg" and "demoLicensePlate.jpg".
The new class is called VNRecognizeTextRequest. Dump this in a playground and give it a whirl:
import Vision
enum DemoImage: String {
case document = "demoDocument"
case licensePlate = "demoLicensePlate"
}
class OCRReader {
func performOCR(on url: URL?, recognitionLevel: VNRequestTextRecognitionLevel) {
guard let url = url else { return }
let requestHandler = VNImageRequestHandler(url: url, options: [:])
let request = VNRecognizeTextRequest { (request, error) in
if let error = error {
print(error)
return
}
guard let observations = request.results as? [VNRecognizedTextObservation] else { return }
for currentObservation in observations {
let topCandidate = currentObservation.topCandidates(1)
if let recognizedText = topCandidate.first {
print(recognizedText.string)
}
}
}
request.recognitionLevel = recognitionLevel
try? requestHandler.perform([request])
}
}
func url(for image: DemoImage) -> URL? {
return Bundle.main.url(forResource: image.rawValue, withExtension: "jpg")
}
let ocrReader = OCRReader()
ocrReader.performOCR(on: url(for: .document), recognitionLevel: .fast)
There's an in-depth discussion of this from WWDC19

Adding my own progress on this, if anyone have a better solution:
I've successfully drawn the region box and character boxes on screen. The vision API of Apple is actually very performant. You have to transform each frame of your video to an image and feed it to the recogniser. It's much more accurate than feeding directly the pixel buffer from the camera.
if #available(iOS 11.0, *) {
guard let pixelBuffer = CMSampleBufferGetImageBuffer(sampleBuffer) else {return}
var requestOptions:[VNImageOption : Any] = [:]
if let camData = CMGetAttachment(sampleBuffer, kCMSampleBufferAttachmentKey_CameraIntrinsicMatrix, nil) {
requestOptions = [.cameraIntrinsics:camData]
}
let imageRequestHandler = VNImageRequestHandler(cvPixelBuffer: pixelBuffer,
orientation: 6,
options: requestOptions)
let request = VNDetectTextRectanglesRequest(completionHandler: { (request, _) in
guard let observations = request.results else {print("no result"); return}
let result = observations.map({$0 as? VNTextObservation})
DispatchQueue.main.async {
self.previewLayer.sublayers?.removeSubrange(1...)
for region in result {
guard let rg = region else {continue}
self.drawRegionBox(box: rg)
if let boxes = region?.characterBoxes {
for characterBox in boxes {
self.drawTextBox(box: characterBox)
}
}
}
}
})
request.reportCharacterBoxes = true
try? imageRequestHandler.perform([request])
}
}
Now I'm trying to actually reconize the text. Apple doesn't provide any built in OCR model. And I want to use CoreML to do that, so I'm trying to convert a Tesseract trained data model to CoreML.
You can find Tesseract models here: https://github.com/tesseract-ocr/tessdata and I think the next step is to write a coremltools converter that support those type of input and output a .coreML file.
Or, you can link to TesseractiOS directly and try to feed it with your region boxes and character boxes you get from the Vision API.

Thanks to a GitHub user, you can test an example: https://gist.github.com/Koze/e59fa3098388265e578dee6b3ce89dd8
- (void)detectWithImageURL:(NSURL *)URL
{
VNImageRequestHandler *handler = [[VNImageRequestHandler alloc] initWithURL:URL options:#{}];
VNDetectTextRectanglesRequest *request = [[VNDetectTextRectanglesRequest alloc] initWithCompletionHandler:^(VNRequest * _Nonnull request, NSError * _Nullable error) {
if (error) {
NSLog(#"%#", error);
}
else {
for (VNTextObservation *textObservation in request.results) {
// NSLog(#"%#", textObservation);
// NSLog(#"%#", textObservation.characterBoxes);
NSLog(#"%#", NSStringFromCGRect(textObservation.boundingBox));
for (VNRectangleObservation *rectangleObservation in textObservation.characterBoxes) {
NSLog(#" |-%#", NSStringFromCGRect(rectangleObservation.boundingBox));
}
}
}
}];
request.reportCharacterBoxes = YES;
NSError *error;
[handler performRequests:#[request] error:&error];
if (error) {
NSLog(#"%#", error);
}
}
The thing is, the result is an array of bounding boxes for each detected character. From what I gathered from Vision's session, I think you are supposed to use CoreML to detect the actual chars.
Recommended WWDC 2017 talk: Vision Framework: Building on Core ML (haven't finished watching it either), have a look at 25:50 for a similar example called MNISTVision
Here's another nifty app demonstrating the use of Keras (Tensorflow) for the training of a MNIST model for handwriting recognition using CoreML: Github

I'm using Google's Tesseract OCR engine to convert the images into actual strings. You'll have to add it to your Xcode project using cocoapods. Although Tesseract will perform OCR even if you simply feed the image containing texts to it, the way to make it perform better/faster is to use the detected text rectangles to feed pieces of the image that actually contain text, which is where Apple's Vision Framework comes in handy.
Here's a link to the engine:
Tesseract OCR
And here's a link to the current stage of my project that has text detection + OCR already implemented:
Out Loud - Camera to Speech
Hope these can be of some use. Good luck!

For those still looking for a solution I wrote a quick library to do this. It uses both the Vision API and Tesseract and can be used to achieve the task the question describes with one single method:
func sliceaAndOCR(image: UIImage, charWhitelist: String, charBlackList: String = "", completion: #escaping ((_: String, _: UIImage) -> Void))
This method will look for text in your image, return the string found and a slice of the original image showing where the text was found

Firebase ML Kit does it for iOS (and Android) with their on-device Vision API and it outperforms Tesseract and SwiftOCR.

Related

Action ML Classifier not giving expected results

I am creating an app which detect the exercises. i trained the model using create ML. i got 100% result in create ML application. But when i am integrating into the application using Vision framework it's always showing only one exercise. i followed the code exactly from Build an Action Classifier with Create ML for creating ml and requesting VNHumanBodyPoseObservation. Followed this for converting VNHumanBodyPoseObservation to MLMultiArray.
Here is the code what i do:
func didOutput(pixelBuffer: CVPixelBuffer) {
self.extractPoses(pixelBuffer)
}
func extractPoses(_ pixelBuffer: CVPixelBuffer) {
let handler = VNImageRequestHandler(cvPixelBuffer: pixelBuffer)
let request = VNDetectHumanBodyPoseRequest { (request, err) in
if err == nil {
if let observations =
request.results as? [VNRecognizedPointsObservation], observations.count > 0 {
if let prediction = try? self.makePrediction(observations) {
print("\(prediction.label), confidence: \(prediction.confidence)")
}
}
}
}
do {
// Perform the body pose-detection request.
try handler.perform([request])
} catch {
print("Unable to perform the request: \(error).\n")
}
}
func makePrediction(_ observations: [VNRecognizedPointsObservation]) throws -> (label: String, confidence: Double) {
let fitnessClassifier = try PlayerExcercise(configuration: MLModelConfiguration())
let numAvailableFrames = observations.count
let observationsNeeded = 60
var multiArrayBuffer = [MLMultiArray]()
for frameIndex in 0 ..< min(numAvailableFrames, observationsNeeded) {
let pose = observations[frameIndex]
do {
let oneFrameMultiArray = try pose.keypointsMultiArray()
multiArrayBuffer.append(oneFrameMultiArray)
} catch {
continue
}
}
// If poseWindow does not have enough frames (45) yet, we need to pad 0s
if numAvailableFrames < observationsNeeded {
for _ in 0 ..< (observationsNeeded - numAvailableFrames) {
do {
let oneFrameMultiArray = try MLMultiArray(shape: [1, 3, 18], dataType: .double)
try resetMultiArray(oneFrameMultiArray)
multiArrayBuffer.append(oneFrameMultiArray)
} catch {
continue
}
}
}
let modelInput = MLMultiArray(concatenating: [MLMultiArray](multiArrayBuffer), axis: 0, dataType: .float)
//
//
let predictions = try fitnessClassifier.prediction(poses: modelInput)
return (label: predictions.label, confidence: predictions.labelProbabilities[predictions.label]!)
}
func resetMultiArray(_ predictionWindow: MLMultiArray, with value: Double = 0.0) throws {
let pointer = try UnsafeMutableBufferPointer<Double>(predictionWindow)
pointer.initialize(repeating: value)
}
I suspect the issue happening while converting VNRecognizedPointsObservation to MLMultiArray Please help me, i am trying to achieve this so hard. Thanks in advance.
Are you running your app on a simulator? Because I had the same issue that the model predicted wrong results when I ran my image classifier app on a iPhone 12 simulator. But the issue was solved when I tried to run the app on a real device. So maybe there is nothing wrong with your model or code, try running it on a real device and see if you get your intended results.

Why cant I find the same keyword more than once in Swift?

Basically Im using the vision framework to find keywords the user searches for in the UITextField but if there is more than one of the same words on a given line it only detects the first word. For example lets say I have an image and the text has "in in in in in in" only the first "in" will get detected and the rest won't. Only way it will detect the other words if they are on separate lines like on the right side of the image. Why does this happen?
Here is a pic explaining the problem and my code:
func detectTextHandler(request: VNRequest, error: Error?) {
guard let results = request.results as? [VNRecognizedTextObservation] else {
return
}
DispatchQueue.main.async {
self.previewView.layer.sublayers?.removeSubrange(1...)
for visionResult in results {
guard let candidate = visionResult.topCandidates(1).first else {
continue
}
let words = candidate.string.split{ $0.isWhitespace }.map{ String($0)}
for word in words {
if let wordRange = candidate.string.range(of: word, options: .caseInsensitive),
let boxObservation = try? candidate.boundingBox(for: wordRange) {
if self.searchTextField.text == word
{
print(word)
self.highlightLetters(box: boxObservation)//highlights words
}
}
}
}
}
}

How to decode from 1D barcode image

Below code is only worked for decoding from a QR code image, not applicable for 1D barcode.
I don't want to use any third-party library.
Is it possible to get any CIQRCodeFeature for Barcode image?
your help is appreciated.
func scanCodeFromImage(image: UIImage) -> String? {
guard let detector = CIDetector(ofType: CIDetectorTypeQRCode, context: nil, options: [CIDetectorAccuracy : CIDetectorAccuracyHigh]), let ciImage = CIImage(image: image), let features = detector.features(in: ciImage) as? [CIQRCodeFeature] else { return nil }
var qrCodeText = ""
for feature in features {
if let message = feature.messageString {
qrCodeText += message
}
}
return qrCodeText
}
No third party libraries are needed for this, Apple provides the AVFoundation framework
Follow this tutorial for a better understanding of how to get it working, all you have to do is change the type of barcode you want to scan, and you can select multiple different types or just one type. This could also be done with images or straight from the camera without using a third party library.

How To Get Bounding Box Data For Created mlmodel With Playground

We created a mlmodel with playground like https://developer.apple.com/documentation/createml/creating_an_image_classifier_model.
Then we used following code to get bounding box data of objects in that mlmodel. But in "results" we can get just prediction values and object names we modeled, even that was exiting but not our aim.
print("detectOurModelHandler (results)") Shows us the all the objects and
prediction values in our mlmodel and it is VNClassificationObservation.
So it is no surprise that we do not have box data.
So the problem is how to create model as VNRecognizedObjectObservation, I think ?
According to https://developer.apple.com/documentation/vision/recognizing_objects_in_live_capture we are supposed to get bounding box data.
But we can not. Even print("detectOurModelHandler 2") is never called like dump(objectBounds).
We call findOurModels in captureOutput by the way. We call it like once in 1 second to test our model at the moment.
lazy var ourModel:VNCoreMLModel = { return try! VNCoreMLModel(for: ImageClassifier().model)}()
lazy var ourModelRequest: VNCoreMLRequest = {
return VNCoreMLRequest(model: ourModel, completionHandler: detectOutModelHandler)
}()
func findOurModels(pixelbuffer: CVPixelBuffer){
let testImage = takeAFrameImage(imageBuffer: pixelbuffer)
let imageForThis = testImage.cgImage
let requestOptions2:[VNImageOption : Any] = [:]
let handler = VNImageRequestHandler(cgImage: imageForThis!,
orientation: CGImagePropertyOrientation(rawValue: 6)!,
options: requestOptions2)
try? handler.perform([ourModelRequest])
}
func detectOurModelHandler(request: VNRequest, error: Error?) {
DispatchQueue.main.async(execute: {
if let results = request.results {
print("detectOurModelHandler \(results)")
for observation in results where observation is VNRecognizedObjectObservation {
print("detectOurModelHandler 2")
guard let objectObservation = observation as? VNRecognizedObjectObservation else {
continue
}
let objectBounds = VNImageRectForNormalizedRect(objectObservation.boundingBox, self.frameWidth, self.frameHeight)
dump(objectBounds)
}
}
})
}
It can not be done using CreateML.
I did not do it yet but it is said a model with bounding data could be created with Turi Create.

Firebase MLKit Text Recognition Error

I'm trying to OCR my image using Firebase MLKit but it fails and return with error
Text detection failed with error: Failed to run text detector because self is nil.
/// Detects texts on the specified image and draws a frame for them.
func detectTexts() {
let image = #imageLiteral(resourceName: "testocr")
// Create a text detector.
let textDetector = vision.textDetector() // Check console for errors.
// Initialize a VisionImage with a UIImage.
let visionImage = VisionImage(image: image)
textDetector.detect(in: visionImage) { (features, error) in
guard error == nil, let features = features, !features.isEmpty else {
let errorString = error?.localizedDescription ?? "No results returned."
print("Text detection failed with error: \(errorString)")
return
}
// Recognized and extracted text
print("Detected text has: \(features.count) blocks")
let resultText = features.map { feature in
return "Text: \(feature.text)"
}.joined(separator: "\n")
print(resultText)
}
}
It looks like you need to keep a strong reference to textDetector, otherwise the detector gets released before the completion block can be called.
Changing your code a bit:
var textDetector: VisionTextDetector? // NEW
/// Detects texts on the specified image and draws a frame for them.
func detectTexts() {
// ... truncated ...
textDetector = vision.textDetector() // NEW
let visionImage = VisionImage(image: image)
textDetector?.detect(in: visionImage) { (features, error) in // NEW
// Callback implementation
}
}
You can also unwrap it to make sure it's not nil after you assign it:
guard let textDetector = textDetector else {
print("Error: textDetector is nil.")
return
}
I hope that helps!
VisionTextDetector is no more supported so you have to use VisionTextRecognizer.
Here is an example code and I hope its helpful
//MARK: Firebase var
lazy var vision = Vision.vision()
// replace VisionTextDetector with VisionTextRecognizer
var textDetector: VisionTextRecognizer?
override func viewDidLoad() {
super.viewDidLoad()
textDetector = vision.onDeviceTextRecognizer()
}
// also instead of using detect use process now
textDetector!.process(image) { result, error in
guard error == nil, let result = result else {
//error stuff
return
}
let text = result.text
self.textV.text = self.textV.text + " " + text
}
}

Resources