Scan the MRZ code of a passport and extract the firstname, lastname, passport number, nationality, date of birth, expiration date and personal numer.

Overview

PassportScanner

Works with 2 and 3 line identity documents.

Build Status Issues Stars Version License Platform Documentation

Git Twitter LinkedIn eMail

What is this

With PassportScanner you can use your camera to scan the MRZ code of a passport. It will extract all data like firstname, lastname, passport number, nationality, date of birth, expiration date and personal numer. Theres is Support for the TD1 and TD3 format (2 or 3 lines)

IMPORTANT NOTICE: SCANNING IDENTITY DOCUMENTS IS IN MOST CASES RESTRICTED BY LAW. OBSERVE THE APPLICABLE LAWS USING THIS TOOL. THE COPYRIGHT HOLDER IS NOT IN ANY WAY LIABLE FOR UNLAWFUL USAGE OF THIS TOOL.

PassportScanner is trying to optimize OCR results by first performing some graphic filters. The exposure filter is dynamic. This means that if the image is dark it tries to light it up and visa versa. As you can see in the demo animation below you will be able to scan a passport about once every 3 seconds.

wait a moment until the .gif below is downloaded...

animated

Building the PassportScanner demo

The current version is tested with Xcode 11.3, Swift 5

  1. Clone the repo to a working directory

  2. CocoaPods is used to manage dependencies. Pods are setup easily and are distributed via a ruby gem. Follow the simple instructions on the website to setup. After setup, run the following command from the toplevel directory of PassportScanner to download the dependencies:

pod install
  1. Open the PassportScanner.xcworkspace in Xcode.

  2. Build and Run the app.

External components for the demo

PassportScanner is using the following components which can be installed using CocoaPods.

  • TesseractOCRiOS Tesseract OCR iOS is a Framework for iOS7+.
  • GPUImage An open source iOS framework for GPU-based image and video processing
  • UIImage-Resize Category to add some resizing methods to the UIImage class, to resize it to a given CGSize — or fit in a CGSize keeping aspect ratio

Using PassportScanner in your own App

'PassportScanner' is now available through the dependency manager CocoaPods. At this moment this can be installed by executing:

[sudo] gem install cocoapods

If you have installed cocoapods you can just add PassportScanner to your workspace by making sure the following lines are in your Podfile:

source 'https://github.com/CocoaPods/Specs.git'
platform :ios, '8.0'
use_frameworks!
pod "PassportScanner"

Add an import at the top of your swift file like this:

import PassportScanner

If you want support for older versions than iOS 8.0, then you can also just copy the Pod folder containing the 2 classes MRZ and PassportScannerController to your app. Also add the above mentioned external components to your podfile.

When PassportScanner is added to your project, then copy the PassportScanner view from the Main.storyboard of the demo app to your own storyboard. After that you can customize that view. Also copy the MyScanViewController.sift to your own project. See the ViewConroller.swift to see how you can initiate the view and get the MRZ data

also copy over the tessdata folder to our app. That folder contains the trained data for tesseract (The OCR engine)

Here is sample code for a PassportScannerController:

protocol ProcessMRZ {
    func processMRZ(mrz:MRZ)
}

class MyScanViewController: PassportScannerController {

    var delegate: ProcessMRZ?

    // the .StartScan and .EndScan are IBOutlets and can be linked to your own buttons

    // For now just start scanning the moment this view is loaded
    override func viewDidLoad() {
        super.viewDidLoad();
        self.debug = true // So that we can see what's going on (scan text and quality indicator)
        self.accuracy = 1  // 1 = all checksums should pass (is the default so we could skip this line)
        self.mrzType = .auto // Performs a little better when set to td1 or td3
        self.showPostProcessingFilters = true // Set this to true to to give you a good indication of the scan quality
        self.StartScan(self)
    }
    
    override func succesfullScan(mrz: MRZ) {
        print("mrz: {\(mrz.description)\n}")
        delegate?.processMRZ(mrz)
        self.dismissViewControllerAnimated(true, completion: nil)
    }

    override func abbortScan() {
        self.dismissViewControllerAnimated(true, completion: nil)
    }
}

Then in your own code you can access this PassportScannerController and process the MRZ data with something like this:

class ViewController: UIViewController, ProcessMRZ {

    @IBOutlet weak var mrzLabel: UILabel!

    override func supportedInterfaceOrientations() -> UIInterfaceOrientationMask {
        return UIInterfaceOrientationMask.Portrait
    }

    override func prefersStatusBarHidden() -> Bool {
        return true
    }

    @IBAction func StartScan(sender: AnyObject) {
        let storyboard: UIStoryboard = UIStoryboard(name: "Main", bundle: nil)
        let scanVC: MyScanViewController = storyboard.instantiateViewControllerWithIdentifier("PassportScanner") as! MyScanViewController
        scanVC.delegate = self
        self.presentViewController(scanVC, animated: true, completion: nil)
    }

    func processMRZ(mrz:MRZ) {
        self.mrzLabel.text = mrz.description
        mrzLabel.sizeToFit()
    }
}

License

PassportScanner is available under the MIT 3 license. See the LICENSE file for more info.

My other libraries:

Also see my other open source iOS libraries:

  • EVReflection - Reflection based (Dictionary, CKRecord, JSON and XML) object mapping with extensions for Alamofire and Moya with RxSwift or ReactiveSwift
  • EVCloudKitDao - Simplified access to Apple's CloudKit
  • EVFaceTracker - Calculate the distance and angle of your device with regards to your face in order to simulate a 3D effect
  • EVURLCache - a NSURLCache subclass for handling all web requests that use NSURLReques
  • AlamofireOauth2 - A swift implementation of OAuth2 using Alamofire
  • EVWordPressAPI - Swift Implementation of the WordPress (Jetpack) API using AlamofireOauth2, AlomofireJsonToObjects and EVReflection (work in progress)
  • PassportScanner - Scan the MRZ code of a passport and extract the firstname, lastname, passport number, nationality, date of birth, expiration date and personal numer.
  • AttributedTextView - Easiest way to create an attributed UITextView with support for multiple links (url, hashtags, mentions).
Comments
  • Add support to scan the MRZ of an ID card

    Add support to scan the MRZ of an ID card

    As reported by @jrodriguezq there is no support for the MRZ for an ID card. I will add this as an enhanecment later.

    Original question:

    The last commit fixes the issue that I had with the camera, now it looks like this: img_3139

    The OCR it's working, but I'm not able to correctly read the MRZ. As you can see, the image appears to be clear enough. The usual case of the mrz captured it's something like this:

    INCHL1D41230409D26<<<<<<<<<<<< 9002063M2002064CHL17409866<2<6 CORTES<CONTRERAS<<SERGIO<JAV ll 12

    This appears to be almost correct, but some numbers and blank characters usually stand in the way and I'm not able to scan the MRZ. Is there any configuration that I'm missing?

    Again, many thanks for the library.

    enhancement 
    opened by evermeer 17
  • Camera Scanning is not clear

    Camera Scanning is not clear

    I am developing Passport Scanner app by referring PassportScanner program (https://github.com/evermeer/PassportScanner). I am testing above code on iPod 5th Gen. But I am facing problem in that, the problem is “camera to scan the MRZ code of a passport”, once it is launched is not coming clear and Pink dots are coming. How to get clear camera scanning, Do I have to change any camera parameter? Please help me to improve camera scanning.

    passportscanner

    opened by krmadhan 10
  • After scanning for some seconds with a Scan quality insufficient : 0.0, I get a

    After scanning for some seconds with a Scan quality insufficient : 0.0, I get a "GPUImageFramebuffer unlock" Exception.

    2016-03-21 17:50:02.556 PassportScanner[9842:4849034] *** Assertion failure in -[GPUImageFramebuffer unlock], /Users/frlobo/Downloads/PassportScanner-master-2/Pods/GPUImage/framework/Source/GPUImageFramebuffer.m:269 2016-03-21 17:50:02.558 PassportScanner[9842:4849034] *** Terminating app due to uncaught exception 'NSInternalInconsistencyException', reason: 'Tried to overrelease a framebuffer, did you forget to call -useNextFrameForImageCapture before using -imageFromCurrentFramebuffer?' *** First throw call stack: (0x18382d900 0x182e9bf80 0x18382d7d0 0x1841a099c 0x100197c80 0x100197e2c 0x184ac447c 0x184ac3510 0x18370d664 0x184ac3210 0x18370d664 0x18851cc24 0x182eb5ae8 0x18371142c 0x18411741c 0x1841d6728 0x101b65bb0 0x101b6b658 0x1837e4bb0 0x1837e2a18 0x183711680 0x184c20088 0x188588d90 0x1001005e4 0x1832b28b8) libc++abi.dylib: terminating with uncaught exception of type NSException

    bug Fixed? 
    opened by kikolobo 9
  • Crash while scanning

    Crash while scanning

    Hi, I encountered the following error:

    2015-12-31 13:47:49.125 PassportScanner[3038:1427113] *** Assertion failure in -[GPUImageFramebuffer unlock], /Users/francis/Downloads/PassportScanner-master/Pods/GPUImage/framework/Source/GPUImageFramebuffer.m:269 2015-12-31 13:47:49.130 PassportScanner[3038:1427113] *** Terminating app due to uncaught exception 'NSInternalInconsistencyException', reason: 'Tried to overrelease a framebuffer, did you forget to call -useNextFrameForImageCapture before using -imageFromCurrentFramebuffer?'

    Apart from that this is fantastic! Thanks for your work.

    Fixed? 
    opened by francisjervis 7
  • Reduce Tesseract delay?

    Reduce Tesseract delay?

    Hi Edwin!

    I've been making tests with this scanner and I noticed every time Tesseract is called to analyze the processed capture (according to the code is every 0.2 secs) cause a delay in the camera view and the view get freeze constantly. I guess Tesseract take time to read the eng.traineddata and that is the cause of the delay.

    Is there some way to reduce that delay and get a camera view with no freezing?

    opened by alejandroruizponce 5
  • Feature request: Disable camera filters in the preview

    Feature request: Disable camera filters in the preview

    Hi everybody, Currently the library show a manipulated video in the camera preview view. The image is black and white, several filters have been applied in order to facilitate the acquisition with the ocr. For some uses this is not acceptable, I'm trying to figure out if it is possible to restore the "standard camera preview" disabling all filters, re applying them in post processing when the frame is passed to the recognition engine.

    Any suggestion in where to start? Thanks!

    opened by punto2018 5
  • Camera freeze in white screen

    Camera freeze in white screen

    The issue can be reproduced as follow:

    • start the ocr by pressing "scan" in the demo
    • the camera view show "some frame" for a while
    • than it "fade" in white nothing can be done to acquire the MRZ

    Tested on iPhone 6S and iOS 12.

    img_0061trim

    opened by punto2018 4
  • How to validate 3-lines MRZ?

    How to validate 3-lines MRZ?

    Hello @evermeer, awesome work!

    I met the challenge - how to validate second line of triple-lined MRZ. I saw the wikipedia but there is no exhaustive info about how to do that. And other question... in your MRZ.swift i saw that you extract passportNumber like two leading letters and other is numbers.. where you found this format? I want the same explanation for document number for triple-lined MRZ's.

    Thanks in advance!

    question Fixed? 
    opened by andymedvedev 4
  • Possible memory leak or normal behavior

    Possible memory leak or normal behavior

    I've just noticed that after each scan the memory allocated is getting 8-10Mb bigger (after about 10 scans memory usage exceeds 100 Mb ). Is it OK, or maybe an issue of Debug Navigator? It seems that when the PassportScannerController dismisses, it has not been fully deallocated. Tesseract cash clean up doesn't help.

    opened by babucha 3
  • added ocr parsing rect setter, added tesseract training data path set…

    added ocr parsing rect setter, added tesseract training data path set…

    Hi, I've added some features:

    • ocr parsing rect setter Now it is possible to set the rect to be used by tesseract for the ocr. It is useful in case you need to overlap a mask to the render view where the mrz is not in the middle of the screen. The ocr area is now visible in red when debug == true Note: the default mask (overlay.png) in the example project is not correct, maybe it now can be removed (or fixed)

    • added tesseract training data path setter It is useful if the training data are not in the main bundle

    added controls in order to allow the use the view without mounting the vc

    • sometimes the developer can't mount a vc but need only the view in its vc hierarchy. In this case the viewDidLoad and viewDidAppear was not well managed.

    Thanks and Enjoy

    opened by punto2018 3
  • objc support, viewcontroller as delegate

    objc support, viewcontroller as delegate

    Changelog: -added support for objective c projects

    • PassportScannerController can be used as delegate, without need to subclass it
    • Added the processed image getter
    • it is already merged with the version 4.0.1
    opened by punto2018 3
  • When scanning a real passport, it has a hard time to scan the MRZ

    When scanning a real passport, it has a hard time to scan the MRZ

    Not sure why. I have both a Canadian and US passport. The Canadian did scan on the first try. I got one shot at the US but subsequent attempts failed. I just get this warning in the console:

    Scan quality insufficient : 0.0
    2020-12-21 12:03:57.630215-0500 PassportScanner[1519:784903] [Unknown process name] CGImageCreate: invalid image alphaInfo: kCGImageAlphaNone. It should be kCGImageAlphaNoneSkipLast
    - Start recognize
    Scan result : bylaw 1w xf I  L
    
    

    I have ideal light, the code appears clearly on the iPhone 11 that I'm using.

    opened by LaurentDaudelin 1
Releases(4.7.0)
Owner
Edwin Vermeer
Edwin Vermeer
computer vision, image processing and machine learning on the web browser or node.

Image processing and Machine learning labs   computer vision, image processing and machine learning on the web browser or node note Fast Fourier Trans

ryohei tanaka 487 Nov 11, 2022
Convolutional Recurrent Neural Networks(CRNN) for Scene Text Recognition

CRNN_Tensorflow This is a TensorFlow implementation of a Deep Neural Network for scene text recognition. It is mainly based on the paper "An End-to-En

MaybeShewill-CV 1000 Dec 27, 2022
Official code for "Bridging Video-text Retrieval with Multiple Choice Questions", CVPR 2022 (Oral).

Bridging Video-text Retrieval with Multiple Choice Questions, CVPR 2022 (Oral) Paper | Project Page | Pre-trained Model | CLIP-Initialized Pre-trained

Applied Research Center (ARC), Tencent PCG 99 Jan 06, 2023
Rest API Written In Python To Classify NSFW Images.

✨ NSFW Classifier API ✨ Rest API Written In Python To Classify NSFW Images. Fastest Solution If you don't want to selfhost it, there's already an inst

Akshay Rajput 23 Dec 30, 2022
A machine learning software for extracting information from scholarly documents

GROBID GROBID documentation Visit the GROBID documentation for more detailed information. Summary GROBID (or Grobid, but not GroBid nor GroBiD) means

Patrice Lopez 1.9k Jan 08, 2023
Python-based tools for document analysis and OCR

ocropy OCRopus is a collection of document analysis programs, not a turn-key OCR system. In order to apply it to your documents, you may need to do so

OCRopus 3.2k Dec 31, 2022
Application that instantly translates sign-language to letters.

Sign Language Translator Project Description The main purpose of project is translating sign-language to letters. In accordance with this purpose we d

3 Sep 29, 2022
TextBoxes++: A Single-Shot Oriented Scene Text Detector

TextBoxes++: A Single-Shot Oriented Scene Text Detector Introduction This is an application for scene text detection (TextBoxes++) and recognition (CR

Minghui Liao 930 Jan 04, 2023
Table recognition inside douments using neural networks

TableTrainNet A simple project for training and testing table recognition in documents. This project was developed to make a neural network which reco

Giovanni Cavallin 93 Jul 24, 2022
Create single line SVG illustrations from your pictures

Create single line SVG illustrations from your pictures

Javier Bórquez 686 Dec 26, 2022
POT : Python Optimal Transport

This open source Python library provide several solvers for optimization problems related to Optimal Transport for signal, image processing and machine learning.

Python Optimal Transport 1.7k Jan 04, 2023
Maze generator and solver with python

Procedural-Maze-Generator-Algorithms Check out my youtube channel : Auctux Ressources Thanks to Jamis Buck Book : Mazes for programmers Requirements P

Joseph 19 Dec 07, 2022
This repository contains the code for the paper "SCANimate: Weakly Supervised Learning of Skinned Clothed Avatar Networks"

SCANimate: Weakly Supervised Learning of Skinned Clothed Avatar Networks (CVPR 2021 Oral) This repository contains the official PyTorch implementation

Shunsuke Saito 235 Dec 18, 2022
[BMVC'21] Official PyTorch Implementation of Grounded Situation Recognition with Transformers

Grounded Situation Recognition with Transformers Paper | Model Checkpoint This is the official PyTorch implementation of Grounded Situation Recognitio

Junhyeong Cho 18 Jul 19, 2022
Dirty, ugly, and hopefully useful OCR of Facebook Papers docs released by Gizmodo

Quick and Dirty OCR of Facebook Papers Gizmodo has been working through the Facebook Papers and releasing the docs that they process and review. As lu

Bill Fitzgerald 2 Oct 28, 2021
Deep learning based page layout analysis

Deep Learning Based Page Layout Analyze This is a Python implementaion of page layout analyze tool. The goal of page layout analyze is to segment page

186 Dec 29, 2022
Fast style transfer

faststyle Faststyle aims to provide an easy and modular interface to Image to Image problems based on feature loss. Install Making sure you have a wor

Lucas Vazquez 21 Mar 11, 2022
DouZero is a reinforcement learning framework for DouDizhu - 斗地主AI

[ICML 2021] DouZero: Mastering DouDizhu with Self-Play Deep Reinforcement Learning | 斗地主AI

Kwai 3.1k Jan 05, 2023
caffe re-implementation of R2CNN: Rotational Region CNN for Orientation Robust Scene Text Detection

R2CNN: Rotational Region CNN for Orientation Robust Scene Text Detection Abstract This is a caffe re-implementation of R2CNN: Rotational Region CNN fo

candler 80 Dec 28, 2021