OCTUS iOS SDK

Octus SDK uses advanced deep learning technologies for accurate and fast ID scanning and OCR. Businesses can integrate the Octus SDK into native iOS Apps which comes with pre-built screens and configurations. The SDK returns the scanned images, extracted data and error codes. And as a safety measure, the SDK does not store any of the personal data or ID images that are scanned.

You can find the release history at Changelog

‼ ATTENTION ‼ → BREAKING CHANGE introduced at Octus SDK v1.6.6. We have introduced a new license format. If you are using versions prior to v1.6.6 and intend to update to v1.6.6 or above please contact support@frslabs.com for an updated license.

Table Of Content

• Prerequisite
• iOS SDK Requirements
• Permission
• Installation
• Usage example
• Octus Result
• Octus Parameters
• Octus Error Codes
• Help

Prerequisite

You will need a valid license to use the Octus SDK, which can be obtained by contacting support@frslabs.com .

Depending on the license - offline or online - you have opted for, the ping functionality to billing servers will be disabled or enabled. For instance, if you have opted for the offline SDK model, then there will be no server ping needed to our billing server to bill you. However, if you have chosen a transaction based pricing, then after each transaction, a ping request will be made to our billing server. This cannot be overrided by the App. A point to note is that if the ping transaction fails for any reason, the whole transaction will be void without any results from the SDK.

Once you have the license , follow the below instructions for a successful integration of Octus SDK onto your iOS Application

Minimum Requirements

• Xcode 13.0
• iOS 13.0+
• Swift 5.0

Permission

In Info.plist file add following code to allow your application to access iPhone's camera: <key>NSCameraUsageDescription</key> <string>Allow access to camera</string>

Installation

CocoaPods

You can use CocoaPods to install Octus by adding it to your Podfile:

platform :ios, '13.0'
source 'https://gitlab.com/frslabs-public/ios/octus.git'
source 'https://github.com/CocoaPods/Specs.git'
use_frameworks!
pod 'Octus','1.7.5'
pod 'TesseractOCRiOS', :git => 'https://github.com/gali8/Tesseract-OCR-iOS.git'
pod 'TensorFlowLiteSwift', '2.6.0'

Save/Edit Netrc settings to install custom pod

You will need a valid netrc credentials to install octus from maven, which can be obtained by contacting support@frslabs.com.

1. Create or edit .netrc file under current user's home directory
2. Write the below lines into that file, replace <YOUR_USERNAME> and <YOUR_PASSWORD> with your credentials which is shared through email and save the file.

machine www.repo2.frslabs.space
login <YOUR_USERNAME>
password <YOUR_PASSOWRD>

3. In terminal enter below command to install the pod pod install or pod update.

4. Connect with physical device to build and run Octus, It will not build/run in simulator due to camera dependency.

To get the full benefits import Octus wherever you import UIKit

import UIKit
import Octus

Supported Tessdata installation

1. Download and drop model.trainneddata in tessdata folder of your project.
2. Congratulations!

Usage example

import Octus

let scanner = IdScannerController(delegate: self)
scanner.modalPresentationStyle = .fullScreen
scanner.licenceKey = "YOUR_LICENCE_KEY"
scanner.isManualScanEnabled = false
scanner.isOrientationFlat = false
scanner.documentType = Document.PAN.rawValue
scanner.documentCountry = Country.in.rawValue
scanner.documentSubType = ScanMode.OCR.rawValue
scanner.documentSide = 1
scanner.isWatermarkNeeded = true/false (CHEQUE)
present(scanner, animated: false)

Handling the result

class  ViewController: UIViewController, IdScannerControllerDelegate {

    /// Tells the delegate that the user scanned a document.
    ///
    /// - Parameters:
    ///   - scanner: The scanner controller object managing the scanning interface.
    ///   - results: The results of the user scanning with the camera. 
    ///   - results.octusResult: return output as NSMutableDictionary.
    /// - Discussion: Your delegate's implementation of this method should dismiss the octus scanner controller.
    func idScannerController(_ scanner: IdScannerController, didFinishScanningWithResults results: IdScannerResults) {
          DispatchQueue.main.async {
            print("OctusResult: ", results.octusResult)
        }      
    }
    
    /// Tells the delegate that the user cancelled the scan operation.
    ///
    /// - Parameters:
    ///   - scanner: The scanner controller object managing the scanning interface.
    /// - Discussion: Your delegate's implementation of this method should dismiss the image scanner controller.
    func  idScannerControllerDidCancel(_ scanner: IdScannerController) {
          DispatchQueue.main.async {
            print("Scan operation cancelled")
        }
    }
    
    /// Tells the delegate that an error occured during the user's scanning experience.
    ///
    /// - Parameters:
    ///   - scanner: The scanner controller object managing the scanning interface.
    ///   - error: The error that occured.
    func idScannerController(_ scanner: IdScannerController, didFailWithError error: Int{        
          DispatchQueue.main.async {
            print("ErrorCode: ", error)
        }
    }  
 }
 

Octus Result

     let resultJson = convertToJson(jsonObject: results.octusResult)
     let resultDict = convertToDictionary(text: resultJson)
   
     let result = resultDict!["OctusData"] as! [String:String]   
     let code = result["code"]
     
     let code = result["code"]
     let docType = result["documentType"]
     let name1 = result["name1"] ?? ""
     let name2 = result["name2"] ?? ""
     let idNumber = result["number"] ?? ""
     let dob = result["dob"] ?? ""
     let yob = result["yob"] ?? ""
     let country = result ["country"] ?? ""
     let expiry = result["expiry"] ?? ""
     let address = result["address"] ?? ""
     let gender = result["gender"] ?? ""
     let issuedBy = result["issuedBy"] ?? ""
     let ifsc = result["ifsc"] ?? ""
     
     let imagePathFace = result["facePath"] ?? "" 
     let imagePathFront = result["frontImagePath"] ?? ""
     let imagePathBack = result["backImagePath"] ?? ""
     let imagePathCrop = result["croppedImagePath"] ?? ""
     
     /// Retrieve image from document directory
            
     let image = getImageFromDocumentDirectory(imagePath: imagePathFace, fileName: "fileName")
     
     `FileNames` - 
         Front Image - "OCTUS_FILE_FRONT.png"
         Back Image - "OCTUS_FILE_BACK.png"
         Face Image - "OCTUS_FILE_FACE.png"
         Crop Image - "OCTUS_FILE_CROP.jpg"
     
      if imagePath.count > 0 {
           let resultImage = getImageFromDocumentDirectory(imagePath: imagePath, fileName: fileName)
           imageView.image = resultImage
        }

    /// Supported Mathods
    
    func convertToJson(jsonObject:NSMutableDictionary) -> String{
        let jsonData: NSData
        do {
            jsonData = try JSONSerialization.data(withJSONObject: jsonObject, options:.prettyPrinted) as NSData
            let jsonString = (String(data: jsonData as Data, encoding: String.Encoding.utf8))!.replacingOccurrences(of: "\\", with: "")
            return jsonString
        } catch _ {
            print ("JSON Failure")
        }
        return ""
    }
    
    func convertToDictionary(text: String) -> [String: Any]? {
        if let data = text.data(using: .utf8) {
            do {
                return try JSONSerialization.jsonObject(with: data, options: []) as? [String: Any]
            } catch {
                print(error.localizedDescription)
            }
        }
        return nil
    }
    
    func getImageFromDocumentDirectory(imagePath : String, fileName: String) -> UIImage {
        let nsDocumentDirectory = FileManager.SearchPathDirectory.documentDirectory
        let nsUserDomainMask    = FileManager.SearchPathDomainMask.userDomainMask
        let paths               = NSSearchPathForDirectoriesInDomains(nsDocumentDirectory, nsUserDomainMask, true)
        if let dirPath = paths.first{
            let frontimageURL = URL(fileURLWithPath: dirPath).appendingPathComponent(fileName)
            let frontimage    = UIImage(contentsOfFile: frontimageURL.path)
            return frontimage ?? UIImage()
        }
        else {
            return UIImage()
        }        
    }
    

Octus Error Codes

Error codes and their meaning are tabulated below

Code	Message
801	Scan timed out
802	Invalid ID parameters passed
803	Camera permission denied
805	License expired
806	License invalid
807	Octus server error
808	Transaction API error
811	QR not detected
812	QR parsing failed
108	Internet unavailable
401	API limit exceeded
429	Too many requests

Octus Parameters

• scanner.licenceKey = "LICENCE KEY" (Required)

  Accepts the Octus licence key as a String

• scanner.isManualScanEnabled = false (Required)

  true - Displays a button used to start the scan when clicked false - Automatically starts scanning as soon as camera preview is ready

• scanner.documentCountry = Country.countryISOcode.rawValue (Required)

  Sets the country associated with the Document.

  For the complete list of supported countries refer ISO_3166-1_alpha-2 format code

• scanner.documentType = Document.value.rawValue (Required)

  Sets the Document which has to be scanned. Possible values are,

  Value	Effect
  Document.PAN	Pan Card
  Document.ADR	Aadhaar Card
  Document.VID	Voter ID
  Document.NID	National ID
  Document.PPT	Passport
  Document.VSA	Visa
  Document.DRV	Driving Licence
  Document.CQL	Cheque Leaf
  Document.GST	GST Form
  Document.IMG_A	Masked Aadhaar Image Capture
  Document.IMG_H	Plain Horizontal Image capture
  Document.IMG_V	Plain Vertical Image capture
  Document.E_MANDATE	E-mandate Document Capture

• scanner.documentSubType = ScanMode.OCR.rawValue (Required)

  Sets the Document Sub Type . Majority of the documents support only ScanMode.OCR.rawValue as a sub type.

  Documents where both ScanMode.OCR.rawValue and ScanMode.BARCODE.rawValue apply are ,

  • Document.ADR

  Documents where only ScanMode.MRTD.rawValue apply are ,

  • Document.PPT
  • Document.VSA

• scanner.isWatermarkNeeded = true/false (Optional) (CHEQUE)

  • Sets the "CANCELLED" watermark to cheque if it is true.
  • Doesn't set "CANCELLED" watermark to cheque if it false.

Possible values for Sub Type are,

Value	Effect
ScanMode.OCR.rawValue	Scans the document in OCR mode
ScanMode.BARCODE.rawValue	Scans the document in QR mode
ScanMode.MRTD.rawValue	Scans the document in MRZ mode
ScanMode.CROP.rawValue	Scans the document in Crop mode

Note: ScanMode.OCR is only supported from iOS version 13 and above. For versions below this, please use our APIs.

• scanner.documentSide = 1 (or 2) (Required)

  Set the document side 1 or 2 according to documentType AND ScanMode

  Value	ScanMode	DocumentSide
  Document.PAN	OCR	1
  Document.ADR	OCR, BARCODE	2, 1
  Document.VID	OCR	2
  Document.PPT	MRTD	1 OR 2
  Document.DRV	OCR	2
  Document.CQL	OCR	1
  Document.GST	OCR	1
  Document.IMG_A	CROP	2
  Document.IMG_H	CROP	2
  Document.IMG_V	CROP	2
  Document.E_MANDATE	OCR	1

  Note: In case of MRTD, DocumentSide = 2 is only applicable for Indian passports with address on the back page.

• scanner.isOrientationFlat = false (Optional) (Defaults to false)

Sets the value for which the scanner should lock the orientation with respect to the scan surface . Possible values are,

Value	Effect
true	Scans only when orientation of the phone(camera) is perpendicular(flat) to the scan surface.
false	Scans ignoring the orientation of the phone(camera) to the scan surface.

NOTE

In order to get the Bank details, an API call has to be made to the server to do the MICR lookup and fetch the details.

Help

For any queries/feedback , contact us at support@frslabs.com