🛂 Passport OCR

🛂 พาสปอร์ต

Welcome to iApp Passport OCR API, an AI product developed by iApp Technology Co., Ltd. Our API is designed to automatically recognize and extract information from scanned passport images with high accuracy and speed. The system specifically focuses on reading the Machine Readable Zone (MRZ) at the bottom of passport pages, extracting crucial information like the holder's name, nationality, date of birth, and more.

Try Demo

Visit our API Portal to test the Passport OCR API with your own images.

Getting Started

1. Prerequisites

   • An API key from iApp Technology
   • Scanned passport images (profile page)
   • Supported file formats: JPEG, JPG, PNG, PDF
   • Maximum file size: 10MB

2. Quick Start

   • Fast processing (1-2 seconds per image)
   • High accuracy text extraction (95.51%)
   • Support for multiple file formats

3. Key Features

   • Extracts information from TD3 type MRZ
   • OCR accuracy scoring and validation
   • Face image extraction in base64 format
   • Support for multiple page PDF documents
   • Handles rotated and skewed images
   • Image segmentation pre-processing for skewed images
   • Supports Thai and English language

4. Security & Compliance

   • GDPR and PDPA compliant
   • No data retention after processing

How to get API Key?

Please visit API Portal to view your existing API key or request a new one.

Example

Profile Page

Request:

curl -X POST https://api.iapp.co.th/passport-ocr/v2/ocr \
-H "apikey: YOUR_API_KEY" \
-F "file=@/path/to/passport.jpg"

Response:

{
    "check_composite": "4",
    "check_date_of_birth": "4",
    "check_expiration_date": "4",
    "check_number": "6",
    "check_personal_number": "3",
    "country": "THA",
    "date_of_birth": "930710",
    "expiration_date": "250304",
    "method": "direct",
    "mrz_type": "TD3",
    "names": "MATHANIDA",
    "nationality": "THA",
    "number": "AC1062346",
    "personal_number": "0012345678913",
    "raw_text": "P<THAPOSHNASWADIWONG<<MATHANIDA<<<<<<<<<<<<<\nAC10623466THA9307104F25030440012345678913<34",
    "sex": "F",
    "surname": "POSHNASWADIWONG",
    "type": "P",
    "valid_composite": true,
    "valid_date_of_birth": true,
    "valid_expiration_date": true,
    "valid_number": true,
    "valid_personal_number": true,
    "valid_score": 100,
    "face": "/9j/4AAQSkZJRgABAQAAAQABAAD/..."
}

Response Fields Explanation

Field	Description	Format	Example
check_composite	Overall check digit for upper and middle machine readable lines	Single digit	"4"
check_date_of_birth	Check digit for date of birth field	Single digit	"4"
check_expiration_date	Check digit for expiration date field	Single digit	"4"
check_number	Check digit for document number	Single digit	"6"
check_personal_number	Check digit for personal number	Single digit	"3"
country	Country code of passport issuance	3 letters	"THA"
date_of_birth	Date of birth	YYMMDD	"930710"
expiration_date	Document expiry date	YYMMDD	"250304"
method	Reading method used	String	"direct"
mrz_type	Type of machine readable zone	String	"TD3"
names	First/given name(s)	String	"MATHANIDA"
nationality	Nationality of holder	3 letters	"THA"
number	Passport number	String	"AC1062346"
personal_number	National ID number	String	"0012345678913"
raw_text	Raw MRZ text from passport	String	"P<THAPOSHNASWADIWONG..."
sex	Gender	Single letter	"F"
surname	Last/family name	String	"POSHNASWADIWONG"
type	Document type	String	"P"
valid_composite	Overall MRZ validity check	Boolean	true
valid_date_of_birth	Date of birth check digit validity	Boolean	true
valid_expiration_date	Expiry date check digit validity	Boolean	true
valid_number	Document number check digit validity	Boolean	true
valid_personal_number	Personal number check digit validity	Boolean	true
valid_score	Overall validity score	Integer 0-100	100
face	Extracted face image	Base64 string	"/9j/4AAQSkZJRgABA..."

PDF File Processing

When processing PDF files, the response format differs slightly:

1. The response will be a JSON array containing results for each page
2. Each page result includes:
   • page: Page number
   • info: Contains all the standard fields as shown above
3. A process_time field shows total processing duration

Example PDF response structure:

[
    {
        "page": 1,
        "info": {
            "sex": "F",
            "number": "AC1062346",
            "expiration_date": "04/03/25",
            "raw_text": "P<THAPOSHNASWADIWONG<<MATHANIDA<<<<<<<<<<<<<\nAC10623466THA9307104F25030440012345678913<34",
            "personal_number": "0012345678913",
            "surname": "POSHNASWADIWONG",
            "nationality": "THA",
            "date_of_birth": "10/07/93",
            "names": "MATHANIDA",
            "country": "THA",
            "face": "iVBORw0...CYII=",
            "inference": "3.888",
            "file_name": "1676365480.3823211_thai_passport_page_1.png",
            "message": "Success",
            "status_code": 200
        }
    },
    {
        "process_time": "4.918s"
    }
]

Features & Capabilities

Core Features

• MRZ (Machine Readable Zone) text extraction
• Face image extraction
• Multiple page PDF support
• Automatic image orientation correction
• Image segmentation for skewed images
• High accuracy validation checks

Supported Fields

• Document type
• Issuing country
• Document number
• Holder's name
• Nationality
• Date of birth
• Gender
• Expiration date
• Personal number
• Face image (base64)
• Raw MRZ text
• Validation scores

API Reference

Recognizing Passport

POST https://api.iapp.co.th/passport-ocr/v2

Performs OCR on passport profile page and extracts MRZ data

Parameters:

• file (required): Image file (JPG, PNG, PDF)
• apikey (required): Your API key
• options: Enable "segmentation" for skewed image processing

Code Examples

Python

import requests

url = "https://api.iapp.co.th/passport-ocr/v2/ocr"
headers = {
"apikey": "YOUR_API_KEY"
}
files = {
"file": open("passport.jpg", "rb")
}

response = requests.post(url, headers=headers, files=files)
print(response.json())

JavaScript

const axios = require("axios")
const FormData = require("form-data")
const fs = require("fs")

let data = new FormData()
data.append("file", fs.createReadStream("passport.jpg"))

let config = {
    method: "post",
    url: "https://api.iapp.co.th/passport-ocr/v2/ocr",
    headers: {
        apikey: "YOUR_API_KEY",
        ...data.getHeaders(),
    },
    data: data,
}

axios(config)
    .then((response) => console.log(response.data))
    .catch((error) => console.log(error))

PHP

<?php
$curl = curl_init();

curl_setopt_array($curl, array(
CURLOPT_URL => 'https://api.iapp.co.th/passport-ocr/v2/ocr',
CURLOPT_RETURNTRANSFER => true,
CURLOPT_POST => true,
CURLOPT_HTTPHEADER => array(
    'apikey: YOUR_API_KEY'
),
CURLOPT_POSTFIELDS => array(
    'file'=> new CURLFILE('passport.jpg')
)
));

$response = curl_exec($curl);
curl_close($curl);
echo $response;

Swift

import Foundation

let url = URL(string: "https://api.iapp.co.th/passport-ocr/v2/ocr")!
let boundary = "Boundary-\(UUID().uuidString)"

var request = URLRequest(url: url)
request.httpMethod = "POST"
request.setValue("YOUR_API_KEY", forHTTPHeaderField: "apikey")
request.setValue("multipart/form-data; boundary=\(boundary)", forHTTPHeaderField: "Content-Type")

let imageData = try! Data(contentsOf: URL(fileURLWithPath: "passport.jpg"))

var body = Data()
body.append("--\(boundary)\r\n")
body.append("Content-Disposition: form-data; name=\"file\"; filename=\"passport.jpg\"\r\n")
body.append("Content-Type: image/jpeg\r\n\r\n")
body.append(imageData)
body.append("\r\n--\(boundary)--\r\n")

let task = URLSession.shared.uploadTask(with: request, from: body) { data, response, error in
    if let data = data {
        let json = try? JSONSerialization.jsonObject(with: data)
        print(json ?? "")
    }
}
task.resume()

Kotlin

import okhttp3.*
import java.io.File

val client = OkHttpClient()

val file = File("passport.jpg")
val requestBody = MultipartBody.Builder()
    .setType(MultipartBody.FORM)
    .addFormDataPart(
        "file",
        file.name,
        RequestBody.create(MediaType.parse("image/jpeg"), file)
    )
    .build()

val request = Request.Builder()
    .url("https://api.iapp.co.th/passport-ocr/v2/ocr")
    .addHeader("apikey", "YOUR_API_KEY")
    .post(requestBody)
    .build()

client.newCall(request).execute().use { response ->
    println(response.body()?.string())
}

Java

import java.io.*;
import java.net.HttpURLConnection;
import java.net.URL;
import java.nio.file.*;

public class PassportOCR {
    public static void main(String[] args) throws IOException {
        String boundary = "Boundary-" + System.currentTimeMillis();
        URL url = new URL("https://api.iapp.co.th/passport-ocr/v2/ocr");
        HttpURLConnection conn = (HttpURLConnection) url.openConnection();

        conn.setRequestMethod("POST");
        conn.setDoOutput(true);
        conn.setRequestProperty("apikey", "YOUR_API_KEY");
        conn.setRequestProperty("Content-Type", "multipart/form-data; boundary=" + boundary);

        try (OutputStream os = conn.getOutputStream()) {
            byte[] fileBytes = Files.readAllBytes(Paths.get("passport.jpg"));

            os.write(("--" + boundary + "\r\n").getBytes());
            os.write(("Content-Disposition: form-data; name=\"file\"; filename=\"passport.jpg\"\r\n").getBytes());
            os.write(("Content-Type: image/jpeg\r\n\r\n").getBytes());
            os.write(fileBytes);
            os.write(("\r\n--" + boundary + "--\r\n").getBytes());
        }

        try (BufferedReader br = new BufferedReader(new InputStreamReader(conn.getInputStream()))) {
            String line;
            while ((line = br.readLine()) != null) {
                System.out.println(line);
            }
        }
    }
}

Dart

import 'package:http/http.dart' as http;
import 'dart:io';

Future<void> scanPassport() async {
    var url = Uri.parse('https://api.iapp.co.th/passport-ocr/v2/ocr');
    var request = http.MultipartRequest('POST', url);

    request.headers['apikey'] = 'YOUR_API_KEY';

    request.files.add(
    await http.MultipartFile.fromPath(
        'file',
        'passport.jpg',
    ),
    );

    var response = await request.send();
    var responseBody = await response.stream.bytesToString();
    print(responseBody);
}

Limitations and Best Practices

Limitations

• Maximum file size: 10MB
• Supported formats: JPG, PNG, PDF
• One passport per image
• Clear, focused images required

Best Practices

• Ensure good lighting conditions
• Avoid glare and reflections
• Center the passport in frame
• Keep the passport flat
• Use high resolution images
• Enable segmentation option for skewed images

Accuracy & Performance

Overall Accuracy

• Current Version (2.0): 95.51%
• Previous Version (1.1): 88.86%

Field-Level Accuracy (Version 1.1)

• Name: 81.93%
• Surname: 90.26%
• Document Number: 84.03%
• Nationality: 88.09%
• Face Detection: 100%

Processing Speed

• Single image: 1-2 seconds
• PDF documents: 2-3 seconds per page
• Additional time when using segmentation

Factors Affecting Accuracy

• Image quality
• Lighting conditions
• Document positioning
• Image resolution
• Document condition
• Image skew angle

History

Version 2.0 (February 2023)

• Increased overall accuracy from 88.86% to 95.51%
• Upgraded entire OCR engine
• Added image segmentation pre-processing for skewed images

Version 1.1 (January 2023)

• Added face extraction to base64
• Added support for slightly tilted images
• Added PDF file support
• Added multiple page PDF support

Version 1.0 (February 2022)

• Initial release
• Support for all country passports using MRZ
• Basic image format support (PNG, JPG, JPEG)

Pricing

AI API Service Name	Endpoint	IC Per Request	On-Premise
Passport OCR [v2.0]	passport_ocr_v1	0.75 IC/Request	Contact
	passport_ocr_v2	0.75 IC/Request