🏦 Bank Statement OCR

1 ICper page

v1.0 Active Jul 2026 POST /ocr/v3/bank-statement/file

Welcome to the Bank Statement OCR API, part of iApp Any Documents OCR by iApp Technology Co., Ltd. Extracts account information and the complete transaction table from Thai bank statements (all major Thai banks). Multi-page PDF statements are combined into a single result.

Try Demo!

Try Our AI Demo

Login or create a free account to use this AI service demo and explore our powerful APIs.

Get 100 Free Credits (IC) when you sign up!

Offer ends December 31, 2025

Create Free AccountLogin

Example Images (Click to try)

API Key

Log in to use your API keys or Create an account

Endpoint URL

cURL Command

Upload Bank Statement

Getting Started

1. Prerequisites

   • An API key from iApp Technology
   • Document file: Bank Statement (รายการเดินบัญชีธนาคาร)
   • Supported file formats: PDF, JPG, JPEG, PNG, WEBP
   • Maximum file size: 50MB

2. Quick Start

   • Single POST with multipart/form-data
   • Structured JSON out, with a per-field confidence score
   • Multi-page PDFs are combined into a single docs entry.

3. Key Features

   • Field-level extraction with confidence scoring
   • Thai and English output via target_lang
   • Optional raw OCR text via return_ocr
   • Handles photographs and scanned PDFs

4. Security & Compliance

   • GDPR and PDPA compliant
   • Secure API endpoints
   • No data retention after processing

How to get API Key?

Please visit API Key Management page to view your existing API key or request a new one.

API Endpoints

Endpoint	Method	Description	Cost
/ocr/v3/bank-statement/file /v3/store/ocr/bank-statement	POST	Extract account details and the full transaction table from Thai bank statements	1 IC per page

Both forms are equivalent. There is no base64 variant for this endpoint — file upload only.

Request

Request Headers

Header	Value	Required
apikey	Your iApp API key	Yes

Request Body (multipart/form-data)

Parameter	Type	Default	Description
file	File (repeatable)	—	Required. Document file. PDF pages are processed automatically.
return_ocr	Boolean	false	Include the raw OCR text in the response.
target_lang	String	th	Language of extracted values: th or en.
include_confidence	Boolean	true	Include a per-field confidence score (0.00–1.00).

cURL

curl -X POST "https://api.iapp.co.th/ocr/v3/bank-statement/file" \
  -H "apikey: YOUR_API_KEY" \
  -F "file=@/path/to/document.pdf" \
  -F "target_lang=th" \
  -F "include_confidence=true"

Response

{
    "docs": [
        {
            "message": "success",
            "raw": {},
            "processed": {
                        "bank_name": "ธนาคารกสิกรไทย",
                        "account_holder_name": "น.ส. สมหญิง ตัวอย่าง",
                        "account_number": "XXX-2-48512-X",
                        "account_type": "เงินฝากออมทรัพย์",
                        "branch": "สาขาซีคอนสแควร์ 2",
                        "period_from": "01-04-2018",
                        "period_to": "20-09-2018",
                        "total_withdrawal_amount": 101797.39,
                        "total_deposit_amount": 101780.19,
                        "transactions": [
                                    {
                                                "date": "10-09-2018",
                                                "description": "ฝากเงินสด ไม่มีสมุดคู่ฝาก",
                                                "withdrawal": null,
                                                "deposit": 100.0,
                                                "balance": 206.14,
                                                "channel": "0929"
                                    }
                        ]
            },
            "confidence": {
                        "bank_name": 0.95,
                        "account_holder_name": 0.95,
                        "account_number": 0.95,
                        "account_type": 0.95,
                        "branch": 0.95,
                        "period_from": 0.95
            },
            "process_ms": 4120
        }
    ],
    "docs_count": 1,
    "process_ms": 4300
}

The iapp-input-pages response header reports the number of billed pages.

Supported Fields

Field	Type	Description
bank_name	String	Bank name
account_holder_name	String	Account holder name
account_number	String	Account number, as printed (including masking X's)
account_type	String	Account type, e.g. เงินฝากออมทรัพย์
branch	String	Branch
period_from	String	Statement period start date
period_to	String	Statement period end date
total_withdrawal_amount	Number	Total withdrawals (THB)
total_deposit_amount	Number	Total deposits (THB)
transactions	Array	Transaction rows — each has date, description, withdrawal, deposit, balance, channel

Error Handling

Status	Meaning
400	Unsupported file extension, or file larger than 50MB
401	Missing or invalid apikey
402	Insufficient credit balance
404	Unknown document type in the URL
422	No file attached to the request
500	Processing error — the response contains an error code to quote to support

Code Examples

Python

import requests

url = "https://api.iapp.co.th/ocr/v3/bank-statement/file"
headers = {"apikey": "YOUR_API_KEY"}
files = {"file": open("document.pdf", "rb")}
data = {"target_lang": "th", "include_confidence": "true"}

response = requests.post(url, headers=headers, files=files, data=data)
result = response.json()
print(result["docs"][0]["processed"])
print(result["docs"][0]["confidence"])

JavaScript

const form = new FormData();
form.append("file", fileInput.files[0]);
form.append("target_lang", "th");
form.append("include_confidence", "true");

const res = await fetch("https://api.iapp.co.th/ocr/v3/bank-statement/file", {
  method: "POST",
  headers: { apikey: "YOUR_API_KEY" },
  body: form,
});
const result = await res.json();
console.log(result.docs[0].processed);

PHP

<?php
$ch = curl_init();
curl_setopt($ch, CURLOPT_URL, "https://api.iapp.co.th/ocr/v3/bank-statement/file");
curl_setopt($ch, CURLOPT_POST, true);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_HTTPHEADER, ["apikey: YOUR_API_KEY"]);
curl_setopt($ch, CURLOPT_POSTFIELDS, [
    "file" => new CURLFile("/path/to/document.pdf"),
    "target_lang" => "th",
]);
$response = curl_exec($ch);
curl_close($ch);
echo $response;

Java

OkHttpClient client = new OkHttpClient();

RequestBody body = new MultipartBody.Builder()
    .setType(MultipartBody.FORM)
    .addFormDataPart("file", "document.pdf",
        RequestBody.create(new File("/path/to/document.pdf"),
            MediaType.parse("application/pdf")))
    .addFormDataPart("target_lang", "th")
    .build();

Request request = new Request.Builder()
    .url("https://api.iapp.co.th/ocr/v3/bank-statement/file")
    .addHeader("apikey", "YOUR_API_KEY")
    .post(body)
    .build();

try (Response response = client.newCall(request).execute()) {
    System.out.println(response.body().string());
}

Swift

let url = URL(string: "https://api.iapp.co.th/ocr/v3/bank-statement/file")!
var request = URLRequest(url: url)
request.httpMethod = "POST"
request.setValue("YOUR_API_KEY", forHTTPHeaderField: "apikey")

let boundary = UUID().uuidString
request.setValue("multipart/form-data; boundary=\(boundary)",
                 forHTTPHeaderField: "Content-Type")

var data = Data()
let fileData = try Data(contentsOf: URL(fileURLWithPath: "/path/to/document.pdf"))
data.append("--\(boundary)\r\n".data(using: .utf8)!)
data.append("Content-Disposition: form-data; name=\"file\"; filename=\"document.pdf\"\r\n\r\n".data(using: .utf8)!)
data.append(fileData)
data.append("\r\n--\(boundary)--\r\n".data(using: .utf8)!)

URLSession.shared.uploadTask(with: request, from: data) { responseData, _, _ in
    if let responseData, let json = String(data: responseData, encoding: .utf8) {
        print(json)
    }
}.resume()

Kotlin

val client = OkHttpClient()

val body = MultipartBody.Builder()
    .setType(MultipartBody.FORM)
    .addFormDataPart(
        "file", "document.pdf",
        File("/path/to/document.pdf").asRequestBody("application/pdf".toMediaType())
    )
    .addFormDataPart("target_lang", "th")
    .build()

val request = Request.Builder()
    .url("https://api.iapp.co.th/ocr/v3/bank-statement/file")
    .addHeader("apikey", "YOUR_API_KEY")
    .post(body)
    .build()

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

Dart

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

final request = http.MultipartRequest(
  'POST',
  Uri.parse('https://api.iapp.co.th/ocr/v3/bank-statement/file'),
);
request.headers['apikey'] = 'YOUR_API_KEY';
request.fields['target_lang'] = 'th';
request.files.add(
  await http.MultipartFile.fromPath('file', '/path/to/document.pdf'),
);

final response = await request.send();
print(await response.stream.bytesToString());

Limitations & Best Practices

Limitations

• Maximum file size 50MB per request
• Supported formats: PDF, JPG, JPEG, PNG, WEBP
• Accuracy depends on image quality — heavily degraded scans, glare and handwriting over printed values reduce field-level accuracy
• Extraction is limited to the fields listed above

Best Practices

• Prefer the original PDF over a photograph where you have one
• Photograph the whole document, square-on and in even lighting
• Use at least 300 DPI for scans
• Check the confidence score per field and route low-confidence values to manual review
• Send multiple documents in one request where supported to reduce round trips

Accuracy & Performance

Field-level extraction accuracy measured on the internal ChindaOCR-M3 golden evaluation set — visually re-verified labels over real Thai documents, including low-resolution photographs and bills partly obscured by handwriting.

Document Type	Docs	Fields	Accuracy
Bank Statement	2	10	100%
Overall, all six document types	25	147	98.6%

Support

Need help? Contact us:

• Email: sale@iapp.co.th
• Phone: 086-322-5858
• Discord: iApp Technology