Receipt OCR API | Atomic APIs

What It Does

The Receipt OCR API uses Tesseract OCR to extract text from receipt and invoice images, then applies intelligent parsing to identify and structure key financial information. The API processes base64-encoded images and returns a comprehensive JSON object containing all extracted data.

Key Features

Multi-format Date Parsing: Automatically detects and converts 6 common date formats to ISO 8601
Line Item Extraction: Parses individual items with descriptions, quantities, and prices
Financial Summary: Extracts subtotal, tax, and total amounts
Payment Detection: Identifies payment methods (cash, credit, debit, etc.)
Currency Recognition: Detects currency symbols and codes
Confidence Scoring: Returns confidence levels for extracted data

Code Examples

                curl -X POST https://api.atomicapis.dev/api/receipt-ocr \
  -H "X-RapidAPI-Proxy-Secret: YOUR_SECRET" \
  -H "Content-Type: application/json" \
  -d '{
    "image": "iVBORw0KGgoAAAANSUhEUgAA...",
    "language": "eng",
    "includeRawText": false
  }'
              

                const response = await fetch('https://api.atomicapis.dev/api/receipt-ocr', {
  method: 'POST',
  headers: {
    'X-RapidAPI-Proxy-Secret': 'YOUR_SECRET',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    image: 'iVBORw0KGgoAAAANSUhEUgAA...',
    language: 'eng',
    includeRawText: false
  })
});

const data = await response.json();
console.log(data);
              

                import requests

response = requests.post(
    'https://api.atomicapis.dev/api/receipt-ocr',
    headers={
        'X-RapidAPI-Proxy-Secret': 'YOUR_SECRET',
        'Content-Type': 'application/json'
    },
    json={
        'image': 'iVBORw0KGgoAAAANSUhEUgAA...',
        'language': 'eng',
        'includeRawText': False
    }
)

data = response.json()
print(data)
              

                using System.Net.Http.Json;

var client = new HttpClient();
client.DefaultRequestHeaders.Add("X-RapidAPI-Proxy-Secret", "YOUR_SECRET");

var request = new
{
    image = "iVBORw0KGgoAAAANSUhEUgAA...",
    language = "eng",
    includeRawText = false
};

var response = await client.PostAsJsonAsync(
    "https://api.atomicapis.dev/api/receipt-ocr", 
    request
);

var data = await response.Content.ReadFromJsonAsync<object>();
Console.WriteLine(data);
              

Request Parameters

Name	Type	Required	Description
`image`	string	Required	Base64-encoded image data. Supports JPEG, PNG, WebP, TIFF, and BMP formats. Maximum size: 10MB.
`language`	string	Optional	OCR language code (e.g., "eng", "deu", "fra"). Default: `"eng"`.
`includeRawText`	boolean	Optional	When set to `true`, includes raw OCR text output in the `rawText` response field. Omit or set to `false` to exclude.

Response Format

200 OK - Success Response

{
  "receipt": {
    "merchantName": "Starbucks Coffee",
    "merchantAddress": "123 Main St, Seattle, WA 98101",
    "date": "2024-01-15",
    "time": "14:32",
    "items": [
      {
        "description": "Grande Caffe Latte",
        "quantity": 1,
        "unitPrice": 4.95,
        "totalPrice": 4.95
      },
      {
        "description": "Blueberry Muffin",
        "quantity": 2,
        "unitPrice": 2.75,
        "totalPrice": 5.50
      }
    ],
    "subtotal": 10.45,
    "tax": 0.87,
    "total": 11.32,
    "paymentMethod": "VISA",
    "cardLastFour": "4242",
    "currency": "USD"
  },
  "confidence": 87.45,
  "rawText": null,
  "processingTimeMs": 1240.35
}

Response Fields

Field	Type	Description
`receipt.merchantName`	string \| null	Name of the merchant or store. Null if not detected.
`receipt.merchantAddress`	string \| null	Address of the merchant. Null if not detected.
`receipt.date`	string \| null	Transaction date in ISO 8601 format (YYYY-MM-DD). Null if not detected.
`receipt.time`	string \| null	Transaction time (e.g., "14:32" or "2:32 PM"). Null if not detected.
`receipt.items`	array	Array of purchased items, each with description (string), quantity (integer), unitPrice (number \| null), and totalPrice (number).
`receipt.subtotal`	number \| null	Subtotal before tax. Null if not detected.
`receipt.tax`	number \| null	Tax amount. Null if not detected.
`receipt.total`	number \| null	Final total amount. Null if not detected.
`receipt.paymentMethod`	string \| null	Payment method detected (VISA, MASTERCARD, AMEX, DISCOVER, DINERS, JCB, DEBIT, CREDIT, or CASH). Null if not detected.
`receipt.cardLastFour`	string \| null	Last four digits of the payment card. Null if not detected.
`receipt.currency`	string \| null	Detected currency code (USD, EUR, GBP, or JPY). Null if not detected.
`confidence`	number	Average OCR confidence score (0-100) from Tesseract, rounded to 2 decimal places.
`rawText`	string \| null	Raw OCR text output (only if includeRawText is true).
`processingTimeMs`	number	Processing time in milliseconds, rounded to 2 decimal places.

Use Cases

Expense Tracking

Build expense tracking apps that automatically extract data from receipt photos. Users can simply snap a picture and have all details captured instantly.

Accounting Integration

Integrate with accounting software like QuickBooks or Xero. Automatically populate expense entries from receipt images without manual data entry.

Receipt Management

Create digital receipt archives with searchable structured data. Enable users to find receipts by merchant, date, amount, or item description.

Build Constraints

Tesseract CLI Shell-out

The API uses Tesseract OCR via command-line interface. Images are processed through temporary files that are securely managed and cleaned up after extraction.

Supported Formats & Size

Accepts base64-encoded images in JPEG, PNG, WebP, TIFF, and BMP formats. Maximum file size is 10MB. Larger images should be optimized before encoding.

Temp File Lifecycle

Temporary files are created during processing and automatically deleted immediately after OCR extraction completes. No images are persisted on the server.

Stateless Operation

The API is completely stateless. Images are discarded immediately after processing. No data is stored or retained between requests.

Error Codes

Error	HTTP Status	Message
`Missing image`	400	The 'image' field is required and cannot be empty.
`Invalid base64`	400	Invalid base64-encoded image data.
`Image too large`	400	Image exceeds maximum allowed size of 10MB.
`OCR error`	400	An error occurred during receipt OCR processing.
`Unauthorized`	401	Invalid or missing X-RapidAPI-Proxy-Secret header.

Receipt OCR