Digital Receipt Detection
Identify digitally created receipts to distinguish them from photographed paper receipts for enhanced processing accuracy and fraud prevention. Taggun's Digital Receipt Detection feature helps optimize your receipt processing workflow by identifying digitally created or edited receipts that may require different handling compared to authentic photographed paper receipts. This is particularly valuable for fraud prevention, promotional campaign validation, and ensuring receipt authenticity.
Key Capabilities
- Digital Document Identification: Detects receipts that are digitally created rather than photographed from paper.
- Multiple Detection Methods: Uses color pattern analysis, text line alignment detection, and white space analysis for comprehensive assessment.
- Confidence Scoring: Provides a confidence score for the likelihood of digital creation.
How It Works
When a receipt image is submitted, Taggun's digital detection system performs the following analyses:
- Hough Line Analysis: Examines text line alignment to detect perfectly horizontal lines typical of digitally typeset documents.
- Color Pattern Analysis: Samples colors across the image to identify limited color palettes characteristic of digital creation.
- White Space Analysis: Measures white space percentage, as digital documents typically contain more uniform white backgrounds than photographed papers.
Based on these analyses, Taggun's system determines whether the receipt is digital and assigns a confidence score.
Setup Process
1. Contact Taggun
Reach out to [email protected] to enable this feature for your account.
2. Use Verbose Endpoints
The digital receipt detection feature is available in all our verbose endpoints.
Note on Receipt Validation APIsDigital Receipt Detection is currently available in both Taggun Data Extraction APIs and Receipt Validation APIs. The feature can be configured per campaign in validation workflows through the
allowDigitalDetection
setting.
Understanding Response Properties
Field Name | Type | Description |
---|---|---|
entities.digitalDetection | object | Contains the digital detection results. |
entities.digitalDetection.data.isDigital | boolean | Indicates whether the receipt appears to be digitally created. |
entities.digitalDetection.data.digitalScore | number | The calculated confidence score indicating the likelihood of digital creation (0.0 to 1.0). |
For Receipt Validation APIs the response will be the same, but not under the entities
property, but instead at the root level.
Example
A user submits a digitally created receipt:

Response when a digital receipt is submitted:
{
"entities": {
"digitalDetection": {
"data": {
"isDigital": true,
"digitalScore": 0.92
}
}
}
}
In this example:
- The receipt is identified as digital (
isDigital: true
). - The digital score is high (0.92), indicating a strong likelihood that this is a digitally created receipt.
Use Cases
- Fraud Prevention: Identify potentially suspicious digitally created receipts in promotional campaigns or expense submissions.
- Campaign Validation: Ensure receipt authenticity by flagging digital receipts that may not meet campaign requirements.
- Processing Optimization: Route digital receipts for additional verification while allowing authentic photographed receipts standard processing.
- Compliance: Meet regulatory requirements that may distinguish between original and digital receipt formats.
FAQs
Q: Will this feature detect all digitally created receipts?
A: The feature is designed to identify receipts with clear digital characteristics such as perfect text alignment, limited color palettes, or high white space ratios. Sophisticated digital recreations may require additional validation methods.
Q: How accurate is the digital detection?
A: While highly accurate for typical digital creation patterns, no system is perfect. We recommend using the digital score to set appropriate thresholds for your use case and combining this feature with other validation tools for comprehensive protection. The default threshold is 0.8, but you can adjust this using the confidence score to better align with the specific risk tolerance and requirements of your use case.
Q: Does this work with all file formats?
A: Yes, digital detection supports multiple formats including PNG, JPEG, GIF, PDF, and HEIC files with automatic format detection.
Best Practices
- Threshold Configuration: Use the digital score to set appropriate thresholds for your specific use case rather than relying solely on the boolean flag.
- Layered Approach: Combine digital detection with other fraud prevention and validation tools for comprehensive receipt verification.
- Campaign Settings: Configure the
allowDigitalDetection
setting appropriately for different validation campaigns based on your requirements. - Manual Review: Implement review processes for receipts flagged as digital, especially in high-stakes validation scenarios.
Start Building
Contact us at [email protected] to enable Digital Receipt Detection on your account.
Updated 8 days ago