<img height="1" width="1" style="display:none;" alt="" src="https://px.ads.linkedin.com/collect/?pid=1611884&amp;fmt=gif">
Skip to content
    Try Now
    Get a Demo
      Try Now
      Get a Demo
      Product

      Fraud Detection

      fraud-detection
      fraud-detection-mobile

      ID Fraud Checks

      ESCREEN DETECTION

      Identifies common spoofing attempts by detecting whether an ID was captured directly—or if it’s a screen image taken from a phone, computer, or tablet.

      PAPER PRINT DETECTION

      Flags IDs printed on paper that shouldn’t be, catching low-effort forgeries unless the document type explicitly allows paper formats.

      PRIMARY FACE DETECTION

      Ensures the main face on the ID is present, clearly visible, and unobstructed—free from blur, glare, hands, or thumbs.

      ID FACES COMPARISON

      Verifies  all face images on the ID (if multiple exist) are consistent with each other and legitimate.

      PHOTO INJECTION DETECTION

      Detects whether the face image on the ID has been digitally altered, swapped, or tampered with.

      FACE-TO-AGE CORRELATION

      Estimates the person’s age in the photo and compares it against the expected age based on date of birth (DOB) and issue date. Flags mismatches that indicate fraud.For example, if an ID was issued on March 1, 2020, to someone born on March 1, 2000, we would expect the face to resemble that of a 20-year-old. If the facial age appears significantly inconsistent, the ID is flagged for potential fraud.

      IMAGE MANIPULATION ANALYSIS

      Reviews the entire ID image for signs of tampering, overlays, or digital alterations

      TEMPLATE MATCHING

      Confirms that the ID matches trusted templates from Vouched’s proprietary library—checking layout, text placement, and structural integrity against known patterns.

      KNOWN FRAUDULENT NAMES CHECK

      Automatically flags names that match known fraudulent identities, including test names used by government agencies or fictional characters.

      KNOWN FACES DATABASE MATCH

      Cross-references the ID face against a database of flagged or synthetic faces, including celebrities, or stock images

      BARCODE VS. FRONT IMAGE VERIFICATION

      Compares barcode data to the front of the ID to detect inconsistencies or mismatched documents.

      BARCODE VS. USER-SUBMITTED NAME CHECK

      Ensures the name encoded in the barcode matches the name submitted by the user—adding another layer of fraud protection.

      SUBMITTED DRIVER’S LICENSE VS DMV RECORDS (OPTIONAL)

      Validates the submitted driver’s license against official DMV records to confirm authenticity.

      Selfie Fraud Checks

      REAL PERSON DETECTION

      Validates that the selfie contains a real human—filterning out non-human objects like pets, props, or backgrounds.

      SELFIE QUALITY ANALYSIS

      Checks for clarity, brightness, and obstructions—ensuring the image isn’t too blurry, dark, or partially covered.

      LIVENESS DETECTION

      Confirms the selfie is live and authentic - not a printed photo, AI-generated image, or deepfake - using biometric analysis.

      SELFIE-TO-ID FACE MATCH

      Matches the selfie to the primary face on the ID and, when applicable, to secondary face images—strengthening identity assurance.

      SELFIE-TO-ID FACE AGE MATCH

      Estimates the person's facial age in the selfie and compares it to the expected age based on the ID’s date of birth—flagging inconsistencies.

      IMAGE METADATA REVIEW

      Analyzes the underlying metadata of the submitted selfie to detect any signs of image manipulation or tampering.

      KNOWN FACED DATABASE MATCH

      Cross-references selfies against a database of known fraudulent faces—flagging celebrities, or stock photos.

      At Vouched we are continuously improving our platform by using human reviews to quality check our AI models, as well as use this information to improve our models. Our biometrics checks comply with ISO/IEC 30107-3 standards.

      Get a Demo