Document fraud checker
Review an ID as a physical object and a digital file.
Upload one image for universal forensic analysis, or up to four related images such as an ID front, ID reverse, selfie, and proof photo. Relevant cross-file checks run automatically without forcing a special mode.
Scene routing and rectification
OpenCV locates a likely document, estimates its geometry, and creates a normalized view for structural checks while preserving the original report.
Independent text extraction
Tesseract and PaddleOCR read printed structure separately. Normalized fields are compared with keyed digests rather than storing raw document text for cross-image checks.
MRZ and barcode consistency
When a supported machine-readable zone or barcode is visible, checksums and normalized fields can be compared with the rest of the document evidence.
Portrait and cross-image checks
If a document portrait and a selfie are detected, face similarity and field consistency are presented as research evidence with clear quality limits.
No automated detector can prove authenticity. AI Screener combines independent signals, shows the evidence, and leaves the decision with a human.
Run the free checkerQuestions to ask of the result
Limits matter as much as the signals.
Can this verify a government ID?
No. It reports structural and forensic evidence to support human review; it does not connect to an issuer database or make a legal identity decision.
Do I need to label the front, back, or selfie?
No. The pipeline routes each image automatically and only runs face and document consistency checks when compatible evidence is present.