Skip to content
SIH Buddyby Ganeev Singh
Dev
All problem statements
SIH26188Worth consideringacceptance 3/5

Al-Based Fake Identity & Document Screening System

Ministry of Home Affairs · Miscellaneous · Software

Excellently specified with the tampering module as the clear differentiator, but you build your own tampered examples so the detector only catches what you made — invest in the forensic detection, lean on deterministic MRZ validation, and be honest that real forgeries are subtler than synthetic ones.

What it actually is

Border checkpoints process thousands of passports and visas by hand, and cannot reliably catch sophisticated forgeries, altered photos, tampered stamps or people using multiple identities. The ask is an AI platform that reads a travel document, checks its fields against rules and databases, detects tampering, verifies the face, and produces a risk score to help officers decide faster.

What to build

A document-screening platform with four modules: OCR extraction pulling the relevant fields from passports, visas, national IDs and permits; document validation checking those fields against official format and rule standards; a tampering-detection module — the core AI — flagging photo replacement, text manipulation and stamp forgery through image-forensics and metadata analysis; and face verification matching the document photo to the bearer, combining into a risk score that assists an officer rather than deciding automatically.

Smallest thing that wins the room

Scan a genuine specimen document and a tampered one, and show the platform extracting the fields, flagging the tampered photo and altered expiry date through image-forensic cues, verifying the face match, and producing a higher risk score for the tampered document.

How crowded this one gets

A guess, projected from the 2025 statements — the last year where both the submission counts and the winners were published.

Moderate150–340 teams expectedroughly 1 in 125–289 wins it

Quieter than 33% of the 226 · #152 of 226 by expected field

A normal-sized field. Your idea has to be good, not miraculous.

Why: defence, intelligence and space bodies drew small fields.

This is a guess, not a fact

Nobody has published 2026’s numbers yet. This is an analysed estimate from last year’s pattern, so please do not take it as the truth — check the live counter on the SIH portal before you decide anything. The range covers the middle half of likely outcomes, so one statement in two lands outside it. Entry closes at 500 ideas per statement, so no range goes past that — a statement that reaches the cap fills and shuts rather than drawing an unlimited crowd. The model reads only three things a team can see before choosing — software or hardware, the theme, and what kind of body posted it — and those explain about a quarter of the variation in last year’s field sizes (R² 0.25 on held-out statements). Trust the band more than the number, and the ordering more than either. It cannot see how good your idea is, which is the part that actually decides it.

The scores

The number is the shorthand. The line under it is the reason.

What you will be writing

  • OCR / MRZ extraction (Tesseract, PassportEye)
  • Image-forensic tampering detection (ELA, noise, copy-move)
  • Document format / MRZ checksum validation
  • Face verification (ArcFace)
  • Metadata analysis
  • Risk-score fusion
  • Document forensics
  • Identity verification
  • Border security

Prior art to read before you start

travel document tampering detection · identity document OCR and validation · face-to-document verification

Analysed by Claude Opus. Every score above is a judgment call with its reasoning attached — kindly cross-check this against the official statement on the SIH portal before your team commits to it.