Why AFIS Is Being Replaced by 'Crime Kundli' — India's Next-Generation Criminal Identification System
From single-biometric fingerprint matching to a seven-parameter digital criminal profile — here's the full story of India's most significant forensic technology transition.
📋 In This Report
- What Is AFIS / NAFIS? A Brief History
- The Limitations That Triggered Change
- What Is Crime Kundli?
- Legal Foundation: The CPI Act 2022
- The Seven Biometric Markers of Crime Kundli
- AFIS vs Crime Kundli — Detailed Comparison Table
- FRS Integration & Real-Time Surveillance
- Timeline of India's Forensic ID Evolution
- Privacy Concerns & Legal Challenges
- Expert Verdict: Is Crime Kundli the Future?
1. What Is AFIS / NAFIS? A Brief History
The Automated Fingerprint Identification System (AFIS) has been the cornerstone of forensic identification in law enforcement for over two decades. At its core, AFIS is a computerized system that stores, analyzes, and compares fingerprint data — automating what was once a painstaking, weeks-long manual process into a task of minutes.
In India, the fingerprint identification journey began with FACTS (Fingerprint Analysis and Criminal Tracing System), co-developed by the National Crime Records Bureau (NCRB) and CMC. The latest version — FACTS 5.0 — used Image Processing and Pattern Recognition to capture, encode, store, and match fingerprints, including latent prints from crime scenes. It stored pattern class, core and delta information, minutiae location, ridge counts, and demographic data like gender and conviction details.
However, a 2018 NCRB report concluded that FACTS 5.0, last upgraded in 2007, had effectively "outlived its shelf life." This paved the way for NAFIS — the National Automated Fingerprint Identification System — launched by the NCRB as a centralized, 24/7 national repository. NAFIS assigns every arrested individual a unique 10-digit National Fingerprint Number (NFN) linked to all FIRs across their lifetime.
🔍 NAFIS By the Numbers (As of October 2024)
- 1.06 crore criminal fingerprint records in the national database
- AFIS operational at NCRB HQ and 22 state headquarters
- 11 states still yet to install the system at the time of AMBIS planning
- First identification via NAFIS: Madhya Pradesh, April 2022 (deceased person)
- Connected to Fingerprint ID Systems of all States and Union Territories
AFIS operates through three main steps: image enhancement (using Gabor filters to sharpen ridge-valley structures), feature extraction (pulling minutiae points — ridge endings, bifurcations, spatial patterns), and candidate matching (generating a ranked list of probable matches with similarity scores). Trained examiners then verify any positive match before it is acted upon.
2. The Limitations That Triggered Change
Despite its contributions, AFIS in India — particularly state-level deployments — accumulated a list of significant shortcomings that law enforcement agencies could no longer ignore:
⚠️ Documented Shortcomings of Legacy AFIS (Per NCRB/AMBIS Analysis)
- Outdated technology: Most state AFIS systems used proprietary encoding and matching algorithms that lacked commonality.
- No interoperability: No interstate or inter-AFIS connectivity — data portability was absent even between different versions of the same vendor's product.
- Single modality: AFIS only matched fingerprints — it could not incorporate face, iris, voice, or other biometric data.
- Fake identity vulnerability: Criminals could change names, addresses, and documents — fingerprints alone could not always conclusively identify someone under a false alias in real-time field operations.
- No mobile integration: Most AFIS platforms were desktop-only, unavailable to field officers on patrol.
- Siloed databases: State systems rarely "talked" to each other, enabling inter-state criminals to evade detection.
- Latent print challenge: Smudged, partial, or degraded crime scene prints returned poor match rates on outdated systems.
In short, AFIS was a single-instrument orchestra in an era demanding a full forensic symphony. The rise of repeat offenders using multiple identities across state borders made the single-biometric approach increasingly inadequate.
3. What Is Crime Kundli?
In June 2025, the Delhi Police officially launched "Crime Kundli" — a comprehensive biometric profiling system being deployed across police stations. The name is deeply evocative: just as an astrological kundli (horoscope) captures every planetary aspect of a person's existence, Crime Kundli builds an exhaustive digital identity for every arrested or convicted individual.
"Crime Kundli is not just a fingerprint database. It is a complete biometric biography of a criminal — designed to ensure there is no escape through aliases, disguises, or forged documents."
— Delhi Police Official Statement, June 2025Rather than querying a single biometric trait, Crime Kundli cross-references multiple data points simultaneously — making false identification exponentially harder. The system is integrated with Delhi Police's broader digital infrastructure, including its Facial Recognition System (FRS) and the national CCTNS (Crime and Criminal Tracking Network & Systems) framework.
Technically, Crime Kundli falls under the broader category of AMBIS — Automated Multi-Modal Biometrics Identification System — the NCRB's planned upgrade to replace siloed AFIS installations with an interconnected, multi-biometric national platform.
4. Legal Foundation: The Criminal Procedure (Identification) Act, 2022
Crime Kundli is not merely a technological update — it is legally mandated. The Criminal Procedure (Identification) Act, 2022 (CPIA 2022), which came into force on 4 August 2022, replaced the 102-year-old Identification of Prisoners Act of 1920 and gave law enforcement sweeping new powers to collect biometric data.
Finger-impressions, palm-print impressions, footprint impressions, photographs, iris and retina scans, physical and biological samples and their analysis, and behavioural attributes including signatures, handwriting, and voice samples — from any person convicted, arrested, or detained under preventive detention law.
The Act empowers the NCRB to store, preserve, share with any law enforcement agency, and destroy records at the national level. At the state level, each state designates its own appropriate agency. Crucially, CPIA 2022 mandated the NCRB to develop Standard Operating Procedures specifying equipment and devices for biometric data collection — directly enabling the design and rollout of Crime Kundli.
5. The Seven Biometric Markers of Crime Kundli
This is the most significant departure from AFIS. While AFIS captured only one biometric modality (fingerprints), Crime Kundli captures seven distinct biometric markers per individual, creating what forensic experts call a "biometric constellation" that is virtually impossible to spoof:
Fingerprints remain the anchor — the NAFIS-assigned NFN continues to serve as the unique identifier. Facial recognition connects to live CCTV feeds and the FRS database, enabling real-time alerts. Iris scans are the most accurate individual biometric identifier with near-zero error rates. Voice patterns add a behavioral dimension useful in telephonic crime investigations. Footprints revive an older forensic tradition with digital precision. Handwriting samples aid in questioned document cases. And tattoo records — a highly underutilized forensic tool — allow identification of unknown bodies and disguised individuals who may have undergone facial alteration.
6. AFIS vs Crime Kundli — Detailed Comparison
| Parameter | AFIS / NAFIS | Crime Kundli (AMBIS) |
|---|---|---|
| Biometric Modality | Single — Fingerprints only (10-print & latent) | Multi-modal — 7 biometric parameters (fingerprint, face, iris, voice, footprint, handwriting, tattoo) |
| Legal Basis | Identification of Prisoners Act, 1920 (colonial-era) | Criminal Procedure (Identification) Act, 2022 — post-independence, comprehensive |
| Database Scope | State-level AFIS + NAFIS national repository (fingerprints only) | Integrated national AMBIS database covering all 7 biometrics with full state-to-national portability |
| Interoperability | Limited; state systems use proprietary algorithms; poor interstate data sharing | Full interstate and inter-agency interoperability; common encoding standards |
| Real-Time Field Use | Primarily desktop-based; limited mobile deployment (M-CCTNS pilots in Karnataka) | Designed for field deployment; integrated with mobile FRS units (DP-DRISTHI vehicle in Delhi) |
| Fake Identity Detection | Moderate — fingerprints can be bypassed if prints are not on record | High — cross-referencing 7 independent biometrics makes aliases nearly impossible |
| Surveillance Integration | Standalone; not connected to live CCTV or facial recognition cameras | Integrated with FRS, CCTV networks, CCTNS — enables real-time alerts and live-feed matching |
| Unknown Body / Victim ID | Limited to fingerprint match against database | Multi-point ID — face, tattoo, footprint, iris can identify unregistered individuals |
| Matching Speed | Fast for fingerprint (seconds to minutes) — varies by system age | Face match: 1 million records in 200 ms; overall multi-modal match remains fast with AI acceleration |
| Data Retention Authority | NCRB + State Fingerprint Bureaux | NCRB (national level) + State-designated agencies; indefinite retention with NCRB oversight |
| Technology Generation | Legacy — many states still on 2000s-era platforms; FACTS 5.0 "outlived shelf life" per NCRB 2018 | Current — AI-based pattern recognition, neural network matching, cloud-connected architecture |
| Criminal Profiling Depth | Biometric ID only; links to case files via CCTNS manually | Automated cross-linking of biometric profile to criminal history, FIRs, modus operandi, and case data |
| Privacy Scrutiny | Moderate — fingerprints considered relatively non-invasive | High — DNA, iris, biological samples raise significant privacy and fundamental rights concerns |
| Current Coverage | 22 state HQs with AFIS; NAFIS holds 1.06 crore records (Oct 2024) | Pilot deployment — Delhi Police leading; expansion to other states in progress |
7. FRS Integration & Real-Time Surveillance — The Game-Changer
One of the most operationally significant aspects of Crime Kundli is its seamless integration with Delhi Police's Facial Recognition System (FRS). The FRS infrastructure can process footage from over 100 cameras simultaneously, performing live-camera face matches against one million records in just 200 milliseconds.
This was dramatically demonstrated when, between September and November 2024, North Delhi Police used FRS technology to arrest 70 individuals accused of theft and snatching — matching suspects against the criminal database in real time. The system operates from a dedicated facility in Civil Lines, managed by a five-officer team.
In February 2026, Delhi Police went further with the launch of DP-DRISTHI (Delhi Police Dynamic Real-time Identification and Surveillance for Human Tracking Initiative) — a mobile surveillance unit equipped with four cameras that automatically compares faces against the database of known offenders while on patrol. This is the street-level embodiment of Crime Kundli in action.
8. Timeline: India's Forensic Identification Evolution
India establishes the world's first systematic fingerprint classification system in Calcutta — the Henry-Galton system. India literally invented modern fingerprint science.
Identification of Prisoners Act enacted — provides legal sanction for collecting fingerprints and photographs of convicts and arrested persons.
National Crime Records Bureau (NCRB) established, absorbing the Central Fingerprint Bureau and becoming the national custodian of criminal identification data.
FACTS 5.0 (India's AFIS) last upgraded — subsequently described by NCRB in 2018 as having "outlived its shelf life."
CCTNS (Crime and Criminal Tracking Network & Systems) project launched, eventually connecting 15,000+ police stations nationwide.
Criminal Procedure (Identification) Act 2022 passed by Parliament, replacing the 1920 Act and legally enabling multi-biometric data collection at scale.
CPIA 2022 comes into force. NCRB mandated to create SOPs for biometric data collection equipment.
Madhya Pradesh becomes first state to use NAFIS for identifying a deceased individual.
NAFIS crosses 1.06 crore criminal fingerprint records. AFIS operational at NCRB HQ and 22 state HQs.
Delhi Police officially launches Crime Kundli — a seven-biometric comprehensive criminal profiling system — the first major implementation of AMBIS principles in India.
Delhi Police rolls out DP-DRISTHI mobile FRS surveillance unit in Shahdara, bringing real-time Crime Kundli-linked identification to street patrol operations.
9. Privacy Concerns & Legal Challenges
The transition from AFIS to Crime Kundli is not without controversy. The Internet Freedom Foundation has called for a moratorium on biometric facial recognition systems in India, citing concerns over accuracy bias, data security, and the absence of robust legal safeguards governing their use.
In a landmark constitutional challenge, students Sahibe Alam and Saurabh Tripathi filed a writ petition in Delhi High Court (WP Crl 672/2026) challenging the CPIA 2022 after their biometrics were collected by police at a peaceful protest at Jamia Millia Islamia — even without any FIR against them. The case raises critical questions:
⚖️ Key Constitutional Questions Raised
- Right to Privacy (Article 21): Is mass biometric collection proportionate and necessary under the Puttaswamy judgment standard?
- Self-incrimination (Article 20(3)): Do voice samples and biological specimens cross the line into communicative evidence?
- Presumption of innocence: A biometric database of all persons who "came into contact" with the criminal justice system risks treating suspects as criminals.
- Data security: The 2024 Telangana Police TSCOP app data breach demonstrated the vulnerability of centralized biometric repositories.
- Scope creep: CPIA 2022 allows biometric collection from protesters and non-convicted individuals, raising mass surveillance concerns.
The Supreme Court's ruling in State of Bombay vs Kathi Kalu Oghad has been cited in defense of the Act — affirming that "non-communicative" evidence (fingerprints, photographs) does not violate Article 20(3). However, experts argue that voice samples and biological specimens may cross that threshold.
10. Expert Verdict: Is Crime Kundli the Future of Forensic ID in India?
From a purely forensic science perspective, Crime Kundli represents a quantum leap in India's criminal identification capability. The seven-biometric model addresses every documented failure of single-modality AFIS — especially the fake identity problem, inter-state criminal mobility, and the inability to identify suspects from partial crime scene evidence alone.
The integration with FRS and CCTV networks transforms identification from a reactive process (matching a suspect after arrest) to a proactive one (flagging a known criminal the moment they appear on a camera). This is the standard toward which advanced policing systems globally — including the FBI's Next Generation Identification (NGI) system — have been moving.
"India invented fingerprint science in 1897. With Crime Kundli, it may yet lead the world in multi-biometric criminal identification — provided the legal and ethical guardrails keep pace with the technology."
— Budding Forensic Expert Editorial Analysis, 2025For forensic science students and practitioners, the message is clear: the future of criminal identification in India is multi-modal, AI-accelerated, and real-time. AFIS will not disappear overnight — NAFIS continues to expand — but it is increasingly the foundation beneath a much larger architecture, not the structure itself.
Mastering the principles of all seven biometric modalities, understanding the CPIA 2022 legal framework, and staying current with AMBIS developments is no longer optional for the next generation of Indian forensic experts. It is essential.
📚 Sources & References
- NCR Guide — "Delhi Police Launches Crime Kundli to Crack Down on Repeat Offenders and False Identities" (June 7, 2025): ncr-guide.com
- National Crime Records Bureau — NAFIS (Wikipedia / NCRB official): wikipedia.org/wiki/National_Crime_Records_Bureau
- GS SCORE — "NAFIS and the History of Fingerprinting in India": iasscore.in
- SIFS India — "Automated Fingerprint Identification System": sifs.in
- Biometric Update — "Indian Police Deploy Portable Fingerprint Biometric Scanners" (Sept 2022): biometricupdate.com
- Biometric Update — "Indian Police Adopt Facial Recognition Despite Risk of Massive Data Breaches" (May 2024): biometricupdate.com
- ID Tech Wire — "Indian Police Expand AI Facial Recognition System, Arrest 70 Suspects in Delhi" (Dec 2024): idtechwire.com
- Drishti IAS — "Criminal Procedure (Identification) Act 2022": drishtiias.com
- Internet Freedom Foundation — "Delhi HC Issues Notice on CPIA 2022 Constitutional Challenge" (March 2026): internetfreedom.in
- Internet Freedom Foundation — "Delhi's Safe City Project and the Expansion of AI-Enabled Surveillance" (Feb 2026): internetfreedom.in
- Outlook India — "Delhi Police Unveils DP-DRISTHI Surveillance Vehicle with FRS" (Feb 2026): outlookindia.com
- NCRB / SlideShare — "AMBIS: Automated Multi-Modal Biometrics Identification System": slideshare.net/slideshow/ambis
- Ministry of Home Affairs — Criminal Procedure (Identification) Act, 2022 Gazette: mha.gov.in (PDF)
- PRS India — "Criminal Procedure (Identification) Rules, 2022": prsindia.org
- IASbaba — "Rules for Identifying Criminals / CPIR 2022": iasbaba.com
