Delhi's Duplicate Image Problem: How the Capital Stacks Up Against Mumbai, London and Seoul

Delhi's land and property records system holds tens of thousands of duplicate scanned images — the same document photographed twice, misfiled under competing survey numbers, or simply uploaded multiple times across disconnected portals. The Delhi government's Revenue Department acknowledged the backlog exists in its 2024-25 annual administrative report, though it did not publish a precise figure for how many records are affected. Civic data researchers who track digitisation efforts across Indian cities say the problem is widespread.

Why does this matter now? The Delhi Metro Rail Corporation is in the middle of Phase 4 expansion work, with corridors cutting through Janakpuri, Tughlaqabad and Aerocity. Acquiring land along those corridors requires clean, deduplicated property records. Disputed or duplicate imagery in the database can freeze compensation payouts, stall construction timelines and drag families into years of tribunal hearings. The pressure to fix the underlying data infrastructure is no longer theoretical.

What Delhi Is — and Isn't — Doing

The Delhi government runs its land records through the Revenue Department's DILRMP portal — the Digital India Land Records Modernisation Programme, a centrally sponsored scheme that launched its current phase in 2020. The portal is meant to serve as a single window for mutation records, jamabandi entries and scanned pattas. Duplicate image entries arise when district offices in places like Mehrauli, Najafgarh and outer Rohini upload documents through both legacy scanning drives and newer mobile-capture workflows, generating two image files for the same underlying record with no automated deduplication check in place.

Mumbai has moved further along. The Maharashtra government integrated an image hash-matching system into its property card portal in late 2023, flagging duplicate scans at the point of upload rather than relying on manual review. Seoul's city administration completed a full deduplication pass of its building permit imagery archive in 2022, reducing storage overhead by roughly 18 percent, according to a Seoul Metropolitan Government technical bulletin published that year. London's Ordnance Survey has used automated duplicate detection in its GeoPlace address database since 2019. Delhi has no equivalent automated layer published or announced as of July 2026.

The closest Delhi has come is a pilot run by the Delhi Urban Shelter Improvement Board, known as DUSIB, which tested optical character recognition on scanned allotment letters for jhuggi-jhopri clusters in areas including Bhalswa and Kalyanpuri in 2023. That pilot was limited in scope and did not include cross-database deduplication of imagery. It has not been publicly expanded since.

The Cost of Inaction

Duplicate imagery in civic databases is not just a storage nuisance. When two image files exist for the same property record, automated matching systems used by banks and home loan processors can throw up verification errors. That means delays for buyers in newer developments around Dwarka Sector 21 and Rohini, neighbourhoods where flat transactions run into the crores and where even a two-week documentation hold carries a real financial cost.

Globally, the problem has accelerated because of digitisation drives undertaken during the COVID-19 years, when governments rushed documents onto servers without quality-control pipelines. New Delhi is not alone — Dhaka, Nairobi and Jakarta all face comparable legacy duplication problems, according to a 2025 assessment by the World Bank's Open Data for Resilience Initiative, which placed Indian state capitals in a middle tier for geospatial data quality.

Officials at the Delhi Secretariat have not announced a dedicated deduplication programme as of this writing. The Revenue Department's digital records team operates out of its ITO headquarters and has been focused primarily on expanding coverage — getting more records online — rather than cleaning what is already there. That priority order may need to shift. Phase 4 land acquisition timelines, combined with growing pressure from the central government's Digital India Mission to improve data quality benchmarks, give the AAP administration a practical reason to act before the problem migrates from an administrative inconvenience into a headline-level infrastructure failure. The window to fix it quietly, before it becomes a political issue, is shrinking.