Delhi's public digital archives contain tens of thousands of duplicate and misattributed images — photographs of the same stretch of Chandni Chowk filed under three different categories, aerial shots of Connaught Place duplicated across at least four municipal portals, heritage photographs of the Red Fort stored in conflicting formats across the Archaeological Survey of India's servers and the Delhi Urban Heritage Foundation's own repository. The problem has been building for years, but a push in early 2026 to digitise civic records ahead of the Delhi Metro Phase 4 corridor documentation drive has finally forced the issue into the open.
The stakes are more practical than they might appear. When planners, journalists, architects and court litigants pull images from government databases to verify land use, building conditions or heritage status, duplicate and mislabelled files produce errors that travel downstream into reports, legal filings and policy decisions. A 2024 audit of civic data quality commissioned under the Smart Cities Mission — a programme that included Delhi as one of its original 20 selected cities — found that image duplication rates in municipal repositories across Indian metros averaged above 30 percent, a figure that urban data specialists have described as unusually high by international standards. Delhi's own duplication rate in that audit was not separately published, but the capital's archive size — among the largest of any Indian city — makes the problem proportionally significant.
What Other Cities Have Done
London's approach is instructive. The Greater London Authority consolidated its public image assets under a single content management system linked to the London Datastore platform, completing a deduplication pass in 2022 that reduced its civic photo archive by roughly 28 percent. Seoul, which manages one of the most aggressively maintained open-data portals among Asian capitals, deployed a perceptual hashing system across the Seoul Metropolitan Government's digital asset library beginning in 2021; city officials credited that programme with cutting redundant files by nearly a third within 18 months. Mumbai, the most direct Indian comparator, launched a deduplication initiative through the Brihanmumbai Municipal Corporation's IT cell in late 2024, targeting records tied to its Development Plan 2034 land-use maps.
Delhi has no equivalent consolidated programme yet. The Delhi government's Department of Information Technology has been developing a unified data governance framework under its Digital Delhi initiative, but image deduplication has not been a stated priority. The Archaeological Survey of India, which maintains custody of photographic records for monuments including Humayun's Tomb, Qutb Minar and Purana Qila, operates its archives largely independently of the Delhi government's own systems, creating a structural gap that no single body currently bridges.
On the Ground in Old Delhi
The friction is visible in practical work. Researchers using the Delhi State Archives facility on Shyam Nath Marg report pulling multiple conflicting image records for the same heritage property — sometimes with different dates, sometimes with incompatible metadata. The problem is particularly acute for Old Delhi neighbourhoods, where rapid physical change over recent decades means that a duplicated image filed under the wrong year can misrepresent current conditions entirely.
The National Informatics Centre, which provides the technical backbone for a range of central and state government digital services, has tools capable of running deduplication protocols at scale, but deployment depends on individual departments commissioning the work. That commissioning has been slow, in part because image archive maintenance sits low on budget priority lists. Delhi's 2025-26 state budget allocated funds for broader e-governance upgrades, but no specific line item for archive deduplication was made public in documents reviewed for this report.
The Metro Phase 4 documentation push may provide the forcing function the issue needs. The Delhi Metro Rail Corporation is building out six new corridors, and its project-documentation requirements — which include systematic photographic records of affected properties along routes including the Janakpuri West to RK Ashram Marg stretch — will generate tens of thousands of new images that will need to be properly filed from the start. How the DMRC's contractors handle image tagging and storage could either add to the existing duplication problem or demonstrate a cleaner model that other agencies then adopt. The corporation's IT procurement for Phase 4 documentation is expected to be finalised before the end of 2026.