Delhi's Duplicate Image Crisis: The Key Decisions That Will Shape the City's Visual Identity

Delhi's municipal and heritage archives are sitting on tens of thousands of duplicate digital images — redundant photographs clogging systems maintained by the Delhi Urban Shelter Improvement Board, the Archaeological Survey of India's northern circle office, and multiple civic portals run by the Municipal Corporation of Delhi. The problem has quietly ballooned over three years of accelerated digitisation drives, and the decisions made in the next six months will determine whether the city's visual public record becomes a usable civic asset or a sprawling archive of noise.

The stakes are higher right now because Phase 4 of the Delhi Metro expansion — covering corridors from Janakpuri West to Krishna Park Extension, and from Tughlaqabad to Aerocity — has triggered a fresh wave of site documentation. Every station box, every heritage-impact survey, every resettlement colony near Dwarka Sector 21 and Saket has generated parallel photo sets uploaded by contractors, government engineers, and third-party consultants, often without any deduplication protocol in place. When photographs are mislabelled or duplicated across databases, land records get muddied, compensation disputes become harder to adjudicate, and heritage protection orders can be challenged on evidentiary grounds.

Where the Bottleneck Is Forming

The MCD's Geographic Information System cell, operating out of its headquarters on S.P. Mukherjee Civic Centre near Connaught Place, currently runs no automated deduplication layer on its image ingest pipeline, according to publicly available tender documents from late 2025. The Delhi Heritage Conservation Committee, which reports to the Urban Development Department at the state secretariat in ITO, has flagged in its annual review that duplicate survey photographs of sites in Mehrauli, Nizamuddin, and the Walled City of Shahjahanabad have complicated at least a dozen file reviews over the past two years. Neither body has yet published a remediation timeline.

The problem is compounded by vendor fragmentation. The city's digitisation work has been split across at least four separate empanelled agencies since 2022, each using different file-naming conventions and metadata standards. When images of, say, a Lodi Garden boundary wall or a lane in Chandni Chowk are uploaded by two vendors on the same project, the duplicate is rarely flagged because the file names differ even when the pixels are nearly identical. Perceptual hashing — a standard deduplication technique used by platforms including Google Photos — is not mandated in any current MCD or DUSIB procurement specification.

What the Next Six Months Must Deliver

Three decisions are unavoidable. First, the MCD needs to amend its image-upload tender specifications before the next round of Metro Phase 4 documentation contracts are awarded, expected in the third quarter of 2026. Without a hash-based deduplication requirement written into the contract, the same problem will replicate itself across the Janakpuri and Tughlaqabad corridors. Second, the Delhi Heritage Conservation Committee must publish a unified metadata standard — something it has been drafting, per its own published minutes from March 2026, but has not yet finalised. Third, a single nodal officer needs to be designated across MCD, DUSIB, and the state Urban Development Department to own the cross-agency deduplication mandate. Right now responsibility is diffuse enough that no single official can be held accountable for inaction.

Practical pressure is building from below, too. Residents near Turkman Gate and around Ballimaran in Old Delhi have used right-to-information requests to access photographic records tied to heritage notifications and development plans. When those records contain dozens of near-identical images filed under different reference numbers, it slows the RTI response process and creates ambiguity about which image is the authoritative record. Civil society groups working on the Yamuna floodplain documentation — a politically charged subject given the ongoing cleanup dispute between the Kejriwal administration and central government agencies — have run into the same wall.

The technical fix is not expensive. Perceptual hash libraries are open-source. The real cost is administrative will: agreeing on standards, rewriting procurement language, and appointing someone to own the outcome. Delhi has done harder things faster when the political incentive was clear. The question now is whether anyone in the ITO secretariat or at Civic Centre decides that a clean visual archive is worth the bureaucratic friction of making it happen.