Camera-read signs and spoken routes for blind riders inside metro stations
TL;DR Stepbeacon guides blind and low-vision riders from a metro station's gate to the right platform and exit. Printed codes on walls and pillars are read by a phone camera from several metres away, and the app speaks out each turn. Metro corporations pay for setup and upkeep, and riders use it free.
A blind rider can cross a road with a cane and a known route, but a big interchange station is a maze of gates, stairs, lifts and platforms that change with each line. And GPS doesn't work underground.
Blindness in India runs at about 2.31% among women and 1.67% among men, and in a Delhi hospital study about 89% of visually impaired people owned a smartphone, yet only 40.2% of those used any specialised app[1].
So the phone is already in the pocket. What's missing is something in the station for it to read.
Stepbeacon has two parts. The first is a set of printed, high-contrast codes fixed at gates, stairs, lifts, platform edges and exits. A phone camera picks them up from several metres away while the rider is walking, with no need to aim precisely. The second is a station map with routes between every pair of points, kept current by the metro's own staff through a simple editor.
The rider opens the app, says "Platform 2 towards the airport", and moves the phone slowly. When a code comes into view the app speaks the distance and direction to the next turn, and it vibrates when the rider is facing the right way. Lift outage announcements show up on the route automatically.
It works for elderly riders and first-time visitors too, which helps the metro justify the spend. No cameras are installed in the station, and the app sends no images off the phone.
Riders use it free. The payer is the metro rail corporation, which has duties under disability law to make stations accessible. Airports and big railway stations are a second set of buyers. I found no public tender price, so my assumption is ₹8–20 lakh to survey and equip a large station, plus an annual maintenance fee of 15–20% of that.
The Rights of Persons with Disabilities Act 2016 requires accessible transport and stations[2], and the government's 2021 draft harmonised accessibility standards set out what public buildings must provide[3]. Phone cameras and on-device recognition are now fast enough to read small codes at walking pace, and one camera-code vendor has already shown this in New York and Barcelona transit[4].
There's no public figure for what Indian metros spend on wayfinding for blind riders. The demand signal is indirect - 89% smartphone ownership among visually impaired people in one Delhi study, against low use of specialised apps[1].
Count stations bottom-up, starting with the busiest interchanges in two cities.
One Spanish vendor sells camera-read colour codes to transit operators in New York, Barcelona and Madrid, with apps in 42 languages, but lists no Indian deployment[4]. Another indoor-navigation app uses a phone camera to locate riders indoors without beacons, for UK and US rail clients[5]. A volunteer video app connects over a million blind users with sighted volunteers through video calls[6], which works but needs a volunteer every time.
The gap is a local operator that surveys Indian stations, works in Indian languages and maintains the codes.
| Difficulty | Moderate (50/100) |
| Starting capital | ₹1–2 Cr |
| Time to launch | 9–12 months |
Reading the codes is the easy part. Selling to a public-sector metro takes a year or more, through pilots and tenders. Keeping maps right as stations change, and getting blind riders to test and trust the routes, both take steady operational work.
No licence is needed to run the app itself. What matters is the accessibility law that makes metros buy and the way they buy.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
AI should never invent a route. The station map, maintained by metro staff, is the source of truth, and if a code isn't recognised the app should say so plainly rather than guess.
Want to work out the AI layer for Stepbeacon? Talk to Yash →
Here's what can go south -
I think this is a worthwhile business if you've sold to metros or public bodies and can wait out tenders. If you won't spend weekends walking stations with blind testers then it is not for you. Before raising, prove that ten blind riders reach the right platform unaided in one busy station, faster than they do today.
My estimate is ₹1–2 Cr to launch, covering the app, station surveys, printed codes and a long sales cycle with metros.
No special licence applies. You need GST, tender registration with the metro and DPDP compliance for location data.
Yes, and foreign vendors already sell this model to transit operators abroad. Check the local disability law, because that's usually what gives the operator a reason to pay.
The free Stepbeacon validation kit has the line-by-line capital breakdown, the effort scoring, the accessibility-law and procurement checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for riders and metro officials, and 5 assumptions to test. Download the kit →
Thinking about building Stepbeacon or something close to it? The make-or-break work is landing the first metro pilot and then getting through a public tender, which is slow going. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.