Geolocation work used to be Google Earth and a lot of patience. In 2026, the toolkit has matured into a stable set of ten sources that, between them, crack most of the photo-geolocation problems an OSINT investigator runs into. Here they are, ranked roughly by how often I open them on a working investigation.
1. Google Earth Pro (still)
Free, desktop, and the historical-imagery slider remains uniquely valuable. If you can narrow the scene to a country or region, the historical-imagery view often pinpoints the time of construction of a building visible in your source photo, which itself dates the photo.
2. Mapillary and KartaView
Crowdsourced street-level imagery, free, and increasingly comprehensive outside Google Street View’s coverage areas. For Eastern Europe, parts of Africa, and rural areas of most countries, Mapillary often has imagery Google doesn’t.
3. Sentinel Hub Playground
Free access to Sentinel-2 imagery with a date slider going back years. For confirming whether a particular event was visible from space on a particular day, fires, military movements, construction, there’s no better free tool.
4. SunCalc
Date and approximate time verification by sun angle. Given the location, the date, and the sun direction visible in the photo (shadows, reflections), SunCalc tells you what time of day it must have been. The reverse, given the location and time, what direction is the sun, narrows compass orientation in seconds.
5. Wikimapia
Crowdsourced annotations on top of Google Maps. Buildings, factories, military bases, and obscure landmarks have descriptions and tags. For verification of “is this building what I think it is,” Wikimapia is often the fastest answer.
6. Yandex Maps and Yandex Images
For Russia, Belarus, Ukraine, and former Soviet states, Yandex provides better street-level coverage and reverse-image-search results than any Western tool. The reverse image search is also useful globally for identifying landmarks and faces, with the ethical caveats that apply to all face-search technology.
7. PeakVisor and PeakFinder
Mountain identification by silhouette. If your source photo has identifiable peaks, PeakVisor’s overlay narrows the camera location quickly. PeakFinder works on iOS and Android with the phone’s camera live.
8. OpenStreetMap with Overpass Turbo
OSM’s data layer is queryable through Overpass Turbo. “Show me every petrol station in this region with this brand” returns results in seconds. For matching identifiable infrastructure (specific brand of cell tower, particular road sign style), OSM beats Google because the data is queryable.
9. Planet Labs (paid, but cheap for journalism)
Daily imagery of the entire Earth at 3-5 metre resolution. Subscriptions are expensive at scale but cheap for a single investigation. When Sentinel’s revisit cycle isn’t fast enough, the event happened on Tuesday and you need imagery from Wednesday, Planet has it.
10. Reverse image search across multiple engines
Google Lens, Yandex Images, TinEye, and Bing Visual Search each return different results. Run all four. The right answer is often in only one of them.
The discipline behind the tools
Tools don’t geolocate photos, investigators do. The discipline that makes the tools work is methodical: identify the most distinctive feature, search for that feature, narrow the region, then triangulate with secondary features. Document each step so the chain of reasoning is reproducible. Skip the documentation step and you’ll publish wrong on the day it matters.