# Street-Level and Map Intelligence

> Techniques for extracting geolocation data from street-level imagery and maps, covering metadata analysis, landmark identification, and spatial correlation methods.

- **URL:** https://hackindex.io/platforms/osint/geolocation-and-imagery-intelligence/street-level-and-map-intelligence
- **Path:** OSINT > Geolocation & Imagery Intelligence
- **Up:** [Geolocation & Imagery Intelligence](https://hackindex.io/platforms/osint/geolocation-and-imagery-intelligence.md)
- **Published:** 2026-08-10
- **Updated:** 2026-08-10

## Overview

Street-level imagery is the final confirmation step for a ground photo. Matching the target image against a panorama from the same viewpoint proves the location beyond the circumstantial visual clues. Coverage varies enormously by country, so knowing which platform covers a given region is half the task.

## Platform Coverage

- Google Street View: the broadest global coverage, with a historical slider showing past captures of the same spot.
- Mapillary: open, crowd-sourced imagery with an API, often covering places Street View does not, including inside private and rural areas.
- KartaView: another open crowd-sourced platform, a useful second source where Mapillary is sparse.
- Yandex Panoramas: the strongest coverage across Russia and much of the former Soviet space.
- Bing Streetside, Baidu Total View in China, and Kakao and Naver in South Korea: regional platforms that fill major gaps in Google coverage.

## Querying Mapillary by Location

Mapillary exposes an API, so you can pull every street image within a bounding box around a coordinate rather than browsing manually. This is useful for coverage that is not in the web map or for scripting a sweep of an area.

```bash
curl -s "https://graph.mapillary.com/images?access_token=$MAPILLARY_TOKEN&bbox=$MIN_LON,$MIN_LAT,$MAX_LON,$MAX_LAT&fields=id,captured_at,computed_geometry,thumb_2048_url&limit=20" | jq '.data[] | {id, captured_at, geom: .computed_geometry.coordinates}'
```

Each result carries its own coordinate and capture time, so you can pick the image closest to your suspected viewpoint and pull its full-resolution frame from the returned URL.

## Matching the Viewpoint

Confirmation is geometric. Line up fixed, slow-changing features between the target photo and the panorama:

- Building edges, storey counts, and roofline shape.
- Permanent street furniture: lamp posts, bollards, hydrants, post boxes.
- Road layout, kerb lines, and the angle between streets at a junction.

Ignore transient elements such as parked cars, foliage, and advertising, which change between captures. Use the historical slider to find a panorama dated close to the target photo when a feature has since changed.

## When There Is No Coverage

Rural areas, closed campuses, and many countries have no street-level imagery at all. Fall back to geotagged photos on mapping and travel platforms, tourist images on photo-sharing sites, and user-contributed pictures attached to map points of interest. These give ground-level views where no panorama exists.

## References

- [Mapillary API Documentation](https://www.mapillary.com/developer/api-documentation) - Query crowd-sourced street imagery by bounding box and pull image frames.
- [Google Street View](https://www.google.com/streetview/) - Broadest global street-level coverage with a historical timeline.
- [KartaView](https://kartaview.org/) - Open crowd-sourced street imagery, useful as a second source.

## Steps

- [OpenStreetMap and Overpass Queries](https://hackindex.io/platforms/osint/geolocation-and-imagery-intelligence/street-level-and-map-intelligence/openstreetmap-and-overpass.md): Leverage OpenStreetMap data and Overpass API queries to extract geospatial intelligence, identify infrastructure, and geolocate targets using open map datasets.
- [Street-Level Imagery Platforms](https://hackindex.io/platforms/osint/geolocation-and-imagery-intelligence/street-level-and-map-intelligence/street-level-imagery-platforms.md): Leverage street-level imagery platforms like Google Street View and Mapillary for OSINT geolocation, landmark identification, and target environment reconnaissance.