Post

OSINT EP05: Images, Geolocation & Metadata

Posted
By Karan Chaudhary
2 min read
OSINT EP05: Images, Geolocation & Metadata

They say a picture is worth a thousand words. In OSINT, a picture is worth a thousand data points. This episode focuses on IMINT (Imagery Intelligence) and GEOINT (Geospatial Intelligence).

1. Reverse Image Search (RIS)

Don’t just look at the image; look for where else it exists.

Different search engines have different “brains.” You must use the right one for the job.

  • Google Images: The jack-of-all-trades. Good for identifying general contexts, but often censors facial search.
  • Yandex Images: The heavy hitter. Unbelievably good at face matching, background landmark detection, and OCR (reading text on billboards in the background).
  • Bing Visual Search: Excellent for Object Detection. If you need to identify a specific brand of sunglasses or a car model, use Bing.
  • TinEye: The “Exact Match” king. Perfect for seeing if an image is original or a stock photo/repost.

Browser Extensions:

  • Search By Image: Adds a context menu to right-click any image and search across all these engines simultaneously.

2. Facial Recognition OSINT

Finding people, not just pictures.

  • Search4Faces: Specialized for VK/Russian databases, but surprisingly effective globally.
  • PimEyes: The most powerful commercial face search. Paid, but highly effective.
  • FaceCheck.id: A strong alternative.

Pro Tip: Bypassing Paywalls/Blur Sometimes services like FaceCheck show a match but blur the source.

  1. Inspect Element (F12) on the blurred image.
  2. Look for the image source URL. It might be a Base64 string (data:image/webp:base64...).
  3. Copy the string and use a Base64 decoder. Sometimes the source URL is hidden within the metadata of that string.

3. Metadata (EXIF Data)

Every digital photo has a “fingerprint” unless it has been scrubbed.

  • What to look for: GPS Coordinates, Device Model (iPhone 14 Pro), Date/Time original.
  • Note: Social media (FB, Insta, Twitter) strips this data. You need original files (uploaded to blogs, forums, or sent via document mode).
  • Tools:

4. Geolocation Tools

Where in the world is this?

AI Assistance:

  • GeoSpy.ai: An AI model trained to guess locations based on vegetation, architecture, and lighting.

Mapping Powerhouses:

  • Google Maps: Street view is essential. Don’t forget the “Time Machine” feature in Street View to see how a building changed over years.
  • Yandex Maps: Excellent for Eastern Europe/Russia. Has unique “Pedestrian” and “heatmap” views.
  • Mappls: The best detailed maps for India.
  • DualMaps: Splits your screen into Street View, Aerial View, and Map View simultaneously.

Street Level Alternatives: If Google didn’t drive there, someone else might have.

5. Manual Geolocation Tips

  1. Look for text: Store signs, billboards, license plates. Translate them immediately.
  2. The Sun: Check shadows to determine cardinal direction (North/South).
  3. Cross-Reference: If the target’s friend tagged them in a coffee shop, but the target didn’t check in… verify the coffee shop’s interior photos on Google Maps Reviews.
  4. Gamify Learning: Play GeoGuessr to train your brain to recognize soil, bollards, and electric poles.
OSINT, IMINT, GEOINT
This post is licensed under CC BY 4.0 by the author.