Remove EXIF Data from Photos with Termux

Every photo you take with your smartphone contains invisible data embedded directly in the image. GPS coordinates accurate to a few meters, device model, operating system version, date and time of capture. The technical term: EXIF metadata.

Anyone who shares a photo without removing this data also shares their location, device and the time of capture. This applies to messengers, email, Nostr, and private servers. Platforms with automatic EXIF stripping are the exception, not the rule.

This script cleans images in the Downloads folder on the device before they are sent anywhere.

1. What EXIF Data Contains

Typical fields in EXIF metadata of a camera photo:

• GPS coordinates (latitude, longitude, sometimes altitude)
• Device manufacturer and model (e.g. Google Pixel 9 Pro)
• Operating system version
• Date and time of capture
• Camera settings (aperture, exposure time, ISO)

Android screenshots generally do not contain GPS coordinates. AI-generated images from ComfyUI or Stable Diffusion carry no EXIF data at all. The main risk is with photos taken directly by the camera app.

2. Setup: Python and Termux

If not done yet, set up access to device storage first:

termux-setup-storage

Confirm the dialog ("Allow"). Then install Python and the Pillow library:

pkg update && pkg upgrade

pkg install python

pip install pillow

Create a scripts folder:

mkdir -p ~/Skripte

3. The Script: exif-cleaner.py

The script reads images from the Downloads folder, creates pixel-accurate copies without embedded metadata, and saves them to a bereinigt/ subfolder. The original files remain untouched.

Create the file:

nano ~/Skripte/exif-cleaner.py

Paste the content:

#!/usr/bin/env python3
"""
exif-cleaner.py - Removes all metadata (EXIF) from images (Termux version)
Cleaned images are saved in a 'bereinigt/' subfolder.
"""

import os
import sys
from pathlib import Path
from PIL import Image

SUPPORTED = {'.jpg', '.jpeg', '.png', '.webp', '.bmp', '.tiff', '.tif'}

def clean_image(src_path, dst_path):
    with Image.open(src_path) as img:
        data = img.tobytes()
        clean = Image.frombytes(img.mode, img.size, data)
        clean.save(dst_path)

def main():
    print("Alien EXIF-Cleaner (Termux)")
    print("-" * 40)

    base = Path.home() / "storage" / "downloads"
    folder = input(f"Subfolder in {base}/ (empty = entire Downloads folder): ").strip()
    src_dir = base / folder if folder else base

    if not src_dir.exists() or not src_dir.is_dir():
        print(f"Folder not found: {src_dir}")
        sys.exit(1)

    images = [f for f in src_dir.iterdir() if f.suffix.lower() in SUPPORTED]

    if not images:
        print(f"No images found in: {src_dir}")
        sys.exit(1)

    dst_dir = src_dir / "bereinigt"
    dst_dir.mkdir(exist_ok=True)

    print(f"\nSource:  {src_dir}")
    print(f"Target:  {dst_dir}")
    print(f"Images:  {len(images)} found\n")

    ok = 0
    errors = 0

    for img_path in sorted(images):
        dst = dst_dir / img_path.name
        try:
            clean_image(img_path, dst)
            print(f"  ok  {img_path.name}")
            ok += 1
        except Exception as e:
            print(f"  ERR {img_path.name} - Error: {e}")
            errors += 1

    print("\n" + "-" * 40)
    print(f"{ok} images cleaned -> {dst_dir}")
    if errors:
        print(f"{errors} errors")

if __name__ == "__main__":
    main()

(Save in nano: CTRL+O, Enter, then CTRL+X)

4. Running the Script

python ~/Skripte/exif-cleaner.py

The script asks for a subfolder within ~/storage/downloads/. Press Enter to process all images in the entire Downloads folder.

Cleaned copies are saved to:

~/storage/downloads/bereinigt/

Supported formats: JPG, PNG, WebP, BMP, TIFF.

5. How the Cleaning Works

The approach is deliberately straightforward: Pillow reads the raw pixel data (img.tobytes()) and builds a new image object from it (Image.frombytes()). This new object contains no EXIF fields, no ICC profiles, no embedded thumbnails. Only the pixel data itself.

To verify that cleaning worked, use exiftool on a desktop:

exiftool cleaned-photo.jpg

Expected output: few or no fields, no GPS.

No invisible data. No location. No device. That is the goal.