Blur license plates in video, automatically

Blurring a license plate in a video means destroying the pixels that carry the registration number on every frame, then writing a new file — not dragging a rectangle over a still image. A plate that is partially visible, angled, motion-blurred or only in frame for half a second must still be caught and covered, or the redaction is incomplete.

This page explains how automatic plate detection makes that possible, why it outperforms manual editing, which real-world scenarios it covers, and how the result qualifies as irreversible anonymization under GDPR. You can blur license plates in a video right now without an account.

How automatic license plate blurring works

Reliable plate redaction in video is a two-step pipeline:

1. Detection identifies plate regions frame by frame using a model trained on the distinctive shapes, proportions and character patterns of number plates across many countries and formats.
2. Deterministic redaction then replaces those pixel regions — using a fixed, auditable process — and re-encodes the result into a new video file.

Detection alone is not enough. Plates move. A dashcam clip at 30 fps shows a plate appearing for under a second across dozens of frames, with motion blur, angle changes and partial occlusions in between. Tracking bridges those gaps: once a plate is identified, its position is interpolated across frames so the blur stays locked on even when the detector momentarily loses it.

Because the plate regions are re-encoded rather than masked, there is no hidden original layer underneath. That is what makes the output genuinely irreversible, not just visually obscured.

Why automatic beats redacting license plates manually

Manual video redaction in an editor is impractical for anything beyond a short clip:

• A two-minute dashcam clip at 30 fps contains 3,600 frames. Keyframing a blur over a moving plate for all of them is hours of work for two minutes of footage.
• Human editors miss plates that are small, partially visible, reflected in shop windows, or only on screen for a fraction of a second — exactly the instances that can expose a registration.
• Many editor "blur" effects are non-destructive overlays stored in the project file. The underlying pixels are untouched and recoverable. That is not anonymization.

Automatic detection and tracking covers every frame consistently, catches background plates and partial views, and writes a file where the original data is provably gone. A task that would take hours becomes minutes of upload-and-wait.

Real situations where you need to blur license plates in video

Dashcam and driving footage

Dashcam clips captured for insurance, safety reporting or incident documentation routinely capture plates of other vehicles — each one a piece of personal data. Before sharing clips with third parties, insurers or on social media, every plate must be removed. Pair plate redaction with face blurring to cover drivers and pedestrians in the same job.

Street-level and mapping video

Organizations collecting video at street level — for urban planning, delivery route surveys or mapping — capture thousands of plates per hour of footage. Manual redaction at that volume is not viable; automated batch processing is the only workable approach.

CCTV and security footage

Car parks, forecourts and building entrances record plates continuously. When that footage needs to be shared for evidence, disclosed under a data access request, or handed to an analyst outside the organization, plates must be redacted before it leaves your control. The same clip may also contain faces — a single Medianonymizer job can handle both.

Don't forget faces and audio

License plates are only one source of identity in driving or street footage. A clip may also show faces of pedestrians and drivers and carry audio with spoken names, addresses or phone numbers — all personal data. Treating the plates alone leaves the rest of the footage out of scope. Medianonymizer can blur plates and faces and mute sensitive audio segments in a single job, so the complete file is anonymized, not just the plates.

Blur license plates in your video now

Upload a video, choose license plate redaction (and add faces, screens or audio if needed), see the exact price before you pay, and download the anonymized file. No account, no subscription, pay only per job.