What is Mosaic and Blur?
Mosaic and blur are techniques for intentionally obscuring part of an image. Mosaic pixelates the area into blocks; blur smoothly mixes neighboring pixels. They are widely used to hide faces in social posts, conceal license plates, or mask personal information in documents — keeping shared images privacy-safe.
When to Use
Use them when posting photos to social media to obscure passers-by or license plates, when sharing screenshots that contain emails or phone numbers, or when posting real-estate photos with neighboring nameplates. Because everything happens in your browser, images are never sent to an external server.
Practical Limits and What This Tool Cannot Do
Mosaic destroys the original pixel data irreversibly, so you cannot recover the original from a saved image — which is also a security benefit. Block size matters: too large and the subject becomes unrecognizable; too small and AI super-resolution may reverse it. Gaussian blur has the same caveat — too small a radius and the outline is still guessable. To prevent text from being read, use a mosaic block size at least twice the character height.
Trade-offs to Keep in Mind
Browser-side mosaic/blur via the Canvas API can take several seconds on very high-resolution images (over 4000 px). Gaussian blur is especially CPU-intensive due to per-pixel convolution. WebGL/WebGPU could enable GPU acceleration, but this tool prioritizes compatibility with the 2D Canvas API. Selection precision also depends on manual dragging, so pixel-perfect masking has limits.
Real-world Q&A
Q: I want to be sure a license plate is hidden. Use a larger mosaic block size (30 px+) and cover the entire plate with some margin. Edges can leak hints, so process a slightly larger area than the target. Q: Mosaic or blur — which should I use? Mosaic is more conventional for official documents and business use. For casual social posts, Gaussian blur often looks more natural.
Trends in Privacy Protection
Higher-resolution smartphone cameras and the rise of social media have increased demand for photo privacy tools. Google Street View and Apple Maps already use automatic face/plate detection plus blur. Since 2023, AI-based "face anonymization" (replacing faces with synthetic ones) and differential-privacy-based image anonymization have been active research areas. Client-side tools like this one represent the original principle — never send images to a server in the first place.