Image Color Vision Simulator
Review how color-coded areas in a chart, screenshot or illustration change under three fixed color vision model simulations. Choose an image, compare the source and result side by side, and download a PNG with the same pixel dimensions. The preview is a design reference, not a diagnosis or a reproduction of an individual’s vision.
Key features
- Apply the model to every image pixel rather than just two selected colors
- Compare fixed protanopia, deuteranopia and tritanopia simulations with an unmodified reference mode
- Use Machado (2009) severity-100 matrices in linear sRGB with explicit gamut clipping
- Retain source dimensions and alpha in the PNG result
- See processing progress, cancel work and regenerate after changing the mode
How to use
- Choose a static PNG, JPEG or WebP image, or load the colored example.
- Select a simulation mode. All three model modes use the fixed severity-100 endpoint.
- Create the preview and compare color-coded regions with the original image.
- Switch modes and generate again to inspect another model.
- Download the current result as a PNG and keep labels or patterns alongside meaningful colors in your design.
Use cases
- Review neighboring series in a chart before sharing it
- Compare color-coded regions in a user interface screenshot
- Inspect a map or infographic for color-dependent distinctions
- Share a model preview during a design review without uploading the source image
Frequently asked questions
Can this diagnose color blindness or show exactly what someone sees?
No. It applies a mathematical model to image pixels. Individual vision, viewing conditions and displays vary. Use the output as a design reference; it is neither a medical test nor a guarantee of any person’s visual experience.
Which model and severity does it use?
The three modes use the fixed severity-100 matrix tables published with Machado, Oliveira and Fernandes (2009), checked against the open-source Colorspacious data. There is no severity slider or interpolated strength. Original mode applies no color simulation.
Why does the color result differ from a simple RGB filter?
The tool decodes sRGB channels into linear values, multiplies the selected matrix, clips channels to the displayable 0–1 range and re-encodes sRGB. Browser decoding and color management also affect input pixels. This is not a contrast score or a conformance check.
What happens to transparency and image size?
The transform leaves alpha unchanged and exports a PNG at the decoded source width and height. The checkerboard is only a transparency guide. For transparent artwork, the eventual background affects appearance; compare a flattened screenshot too when that background matters.
Are original bytes, metadata or wide-gamut profiles preserved?
No. The browser decodes the image into an sRGB canvas and creates a new PNG. Original file bytes, EXIF and embedded color profiles are not retained. Even Original mode is a browser-decoded PNG export, not a byte-identical copy.
Why is the download cleared when I change modes?
A result belongs to its source image and selected mode. Changing either cancels stale work and clears the old download. If PNG encoding is already in progress, its result is discarded and the next job waits until that operation releases its resources.
Privacy
Selected images and simulation settings are processed in the current page’s browser memory. This tool does not upload those images or settings and does not automatically store them. Choosing download saves the resulting PNG to your device.
Comments & questions