Split an image into visual cryptographic shares that reveal the secret only when overlaid. Based on Naor-Shamir scheme.
This tool splits an image into two shares of random noise that decrypt by being stacked, not by computation, based on the Naor-Shamir visual cryptography scheme. Every pixel expands into a 2x2 block; white pixels get identical blocks in both shares, black pixels get complementary ones, and the choice is randomized per pixel. Overlay the shares - any black subpixel wins - and the original reappears at half contrast. The surprise: either share alone is meaningless noise, and there is no key to leak or lose.
A concrete input and expected output from the current implementation.
Input
A 2x2 black-and-white image, pixels in row-major order: white, black, black, white (uploaded as an image file).
Expected output
Two 4x4 shares (B = black, W = white subpixel). One valid random run: Share 1: B W B W / W B W B / W B W B / B W B W. Share 2: B W W B / W B B W / B W W B / W B B W. Overlaid (OR, black wins): B W B B / W B B B / B B W B / B B B W - a gray 2x2 block top-left and bottom-right, solid black top-right and bottom-left, reproducing the original.
Each input pixel becomes a 2x2 block, so a 2x2 image yields two 4x4 shares. The example pairs identical diagonal blocks for white pixels and complementary blocks for black pixels, so the OR overlay shows two black subpixels per white pixel (read as gray) and four per black pixel (solid black). Any random choice of diagonals gives an equivalent result.
Every source pixel becomes a 2x2 block in each share, so each share is twice the source's width and height. White pixels are encoded with identical blocks in both shares; black pixels with complementary blocks. Which of the two diagonal patterns is used is chosen randomly, independently per pixel, which is what makes each share look like uniform noise.
Reconstruction is a logical OR: wherever either share has a black subpixel, the stack shows black. Black pixels overlay to four black subpixels; white pixels to two black and two white, which the eye reads as gray. Because pattern choice is uniform and random, a single share is statistically identical whether the secret pixel was black or white - it carries zero information about the image.
With the standard 2-of-2 construction, yes: for every pixel, both diagonal patterns are equally likely in one share regardless of whether the secret pixel was black or white, so the share contains no information about the image - no filter, stretch, or statistical analysis can extract it. That guarantee assumes the tool's randomness is genuinely random and not predictable.
The classic scheme is 2-of-2: exactly two shares, both required. Generalizations exist - k-of-n visual cryptography, where any k of n transparencies reconstruct the image - but they cost pixel expansion: shares grow several times larger and contrast drops further. Whether this tool offers such variants is something to verify before planning a multi-party split.
The example and behavioral notes were checked against the browser implementation. Standards and primary references below define the relevant format, formula, or platform behavior.
Generate strong, random passwords with customizable length, character sets, and complexity requirements.
Analyze password strength with entropy scoring, crack-time estimates, and improvement suggestions.
Create cryptographically random strings for API keys, tokens, salts, and unique identifiers.
Create QR codes for URLs, text, Wi-Fi credentials, vCards, and more with customizable size and colors.
Scan and decode QR codes from uploaded images or your device camera to extract embedded data.
Compute MD5, SHA-1, SHA-256, and SHA-512 checksums to verify file integrity and detect tampering.