How to match the colours of two photos
You have two photos: one whose colours you love, and one that needs to look like it. Maybe they're from different cameras, different days, or one is simply a look you want to borrow. Here are the three approaches that actually exist, and how to tell whether a match worked.
Route 1: match by eye
Open both images side by side and adjust white balance, tone curve and saturation until they agree. This works, and it teaches you to see colour — but it has two structural problems. First, human colour perception adapts within seconds: after staring at a warm image, your eyes recalibrate and "matched" drifts warm. Second, the result isn't reusable — next pair, you start again.
Route 2: histogram matching
The classic automatic method: reshape each colour channel of your photo so its distribution matches the reference's. Done naively in RGB it often works for tone but produces hue errors, because the channels are adjusted independently — pushing red without regard to what green and blue are doing shifts hues sideways. It also matches distributions, not content: if the reference is half sky and yours isn't, the match is skewed by composition.
Route 3: fitted transform in a perceptual space
The refinement that fixes most of Route 2's problems: convert both images into a perceptual colour space (one where lightness and colour are separate, and equal distances look equally different), match tone there, and fit the chroma shift as a joint operation rather than per-channel — including how the colour shift varies with brightness, which is what split-toned looks are made of. Because the transform is fitted between your specific pair, it can also be exported (as a LUT or preset) and reused on similar material.
How to judge a match
- Check neutrals first: greys and whites reveal casts immediately. If the neutrals agree, the match is mostly there.
- Check skin if there is any — the hue band viewers are most sensitive to.
- Look at the extremes: deep shadows and near-highlights are where automatic matches most often clip or crossover.
- Measure it if you can: a colour-difference metric like ΔE2000 puts a number on the remaining gap instead of trusting adapted eyes.
Tonemara does all of this in your browser in one click: drop a reference
image and your own photo, preview the match free, and export a 33×33×33 .cube
LUT plus a Lightroom-compatible .xmp preset. Your images never leave your device.