What this does
It turns a photo or drawing into a fuse bead pattern: the picture rebuilt from beads, using a real brand’s colour range, laid out on standard 29 × 29 pegboards. You choose the beads, the size and the most colours you want to buy; the preview shows the result bead by bead, holes and all, with the pegboard edges drawn in. The download is a pattern you can work from: an overview with a map of the boards, a shopping list with every colour’s code and count, and a page for each pegboard.
Everything happens in your browser. The photo is never uploaded.
How to use it
- Drop a JPEG, PNG or WebP onto the tool, or press “Choose an image”.
- Choose your beads: Perler, Hama Midi, Artkal S or Nabbi.
- Set “Beads across”. The height follows the picture, and the line under the controls counts the beads and the pegboards. 29 across is one board; 58 is two.
- Set the most colours you want to buy. Fewer is cheaper and simpler; more is closer to the photo.
- Download the PDF pattern. Print it, put a page under a clear pegboard or beside it, and work board by board.
Choosing the colours, measured
When you cap the colours, something has to decide which beads make the cut. The obvious rule is to keep the ones the picture would use most. This tool instead groups the picture’s colours into as many clusters as you allow and takes the nearest bead to each, which keeps small but distinctive colours, like eyes, that a most-used rule spends on background shades.
We measured both on the 24 standard Kodak test photographs, 58 Perler beads across, judging each bead against the true colour of its square by CIEDE2000, the industry’s standard formula for how different two colours look. Capped at eight colours, the clustering choice was 5% closer to the photographs (10.73 against 11.30) and better on 18 of the 24. Capped at sixteen, the two methods tie, so the choice matters most when you are keeping the colour count low.
The same test explains why bead art is an approximation however many colours you allow: with all 103 Perler colours available, the average difference was still 9.8. Photographs contain colours no bead is made in, so the pattern gets the nearest bead, and the eye forgives it at a distance.
The pattern
The PDF opens with an overview of the whole picture with the pegboards marked, A1, A2, B1 and so on. Next comes the shopping list: each colour’s symbol, code, name and bead count, most used first. Then each pegboard gets its own page, with every square showing its colour and its symbol, gridlines heavier every five so you can count along a row, and coordinates that count across the whole picture, so a bead is findable without counting from a corner. Under each board is what that board alone needs.
Where the bead colours come from
The colour ranges come from the beadcolors project, an open dataset of bead colours sampled by hobbyists from real beads, published under the MIT licence by maxcleme and its contributors. They are good approximations, not manufacturer measurements, and a bead’s colour shifts with the light anyway. Clear, glitter, glow and metallic beads are left out by default, because one colour value cannot describe a bead you can see through or that sparkles; tick the box to allow them.
Limitations
5 mm beads on square pegboards. Mini beads and shaped boards are not covered.
Approximate colours. As above: buy by code, and expect the finished piece to look slightly different from the screen.
Transparency becomes white. Transparent parts of a PNG are treated as white beads.