Tile Matchmaker

Vein matching, wall layout and cut plans

Sources 0
Cells 0
Tiles 0
Patterns 0
Tile vein matching

Lay out veined tiles before anything is set.

Photograph the tiles, set the arrow on each, press Run. Every tile is compared with every other and the whole wall is laid out for the best overall match: veins that carry on, quiet areas that flow together. You swap, lock and re-solve, then print the sticker sheet.

A real wall, solved

Ryan's Convent Glen back wall: 21 porcelain tiles with marble veins, cut from a photo of the finished wall.

Play: Beat the Installed Wall

The same 21 tiles, shuffled. Swap them until the veins flow, then compare your wall with the wall as it was installed. Tiles saved to the tile archive become new walls to play.

▶ Play

Your job, four steps

  1. Photos: lay the tiles out face up and photograph them (Sources), then mark the tiles (Perspective).
  2. Arrows: set which way the arrow on the back of each tile points (Wall match).
  3. Run: the app reads the veins and lays out the whole wall; swap, lock and re-solve until it reads right.
  4. Print: sticker sheet, inventory sheet and cut list for the install.

No fake certainty

Every number comes from the photo's pixels or your measurements, and carries its grade. Joints the photos say little about are marked so you check them by eye.

Sources

Add photos of your tiles, then tap Find Tiles. Best: one slab per shot from above, or a layout with a gap between tiles.

No photo yet. Add one with Upload Image, Camera or Photos.

Calibrate

Set the real scale from a scan card, paper or known object in the photo. Without it every size is an estimate.

Folder

The job's working folder: project info, finished wall sizes and the cut calculator. The wall sizes feed the layout planner.

Find the tiles in your photo

Press Find Tiles. Fix any outline that is wrong, then Accept Cells to add the tiles.

Perspective Builder Canvas

Inventory

Accepted detections become durable tile records with source, scan label, evidence grade, and crop data.

Wall match

Add your tiles, set the arrows and press Run: every tile is scored against every other, and the whole wall is laid out for the best overall match. Swap, turn and lock tiles, then re-solve.

Vein Analysis

Real outer-edge signatures, ring-profile evidence, vein vectors, and the current Frangi-like ridge analysis are available.

Outer perimeter evidence is sampled from pixels. The current inner-band profile is an estimated fallback until calibrated one-inch inward sampling is implemented.

Matching

Physical rules, orientation policy, perimeter scoring, inner-band scoring, growth consistency, and human-readable explanations are wired into the current engine.

Matching never overrides hard physical or manufacturer constraints. Inner-band confidence remains limited until measured one-inch evidence replaces the estimated fallback.

Layout

Wall planning, placement labels, balanced cut planning, sliver warnings, and wall-flow scoring are active. Special geometry such as Alcove is being recovered into the shared engine.

Final label grammar: B 1^↑3

Sandi Logic

How the Sandi tile matching system thinks, in plain words.

1. The core problem

A layout of porcelain tiles with marble veins is not a normal image-matching problem. The computer is not just looking for similar pixels. It has to understand that every tile is a physical object, every vein is a directional structure, and every edge may either connect beautifully to another tile or break the flow completely.

Add to that the brutal reality of tile work: every tile is unique, every wall is a slightly different size, every cut piece changes the math, and once you start cutting tiles you cannot undo it. So matching has to happen before the saw runs.

2. The creative breakthrough: the Sandi Decimal System

Instead of treating each tile as a blob, Sandi encodes the four edges of each tile as a directional event stream — position along the edge, angle of the vein crossing the edge, thickness, and strength. Each event becomes a decimal number. A tile becomes a short string of decimal events. That string is the tile's fingerprint.

3. Why this matters

Two tiles whose veins continue smoothly across their shared edge will produce mirror-image edge events. A connection that looks beautiful is also mathematically scoreable.

4. The matching equation

if (
  positionError  <= tolerance.position &&
  angleError     <= tolerance.angle &&
  thicknessError <= tolerance.thickness &&
  strengthError  <= tolerance.strength
) {
  connectionValid = true;
}

The weighted score gives position and angle the most importance:

score =
  0.45 * positionScore +
  0.30 * angleScore +
  0.15 * thicknessScore +
  0.10 * strengthScore

5. The struggle of teaching this to a computer

Computer vision wants to see pixels. Sandi forces it to see structure. The hard part was making the system stop pattern-matching colours and start tracking vectors. v15→v19 was about this fight. v18 introduced cut-aware vein vectors. v30.3 restored the projection pipeline cleanly.

6. The physical workflow

  1. Lay out all candidate full tiles face-up.
  2. Photograph each tile at known scale (use a Tile Match Maker scan card).
  3. App extracts perimeter-band signature + inner-1-inch vein vectors.
  4. App proposes wall layouts that minimize sliver cuts and bad corner wraps.
  5. Cut tile, mount, photograph the cut piece. App records the cut fragment as its own entity, linked to its parent tile.

7. Full tile vs. cut fragment

This is the single most important distinction in v18+. A finished-wall extraction may document only the visible cut tile. It should not automatically become a new full tile identity. Perspective equality and visible fraction must support it before promoting.

8. Database-ready structure

Every tile produces a record with: ID, source image, perimeter signature, vein vector list, perspective equality score, visible fraction, parent tile (if cut), role (hero/connector/buffer/terminal/reject), and the QR-Q payload. This is what the upcoming memory layer will store.

9. Main beta goal

Match leftover cut pieces to their parent tiles, score wall layouts, and produce QR-Q payloads that travel with the tile from inventory → wall → photo record. Beta is judged on whether contractors can trust the outputs.

10. Rule: no fake certainty

Tile Match Maker never invents matches or measurements. Every output carries an evidence grade (L0 → L4). If the system is unsure, it asks. If you press it to commit without enough evidence, it refuses. This is the charter that separates this from every other vision toy.

11. The complete decision loop

capture → calibrate → project → extract veins
   ↓
score perimeter signature
   ↓
suggest matches (with confidence)
   ↓
show continuation line
   ↓
user accepts, changes, or rejects
   ↓
save as training data

Sandi / QR-Q

Tile evidence can be encoded through Sandi, translated into QR-Q star-injector payloads, rendered as a star map, and exported.

No QR-Q payload yet.

Pattern Library

Local pattern memory, import/export, conflict handling, and training-event storage are wired. Cloud sync remains an adapter boundary, not a claimed live service.

Local-first pattern memory makes approved tile evidence reusable without requiring cloud connectivity.

Export / Import

Project save/load, install-plan generation, sticker sheets, print view, and pattern import/export modules are loaded.

Required outputs remain: project JSON, pattern library JSON, Sandi signatures, QR-Q payloads, and install plan.

Settings

Project defaults, measurement grades, wall dimensions, tile size, and storage preferences.

Toolbox