Single-view planar metrology · Android · Free

Measure with your phone.
One photo. Real numbers.

Photograph a flat surface with a red-white reference stick. Mark all four corners of the object and all four corners of the stick. Metrologist returns width, height, area, diagonal, and angles. No internet needed. Ever.

Three steps from photo to measurement

STEP 01

Photograph the surface

Place your reference stick with red ends and a checker border flat against the wall, door, or window. Take a clear photo with even lighting and a moderate viewing angle.

STEP 02

Mark corners + stick

Drag the green box around your object's four corners and the red box around the reference stick. Pinch-zoom and the built-in magnifier let you land precisely on each point. Enter the stick's length and width in the app settings.

STEP 03

Read the results

Width, height, area, diagonal, and corner angles, with a confidence score. No server call. No account. Works entirely offline.

Built for people who care about accuracy

Real-world units

Results in centimetres, inches, or metres. Set once, never touched again.

Works offline

All computation happens on-device. No internet, no cloud, no account required.

Confidence score

Every result has a confidence score. You know immediately whether to trust the number or retake the photo.

Walls, doors, windows

Any flat planar surface. The geometry engine handles the perspective math; you just tap.

Pinch-zoom magnifier

A built-in magnifier for precise corner placement, even on small details in the photo.

100% unit-tested engine

The Kotlin measurement engine is unit-tested end to end. Results you can rely on, not just hope for.

How close does it get?

±1–5%
Typical error range
92% conf.

Single-view planar metrology works from one photograph. No laser, no tape. Error typically falls between 1 and 5%, depending on photo quality, perspective angle, and how carefully you place the reference stick. For most DIY and trade tasks (cutting materials, ordering tiles, quoting jobs) that's more than enough. The on-screen confidence score tells you when a reading is solid and when it's worth retaking the photo.

Free, transparent, and open

Apache-2.0 License

Metrologist is open-source. The measurement engine, the UI, the tests: all on GitHub. Use it, fork it, learn from it. Contributions welcome.

Built by Cocode · Created by Babak Bandpey · View on GitHub →