SAAF

Sensing and Accounting of Aquatic Floatables

Request a demo
A beach-cleaning tractor and crew seen straight down on a city beach at sunrise

Impact

A numberyou can act on.

Interventions get planned without baseline data. A survey gives them one: how much waste, what kind, where it gathers and when.

100%
Coverage, not samplingA transect walk samples a strip and extrapolates. A drone pass records every item in the flight area.
~1 s
Per 4K frameA morning's flying is processed the same day.
6
Classes itemisedCounts broken out by material, so glass and plastic are budgeted separately.

What you get back

Four things, every survey.

Nothing is summarised away. The number, the frame it came from, and the frame kept for next time.

An aerial frame with detected litter boxed by the model
01

Annotated frames

Every detection drawn on the original imagery, so any number can be traced back to the pixel it came from.

A plastic bottle on sand with dimension callouts drawn over it
02

Per-class totals

Count, footprint area and estimated mass, broken out by material. Glass and plastic are budgeted separately.

A survey grid laid out on white sand beside clear shallows
03

Survey totals

The same figures for the whole flight area, in one table you can hand to a committee.

Volunteers collecting litter along a mangrove riverbank, seen from above
04

The imagery itself

Kept, so a later survey of the same stretch is directly comparable to this one.

Repeatability

The same method, flown again, is the point.

A single survey says how much waste is on a shore today. The same method flown again at other tides and in other seasons says how it moves, and whether anything you did worked, including work far upstream. Because the pipeline is fixed, same model, same scale calculation, same reference weights, two surveys of the same stretch differ only by what actually changed on the ground.

Survey 1 · before cleanupillustrative
Survey 2 · after cleanupillustrative
Survey 3 · next seasonillustrative
Shapes only. Your numbers come from your imagery, in kilograms per class.

Corrections feed back

Where it is wrong, a person fixes the box.

The correction feeds back into the model, which improves on the imagery it is actually being used on rather than on a benchmark that resembles it.

  1. 01Fly
  2. 02Detect
  3. 03Correct
  4. 04Improve

What this is not

It is a measurement instrument, not a cleanup service.

It tells you how much waste is reaching the water, what kind it is, where it gathers and roughly what it weighs. NGOs plan clean-ups with it, municipalities target collection, researchers study leakage, donors see verified impact. Removing the waste is still work for people with bags.

Bring usa waterway.

We process a sample survey and show you the output before you commit to anything.

Request a demo