alterBiota
Back to blog

Image-based aggregate gradation on any device

How smartphones and tablets turn aggregate QC into a five-minute task, and how digitalΔggregate converts photos into sieve-equivalent gradation plus shape and texture data.

alterBiota · July 16, 2026

A conventional sieve analysis is accurate, but it's slow, and for most producers it only happens occasionally, not often enough to catch drift between tests. What if the same gradation insight came from a photo you could take at the stockpile, the lab bench, or the site, and came back in minutes?

That's the shift image-based gradation makes.

From a photo to a gradation curve

The workflow with digitalΔggregate is deliberately simple:

  1. Capture: photograph the aggregate with a phone or tablet, using a scale reference chip so the system knows the true particle size.
  2. Upload: images go to the cloud through the mobile app.
  3. Analyze: computer vision segments every particle and measures it.
  4. Result: gradation and shape data post back to a web app, typically in under five minutes.

No oven, no nested sieves, no re-weighing, and it can be done by anyone on the team, not just a lab technician. Results are also traceable: the source photos stay with the analysis, so you can look back at them whenever a number is questioned. Standard sieving has to trust that someone wrote down the right value—errors can never be checked after the fact.

<5 min

From photo to result

Any device

Phone or tablet capture

SievEQ

Sieve-equivalent gradation

Cloud

Processed and posted to web app

What the vision system measures

Computer vision sees each particle individually, so it captures far more than mass on a screen. For every particle it can derive geometry that describes size and shape:

  • Feret Min / Feret Max: the smallest and largest caliper widths
  • Aspect ratio: Feret Max ÷ Feret Min — how stretched the particle is relative to an equidimensional shape
  • Circularity: how close the outline is to a circle
  • Roughness: surface texture, a property that has not been measured any other way

From the size distribution it builds SievEQ, a sieve equivalent that acts as a predicted proxy for a conventional sieve analysis (in the spirit of ASTM C136 / EN 933-1). Flat-and-elongated ratios can also be assessed, directionally similar to ASTM D4791.

The takeaway

A sieve gives you one thing: mass retained by size. An image gives you size, shape, and texture, from the same five-minute capture.

Why "any device" is the point

Because the capture is just a series of photos, the barrier to testing drops to near zero. A pile can be checked on the way past. A grab sample can be run at the bench or tailgate. A questionable delivery can be documented on the spot. When testing is this easy, it happens far more often, and frequency is what actually catches drift.

Validation against physical sieves has shown SievEQ tracking conventional results closely across a range of customer aggregates, so the speed doesn't come at the cost of trust in the number.

Where it fits

Image-based gradation doesn't replace your standards knowledge; it removes the friction that keeps you from measuring often. Pair it with the plant-side story in inline vs mobile imaging, and see why frequency matters in aggregate variability.

See it on your own aggregate

Send us a sample or book a demo and we'll show gradation and shape results from a phone photo in minutes.