A phone camera can see a lot about a wheat sample. It cannot see what makes bread rise. We think a grading tool should be as clear about its blind spots as about its results.
Where GrainGrade stands: GrainGrade is in development. Counting is built and measured; guided capture, the report, defect models and lot trends are on the roadmap, and figures other than the counting error are design targets, not results.
Ask a baker what they want from wheat and they will talk about flour that absorbs water well, dough that stretches without tearing and loaves that rise. Ask a mill buyer at the gate and they will talk about broken grains, foreign matter and moisture. Both are describing wheat quality. Only part of it is visible in a photograph.
A top-down photo of a single layer of grains, with a scale marker in frame, can show a great deal. GrainGrade is being built to count and measure every grain in millimetres and to classify seven kinds of defect:
Those are the defects that FCI FAQ limits are written around, and they are the ones that cause most value cut disputes at the gate.
The product page states it plainly: the camera cannot see moisture, protein, gluten, aroma or insects hidden inside a grain. That list matters more than it looks.
| Property | Why the camera misses it | What to use instead |
|---|---|---|
| Moisture | Water content does not show on the surface | A moisture meter; GrainGrade takes the reading as an entry |
| Protein and gluten | These are inside the grain and decide baking quality | A laboratory test |
| Aroma | A smell, not a shape or colour | A trained person’s check |
| Hidden insects | An infestation without an outside hole looks like sound grain | Lab methods or physical inspection |
For a baker, protein and gluten are close to the whole point. A sample can grade perfectly on every visible defect and still make disappointing bread. A photo grade tells you the lot is clean and sound; it does not tell you it will bake well.
It would be easy to leave the limits in a help page and let the grade speak for itself. We think that is the wrong call. A report that shows a confident FAQ grade without mentioning moisture invites people to treat it as a complete assessment. So the design is that moisture is entered from a meter reading, because FCI grading needs it, and every report says plainly what the camera cannot see.
This is the same principle as refusing a bad photo. A tool that admits what it cannot do is easier to trust on what it can do.
Once guided capture and the report are built, we see photo grading as the fast, consistent first check, not the only one:
For the full list of parameters and how each is tested, read wheat quality parameters: what is checked and why. For the intake process at the gate, see wheat intake at the mill gate. More is on the GrainGrade guides page. GrainGrade is open for pilot conversations while the remaining phases are built.
No. Moisture is not visible on the surface. GrainGrade is designed to take moisture from a meter reading, because FCI grading needs it, and every report says so.
Not necessarily. Protein and gluten decide baking quality and are inside the grain, so they need a laboratory test.
GrainGrade is in development. Counting is built and measured; guided capture, the report and defect models are being built in phases, and a pilot is open.
See it on your own data. GrainGrade — Wheat graded from a photo. Book a 30-minute working session with an engineer.
