InsightsFood and grain

Which rice defects can be sorted, and which cannot?

Direct answer

Visible ones. Yellow and off-colour grains, spotted kernels, chalky grains, broken pieces, dark specks, husk fragments and stones are all separable by visible-light inspection. What cannot be sorted is anything that is not visible: appearance does not confirm the cause of discolouration, and no food-safety diagnosis is implied.

What visible inspection reaches

Rice is a small, fast-flowing stream, and most of what a buyer downgrades it for is visible. The classes that can be defined as rejects during a test include:

Yellow grains, outside the agreed recipe boundary.
Off-colour grains, selected during sample comparison.
Spotted or discoloured kernels defined as rejects.
Dark grains and fragments that exceed the selected tolerance.
Chalky grains, selected for removal after sample comparison.
Damaged and broken grains matching the tested visual reject pattern.
Immature grains, defined by visible colour, opacity and shape signals.
Foreign material: stones, husk fragments, plant stems and packaging debris.

The list is a menu, not a promise. Which of these become reject classes for your stream is set on representative accepted and rejected samples for your variety, origin and market grade.

Two panels. On the left, the classes that can be defined as reject classes, each drawn: yellow grain, off-color grain, spotted kernel, dark speck, chalky grain, broken grain, husk fragment and stone. On the right, what cannot be determined from appearance: why the grain is discolored, whether it is safe to eat, and anything not visible on the surface — a magnifier over a discolored grain returns a question mark.

Where the line is

The limit is not a resolution limit. It is that visible-light inspection evaluates appearance, and appearance is not always the property you actually care about.

Two published statements set the boundary precisely. Visible appearance does not confirm the cause of discolouration — a grain can be rejected for looking wrong without that saying why it looks wrong. And only visible damage can be evaluated by visible-light inspection, so nothing is claimed about food-safety diagnosis or invisible contamination detection.

That distinction matters commercially. A mould-like discolouration can be defined as a visual reject class by sample review; it is not a mycotoxin test, and the two should not be confused.

One recipe does not cover every rice

White, brown, parboiled, glutinous and black rice do not present the same problem. Parboiled rice carries a warm yellow or amber tone from processing, so its normal colour sits close to excessive yellowing — the accepted window has to be drawn in a different place than it would be for white rice.

Chalkiness is another case: it may need a separate recipe from ordinary colour sorting rather than being folded into the same decision. Material testing defines what counts as accepted product and what counts as reject for each origin, variety and market grade.

Two color scales running from pale to deeply yellow. For white rice the accepted window sits at the pale end, so the boundary against excessive yellowing is drawn early. For parboiled rice, whose normal color is a warm amber from processing, the accepted window sits far to the right and close to that same boundary.

What shifts detectability

Contrast, stream depth and camera lighting affect how reliably a defect class is seen. A thin, evenly presented stream under tuned lighting is a different problem from a deep, surging one, even with identical rice.

Handling is the other half. Rice breaks, so material handling has to reduce breakage risk — a sorter that creates broken grains while removing them is not a net gain.

Where this stops applying

Visible appearance does not confirm the cause of discolouration. A visual reject class is a visual reject class.
Only visible damage can be evaluated by visible-light inspection. No food-safety diagnosis or invisible-contamination detection is claimed.
Accuracy, purity, throughput, product loss, cost and certification claims are omitted from published material and come from tested results.
Chalkiness may need a separate recipe rather than being handled by ordinary colour sorting.
What counts as accepted product and what counts as reject is defined per rice origin, variety and market grade during testing.

Common follow-ups

Can a rice sorter remove yellow grains?

Yes, as a defined reject class. The yellow threshold is set with representative accepted and rejected samples, because the boundary between acceptable and excessive yellowing depends on the variety and the market grade.

Can chalky grains be sorted?

Yes, after sample comparison. Chalkiness may need a separate recipe from ordinary colour sorting rather than being handled inside the same decision.

Can optical sorting detect mould or contamination in rice?

It can define visible mould-like discolouration as a reject class through sample review. It is not a food-safety diagnosis, and no claim is made about invisible contamination.

Does a rice sorter break grains?

Material handling has to be designed to reduce breakage risk, because rice is fragile. Broken grains are also themselves a reject class that can be defined during testing.

Does one recipe work for all rice?

No. White, brown, parboiled, glutinous and black rice present differently, and parboiled rice in particular has a normal colour close to excessive yellowing. Recipes are defined per origin, variety and market grade.