Vibe Imaging Analytics
Research
Talk to Our Team
Rice Research

Rice Phenotyping for Breeding and Research

Measure morphology, color, chalkiness, and kernel integrity across lines, trials, and environments. Generate per-kernel images and structured outputs for statistical analysis in R or Python.

Morphology
Color
Chalkiness
Kernel integrity
Trait distributions
Discuss Your Rice ProjectView Rice Publications
Rice preview

Quantitative rice phenotyping for research programs

Breeding and research teams often need more than a single average for an entire sample. Vibe measures individual kernels so researchers can compare complete trait distributions, identify outliers, and quantify variation within and between lines.

A consistent imaging protocol makes results easier to compare across operators, seasons, locations, and experimental treatments.

What can your program investigate?

Compare breeding lines

Quantify differences in kernel morphology, chalkiness, color, integrity, and trait distributions between lines.

Evaluate G×E and treatment effects

Apply a consistent measurement protocol across locations, seasons, replicates, and experimental treatments.

Verify grain phenotype after selection

Evaluate whether selecting for resistance, tolerance, biofortification, or another target trait affected the resulting grain phenotype.

Build reusable research datasets

Export per-kernel images, measurements, classifications, sample summaries, and metadata for statistical analysis or model development.

Measurements, defined

Three milled rice kernels of decreasing length side by side, labelled long grain, medium grain and short grain, each with a length measurement guide
Milled rice kernels compared at long, medium, and short grain lengths.

Kernel morphology

Measure length, width, area, aspect ratio, shape, and size distributions for individual kernels.

  • Length and width
  • Length-to-width ratio
  • Long / medium / short class
Three milled rice kernels showing a chalky, opaque white patch in the center of each kernel against the translucent surrounding endosperm
Milled rice kernels with an opaque chalky patch in the endosperm.

Chalkiness and translucency

Quantify visible chalkiness and translucency using a rice-specific configuration validated for the study material.

  • Per-kernel chalky / non-chalky
  • Configurable area threshold
  • Translucency scoring
Three milled rice kernels of decreasing length, showing the range from a whole kernel down to a small broken fragment
Milled rice kernels ranging from whole to a small broken fragment.

Broken and intact kernels

Separate intact and broken kernels using configured dimensional rules and review the result at the individual-kernel level.

Three milled rice kernels side by side, labelled hard milled, well milled and reasonably well milled, showing progressively more retained bran colour
Milled rice kernels at three milling grades, showing progressively more retained bran.

Color and appearance

Compare per-kernel color values, appearance distributions, and validated visual categories across samples.

  • L*a*b* color values
  • Milling / appearance categories
  • Sample-level color distributions
Three rice kernels side by side; the shorter kernels on either side are labelled expected type, and a longer kernel in the middle, circled, is labelled other type
Rice kernels of the expected grain type alongside a longer, off-type kernel.

Purity and off-types

Quantify visually distinct material and user-defined off-types according to the study's classification criteria.

Three rice kernels side by side, labelled red rice, paddy kernel and damaged kernel, showing red bran, an unhulled hull, and a dark damaged patch respectively
Rice kernels showing red bran, an unhulled paddy kernel, and a heat- or water-damaged kernel.

Trait distributions

Compare complete distributions, sample summaries, variation, and outliers instead of relying only on one average value.

Per-kernel records
Sample summaries
Images
CSV and Parquet
Analysis metadata

From sample to research-ready data

Step 1

Scan the sample

Step 2

Measure and classify traits

Step 3

Compare lines, trials, or treatments

Step 4

Export data, images, and metadata

The workflow preserves the connection between each kernel, its image, its measurements, and its sample metadata.

Compare lines, trials, and locations

Review per-kernel measurements, sample summaries, distributions, and classifications in the Vibe software. Export the results for independent statistical analysis and reproducible figures.

Vibe QM3i Statistical Report screen for a rice sample, showing a grain-type breakdown by size, a colour/defect analysis, a length distribution chart, and general sample informationExample QM3i Statistical Report screen for one rice sample.
  • Per-kernel measurements
  • Sample-level summaries
  • Trait distributions
  • Reviewable kernel images
  • CSV, Parquet, images, and metadata
Illustrative example

Grain-type breakdown for one anonymized rice sample, by size class:

Type% of weight# of kernels% of kernelsMinMax
Long18.1010417.453.003.50
Medium52.6429749.832.003.00
Short3.34254.191.002.00
Broken5.546811.41——
Size Outlier19.289515.94——
Total100.00596100.00——

Published rice research using Vibe

See how research teams have used Vibe measurements in breeding, grain characterization, biofortification, and cultivar comparison.

Marker-assisted forward breeding to develop a drought-, bacterial-leaf-blight-, and blast-resistant rice cultivar preview
International Rice Research Institute (IRRI) / ICAR-Indian Institute of Rice Research

Marker-assisted forward breeding to develop a drought-, bacterial-leaf-blight-, and blast-resistant rice cultivar

What Vibe measured: Grain length and breadth, to confirm grain dimensions were preserved alongside the introgressed resistance and drought-tolerance traits.

Read the publication
Engineering Herbicide-Tolerance Rice Expressing an Acetohydroxyacid Synthase with a Single Amino Acid Deletion preview
Shanghai Academy of Agricultural Sciences

Engineering Herbicide-Tolerance Rice Expressing an Acetohydroxyacid Synthase with a Single Amino Acid Deletion

What Vibe measured: 1000-grain weight, grain width, and grain length, to confirm the mutation did not compromise grain quality.

Read the publication
The acceptance of zinc biofortified rice in Latin America: A consumer sensory study and grain quality characterization preview
Wageningen University & Research / HarvestPlus (CIAT)

The acceptance of zinc biofortified rice in Latin America: A consumer sensory study and grain quality characterization

What Vibe measured: Grain length, length-breadth ratio, and chalkiness of biofortified and control varieties.

Read the publication
Developing Quick Screening Method to Identify Rice Cultivars with Unique Aromatic Features preview
University of Arkansas

Developing Quick Screening Method to Identify Rice Cultivars with Unique Aromatic Features

What Vibe measured: Kernel length, width, length-to-width ratio, and colour on brown rice across 126 genotypes.

Read the publication

Example research workflows

Varietal Purity and Population Composition preview

Varietal Purity and Population Composition

Quantify off-types and population composition in mixed rice samples using image-based segmentation, morphology, and colour classification.

View more rice workflows

Methods and study configuration

Vibe measures visible rice-kernel morphology, color, integrity, and configured image-based characteristics at individual-kernel and sample levels.

Configure the workflow using representative material, consistent sample presentation, and reference categories appropriate to the study.

7 CFR Part 868ISO 7301Codex Standard 198-1995

Frequently asked questions

Yes. Every kernel in a sample keeps its own record — measurements, classification, and image — exportable as CSV or Parquet with sample metadata, ready for R, Python, or another statistics tool.

Yes. The same imaging protocol applies to every sample, so per-kernel distributions from different lines, trials, seasons, or locations can be compared directly rather than only as single averages.

Yes. Kernels are imaged, not consumed or altered, so the same sample remains available for further testing afterward.

Yes. Chalkiness thresholds, size classes, and off-type or purity criteria can be configured to match a study's own definitions instead of a fixed default.

Vibe measures visible appearance, morphology, color, and kernel integrity.

Planning a rice breeding or phenotyping project?

Tell us which traits you need to measure, how many samples you expect to analyze, and how you plan to use the results. We can help you evaluate whether the Vibe workflow fits your study.

Discuss Your Rice ProjectRequest a Sample Analysis