Soybean Quality Evaluation Using Machine Vision

This study presents a proof-of-concept method for identifying materials other than grain (MOG) during soybean harvest. Images of grain samples were taken using front-lighting and back-lighting techniques, and various image processing techniques were used to identify different dockage fractions, namely split, contaminated, and defective beans and stems/pods. The image background was segmented using the HSI color space model, followed by the identification of the dockage fractions using median filtering, morphological operations, watershed transformation, and component labeling based on projective area and circularity. Split, contaminated, and defective beans and stems/pods were accurately identified by the algorithms with 96%, 75%, 98%, and 98% accuracy, respectively.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-14
DOI
https://doi.org/10.5281/zenodo.24217265
Primary Topic
Smart Agriculture and AI
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Soybean Quality Evaluation Using Machine Vision

Samruddhi Dhanaji Ilage, Siddhi Santosh Mirajkar, Dr. Sapna Shrama
Zenodo (CERN European Organization for Nuclear Research)
Smart Agriculture and AI
article

Soybean Quality Evaluation Using Machine Vision

Samruddhi Dhanaji Ilage, Siddhi Santosh Mirajkar, Dr. Sapna Shrama
article en

Abstract

This study presents a proof-of-concept method for identifying materials other than grain (MOG) during soybean harvest. Images of grain samples were taken using front-lighting and back-lighting techniques, and various image processing techniques were used to identify different dockage fractions, namely split, contaminated, and defective beans and stems/pods. The image background was segmented using the HSI color space model, followed by the identification of the dockage fractions using median filtering, morphological operations, watershed transformation, and component labeling based on projective area and circularity. Split, contaminated, and defective beans and stems/pods were accurately identified by the algorithms with 96%, 75%, 98%, and 98% accuracy, respectively.

Zenodo (CERN European Organization for Nuclear Research)
Smart Agriculture and AI
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon