A new approach in sorting of mixed crops by acoustics

The agricultural cultivation of mixed crops, often a mixture of legumes and non-legumes, has become significantly more important in recent years in both organic and non-organic farming. In practice, optical sorting is currently often used for sorting batches in the post-harvest. These systems work reliably and quickly. However, the optical sensors are very sensitive to environmental influences and comparatively expensive. As a result, these systems are generally only used on large farms, where they are stationary and in a clean environment. For the above-mentioned reasons, it is hardly conceivable for optical sorters to be used by smaller producers or for inter-company use. Against this background, the aim of the studies presented here was to investigate the potential of acoustic methods for sorting different legumes and grain in the post-harvest. In a laboratory experiment, the sound of the impact of grains of two legumes, pea and field bean, as well as different cereals, rye, wheat, barley and oats has been investigated. The grains were dropped in a precisely defined height and time sequence from a conveyor belt onto a softly suspended impact plate of defined geometry and density. The corresponding noises were recorded with structure-borne sound sensors attached to the plate. The data were analyzed in both the time and frequency range. Features from the time signals and the information from the spectra were used as input parameters for various statistical analyses, such as discriminant analysis, nominal statistical regression and bootstrap forest. For the validation set, a generalized R² of 0.94 was achieved with a mis-classification-rate of 0.18. The results show that the recognition rates were lower for barley, rye and wheat, where field bean, pea and oat are detected with an accuracy of 100%. Ultimately, with regard to the sorting of mixtures commonly used in practice, consisting of a legume and a cereal, the results show for all possible combination of the two legumes and the four cereals a consistent separation of 100%.

Authors

Publication Details

Journal
Universitätsbibliothek Kassel
Published
2026-02-24
DOI
https://doi.org/10.17170/kobra-2026031011991
Primary Topic
Spectroscopy and Chemometric Analyses
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

A new approach in sorting of mixed crops by acoustics

Hubertus Siebald, Boris Kulig, Simon Deuchert, Marius Reith et al.
Universitätsbibliothek Kassel
Spectroscopy and Chemometric Analyses
article

A new approach in sorting of mixed crops by acoustics

Hubertus Siebald, Boris Kulig, Simon Deuchert, Marius Reith, Torben Schmaus, Ino Winterhoff, Oliver Hensel
article en

Abstract

The agricultural cultivation of mixed crops, often a mixture of legumes and non-legumes, has become significantly more important in recent years in both organic and non-organic farming. In practice, optical sorting is currently often used for sorting batches in the post-harvest. These systems work reliably and quickly. However, the optical sensors are very sensitive to environmental influences and comparatively expensive. As a result, these systems are generally only used on large farms, where they are stationary and in a clean environment. For the above-mentioned reasons, it is hardly conceivable for optical sorters to be used by smaller producers or for inter-company use. Against this background, the aim of the studies presented here was to investigate the potential of acoustic methods for sorting different legumes and grain in the post-harvest. In a laboratory experiment, the sound of the impact of grains of two legumes, pea and field bean, as well as different cereals, rye, wheat, barley and oats has been investigated. The grains were dropped in a precisely defined height and time sequence from a conveyor belt onto a softly suspended impact plate of defined geometry and density. The corresponding noises were recorded with structure-borne sound sensors attached to the plate. The data were analyzed in both the time and frequency range. Features from the time signals and the information from the spectra were used as input parameters for various statistical analyses, such as discriminant analysis, nominal statistical regression and bootstrap forest. For the validation set, a generalized R² of 0.94 was achieved with a mis-classification-rate of 0.18. The results show that the recognition rates were lower for barley, rye and wheat, where field bean, pea and oat are detected with an accuracy of 100%. Ultimately, with regard to the sorting of mixtures commonly used in practice, consisting of a legume and a cereal, the results show for all possible combination of the two legumes and the four cereals a consistent separation of 100%.

Universitätsbibliothek Kassel
Openalex Percentile: Top 23%
Spectroscopy and Chemometric Analyses
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.