Related Experiment Video
Updated: May 13, 2026
11:49
Cereal Crop Ear Counting in Field Conditions Using Zenithal RGB Images
Published on: February 2, 2019
An ultrasonic system for weed detection in cereal crops
Dionisio Andújar1, Martin Weis, Roland Gerhards
1Department of Weed Science, Institute for Phytomedicine, University of Hohenheim, Otto-Sander-Straße 5, 70599 Stuttgart, Germany. Andujar@uni-hohenheim.de
Sensors (Basel, Switzerland)
|February 28, 2013
Summary
A new ultrasonic sensor method accurately detects weed infestations in cereal crops by measuring plant height differences. This technology offers a cost-effective approach to site-specific weed management for farmers.
Area of Science:
- Agricultural Engineering
- Precision Agriculture
- Remote Sensing
Background:
- Site-specific weed management is crucial but hindered by the lack of accessible, easy-to-use sensor systems for farmers.
- Current sophisticated sensor systems require additional effort for handling and integration into existing management practices.
Purpose of the Study:
- To evaluate the applicability of a novel sensor-based weed detection method for cereal crops.
- To assess the potential of using ultrasonic distance sensors for determining plant height and biomass as indicators of weed infestation.
Main Methods:
- An ultrasonic distance sensor was employed to measure plant heights in a winter wheat field with varying weed densities and types (grass and broad-leaved weeds).
- Measurements were correlated with weed counts, dry matter biomass, and RGB image analysis (including Excess Green Index - EGI) to assess crop and weed coverage.
- Multiple regression and discriminant analysis were used to evaluate the relationship between ultrasonic readings and weed infestation levels.
Main Results:
- Ultrasonic readings showed significant height differences between weed-infested and weed-free zones.
- Weed density and biomass directly influenced the ultrasonic measurements.
- The system achieved up to 92.8% success in distinguishing between weed-infested and non-infested areas.
Conclusions:
- Ultrasonic distance sensing is a viable, cost-effective method for detecting weed infestations in agricultural fields.
- This technology can support site-specific weed management and potentially reduce the costs associated with weed control.
