Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Light Acquisition

Light Acquisition

In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

UAV-Based RGB Imagery for Hokkaido Pumpkin <i>(Cucurbita max.)</i> Detection and Yield Estimation.

Sensors (Basel, Switzerland)·2020

See all related articles

Search research articles

Related Experiment Video

Updated: Jul 1, 2025

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

Weed Detection from Unmanned Aerial Vehicle Imagery Using Deep Learning-A Comparison between High-End and Low-Cost

Anna Teresa Seiche¹, Lucas Wittstruck¹, Thomas Jarmer¹

¹Institute of Computer Science, Osnabrück University, 49090 Osnabrück, Germany.

Sensors (Basel, Switzerland)

|March 13, 2024

Summary

A new low-cost multispectral camera system shows potential for automated weed control in precision farming. This system, evaluated against a high-end sensor, achieved 76% F1-score for weed detection, enabling sustainable agriculture.

Keywords:

Raspberry Pi U-Net drone precision farming self-built weed detection

More Related Videos

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy

Published on: February 25, 2021

Flying Insect Detection and Classification with Inexpensive Sensors

Flying Insect Detection and Classification with Inexpensive Sensors

Published on: October 15, 2014

Related Experiment Videos

Last Updated: Jul 1, 2025

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy

Published on: February 25, 2021

Flying Insect Detection and Classification with Inexpensive Sensors

Flying Insect Detection and Classification with Inexpensive Sensors

Published on: October 15, 2014

Area of Science:

Agricultural Engineering
Remote Sensing
Computer Vision

Background:

Increasing global crop demand necessitates sustainable agriculture under challenging climate conditions.
Effective weed management is crucial for minimizing crop yield loss and reducing environmental impact through targeted herbicide application.
Accessible and reliable weed detection systems are vital for widespread adoption of precision farming techniques.

Purpose of the Study:

To introduce and evaluate a self-built, low-cost multispectral camera system for weed detection.
To compare the performance of the low-cost system against a high-end MicaSense Altum system.
To assess the potential of the developed system for enabling automated weed control and sustainable precision farming.

Main Methods:

Development of a low-cost, multispectral camera system.
Acquisition of Unmanned Aerial Vehicle (UAV) datasets in maize fields using both the low-cost and Altum systems.
Pixel-based weed and crop classification using a U-Net deep learning model on the collected datasets.
Generation of training and testing data through index-based thresholding and manual annotation.

Main Results:

The low-cost system achieved an F1-score of 76% for weed classification, compared to 82% for the Altum system.
Recall values were 68% for the low-cost system and 75% for the Altum system.
The low-cost system demonstrated a precision of 90%, with minor oversegmentation issues observed for small or overlapping weeds.

Conclusions:

The developed low-cost multispectral camera system shows significant potential for automated weed control applications in precision farming.
Despite minor misclassifications, the system's high precision indicates its viability for practical use by a wider range of end-users.
Further research into spectral properties and real-time processing on different crops is recommended to enhance system capabilities.