Weed Detection from Unmanned Aerial Vehicle Imagery Using Deep Learning—A Comparison between High-End and Low-Cost Multispectral Sensors
Problem
Reliable, affordable weed detection from drone imagery is needed so herbicides can be targeted rather than applied broadly across maize fields.
Why it matters
Targeted herbicide use reduces environmental pollution and crop yield loss caused by weeds.
What the AI did
A U-Net performed pixel-based weed and crop classification on UAV multispectral imagery, comparing a self-built low-cost camera against the high-end MicaSense Altum system.
Results
Weed-class F1-score of 82% (Altum) and 76% (low-cost system); recall values of 75% and 68% respectively
Impact
A low-cost sensor approached the high-end system's weed-detection performance, though with more misclassifications.
Record
| Venue | Sensors |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Anna Teresa Seiche, Lucas Wittstruck, Thomas Jarmer |
| Citations | 29 (OpenAlex, at import time) |
| References | 68 |
| Model or method | U-Net |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/s24051544 |
| Open access | gold |
| Keywords | weed detection, UAV, multispectral, U-Net, deep learning, herbicide |
Importance rating
significant — thousands to millions affected
Abstract states targeted herbicide use reduces environmental pollution and controls weeds, a major cause of yield reduction; sustainable cultivation is called essential under challenging climate conditions.
More in Climate & Environment
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | Learning skillful medium-range global weather forecasting | 2023 | 1,416 |
8
|
| 2 | Accurate medium-range global weather forecasting with 3D neural networks | 2023 | 1,631 |
8
|
| 3 | FuXi: a cascade machine learning forecasting system for 15-day global weather forecast | 2023 | 355 |
8
|
| 4 | Skilful nowcasting of extreme precipitation with NowcastNet | 2023 | 451 |
8
|
| 5 | Forecasting Renewable Energy Generation with Machine Learning and Deep Learning: Current Advances and Future Prospects | 2023 | 378 |
8
|