Machine learning algorithms applied to weed management in integrated crop-livestock systems: a systematic literature review.
In recent times, there has been an environmental pressure to reduce the amount of pesticides applied to crops and, consequently, the crop production costs. Therefore, investments have been made in technologies that could potentially reduce the usage of herbicides on weeds. Among such technologies, Machine Learning approaches are rising in number of applications and potential impact. Therefore, this article aims to identify the main machine learning algorithms used in integrated crop-livestock systems for weed management. Based on a systematic literature review, it was possible to determine where the selected studies were performed and which crop types were mostly used. The main research terms in this study were: "machine learning algorithms" + "weed management" + "integrated crop-livestock system". Although no results were found for the three terms altogether, the combinations involving "weed management" + "integrated crop-livestock system" and "machine learning algorithms" + "weed management" returned a significant number of studies which were subjected to a second layer of refinement by applying an eligibility criteria. The achieved results show that most of the studies were from the United States and from nations in Asia. Machine vision and deep learning were the most used machine learning models, representing 28% and 19% of all cases, respectively. These systems were applied to different practical solutions, the most prevalent being smart sprayers, which allow for a site-specific herbicide application.
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Artigo de periódico biblioteca |
| Language: | eng |
| Published: |
2024-04-23
|
| Subjects: | Weed prevention, Image processing, Inteligência artificial, Processamento de imagem, Erva Daninha, Weed control, Artificial intelligence, |
| Online Access: | http://www.alice.cnptia.embrapa.br/alice/handle/doc/1163826 https://doi.org/10.51694/AdvWeedSci/2024;42:00004 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|