EUVIMEDEUVIMED entdecken
Crossref

Deep Learning for Precision Agriculture: Post-Spraying Evaluation and Deposition Estimation

Harry Rogers, Tahmina Zebin, Grzegorz Cielniak, De La Iglesia, Beatriz, Ben Magri

UEA Digital Repository (University of East Anglia) · 24.09.2024

DOI: 10.48550/arxiv.2409.16213 ↗

Kurz gefasst

Fachartikel in der Zeitschrift UEA Digital Repository (University of East Anglia), erschienen am 24.09.2024. Verfasst von Harry Rogers und 5 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Precision spraying evaluation requires automation primarily in post-spraying imagery. In this paper we propose an eXplainable Artificial Intelligence (XAI) computer vision pipeline to evaluate a precision spraying system post-spraying without the need for traditional agricultural methods. The developed system can semantically segment potential targets such as lettuce, chickweed, and meadowgrass and correctly identify if targets have been sprayed. Furthermore, this pipeline evaluates using a domain-specific Weakly Supervised Deposition Estimation task, allowing for class-specific quantification of spray deposit weights in μL. Estimation of coverage rates of spray deposition in a class-wise manner allows for further understanding of effectiveness of precision spraying systems. Our study evaluates different Class Activation Mapping techniques, namely AblationCAM and ScoreCAM, to determine which is more effective and interpretable for these tasks. In the pipeline, inference-only feature fusion is used to allow for further interpretability and to enable the automation of precision spraying evaluation post-spray. Our findings indicate that a Fully Convolutional Network with an EfficientNet-B0 backbone and inference-only feature fusion achieves an average absolute difference in deposition values of 156.8 μL across three classes in our test set. The dataset curated in this paper is publicly available at https://github.com/Harry-Rogers/PSIE

Schlagwörter

precision sprayingexplainable AIsemantic segmentationweakly supervised segmentationclass activation mappingfully convolutional networkEfficientNetB0lettuceStellaria media

Zitieren

APA

Rogers, H., Zebin, T., Cielniak, G., Iglesia, D. L., Beatriz, & Magri, B. (2024). Deep Learning for Precision Agriculture: Post-Spraying Evaluation and Deposition Estimation. UEA Digital Repository (University of East Anglia). https://doi.org/10.48550/arxiv.2409.16213

Vancouver

Rogers H, Zebin T, Cielniak G, Iglesia DL, Beatriz, Magri B. Deep Learning for Precision Agriculture: Post-Spraying Evaluation and Deposition Estimation. UEA Digital Repository (University of East Anglia). 2024. doi:10.48550/arxiv.2409.16213

BibTeX
@article{Rogers2024,
  author = {Rogers, Harry and Zebin, Tahmina and Cielniak, Grzegorz and Iglesia, De La and Beatriz and Magri, Ben},
  title = {Deep Learning for Precision Agriculture: Post-Spraying Evaluation and Deposition Estimation},
  journal = {UEA Digital Repository (University of East Anglia)},
  year = {2024},
  doi = {10.48550/arxiv.2409.16213}
}
Kostenlos starten →