Effects of Foreground Augmentations in Synthetic Training Data on the Use of UAVs for Weed Detection

Simon Hallösta, Mats Ingemar Pettersson, Mattias Dahl
Proceedings of the 5th Northern Lights Deep Learning Conference ({NLDL}), PMLR 233:81-88, 2024.

Abstract

This study addresses the issue of black-grass, a herbicide-resistant weed that threatens wheat yields in Western Europe, through the use of high- resolution Unmanned Aerial Vehicles (UAVs) and synthetic data augmentation in precision agriculture. We mitigate challenges such as the need for large labeled datasets and environmental variability by employing synthetic data augmentations in training a Mask R-CNN model. Using a minimal dataset of 43 black-grass and 12 wheat field images, we achieved a 37% increase in Area Under the Curve (AUC) over the non-augmented baseline, with scaling as the most effective augmentation. The best model attained a recall of 53% at a precision of 64%, offering a promising approach for future precision agriculture applications.

Cite this Paper

BibTeX
@InProceedings{pmlr-v233-hallosta24a, title = {Effects of Foreground Augmentations in Synthetic Training Data on the Use of {UAV}s for Weed Detection}, author = {Hall{\"o}sta, Simon and Pettersson, Mats Ingemar and Dahl, Mattias}, booktitle = {Proceedings of the 5th Northern Lights Deep Learning Conference ({NLDL})}, pages = {81--88}, year = {2024}, editor = {Lutchyn, Tetiana and Ramírez Rivera, Adín and Ricaud, Benjamin}, volume = {233}, series = {Proceedings of Machine Learning Research}, month = {09--11 Jan}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v233/hallosta24a/hallosta24a.pdf}, url = {https://proceedings.mlr.press/v233/hallosta24a.html}, abstract = {This study addresses the issue of black-grass, a herbicide-resistant weed that threatens wheat yields in Western Europe, through the use of high- resolution Unmanned Aerial Vehicles (UAVs) and synthetic data augmentation in precision agriculture. We mitigate challenges such as the need for large labeled datasets and environmental variability by employing synthetic data augmentations in training a Mask R-CNN model. Using a minimal dataset of 43 black-grass and 12 wheat field images, we achieved a 37% increase in Area Under the Curve (AUC) over the non-augmented baseline, with scaling as the most effective augmentation. The best model attained a recall of 53% at a precision of 64%, offering a promising approach for future precision agriculture applications.} }
Endnote
%0 Conference Paper %T Effects of Foreground Augmentations in Synthetic Training Data on the Use of UAVs for Weed Detection %A Simon Hallösta %A Mats Ingemar Pettersson %A Mattias Dahl %B Proceedings of the 5th Northern Lights Deep Learning Conference ({NLDL}) %C Proceedings of Machine Learning Research %D 2024 %E Tetiana Lutchyn %E Adín Ramírez Rivera %E Benjamin Ricaud %F pmlr-v233-hallosta24a %I PMLR %P 81--88 %U https://proceedings.mlr.press/v233/hallosta24a.html %V 233 %X This study addresses the issue of black-grass, a herbicide-resistant weed that threatens wheat yields in Western Europe, through the use of high- resolution Unmanned Aerial Vehicles (UAVs) and synthetic data augmentation in precision agriculture. We mitigate challenges such as the need for large labeled datasets and environmental variability by employing synthetic data augmentations in training a Mask R-CNN model. Using a minimal dataset of 43 black-grass and 12 wheat field images, we achieved a 37% increase in Area Under the Curve (AUC) over the non-augmented baseline, with scaling as the most effective augmentation. The best model attained a recall of 53% at a precision of 64%, offering a promising approach for future precision agriculture applications.
APA
Hallösta, S., Pettersson, M.I. & Dahl, M.. (2024). Effects of Foreground Augmentations in Synthetic Training Data on the Use of UAVs for Weed Detection. Proceedings of the 5th Northern Lights Deep Learning Conference ({NLDL}), in Proceedings of Machine Learning Research 233:81-88 Available from https://proceedings.mlr.press/v233/hallosta24a.html.

Related Material