AgriSeg Dataset: enabling AI-driven Robotics in Precision Agriculture
AgriSeg Dataset: enabling AI-driven Robotics in Precision Agriculture
Precision agriculture has emerged as a valid alternative to traditional farming technologies in the last two decades. Four fundamental requirements can be associated with this technology, such as increasing productivity, allocating resources reasonably, adapting to climate change, and reducing food waste. Precision agriculture is often led by trending and latest technology, such as mobile robotics, Artificial Intelligence (AI), the Internet of Things (IoT), and computer vision, which allow enhanced data collection and real-time monitoring of the crops, improving decision-making with the sake of reducing the overall environmental impact. Autonomous mobile systems, joined with AI tools and Deep Learning (DL), can provide a significant enhancement and competitive advancement in several agricultural tasks, reducing human labor and improving operational safety.
The focus on creating synthetic data originates from the necessity of having sizable and varied datasets in order to efficiently train and optimize deep learning models. Through modeling different situations in farming environments, we hope to build an extensive artificial dataset that reflects various conditions encountered in row crops. This method reduces the limitations related to data collecting in real-world agricultural situations by speeding up the training process and lowering reliance on hand-labeled images.
Synthetic data: can be used to train the Deep Semantic Segmentation models. For each crop different subset are provided.
Real Dataset: used for testing purposes to validate the quality of synthetic data through generalization testing of Deep Semantic Segmentation models.
Please if use our data consider citing our related works:
Martini, M., Ambrosio, M., Navone, A., Tuberga, B., & Chiaberge, M. (2024). Enhancing visual autonomous navigation in row-based crops with effective synthetic data generation. Precision Agriculture, 1-22.
Martini, M., A. Eirale, B. Tuberga, M. Ambrosio, A. Ostuni, F. Messina, L. Mazzara, and M. Chiaberge. "Enhancing navigation benchmarking and perception data generation for row-based crops in simulation". Precision agriculture ’23. Leiden, The Netherlands: Wageningen Academic, 2023. https://doi.org/10.3920/978-90-8686-947-3_56 Web.
Angarano, S., Martini, M., Navone, A., & Chiaberge, M. (2023). Domain Generalization for Crop Segmentation with Knowledge Distillation. arXiv preprint arXiv:2304.01029.
Navone, A., Martini, M., Ostuni, A., Angarano, S., & Chiaberge, M. (2023, September). Autonomous navigation in rows of trees and high crops with deep semantic segmentation. In 2023 European Conference on Mobile Robots (ECMR) (pp. 1-6). IEEE.
MEET US
Corso Francesco Ferrucci, 112
Torino, 10141
-
CONTACT US
Office: +39 011 0903552
Lab: +39 011 0903551
Mail: pic4ser@polito.it