Mask the Target: Mask R-CNN-Based Approach for Open-Range Individual Cattle Segmentation
DS Journal of Artificial Intelligence and Robotics · 2023
DOI: 10.59232/air-v1i2p102 ↗Kurz gefasst
Fachartikel in der Zeitschrift DS Journal of Artificial Intelligence and Robotics, 2023. Verfasst von Rotimi-Williams Bello und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Targeting individual cattle in an open-range has been a computer vision problem with great agricultural implications for animal health monitoring, behavior monitoring, and economic benefit. In this paper, the individual cattle targets in the open-range are segmented using Mask R-CNN’s detect and segment approach. We evaluate the performance of the model proposed in this study using the Intersection Over Union (IOU) threshold of 0.5, Average Precision (AP) and mean Average Precision (mAP). The results of the experiment conducted with the Acquired dataset in this study show that the proposed model achieves an accuracy of 95%, thereby affirming the potential of the Mask R-CNN model to perform competitively with any other existing object detection and instance segmentation models for cattle target segmentation with high accuracy and average precision.
Zitieren
APA
Bello, R., & Oladipo, M. A. (2023). Mask the Target: Mask R-CNN-Based Approach for Open-Range Individual Cattle Segmentation. DS Journal of Artificial Intelligence and Robotics. https://doi.org/10.59232/air-v1i2p102
Vancouver
Bello R, Oladipo MA. Mask the Target: Mask R-CNN-Based Approach for Open-Range Individual Cattle Segmentation. DS Journal of Artificial Intelligence and Robotics. 2023. doi:10.59232/air-v1i2p102
BibTeX
@article{Bello2023,
author = {Bello, Rotimi-Williams and Oladipo, Mojisola Abosede},
title = {Mask the Target: Mask R-CNN-Based Approach for Open-Range Individual Cattle Segmentation},
journal = {DS Journal of Artificial Intelligence and Robotics},
year = {2023},
doi = {10.59232/air-v1i2p102}
}
EUVIMED