Deep Learning based Animal Detection and Tracking in Drone Video Footage

Jintasuttisak, Thani; Leonce, Andrew; Shah, Moayyed Sher; Khafaga, Tamer; Simkins, Greg; Edirisinghe, Eran A.

doi:10.1145/3532213.3532280

Cited by 8 publications

(2 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, YOLOv7 outperformed PP-YOLO, YOLOX, YOLOR, Scaled-YOLOv4, YOLOv5, DETR, Deformable DETR, DINO-5scale-R50, and Vit-Adapter-B. Among the numerous works that utilized YOLO algorithms for agricultural tasks are (Hatton-Jones et al, 2021;Schütz et al, 2021;Jintasuttisak et al, 2022;Siriani et al, 2022). Hence, the objective of this paper is to apply YOLOv7 to livestock image detection and segmentation tasks in cattle grazing behavior, monitor and intrusions.…”

Section: Moradeyo O M; Olaniyan As; Ojoawo Ao; Olawale Ja; Bello Rwmentioning

confidence: 99%

YOLOv7 Applied to Livestock Image Detection and Segmentation Tasks in Cattle Grazing Behavior, Monitor and Intrusions

Moradeyo¹,

Olaniyan²,

Ojoawo³

et al. 2023

jasem

View full text Add to dashboard Cite

You only look once (YOLO) is a state-of-the-art, real-time object detection system. YOLO version 7 (YOLOv7) model is a variant of YOLO. The objective of this paper is to apply YOLOv7 to livestock image detection and segmentation tasks in cattle grazing behavior, monitor and intrusions. Data obtained revealed that YOLOv7 performs better in terms of speed and accuracy with a mAP of 0.95 than the baseline techniques.

show abstract

Section: Moradeyo O M; Olaniyan As; Ojoawo Ao; Olawale Ja; Bello Rwmentioning

confidence: 99%

YOLOv7 Applied to Livestock Image Detection and Segmentation Tasks in Cattle Grazing Behavior, Monitor and Intrusions

Moradeyo¹,

Olaniyan²,

Ojoawo³

et al. 2023

jasem

View full text Add to dashboard Cite

show abstract

“…Deep learning (DL) shows great potential in BV hole detection, benefiting from the application of automatic deep convolutional feature extraction (LeCun et al, 2015;Mountrakis et al, 2018;Liu et al, 2020). Recent studies using DL and UAS images have been widely applied to small object detection, such as detecting birds (Hong et al, 2019) and mammals in the wild (Kellenberger et al, 2018;Peng et al, 2020;Jintasuttisak et al, 2022), identifying weeds (Etienne et al, 2021), counting small plants (Oh et al, 2020), and extracting vehicles (Xu et al, 2016;Amato et al, 2019). So far, four previous studies have employed UAS images and DL in rodent hole detection.…”

Section: Introductionmentioning

confidence: 99%

Rodent hole detection in a typical steppe ecosystem using UAS and deep learning

Wang

Liu

et al. 2022

Front. Plant Sci.

View full text Add to dashboard Cite

IntroductionRodent outbreak is the main biological disaster in grassland ecosystems. Traditional rodent damage monitoring approaches mainly depend on costly field surveys, e.g., rodent trapping or hole counting. Integrating an unmanned aircraft system (UAS) image acquisition platform and deep learning (DL) provides a great opportunity to realize efficient large-scale rodent damage monitoring and early-stage diagnosis. As the major rodent species in Inner Mongolia, Brandt’s voles (BV) (Lasiopodomys brandtii) have markedly small holes, which are difficult to identify regarding various seasonal noises in this typical steppe ecosystem.MethodsIn this study, we proposed a novel UAS-DL-based framework for BV hole detection in two representative seasons. We also established the first bi-seasonal UAS image datasets for rodent hole detection. Three two-stage (Faster R-CNN, R-FCN, and Cascade R-CNN) and three one-stage (SSD, RetinaNet, and YOLOv4) object detection DL models were investigated from three perspectives: accuracy, running speed, and generalizability.ResultsExperimental results revealed that: 1) Faster R-CNN and YOLOv4 are the most accurate models; 2) SSD and YOLOv4 are the fastest; 3) Faster R-CNN and YOLOv4 have the most consistent performance across two different seasons.DiscussionThe integration of UAS and DL techniques was demonstrated to utilize automatic, accurate, and efficient BV hole detection in a typical steppe ecosystem. The proposed method has a great potential for large-scale multi-seasonal rodent damage monitoring.

show abstract