Wildlife Real-Time Detection in Complex Forest Scenes Based on YOLOv5s Deep Learning Network

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(8), P. 1350 - 1350

Published: April 11, 2024

With the progressively deteriorating global ecological environment and gradual escalation of human activities, survival wildlife has been severely impacted. Hence, a rapid, precise, reliable method for detecting holds immense significance in safeguarding their existence monitoring status. However, due to rare concealed nature existing detection methods face limitations efficiently extracting features during real-time complex forest environments. These models exhibit drawbacks such as slow speed low accuracy. Therefore, we propose novel model called WL-YOLO, which is designed lightweight This built upon deep learning YOLOv5s. In introduce feature extraction module. module comprised deeply separable convolutional neural network integrated with compression excitation modules backbone network. design aimed at reducing number parameters computational requirements, while simultaneously enhancing representation Additionally, introduced CBAM attention mechanism enhance local key features, resulting improved performance WL-YOLO natural where high concealment complexity. achieved mean accuracy (mAP) value 97.25%, an F1-score 95.65%, 95.14%. results demonstrated that this outperforms current mainstream models. compared YOLOv5m base model, reduces by 44.73% shortens time 58%. study offers technical support protecting intricate environments introducing highly efficient advanced model.

Language: Английский

---

UAV equipped with infrared imaging for Cervidae monitoring: Improving detection accuracy by eliminating background information interference

Ecological Informatics, Journal Year: 2024, Volume and Issue: 81, P. 102651 - 102651

Published: May 24, 2024

Wild Cervidae(deer and their relatives) play a crucial role in maintaining ecological balance are integral components of ecosystems. However, factors such as environmental changes poaching behaviors have resulted habitat degradation for Cervidae. The protection wild Cervidae has become urgent, monitoring is one the key means to ensure effectiveness protection. Object detection algorithms based on deep learning offer promising potential automatically detecting identifying animals. when those used inference unseen background environments, there will be significant decrease accuracy, especially situation that certain type images collected from single scene algorithm training. In this paper, two-stage localization classification pipeline proposed. effectively reduces interference enhances accuracy. first stage, YOLOv7 network designed locate UAV infrared images, while implementing improved bounding box regression through α-IoU loss function enables more accurately. Then, Cevdidae objects extracted eliminate information. second named CA-Hybrid, Convolutional Neural Networks(CNN) Vision Transformer(ViT), well Channel Attention Mechanism(CAM) expression features, constructed accurately identify categories. Experimental results indicate method achieves an Average Precision (AP) 95.9% location top-1 accuracy 77.73% identification. This research contributes comprehensive accurate Cervidae, provides valuable references subsequent UAV-based wildlife monitoring.

Language: Английский

---

Drivers of human–tiger conflict risk and potential mitigation approaches

Ecosphere, Journal Year: 2024, Volume and Issue: 15(7)

Published: July 1, 2024

Abstract Human–wildlife conflict has become a significant challenge for conservationists, particularly in areas where endangered species, such as large carnivores, are recovering. If we fail to keep balance between the interests of humans and wildlife, human–wildlife can have adverse outcomes. However, drivers conflict, how mitigate often poorly understood. In this study, aimed explore possible causes potential mitigating approaches human–tiger risks through spatiotemporal niche partitioning. Based on data from reports Amur tiger ( Panthera tigris altaica ) preying cattle camera trap detection 2014 2019 Hunchun, Northeast China, predicted occurrence created risk maps encounters. We found that was positively driven by prey distribution negatively pastures used domestic grazing. Livestock increasingly predated with limited preferred prey, is, wild pig Sus scrofa sika Cervus nippon ), closer proximity cattle‐grazing land. On basis our models, divided utilized human tigers into low‐, medium‐, high‐risk across multiple scales. suppose scale partitioning management might effectively reduce encounters, prompt harmonized coexistence tigers, provide new solutions other experiencing conflicts.

Language: Английский

---

An Efficient Illumination Invariant Tiger Detection Framework for Wildlife Surveillance

Lecture notes in networks and systems, Journal Year: 2024, Volume and Issue: unknown, P. 173 - 182

Published: Jan. 1, 2024

Language: Английский

---