International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 08 | Aug 2023
p-ISSN: 2395-0072
www.irjet.net
Wildlife Identification using Object Detection using Computer Vision and YOLO Prasham Mehta1, Keval Shah2, Rohit Raval3, Manan Shah4 1Information Technology, Atharva College of Engineering 2Computer Science (A.I), NMIMS 3Computer Science (A.I), NMIMS 4Computer Science (A.I), NMIMS
--------------------------------------------------------------------------***---------------------------------------------------------------------------assessing habitat utilization, and informing conservation Abstract strategies. Over the years, technological advancements in computer vision and object detection have provided powerful tools for automating the process of wildlife monitoring. These tools enable efficient and accurate identification, tracking, and analysis of wildlife species, contributing to conservation efforts worldwide
Wildlife monitoring plays a vital role in conservation efforts by providing insights into population dynamics, habitat utilization and species behavior. This research paper aims to explore the application of the YOLO (You Only Look Once) algorithm a cutting-edge object detection framework, in wildlife monitoring and conservation. The primary focus of this study is to evaluate the effectiveness of the YOLO algorithm in identifying and tracking wildlife species, analyzing their behavior, and supporting conservation efforts. By utilizing CCTV cameras placed in wildlife habitats the YOLO algorithm enables real-time detection, tracking, and classification of wildlife objects thereby providing researchers and conservationists with data. To conduct this research we employed a methodology that involved training the YOLO model using wildlife datasets[7]. This training enabled the model to recognize and classify species accurately. Furthermore, we optimized the model’s performance before deploying it on CCTV camera feeds allowing for the monitoring of wildlife populations. The YOLO algorithms' efficient video frame processing capabilities ensure real-time object detection enabling access to insights about species presence, behavior patterns, and potential threats. The utilization of the YOLO algorithm has proven beneficial as it enables realtime identification of elusive species. This technological advancement plays a role, in providing information, for conservation efforts surrounding these species.
In recent years, the emergence of the YOLO (You Only Look Once) algorithm has revolutionized object detection in computer vision. YOLO stands out for its realtime capabilities, allowing for the rapid processing of video frames and providing immediate insights into the presence and behavior of wildlife species[8]. By leveraging this algorithm, researchers and conservationists can harness the power of object detection to monitor wildlife populations in real time, leading to timely decision-making and effective conservation strategies. The primary objective of this research paper is to investigate and evaluate the application of the YOLO algorithm in wildlife monitoring and conservation. By deploying CCTV cameras strategically placed in wildlife habitats, the YOLO algorithm can be leveraged to analyze video feeds and detect wildlife objects efficiently. The algorithm's ability to identify and classify multiple objects simultaneously enables researchers to obtain accurate and reliable information on species presence and distribution.
INTRODUCTION
The methodology employed in this study involves training the YOLO model on annotated wildlife datasets, ensuring the algorithm's ability to recognize and classify different species accurately[4]. The training process involves optimizing model parameters, selecting appropriate loss functions, and fine-tuning the network architecture. The trained model is then deployed on CCTV camera feeds, enabling continuous monitoring of wildlife populations.
Wildlife monitoring is essential for understanding the dynamics of animal populations,
One of the key advantages of the YOLO algorithm is its real-time performance, enabling immediate detection
Keywords— Object detection, Wildlife monitoring, Conservation, YOLO algorithm, Computer vision, Behavior analysis, Human-wildlife conflicts, Real-time monitoring.
© 2023, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 246