返回Google圖書搜尋

A Computer Vision-Based Object Localization Model for Endangered Wildlife Detection

Arunabha Mohan RoyJayabrata BhaduriTeerath KumarKislay Raj

出版	SSRN, 2023
URL	http://books.google.com.hk/books?id=gGH-zwEACAAJ&hl=&source=gbs_api

註釋Objective. With climatic instability, various ecological disturbances, and human actions threaten the existence of various endangered wildlife species. Therefore, up-to-date accurate, and detailed detection plays an important role in protecting biodiversity losses, conservation, and ecosystem management. Current state-of-the-art wildlife detection models, however, often lack superior feature extraction capability in complex environments, limits the development of accurate and reliable detection models. Method. To this end, we present WilDect-YOLO, deep learning (DL)-based automated high-performance detection model for real-time endangered wildlife detection. This model introduces a new residual block in the CSPDarknet53 backbone for strong and discriminating deep spatial features extraction and integrates DenseNet to improve preserving critical feature information. To enhance receptive field representation, preserve fine-grain localized information, and improve feature fusion, Spatial Pyramid Pooling (SPP) and modified Path Aggregation Network (PANet) have been implemented that provide superior detection under various challenging environments. Results. Evaluating the model performance in a custom endangered wildlife dataset considering high variability and complex backgrounds, WilDect-YOLO obtains a mean average precision (mAP) value of 96.89 %, F1-score of 97.87 %, and precision value of 97.18 % at a detection rate of 59.20 FPS outperforming current state-of-the-art models. Significance. The present research provides an effective and efficient detection model addressing the shortcoming of existing DL-based wildlife detection models by providing highly accurate species-level localized bounding box prediction. Current work constitutes a step towards a non-invasive, fully automated animal observation system in real-time in-field applications.