基于深度学习的动物检测识别方法

Animal Detection and Recognition System Based on Deep Learning

摘要: 针对当前野生动物图像识别面临的数据量大、分析效率低等问题，本文提出并实现了一种基于YOLOv8深度学习模型的动物检测识别系统。本研究首先采集并标注了包含10类野生动物的3000余张图像构建数据集；随后，利用该数据集对YOLOv8模型进行训练与评估，模型在测试集上取得了0.957的mAP@0.5值，表现出较高的检测精度；最后，基于Django框架开发了集用户管理、图像/视频/实时摄像头检测、识别结果统计与管理等功能于一体的Web系统。测试结果表明，该系统能够快速准确地识别多种野生动物，为生态监测与动物保护提供了有效的技术支持。

Abstract: Aiming at the problems of large data volume and low analytical efficiency in current wildlife image recognition, this paper proposes and implements an animal detection and recognition system based on the YOLOv8 deep learning model. This study first collected and annotated over 3000 images containing 10 categories of wildlife to construct a dataset. Subsequently, the YOLOv8 model was trained and evaluated using this dataset, achieving an mAP@0.5 of 0.957 on the test set, demonstrating high detection accuracy. Finally, a web system integrating user management, image/video/real-time camera detection, and recognition result statistics and management was developed based on the Django framework. Test results show that the system can quickly and accurately identify various wild animals, providing effective technical support for ecological monitoring and animal protection.