Enhancing Wildlife Monitoring: Real-Time Alerts Through Email and Messaging

Wildlife conservation faces increasing challenges due to habitat encroachment and human-wildlife conflicts. Ensuring real-time detection of wildlife near human settlements is crucial for protecting both animal and human lives. Traditional wildlife monitoring techniques, such as manual patrolling and fixed camera traps, often suffer from delayed response times and limited coverage. These limitations restrict the ability to implement proactive intervention strategies effectively. Conventional methods primarily rely on human observation or cloud-based object detection models, leading to high operational costs and latency issues, especially in remote environments. This work presents a real-time wildlife detection system based on the YOLOv11 model, integrated with automated SMS and email alert mechanisms. The system operates on an edge device, minimizing dependency on cloud infrastructure and ensuring low-latency performance. Simulation experiments were conducted using a dataset comprising 6,993 wildlife images representing elephants, boars, monkeys, and humans under varying conditions. The detection model was evaluated using precision, recall, F1-score, and mean Average Precision (mAP) metrics. The system was developed using Python with TensorFlow, OpenCV, and Twilio-Gmail APIs for real-time processing and communication. Experimental results demonstrated a mAP of 81.6% at an IoU threshold of 0.5, precision up to 86%, recall up to 75%, and an alert generation time of less than 0.5 seconds with a 98% delivery success rate. In conclusion, the proposed approach offers a robust, real-time, and scalable solution for wildlife monitoring, significantly improving response times and enhancing the effectiveness of conservation efforts.