Endemic species are under threat from various factors, including habitat destruction, illegal hunting, and climate change, which necessitate urgent and effective monitoring solutions. This research introduces an advanced AI system for real-time threat detection and response, utilizing the YOLOv8 algorithm. The system is specifically developed for the protection of endemic species. It combines the strengths of YOLOv8 with enhancements like a multi-scale detection module to tackle the challenges of identifying small or camouflaged species and threats across different ecological environments. In the proposed study, a customized dataset that is developed through the application of the Histogram of Oriented Gradients (HOG) technique along with the Firefly Algorithm is employed. This approach facilitates the efficient fine-tuning of all images within the dataset, enhancing the overall effectiveness of the analysis. The Roboflow platform is used for training, validation, and testing the customized dataset for real-time object detection. YOLOv8 achieves 97.9% mAP, 94.7% precision, and 91.9% recall. Threats are recorded in a Blockchain ledger and sent to Twilio SMS alert system, making it cost-effective and efficient. The proposed framework offers high accuracy, precision, and recall, minimizing false alarms and facilitating quick interventions, making it suitable for smart environmental management systems and biodiversity conservation.

YOLOv8 Roboflow Object Detection Blockchain Technology Twilio