Creating Alert Message Based On Wild Animals Activity Detection Using Hybrid Deep Neural Network

The issue of wildlife attacks is becoming an increasing concern for rural communities and forestry workers.To monitor the movements of wild animals, surveillance cameras and drones are frequently used.However, an effective system is needed to identify the species, track its movement, and provide its location details.Alert notifications can then be sent to ensure the safety of individuals and forestry personnel.Although computer vision and machine learning techniques are commonly applied for animal detection, they are often costly and complex, leading to suboptimal outcomes.This project introduces a Hybrid Visual Geometry Group (VGG)-19 combined with a Bidirectional Long Short-Term Memory (Bi-LSTM) network to detect animals and trigger alerts based on their activities.These alerts are transmitted as Short Message Service (SMS) to the local forest authority for immediate action.The proposed system demonstrates significant improvements in performance, achieving an average classification accuracy of 98%, a mean Average Precision (mAP) of 77.2%, and a Frame Per Second (FPS) rate of 170.The model was evaluated both qualitatively and quantitatively with 40,000 images from three distinct benchmark datasets containing 25 categories, reaching a mean accuracy and precision exceeding 98%.This system provides a dependable solution for delivering precise animal-related information and safeguarding human lives.