Generating Messages applying Hybrid Deep Neural Networks to Discover Natural Animal Movements

Ensuring the safety of individuals in rural and forest-adjacent regions requires effective monitoring of wildlife movements. Traditional surveillance methods such as drones and stationary cameras are insufficient without intelligent detection and alert mechanisms. This paper introduces a hybrid deep learning framework that integrates the VGG-19 convolutional neural network with a Bidirectional Long Short-Term Memory (Bi-LSTM) model to accurately identify animal species and track their movements. The proposed system automatically generates alerts and transmits them via Short Message Service (SMS) to forest authorities, enabling timely responses. Trained on 40,000 images across 25 animal classes, the system achieves a classification accuracy of 98%, a mean Average Precision (mAP) of 77.2%, and a processing speed of 170 frames per second. This solution demonstrates superior efficiency in recognizing and localizing animal activity, significantly enhancing human safety and wildlife monitoring efforts.