Unified Detection Framework for Robbery Events: Integrating YOLOv8, Fast R-CNN, and RetinaNet with Explainable AI Validation and Real-time Android Application Integration

This research proposes a Unified Detection Framework for Robbery Events by combining YOLOv8, Fast R-CNN, and RetinaNet models. The framework incorporates Explainable AI Validation for transparency and real-time detection capabilities. Additionally, an Android application integration is implemented for practical deployment. Through experimentation and evaluation, the effectiveness of the framework in detecting robbery events is demonstrated. The fusion of multiple object detection models enhances accuracy and efficiency, while the integration of explainable AI validation ensures interpretability of results. The proposed framework shows promising results for addressing security concerns in real-world scenarios.