AI-Driven Anomaly Detection in Fusing Audio and Video Surveillance Systems Based on LSTM Model
Puspita Dash, Sridharan Brindha, V. Ranetha, A. T. Hari Sowmiyaa · 2025
The proposed work is a novel approach to enhancing anomaly detection in map-based datasets through the utilization of a spatial ensemble technique combined with deep learning models. By leveraging the strengths of both spatial ensemble methods and deep learning algorithmsthe proposed framework aims to improve the accuracy and efficiency of anomaly detection in spatial datasets. The spatial ensemble approach enables the integration of diverse spatial information sources to enhance anomaly detection performance while deep learning models provide the capability to extract complex patterns and relationships from the data. Through a series of experiments on realworld map-based dataset, the proposed approach demonstrates superior anomaly detection performance compared to traditional methods. The integration of spatial ensemble techniques with deep learning algorithm offers a promising solution for detecting anomalies in spatial datasets with high accuracy and reliability making it a valuable tool for various applications in geospatial analysis, environmental monitoring and urban planning.