Buildings fulfill a multitude of societal needs, serving as shelters, ensuring security, and providing living and working spaces. However, ensuring the safety of these spaces, particularly homes, remains a significant challenge due to limitations in traditional intrusion and theft detection methods. These limitations includes: the inability of the system to detect an act of intrusion without human input, the inability of system to report the act of intrusion in real time manner. This research proposes a novel approach to address this challenge by leveraging deep learning-based automatic image classification systems, utilizing artificial intelligence (AI) techniques, specifically Deep Neural Networks (DNN). The research aims to investigate the effectiveness of deep learning models for intrusion and theft detection in commercial and residential buildings. , the study intends to develop and evaluate prototype systems capable of automatically detecting intrusions and thefts using footage captured by the camera within building premises. In addition to software components, the research will leverage cloud computing infrastructure for resource-intensive tasks such as storage and optimization. Hardware components such as cameras will be employed for data collection, enabling the training and testing of deep learning models on real-world datasets

Deep Neural Networks