Modern battlefield environments require intelligent surveillance systems capable of real-time visual analysis and autonomous threat detection. Traditional monitoring systems rely heavily on manual observation or simple motion-trigger mechanisms, which lack intelligent image interpretation and identity verification capabilities. This paper presents the design and implementation of an AI-driven IoT-based battlefield surveillance system primarily focused on real-time image and video processing. The proposed system continuously captures live video streams using a camera module and processes frames through a machine learning-based classification model to identify authorized personnel and potential intruders. Deep learning techniques are employed for feature extraction and facial or object recognition, enabling accurate intrusion detection with reduced false alarms. Upon detecting unauthorized presence, the system triggers automated alerts, email notifications, and secure logging through IoT connectivity. The architecture supports edge-based processing to reduce latency and enhance operational efficiency in remote battlefield environments. Experimental validation demonstrates improved detection accuracy, faster response time, and enhanced situational awareness compared to conventional surveillance approaches.
K.Kavin et al. (Sun,) studied this question.