Bullying action detection and recognition algorithm based on millimeter-wave radar

With the continuous advancement of technology, millimeter-wave radar-based motion detection has shown great potential in security and surveillance. This paper presents a bullying behavior detection algorithm that analyzes radar signals to identify and assess bullying in real time. The algorithm integrates signal processing, machine learning, and pattern recognition to extract features from reflected radar signals and classify behaviors. Dynamic movement data is captured, and motion features are extracted for recognition using an enhanced feature selection and classification method. Extensive testing demonstrates that the algorithm efficiently and accurately detects bullying behaviors in various scenarios, with strong anti-interference and high recognition accuracy. This research provides a novel approach for intelligent security systems based on millimeter-wave radar, with promising practical applications.