Implementation of elderly health monitoring system based on motion target recognition using FPGA

The development of motion target recognition technology has opened up prospects for its application in health and safety monitoring. To meet the needs of small-scale, home-based, and real-time applications in health monitoring systems, this study designs a motion target recognition system based on FPGA. The system uses a CMOS camera as the video capture device, SDRAM as the video stream cache and storage medium, and takes advantage of FPGA’s parallel processing capabilities. By employing pipeline for modular design, the system performs gray-scale conversion, frame difference module, and morphological processing on the video stream, enabling real-time tracking and monitoring of image acquisition and target recognition, and determining whether the elderly person is in a state of safety risk such as falling or sitting for a long time. The results and warnings are then transmitted through UART Wi-Fi to a wireless router, reaching family members' mobile phones or community hospital terminals through the internet, in order to develop an effective, stable, miniaturized, and low-power home-based elderly health monitoring system.