An intelligent humidity sensing system for human behavior recognition
Huabin Yang, Qiming Guo, Guidong Chen, Yuefang Zhao, Meng Shi, Na Zhou, Chengjun Huang, Haiyang Mao
Abstract
An intelligent humidity sensing system has been developed for real-time monitoring of human behaviors through respiration detection. The key component of this system is a humidity sensor that integrates a thermistor and a micro-heater. This sensor employs porous nanoforests as its sensing material, achieving a sensitivity of 0.56 pF/%RH within a range of 60-90% RH, along with excellent long-term stability and superior gas selectivity. The micro-heater in the device provides a high operating temperature, enhancing sensitivity by 5.8 times. This significant improvement enables the capture of weak humidity variations in exhaled gases, while the thermistor continuously monitors the sensor's temperature during use and provides crucial temperature information related to respiration. With the assistance of a machine learning algorithm, a behavior recognition system based on the humidity sensor has been constructed, enabling behavior states to be classified and identified with an accuracy of up to 96.2%. This simple yet intelligent method holds great potential for widespread applications in medical assistance analysis and daily health monitoring.