Research On Human Position Determination Based On Infrared Technology

The continuous development of science and technology has promoted the further improvement of people's living standards,especially in recent years,the rapid development of smart home,so that people's way of life has undergone tremendous changes.As an important part of smart home,human positioning technology plays an indispensable role.Determining the location of people in the home environment has always been the focus and difficulty in the field of smart home.Only by accurately detecting and locating the number and location of human body in the environment,can the advantages of automation of smart home be better brought into play.Therefore,more and more attention has been paid to the research of location determination of human body.Nevertheless,there are some shortcomings in the commonly used positioning technology,which can not meet the requirements of low power consumption and high recognition in smart home.In this paper,two different infrared sensors are synthesized,and a new method of human body location is proposed,which improves the existing technology,improves the recognition ability and accuracy of the location system for human location and number,and hopes to better meet the requirements of smart home and promote the development of smart home.The method of human body location proposed in this study is to synthesize two infrared sensors: D6T-8L-06 infrared sensor and D203 S pyroelectric sensor to form an infrared detection system to work together to determine the position of human body,and to test the accuracy of the system through experiments.Firstly,infrared radiation sources in the surrounding environment are detected by the detection system composed of D6T-8L-06 and D203 S infrared sensors.The acquired data are stored in the memory of STM32F103ZET6.The information acquired by D6T-8L-06 detectors is clustered by K-means to determine the location of radiation sources.Then,the different radiation sources in the surrounding environment are confirmed by morphological analysis,and the elimination part is eliminated.The interference of non-human heat source can obtain the position information of human body.Two different methods of feature extraction are used to extract the information collected by D203 S pyroelectric infrared sensor: feature extraction based on FFT and feature extraction based on WPE.Then two different featureextraction methods are classified and recognized by Fisher classifier.Finally,two kinds of infrared information are fused to determine the recognition results.The experimental results show that the infrared system can collect the heat source data,cluster the data collected by D6T-8L-06 sensor and discriminate morphologically,and then filter out the interference of most non-human heat sources in the environment,so as to realize human body localization.After extracting WPE features from the information obtained by D203 S pyroelectric infrared detector,the interference heat source in the experimental environment can be effectively filtered by Fisher linear classifier,after weighted fusion of data layers,the recognition accuracy can reach 97.17%.The expected effect was achieved.Indoor human body positioning technology has developed rapidly in recent years,and has a good development prospects.As an important part of indoor positioning technology,infrared detection still needs further exploration and research in many aspects.The method proposed in this paper improves the existing indoor positioning technology,improves the positioning system's ability to identify the location and number of people and recognition accuracy,can better meet the requirements of smart home,and is expected to promote the development of smart home equipment.