Design And Application Of Gesture Recognition System Based On Infrared Sensor

With the continuous progress of science and technology,gesture recognition interaction technology has gradually become a research focus and difficulty.Because of its natural and intuitive input mode,it has brought users a harmonious and natural interaction.Traditional gesture recognition is mainly based on portable sensors or computer vision.The first type requires users to bring professional devices,and the human-computer interaction experience is not "natural" and the application scope is narrow.The latter improves the user experience,but is affected by various environmental factors or the accuracy of equipment and instruments,resulting in strict usage requirements and complex recognition algorithms.Therefore,this paper proposes a gesture recognition method based on infrared sense to solve the above problem.This method collects the feature threshold of the gesture to achieve the purpose of gesture recognition.In order to improve the accuracy of gesture recognition,this paper combines the convolutional neural network(CNN)and long short term memory(LSTM)hybrid models to classify and recognize rotating gestures.Finally,the gesture recognition system designed in this paper is implemented on the household appliances.The main research content is as follows:First,in order to design a gesture recognition system based on infrared sensor,the infrared array model is constructed on the principle of infrared sensor and gesture recognition,and the gesture types are defined.Secondly,in order to obtain the different data of different gesture movements,a gesture recognition module based on infrared sensor is designed,and a gesture data acquisition platform is built based on this module to realize the data visualization of different gesture movements.Thirdly,in order to improve the accuracy of gesture recognition,the length of the extracted data is unified and normalized to achieve the consistency of the sampled data.To address the issue of insufficient accuracy in gesture classification and recognition,a feature threshold classification algorithm and a CNN LSTM network model classification algorithm are designed to classify and recognize different types of gestures.Finally,on the basis of the above work,a gesture recognition system based on infrared sensor is completed,and this system is applied to the range hood to achieve non-contact gesture recognition on the range hood.In the application of this system,the functional control board of the range hood is connected to the gesture recognition module,the purpose of controlling the range hood through non-contact gesture is achieved,and the control is smooth and natural in practical application.