Method Of Finger Motion Recognition Based On PVDF Sensor Array

Finger motion recognition is one of the key technologies of human-computer interaction based on gestures.Today,there are two main types of method of finger motion recognition recognition,one is finger motion recognition based on vision sensors,and the other is finger motion recognition based on motion sensors.These two methods are the main research hotspots in the field of gesture recognition,but they cannot be widely used because of various problems and limitations.Firstly,the gesture recognition based on vision sensors generally has the problem that the application environment is limited,and there are high requirements for factors such as ambient light and background.Secondly,the gesture recognition based on motion sensors is generally designed as a device to detect the gesture of finger movement.For example,data gloves or thumbwheel mouse,they often require additional switching devices,while also completely occupying the user's hand.In order to solve the problem of hand occupation,during the work of this dissertation,we proposed a method to indirectly recognize finger motions by detecting motion signals of wrist muscles with motion sensors,where the motion sensors are pressure sensors made of polyvinylidene fluorid(PVDF)film.The main work content and innovations of this dissertation have the following three aspects:1.The discrete PVDF sensor array for sensing wrist motion signals was researched.In this dissertation,through analyzing the physiological of the hand and experimental verification,we studied a sensor array structure that can fully sense the wrist motion signals based on PVDF sensors.Then we designed a acquisition system based on the PVDF sensor array,and made a wearable collection device.2.The miniaturization technology of PVDF sensor and its array are studied.Firstly,the structure of miniaturized PVDF sensor unit is improved,and the preparation process of the miniaturized PVDF sensor are studied.Secondly,the performance detection technology of the sensor is studied,and the experimental platform is built.Finally,the structure of the miniaturized sensor array is designed preliminarily,and made a wearable collection device.3.A set of finger motion recognition algorithm for multi-channel PVDF sensor array was proposed.The algorithm includes an algorithm for signal preprocessing,an algorithm for extracting finger motion fragments,a method for constructing feature vectors,a method for establishing a finger motion database,and a four-layer BP neural network.We used this algorithm with the PVDF sensor array to complete thefinger motion recognition experiment.The result shows that the use of the PVDF sensor array to detect wrist motion signals and the finger motion recognition algorithm can effectively achieve the detection of finger motion and its semantic recognition.Through the research during the dissertation work,the wrist sensor array based on the PVDF motion detection sensor we proposed can effectively realize finger motion recognition through wrist detection.It is not only convenient and comfortable to wear,but also realizes the complete liberation of the user's hand.It will be a mobile terminal device that is competitive in the field of human-computer interaction.