Research Of Hand Gesture Capture System Based On Inertial Sensor

Gesture is a nature and intuitive, easy to learn and effective means of communicationwhich is widely used in people’s daily life, it has strong expressive ability, with thedevelopment of virtual reality technology, virtual hand has become a new kind ofhuman-machine interface application, which can complete the perception of exchange andother functions during the interactive process by obtaining the hand motion information inreal-time, all these effectively broaden the human-machine interaction channel, and has abroad application prospect in sports, medicine and military. In recent years, with thedevelopment of microelectronic technology and inertial navigation technology, inertial sensorcan realize the detection of the motion information and measure or track the attitudeinformation of hand accurately, which bright new basic tools and application environment forcapturing the gesture information. Also it has the advantages of simple structure design, lowcost, convenient installation and etc. In this article, combining with the strap down inertialnavigation and data fusion theory, we designed a kind of under illumination constrains, stronganti-interference ability of all attitude gesture capture system based on micro inertial sensortechnology. The gesture capture system consists of multiple sensor nodes and a convergencecontrol nodes, the sensor node consists mainly of three axis accelerometer, three axisgyroscope and three axis magnetometer, which is responsible for the original data collectionand easy treatment of each finger position in the process. Control node collects the data ofeach sensor node, and finally sent the packed data to the host computer through the wirelesstransmission mode, in the host computer, we can realize the data driving the virtual handmodel through further analysis on the hand gesture data, finally we can realize the capture anddisplay for the real hand gesture.In this article, so as to select the appropriate sensors, we conducted the research bystudying the finger motion in3D space, and compared the working principle and performanceparameters of the sensors, furthermore, it carried on the analysis of the possible error of eachsensor. For example, it made a detailed analysis of the data jitter and skew of the accelerationsensor, the accumulated error of gyroscope, and putted forward the effective solution. It alsomade detailed schematic design for the sensor node and sink node which deployed in the hand,and designed the system data transmission protocol. At the same time, it carried on theanalysis to the deployment location of sensor nodes in the fingers, and completed the fusionprocessing of multi sensor data based on the basis of inertial navigation technology, finally theattitude information was outputted by quaternion. In addition, through the analysis of handbiological structure, which built the virtual hand joint model based on three-dimensional perspective by combining with the VS2012development environment in the OpenGLgraphics library and3DS MAX software, at last, we realized the capture and reproduce of themovement gesture through the virtual model initialization calibration and data driven by theattitude data calculation.It proved that the hand gesture capture system has stable and reliable workingperformance through the experiment, which can provide new references for future researchand design of gesture capture system, also it has certain innovation and application value.