A Recognition Method For Hand Motion Direction Based On Charge Induction

With the development of science and technology,human beings have stepped into the era of electronic products with computers as the core.At the same time,smart home technology is also accompany also.Hand gesture is a more natural,comfortable and effective human-computer interaction method in the field of smart home as it can operate various devices conveniently.Therefore,many scientists are researching in the detection and recognition of hand motion.The representation of the existing technology includes data glove,acoustic technology and machine vision technology,etc.Although the data glove technology has good real-time performance and high accuracy,users need to wear bondage equipment all the time,which is lack of comfort and natural feeling.The acoustic technology,is not affected by light and does not require users to wear device on the hand,but the detection distance is less than 10 cm and the control range is limited.Although the technology based on machine vision can recognize various specific gestures with high accuracy,it is more or less limited by obstruction and complex light conditions,so it requires expensive equipment and complex algorithms to ensure the accuracy of recognition.Actually,in some applications,such as the smart home,it is often possible to control multiple functions of the device with simple motion of the hand rather than recognizing precise gestures.Therefore,the aim of this paper is to propose a method of hand motion direction recognition based on charge induction in order to find a costefficient interaction technology for intelligent control.Firstly,the theoretical derivation of charge detection for moving target is introduced and experiment verification of hand motion is given.Then,the direction recognition model is established and the simulation and accuracy analysis is illustrated,which is found that the detection error is decreased with the decrease of the time measurement error and the motion velocity.Finally,a hand motion recognition system based on charge induction was established,including hardware design and software design and the experiments for hand motion direction recognition are carried out under conditions of full occlusion,darkness and body motion interference.Without the need of wearing device on the hand,the proposed method can achieve non-contact recognition for hand motion direction under the conditions of full occlusion and darkness.For the four wave directions,the method works well when the humidity is 30 %-70 %,and the detection error is no more than 2.5 %.In addition,when there is human motion interference,the recognition accuracy of the four directions is nearly 97 %.