Design Of Plantar Type Recognition Based On Intelligent Insole

Based on intelligent insole type recognition design,tracking the frontier technology involved in gait recognition at home and abroad,the gait recognition technology based on gait pose and the gait recognition technology based on plantar pressure are studied.The gait recognition technology based on gait and attitude feature is easy to be interfered in the actual action.In the scene with large flow of people,the image is easily occluded,and the remote image is easily disturbed by the cloudy day and rain and smoke when the image is acquired.Factors such as video image alignment inconsistencies can occur in the identification of health and injured or weight-bearing persons,resulting in a decrease in the accuracy of identification,and the use of human walking due to physical condition,mood,posture,lifestyle,and so on.With the development of modern electronic information technology and sensor technology,more accurate plantar pressure acquisition data can be obtained through the development of modern electronic information technology and sensor technology,which leads to different plantar pressure,and has unique and stable characteristics.The intelligent insole system of gait recognition based on sole pressure is designed and the experiment and analysis of recognition of pes cavus is carried out.The design of intelligent insole is studied.The intelligent insole equipment is miniaturized,and the number of electronic components is reduced by channel reuse.The detail of the analog circuit is analyzed,including the conversion from analog to digital,the selection of the best sampling rate,and the design of the instrument insole with pressure and inertia sensors.Therefore,on the basis of the research on the existing gait recognition technology,the foot pressure characteristic analysis,convolutional neural network analysis and the hardware circuit design of the intelligent insole are carried out.On the basis of the experimental data collection and classification,the hardware circuit of the intelligent insole is designed.The research on the recognition of pes cavus by using intelligent insole is carried out.The results show that the accuracy of plantar type recognition based on convolution neural network is 96% and the expected effect is achieved.The main contents of this thesis are as follows: firstly,the types of gait recognition and the role of gait recognition are analyzed.Describes gait recognition techniques of other biometric features such as face,fingerprint,iris,and their advantages and disadvantages,and then analyzes the identification techniques of gait recognition compared with other biometric features such as face,fingerprint,iris,and so on,and analyzes the technology of gait recognition compared with other biometric features such as face,fingerprint and iris.It has the advantages of non-contact,non-invasion,easy to perceive,difficult to hide and difficult to disguise.Two main recognition methods of gait recognition,one based on gait posture and the other through gait recognition by foot pressure,are analyzed.The main characteristics and development potential of plantar pressure identification are analyzed.The first kind of gait recognition has developed well and has been applied in some places with higher safety level.However,gait recognition based on plantar pressure is rarely used at present,but it can more reflect the biological characteristics of human body.The information obtained is more accurate and difficult to hide.Now gait information can be used in sports training to help athletes to improve their performance,to help criminal investigation,to identify criminal suspects through gait;Help people who have had foot surgery or inconvenient legs and feet to learn about their health,and so on.Secondly,according to the design requirements of the intelligent insole based on plantar pressure,the plantar pressure distribution is carried out,and the plantar pressure information is collected and divided into four parts,six parts and eight parts by dividing the plantar pressure information into four parts,six parts and eight parts,and the pressure information of the foot is divided into four parts,six parts and eight parts.Then the plantar information of each area is collected separately.The study of foot pressure support,foot joint structure,foot pressure formation process,for intelligent insole sensor selection,motion data acquisition basis.Thirdly,the theory principle,structure and function of each module are analyzed.The intelligent insole including control system,data transmission system and data processing system is designed and made.After completing the design of hardware circuit,the hardware circuit is tested.Inertial sensor and pressure sensor are calibrated to realize the data acquisition of plantar pressure using intelligent insole.Finally,the designed hardware is used in the recognition of the pes cavus,and a portable system is designed for the detection of foot deviation.The one-dimensional convolution neural network is used to extract the feature and classify the data,and the trained model is used to identify the foot deviation.The data proves that the method of foot type recognition based on convolution neural network is feasible.The accuracy of recognition of pes cavus can reach 96%.