Research Of Static Gait Recognition Based On Distribution Of Plantar Pressure

In recent years, gait recognition has become more and more popular in the field of biological characteristics, which is based on characteristics of human walking for identity recognition. At present, most of research on gait recognition is realized by computer vision. However, the method of distribution of plantar pressure is just used for data analysis. In addition, the target is mainly concentrated on medical diagnosis, analysis of curative effect and physical exercise. The esearch of distribution of plantar pressure for identification is less. Therefore, this thesis did some research about gait recognition.The whole process of gait recognition includes data collection, data preprocessing, feature extraction, classification and recognition, among which feature extraction is the most important. This thesis mainly adopted two kinds of methods for feature extraction on the distribution of plantar pressure and then used support vector machine for classification one by one.The distribution of plantar pressure is non-linear, so the stability and robustness of feature will be very low if extract feature directly for gait recognition. Firstly, a method of gait recognition based on improved characteristics of spectrum and distance of foot edge is proposed in the third chapter of this thesis. Secondly, according to the fact that the shape of object is an important feature of object, but many researchers just used information of planar pressure or contour of foot edge, a novel method of static gait recognition based Fourier descriptor and singular value decomposition is proposed in the fourth chapter of this thesis. The novel method combines Fourier descriptor and singular value decomposition as a new feature of information of plantar pressure and foot edge for gait recognition. Finally, this thesis classified the extracted feature by support vector machine and listed the analysis of experimental results, which provides reference for the future research of gaitrecognition based on distribution of plantar pressure.