Support vector machine-based pedestrian gait classifying system and method

Abstract

The invention discloses a support vector machine-based pedestrian gait classifying system and a support vector machine-based pedestrian gait classifying method. The system comprises a data acquiring module, a data processing module and an SVM (Storage Virtualization Management) module, wherein the SVM module is provided with a storage medium; the data acquiring module acquires gait training data and transmits the gait training data to the data processing module to be processed; the data processing module outputs the input quantity of the acquired gait training data to the SVM module to be studied to obtain an SVM classifier which is then stored in the storage medium; during real-time test, after real-time gait information is acquired in real time by the data acquiring module, data processing is preformed by the data processing module to acquire the input quantity of the test data; the input quantity of the test data is classified by the SVM module according to the SVM classifier to obtain real-time gait type. A static interval in a pedestrian moving process can be precisely detected by the system and the method, and the positioning precision is improved.

Description

technical field [0001] The present invention relates to a technique in the technical field of inertial navigation, in particular to a system and method for realizing real-time classification of pedestrian gait by using a Support Vector Machine (Support Vector Machine, SVM). Background technique [0002] The purpose of gait classification is to classify pedestrian gait into walking and running categories, and use more reasonable and efficient detection methods for different gaits in pedestrian inertial navigation. In the field of pedestrian inertial navigation, the static detection method for pedestrian walking is not suitable for running gait, so the running positioning accuracy under the same detection method is low. Effective identification of running gait will help to adopt more targeted static detection methods and improve the positioning accuracy of pedestrian inertial navigation algorithms. [0003] SVM is a linear classifier based on statistical learning theory. Its...

AI Technical Summary

Problems solved by technology

Therefore, the analysis of data in the time domain by the above-mentioned technology is not representative enough, and the results obtained are not accurate enough.

Images