A preprocessing method of walking gait signal based on mobile phone acceleration sensor

Abstract

The invention discloses a walking gait signal preprocessing method based on an acceleration sensor of a mobile phone. Abnormal signals and burrs are processed by a Gaussian smooth iterative method, and then the acceleration sensor signals of the swing period and the support period of the walking gait are extracted by the trend ergodic method. It can accurately identify the single stride cycle of the continuous walking gait data, and segment the continuous walking gait signal data into uniform length data, and group the data into uniform length data. package storage. The method of the invention solves the problems in the prior art that the amplitude threshold method and the window extremum method are used to process the walking gait signal, and the existing parameters such as amplitude and period vary greatly, the threshold and the window width are difficult to select, and the selection is unreasonable. Threshold value and window width affect the accuracy of gait signal cycle segmentation and have poor effect on data mining.

Description

technical field [0001] The invention belongs to the technical field of walking gait signal preprocessing, and in particular relates to a walking gait signal preprocessing method based on an acceleration sensor of a mobile phone. Background technique [0002] Gait recognition is the recognition of human identity through the analysis of human or animal gait. It has the advantages of non-contact and long-distance, not easy to camouflage, and non-invasive. It has become a new biometric recognition technology. [0003] Gait refers to a periodic form and appearance of people or animals moving forward through limb movements, and is a very complex behavioral feature. At present, the acquisition of gait features through acceleration sensors is one of the most important detection methods. Portable devices such as mobile phones and wristbands generally have built-in acceleration sensors, which have attracted attention due to their high popularity, high frequency of use, and no addition...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the defects and deficiencies of the above-mentioned prior art threshold or window width selection difficulty, to provide a walking gait signal preprocessing method that does not require preset values ​​such as threshold and window width, through Gaussian smoothing iteration Abnormal signals and glitches are dealt with by using the trend traversal method, and then the trend and law of the acceleration sensor signal in the swing period and support period of the walking gait are extracted by the trend traversal method. The technical pain points of the threshold method and the window extremum method can accurately identify a single stride cycle of continuous walking gait data, and interpolate the continuous walking gait signal data into uniform length data, and store them in a package

Images