Quality evaluation method for wavelet de-noising in gait recognition
TU Bin-bin
GU Li-hua
XU Hui
Abstract：In order to overcome the difficulties in gait analysis,cycle division and feature extraction for gait acceleration singals due to the fact that the acquired signals contain a large number of higher frequency noises such as peaks or mutation,a new method,where the quality evaluation for the wavelet de-noising was adopted to select the optimal parameters for the wavelet de-noising of acceleration signals, was proposed.According to the geometric and physical meanings of quality evaluation indexes,the paramenter selection for both decomposition and reconsitution levels of optimal wavelet in the wavelet de-noising were guided by the evaluation index value.When the evaluation index took the minimum value,the correspongding decomposition level was optimal,and the used wavelet basis funtion was also optimal.The results show that the proposed method can provid the theoretical basis for the selection of both optimal wavelet basis funtion and decomposition levels for the wavelet de-noising of gait signals in the field of gait recognition.
Keywords：gaitgait recognitionacceleration singalwavelet de-noisingquality evaluationwavelet basis functiondecomposition and reconstructionoptimal parameter
Publication Date：2017-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：6( 61-66 )