Foot-to-ground acting force-based gait feature extraction method and gait identification system

Abstract

The invention discloses a foot-to-ground acting force-based gait feature extraction method and a foot-to-ground acting force-based gait identification system. The system comprises a gait data acquisition module and a gait data processing and identification module, wherein the gait data acquisition module acquires the foot-to-ground acting force during walking in a hidden and non-intrusive way; and the gait data processing and identification module extracts gait features and performs gait identification from the foot-to-ground acting force. The gait feature extraction method comprises the following steps of: denoising the foot-to-ground acting force by a wavelet transform hard threshold method; detecting a key point and calculating gait time-domain features by a first-order difference algorithm; and representing the gait frequency-domain features by adopting a wavelet decomposition coefficient after performing waveform alignment on the foot-to-ground acting force. When the system is used for identifying, an optimal gait feature set is selected by a fuzzy C-means algorithm; and the classification or the identification can be performed on the gait based on a support vector machine. The gait information acquired by the method and the system is natural, true and not easy to simulate; the extracted gait features are full and accurate; and the requirements on gait analysis, classification and long-distance gait identification can be met better.

Description

technical field [0001] The invention belongs to the technical field of pattern recognition, in particular to a gait feature extraction method and a gait recognition system based on foot-ground force. Background technique [0002] Gait recognition uses people's walking posture to realize the identification of personal identity or the classification and recognition of physiological, pathological and psychological characteristics of gait. That is, biological characteristics can be carried with you, do not require memory, and are unique and difficult to replicate. At the same time, gait recognition has unique advantages that other biometric recognition technologies do not have, that is, gait information can be collected in a long-distance, concealed and non-contact manner, and gait is difficult to hide and camouflage. It is a long-distance biometric recognition technology. With this, gait recognition has broad application prospects, especially in the identification or authentic...

AI Technical Summary

Problems solved by technology

In addition, in gait-based identification and gait classification or recognition applications based on physiological, pathological and psychological characteristics of gait, gait features are the key, and most of the current gait feature extraction methods focus on image sequences. Lie transform, wavelet transform, and Radon transform characterize the kinematic characteristics of gait, but in fact gait also has dynamic characteristics, but because the existing technology fails to extract gait features from multiple angles and levels, it is impossible to comprehensively Accurately extract gait features that reflect gait kinematics and dynamics

[0004] Chinese Patent No. ZL01144157.7 discloses a method for identifying people through gait image sequences of human walking, but this method cannot perfectly solve the problems of complex background, occlusion and noise interference due to the use of ordinary cameras to capture gait images , cannot be applied to nighttime outdoors without light, and does not consider the problem of deliberately imitating gait

Chinese Patent Application Publication No. CN101251894A discloses a gait recognition method and system that can be used day and night, because it is still based on the gait image recognition method, it still cannot deal with the problems of deliberate imitation and occlusion

Chinese Patent No. ZL200410014352.9 discloses a multi-axis force platform array composed of four six-axis force platforms, and a method for obtaining dynamic gait information when the human body walks through the platform array. The platform array can be used for human balance ability, Testing and analysis of coordination and nervous system function, but does not involve methods of gait feature extraction and gait recognition

However, this personal identification system has two deficiencies: one is that the acoustic oscillation when the microphone is placed on the human body changes with the position of the microphone on the human body; It is difficult to accurately extract the characteristics of the gait waveform

Patent Application International Publication No. WO / 2004 / 040501 discloses a method that uses the peak amplitude of the waveform in the specified frequency band as a reference index, and determines the gait waveform of one step and extracts the gait waveform from the electrical signal detected by the electric field displacement detector. The characteristic method and personal identification system, this method is not affected by the charge interference between the left and right legs and the installation position of the sensor on the body, but because electromagnetic interference will seriously affect the accuracy of detecting electrical signals, it is not suitable for complex In the environment of electromagnetic interference

Moreover, in the technical solutions disclosed in U.S. Patent US 2002 / 0107649A1 and International Publication No. WO / 2004 / 040501, its gait information is still collected by sensors placed on the body, and it is still a contact measurement method. Perceived, it is easy for criminals to adopt anti-reconnaissance methods to muddle through or be provoked to create terror among the crowd

Images