Multi-channel time sequence gait analysis algorithm based on direct feature extraction

Abstract

The invention discloses a multi-channel sequential gait analysis algorithm based on direct feature extraction, and relates to the field of gait analysis methods.The method comprises the following steps: shooting gait video information of a plurality of subjects, segmenting all the video information into single separated gait periods, and adjusting the single separated gait periods to be the same length; calculating the distance of the standardized gait parameter Sz, and clustering the marked training data by using a k-nn algorithm and a parameter distance equation; classifying gait phases andreconstructing the gait phases, selecting a point with the maximum correlation coefficient in the current time sequence as a gait event, re-segmenting the gait according to the point, and iterating the above operations again: calculating the correlation coefficient between adjacent gaits according to the gait classification segmented by the gait event obtained by the previous iterative calculation, and finding out the point of the global maximum correlation coefficient as the gait event to optimize the gait classification of the last iteration until the classification converges. The method caneffectively improve the extraction efficiency and the calculation efficiency.

Description

technical field [0001] The invention relates to the field of gait analysis methods, in particular to a multi-channel time-series gait analysis algorithm based on direct feature extraction. Background technique [0002] With the aging of the population, the number of patients with common diseases of the elderly, such as Parkinson's syndrome, has increased sharply. When the disease is severe, the patient will completely lose the ability to exercise, which will bring great financial and nursing pressure to the patient and family. , early diagnosis and treatment of Parkinson's syndrome can effectively alleviate the deterioration of the condition. [0003] As an inherent physiological feature of the human body, gait can play a role in many fields such as intelligent prosthetics, intelligent monitoring, clinical medicine, rehabilitation therapy, and motion analysis. Through gait detection and recognition, the behavior of the moving human body can be analyzed, and then the abnorma...

AI Technical Summary

Problems solved by technology

This method is accurate in measuring gait, but it is costly and requires a professional measurement site; there are also some systems that use inertial sensors and force plates, which require accurate sensors for calibration and synchronization, and have high requirements for sensors ,higher cost

[0006] At present, there is a kind of method based on vision. This method extracts the outline or skeleton movement of the subject's video, and then uses the mobile platform to calculate the corresponding results. This method not only has high requirements for the equipment to extract the movement. , and the computing power of the mobile platform is high. With the development of science and technology, the existing portable depth camera can effectively estimate the human body posture, thereby greatly reducing the calculation of image processing and reducing the computing power of the mobile platform. , so that the vision-based gait analysis device has been effectively developed, but the existing devices have low extraction efficiency and calculation efficiency

Images