Human posture discrimination method based on support vector machine

Abstract

The invention provides a human posture discrimination method based on a support vector machine. The method comprises steps: a sliding window mechanism is used to segment sample data, and original dataare divided to a plurality of subsets; a three-axis combined acceleration mean value, a three-axis combined acceleration standard deviation, a covariance and a skewness value of each subset are calculated as feature vectors of the corresponding subset, and a classification label value is set for each subset; according to the feature vectors, each corresponding fuzzy factor is calculated; the fuzzy factor and the classification label value are used to train the support vector machine, and the optimal hyperplane decision function for posture classification is acquired; and with the well-trainedsupport vector machine as a classifier, re-sampled human body three-axis acceleration data are classified. A data-oriented machine learning method is adopted to falling posture discrimination, the traditional method of adopting a two-norm based on the human body three-axis acceleration values as a threshold to judge whether a person fall or not is replaced, and the detection rate is enhanced.

Description

Technical field [0001] The invention relates to the field of human body posture discrimination, in particular to a human body posture discrimination method based on a support vector machine. Background technique [0002] With the gradual aging of the population in various parts of the world, the entire international community is paying more and more attention to the problem of aging. Falling is a frequent accident of the elderly, which seriously affects the physical and mental health and even threatens the safety of life. According to a report by the National Security Council of the United States, among people over 65, deaths caused by falls rank first among all accidental deaths, accounting for 33% of accidental deaths in this age group. In the traditional human body gesture recognition method, the two-norm based on the three-axis acceleration value of the human body is generally used as the threshold to judge whether the person falls. This method has poor measurement accuracy ...

AI Technical Summary

Problems solved by technology

In the traditional method for human posture recognition, the two-norm based on the three-axis acceleration value of the human body is generally used as the threshold to judge whether the person has fallen. This method has poor measurement accuracy and is prone to false alarms.

Images