A human body gait recognition method based on micro-Doppler features and a support vector machine

Abstract

The invention discloses a human body gait recognition method based on micro-Doppler features and a support vector machine. The method comprises the following steps: S1) collecting posture data of a human body when marching through a radar; S2) carrying out analysis on the posture data through a time-frequency analysis tool to obtain a corresponding time-frequency picture; S3) extracting bandwidth characteristics and bias characteristics from the time-frequency picture, wherein the bandwidth characteristics indicate span range of positive and negative micro-Doppler frequency caused by gait, and the bias characteristics indicate deviation of the positive and negative micro-Doppler frequency with respect to center frequency, bias value indicating symmetry of swing arms; and S4) inputting the bandwidth characteristics and the bias characteristics to the support vector machine for posture recognition to determine the posture corresponding to the posture data. The method has the following advantage: the method for human body gait classification and recognition through radar data acquisition and micro-Doppler feature extraction is high in gait classification accuracy.

Description

technical field [0001] The invention relates to the field of radar technology and pattern recognition, in particular to a human gait recognition method based on micro-Doppler features and a support vector machine. Background technique [0002] The recognition of human body posture and gait is an emerging technology, and its importance is becoming increasingly apparent. Correctly and quickly identifying the posture and gait of the target human body is of great significance for judging whether the target individual is friendly, judging its threat level, and providing effective evidence for countermeasures. Related needs widely exist in anti-terrorism, security, monitoring and other fields. [0003] Most of the real-time target posture or gait recognition systems that have been put into practical use are based on video surveillance and image processing technology. The recognition accuracy of image-based recognition systems is limited, the recognition process has a large amount ...

AI Technical Summary

Problems solved by technology

[0003] Most of the real-time target posture or gait recognition systems that have been put into practical use are based on video surveillance and image processing technology. The recognition accuracy of image-based recognition systems is limited, the recognition process has a large amount of calculation, and is greatly affected by environmental factors such as lighting. big error

In addition, the image-based human gait recognition system does not have the ability to work around the clock, which limits its application range

Images