Behavior detection method and system based on wireless radio frequency signals

Abstract

The invention discloses a behavior detection method and system based on a wireless radio frequency signal. The method comprises the following steps: acquiring a to-be-detected object to obtain WiFi information; processing the WiFi information to obtain detection data capable of inputting a time sequence Unet; constructing and training a time sequence Unet to obtain a time sequence Unet model; and inputting detection data into the time sequence Unet model to obtain a detection result. The method for detecting the behavior event in the N-to-N mode is realized, namely, a section of wireless radio frequency signal with the length of N is input, and an action category corresponding to each sampling moment is output; the behavior type can be detected, the starting and ending time of the behavior can also be detected, and the method can be applied to downstream tasks sensitive to action time.

Description

technical field [0001] The invention belongs to the technical field of the Internet of Things and artificial intelligence, and in particular relates to a behavior detection method and system based on wireless radio frequency signals. Background technique [0002] Nowadays, camera-based solutions are generally used to study human behavior recognition tasks, such as fall detection, gesture recognition, and typing recognition. There are two main disadvantages: First, when the camera is used in a home scene, it is easy to cause users to worry about important privacy leaks such as faces. Second, the camera is not penetrating. When the user is behind obstacles or walls, camera-based solutions cannot be used for user action recognition. Therefore, the camera solution has more restrictions. [0003] However, the existing mode of behavior recognition tasks based on radio frequency signals also has deficiencies. It is manifested as an "N to 1" mode with a coarser time granularity, t...

AI Technical Summary

Problems solved by technology

There are two main disadvantages: First, when the camera is used in a family scene, it is easy to cause users to worry about important privacy leaks such as faces.

Secondly, the camera is not penetrating, and when the user is behind obstacles or walls, the camera-based solution cannot be used for the user's action recognition

Therefore, the camera solution has more restrictions

[0003] However, the existing mode of behavior recognition tasks based on radio frequency signals also has deficiencies. It is manifested as an "N to 1" mode with a coarser time granularity, that is, inputting a radio frequency signal sequence of length N and outputting the sequence The corresponding action category, but cannot detect the user's action category at each sampling moment within a period of time

Images