A self-adaptive kicking state recognition method and device based on a nine-axis sensor

Abstract

The invention discloses an adaptive kicking state recognition method based on a nine-axis sensor and a device thereof. The recognition method includes the following steps: first, bind the nine-axis sensor to the user's calf, initialize data and collect the user's movement The latest frame of data includes three-dimensional attitude angle, three-axis acceleration and three-axis angular velocity data; secondly, the calculated combined angular velocity and three-axis acceleration are used together to calculate the reference plane, calculate the kick mark KICK, and determine and output the current state as "Tend to kick", "in kick", "keep kick" and "not kick" four states and the corresponding kick duration, at last, the data is updated and stored, the present invention The recognition method and its device process the calf movement information, which can further modularize the movement of the lower limbs. It has the advantages of fast response, higher accuracy and strong self-adaptation, and has the ability of self-adaptive learning. Differentiate identification.

Description

technical field [0001] The invention belongs to the field of human motion detection, in particular to a method of acquiring triaxial angular velocity, triaxial acceleration and attitude angle information of the thigh during human motion, and proposes an adaptive kicking state recognition based on a nine-axis sensor according to the data method and apparatus. Background technique [0002] With the popularization of computers and the development of virtual games, the new human-computer real-time interaction has gradually become a new hotspot in computer technology, and the ability to instantly recognize human motion posture as a new human-computer interaction method has been paid more and more attention. Among them, the recognition of lower extremity movements has a wide range of application prospects, and has positive research significance in the fields of health monitoring, rehabilitation, and virtual sports fitness games. [0003] The knee joint of the human body divides t...

AI Technical Summary

Problems solved by technology

[0003] The human knee joint divides the leg into two main parts, the thigh and the calf. The thigh determines the general trend of the lower limb movement, and the calf refines the movement of the lower limb. Most of the existing leg movement recognition technologies are It processes the motion information of the thigh or the sole of the foot. The present invention directly processes the motion information of the calf. In the existing motion recognition methods based on nine-axis sensors, most of the motion recognition methods in the past time interval The motion state is characterized by feature extraction and then classified. This scheme needs to be recognized relatively reliably after the recognized action is basically completed. For example, the recognition of kicking action. The data is fused, and then features are extracted for discrimination and recognition. Such solutions generally have shortcomings such as untimely response, insufficient accuracy, poor self-adaptation, and inability to distinguish different individual exercise habits.

Images