Lower Limb Motion Recognition Method Based on Multiscale Gauss-Markov Random Field Model
A random field model and motion recognition technology, applied to pattern recognition in signals, character and pattern recognition, computer parts, etc., can solve the problems of poor reliability and low accuracy of recognition technology
Active Publication Date: 2019-07-19
HEFEI INSTITUTES OF PHYSICAL SCIENCE - CHINESE ACAD OF SCI
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] In order to overcome the problems of poor reliability and low accuracy of the existing motion pattern recognition technology in the auxiliary motion system, the present invention proposes a lower limb motion recognition method based on a multi-scale Gauss-Markov random field model, in order to extract stable The local characteristics of the signal can overcome the influence of signal noise caused by motion instability, thereby improving the recognition accuracy and prediction reliability of the lower limb motion pattern, and providing technical support for the stable control of the auxiliary motion system
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0051] The following describes the specific implementation of the present invention in conjunction with the accompanying drawings and examples. In this embodiment, a method for automatic recognition of human lower limb movement patterns based on a multi-scale Gauss-Markov random field model, the overall process is as follows figure 1 As shown, the collected motion data is first preprocessed and the data feature map is constructed, and then the data feature map is decomposed by multi-scale wavelet and the parameters of the Gaussian model are initialized by using the C-means clustering algorithm, and the observation field of each scale image is iteratively updated The parameters of the Gaussian model are combined with the segmentation results of the data feature maps of all scales, and the segmentation boundaries are processed according to the voting principle. Finally, the classification results are matched with the standard sets of each motion mode to determine the mode attribu...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a method for automatic identification of motion patterns of human lower limbs based on a multi-scale Gauss-Markov random field model, comprising: 1 collecting raw data of C kinds of motion patterns of human lower limbs, preprocessing the raw data, and constructing data Feature map; 2 Multi-scale decomposition of signal and modeling of characteristic field; 3 Set the number of iterations Q, iteratively update the Gaussian model parameters {μ c , σ c , π c}; 4 Match the classification result with the standard set of each motion mode to judge the mode attribution, and eliminate the problem of blurred boundaries of the temporally segmented regions of different motion modes according to the voting criteria. The invention can extract stable signal local features, overcome the influence of signal noise caused by motion instability, thereby improving the recognition accuracy and prediction reliability of lower limb motion patterns, and providing technical support for the stable control of auxiliary motion systems.
Description
technical field [0001] The invention relates to the field of pattern discrimination of human body motion parameter measurement signals, in particular to the prediction and recognition technology of human lower limb daily behavior motion intention. Background technique [0002] The automatic recognition of human lower limb motion mode is one of the key technologies to realize the active control and efficient human-computer interaction of intelligent prosthetics and assisting disabled exoskeleton robot systems. In complex daily behavioral sports scenes, how to keenly obtain various motion intentions of the human body, and realize the active assist control of the robot system according to the real-time motion intentions of the human body, so as to achieve the robot system to quickly and accurately adapt to the human body's response and realize natural motion mode, which becomes the difficulty to be solved by non-motion hysteresis control. The existing motion intention recognit...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00
CPCG06V40/23G06F2218/02G06F2218/12
Inventor 王玉成孔令成叶晓东王众辉
Owner HEFEI INSTITUTES OF PHYSICAL SCIENCE - CHINESE ACAD OF SCI