A Human Behavior Data Segmentation Method Based on Mahalanobis Distance
A Mahalanobis distance and data segmentation technology, which can be used in instruments, computing, character and pattern recognition, etc., and can solve problems such as accurate segmentation of difficult boundaries.
Active Publication Date: 2020-11-17
XIAN UNIV OF TECH
View PDF7 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Although the sliding window model has a good recognition effect in single independent action recognition, it is difficult for this method to accurately segment the boundary between basic actions and transition actions between adjacent basic actions in the process of multi-behavioral patterns.
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
[0032] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.
[0033] A kind of human behavior data segmentation method based on Mahalanobis distance of the present invention, specifically implement according to the following steps:
[0034] Step 1. Select a sliding window of size h to segment the sensor data, and calculate the Mahalanobis distance between the data of each window, and divide each window into a time period action set according to the criteria, and implement according to the following steps:
[0035] Step 1.1. Select a sliding window with a length of h to segment the sensor sample data matrix S along the time direction, then the segmented sample data matrix can be expressed as S=[S 1 ,S 2 ,,S k ] T ;
[0036] Step 1.2: After step 1.1, use the Mahalanobis distance to calculate the similarity between these action segments. The larger the Mahalanobis distance, the smaller the similarity, ...
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 behavioral data segmentation method based on the Mahalanobis distance: step 1, select a sliding window with a size h to segment the sensor data, and calculate the Mahalanobis distance between each window data, and divide each The windows are divided into a time period action set; step 2, after the completion of step 1, find out all the basic actions on the basis of the time period action division, and use similarity analysis to process the hidden basic actions; step 3, after step 2 After all the basic actions are determined, the transition action and the disturbance process are distinguished according to whether two adjacent basic actions are the same, and finally all the basic actions and transition actions can be determined. A behavior data segmentation method based on the Mahalanobis distance in the present invention divides human behavior into two types, basic action and transitional action, and realizes accurate segmentation of basic action and transitional action data in the process of multi-behavioral modes.
Description
technical field [0001] The invention belongs to the technical field of human behavior recognition, and in particular relates to a method for segmenting human behavior data based on Mahalanobis distance. Background technique [0002] Human behavior is a continuous process. The human behavior data obtained from the sensor network is a continuous data stream. According to which rules the sample data of this continuous process can be reasonably divided into individual data segments for behavior recognition. The key problem to be solved by data segmentation. Data segmentation is to divide the continuous process into several small time segments, and each time segment is called a window. In other words, data segmentation is the process of discretizing sensor data at continuous time points. Behavior recognition uses window as the basic processing unit. Too little or too much behavior data contained in the window will affect the final recognition effect. Therefore, data segmentation...
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): G06K9/62G06K9/00
CPCG06F2218/12G06F2218/08G06F18/22
Inventor 李军怀田玲王怀军于蕾王侃安洋
Owner XIAN UNIV OF TECH