Human action recognition method based on sparse tensor local fisher discriminant analysis algorithm

A discriminant analysis and behavior technology, applied in character and pattern recognition, computing, computer components and other directions, can solve the problem of destroying the spatial correlation of images, and achieve the effect of improving the recognition rate of human behavior, ensuring sparsity, and good classification.

Active Publication Date: 2022-08-05
HOHAI UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Due to the eigenvalue decomposition of a large matrix, this operation will not only cause a huge computational cost, but also destroy the original spatial correlation of the image

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Human action recognition method based on sparse tensor local fisher discriminant analysis algorithm
  • Human action recognition method based on sparse tensor local fisher discriminant analysis algorithm
  • Human action recognition method based on sparse tensor local fisher discriminant analysis algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0094] The technical solutions of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0095] A method for identifying human behavior based on a sparse tensor local Fisher discriminant analysis algorithm described in the present invention includes the following steps:

[0096] S1: Obtain the silhouette image sequence according to the Weizmann human behavior database, and construct tensor samples. 10 different actions in the human behavior database correspond to 10 types of tensor samples, and construct a training sample set and a test sample set according to the tensor samples ;

[0097] S2: The sparse tensor local Fisher discriminant analysis algorithm (STLFDA) is used to train according to the training sample set to obtain a sparse projection matrix group; the sparse projection matrix group can project the original tensor samples from a high-dimensional space to a low-dimensional tensor quantum space, In order ...

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

No PUM Login to view more

Abstract

The invention discloses a method for recognizing human behavior based on sparse tensor. The method includes the following steps: obtaining human behavior silhouette sequence construction samples, each sample is represented by a third-order tensor; using the sparse tensor local Fisher discriminant analysis to train Sparse projection matrix group; use the obtained sparse projection matrix group to project training samples and samples to be tested into tensor subspaces; calculate the similarity between samples to be tested and training samples, and construct a nearest neighbor classifier based on tensor distance Identify the samples to be tested in the low-dimensional space. The tensor local Fisher discriminant analysis algorithm of the present invention transforms the problem of finding eigenvalues ​​and eigenvectors into a series of linear regression problems, which not only satisfies the objective of the tensor local Fisher discriminant analysis, but also ensures the sparseness of the obtained projection matrix .

Description

technical field [0001] The invention relates to the technical field of feature extraction of human action silhouette sequences, in particular to a human action recognition method based on a sparse tensor local Fisher discriminant analysis algorithm. Background technique [0002] The basic idea of ​​extraction methods based on algebraic features is to project the original samples into subspaces to form algebraic features. Representative methods include principal component analysis (PCA), linear discriminant analysis (LDA) and methods based on manifold learning. Principal Component Analysis (PCA) is an unsupervised feature extraction method that aims to maximize the variance of the projected subspace by maximizing the trace of the covariance matrix. The physical meaning is to make the distance between all samples as large as possible in the projected subspace. Linear discriminant analysis (LDA), when embedding discriminant information, can ensure that the inter-class dispersi...

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
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06V40/20G06V10/764G06K9/62
CPCG06V40/20G06F18/24147
Inventor 卢雨彤韩立新
Owner HOHAI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap