Multi-view image classification method based on hierarchical graph enhanced stacked auto-encoder

A technology of stacking self-encoders and self-encoders, which is applied in the field of multi-view image classification based on graph enhancement, can solve problems such as the inability to extract multi-view image features, and achieve the effect of maintaining geometric structure, balancing complementarity and consistency

Active Publication Date: 2022-08-09
JIANGSU UNIV
View PDF3 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is: in order to solve the technical problems in the above-mentioned prior art that the features of each view of the multi-view image cannot be extracted and the learned features are fused, the present invention provides a stacked autoencoder based on layered graph enhancement The multi-view image classification method of

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
  • Multi-view image classification method based on hierarchical graph enhanced stacked auto-encoder
  • Multi-view image classification method based on hierarchical graph enhanced stacked auto-encoder
  • Multi-view image classification method based on hierarchical graph enhanced stacked auto-encoder

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0054] like figure 1 As shown, this embodiment provides a multi-view image classification method based on a layered graph enhanced stacked autoencoder, including the following steps:

[0055] Step S1, sample collection

[0056] Collect multi-view samples χ={X (1) , X (2) ,...,X (V) }, and normalize it;

[0057] in N is the number of samples, d v is the dimension of the vth view, and V represents the number of views;

[0058] Step S2, build a model

[0059] Build an autoencoder network model, which includes an autoencoder and a fully connected neural network; let the parameters of the vth view in the autoencoder be The parameters in the fully connected neural network are Initialize the parameters of all views in the autoencoder and parameters in a fully connected neural network and public representation H:

[0060] where l represents the lth layer of the autoencoder, L represents the total number of layers of the autoencoder; m represents the mth layer of the f...

Embodiment 2

[0089] This embodiment also provides a multi-view image classification system based on a layered graph enhanced stacked autoencoder, including a sample collection module, a model model building, a model training module, and a real-time classification module, specifically:

[0090] The sample collection module is used to collect multi-view samples χ={X (1) , X (2) ,...,X (V) }, and normalize it;

[0091] in N is the number of samples, d v is the dimension of the vth view, and V represents the number of views;

[0092] Build a model model for building an autoencoder network model. The autoencoder network model includes an autoencoder and a fully connected neural network; let the parameters of the vth view in the autoencoder be The parameters in the fully connected neural network are Initialize the parameters of all views in the autoencoder and the parameters in the fully connected neural network and public representation H;

[0093] where l represents the lth layer...

Embodiment 3

[0120] This embodiment also provides a computer device, including a memory and a processor, where a computer program is stored in the memory, and when the computer program is executed by the processor, the processor enables the processor to execute the above-mentioned multi-view image based on the layered graph enhanced stacked autoencoder The steps of the classification method.

[0121] Wherein, the computer device may be a desktop computer, a notebook computer, a palmtop computer, a cloud server and other computing devices. The computer device can perform human-computer interaction with the user through a keyboard, a mouse, a remote control, a touch pad or a voice control device.

[0122] The memory includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or D interface display memory, etc.), random access memory (RAM), Static random access memory (SRAM), read ...

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 multi-view image classification method based on a hierarchical graph enhancement stacked auto-encoder, and relates to a multi-view image classification method and system based on graph enhancement. The objective of the invention is to solve the technical problem that features of each view of a multi-view image cannot be extracted and learned features cannot be fused in the prior art. The invention provides a multi-view image classification method and system based on a hierarchical graph enhanced stacked auto-encoder. The geometric structure of multi-view data and complementarity and consistency among different views are considered; a layered graph structure is introduced into an auto-encoder to learn representation of a specific view, and local and non-local geometric structures of multi-view data are kept; after the feature representation of each view with geometric structure characteristics is learned, each single view is reconstructed by using a full-connection neural network, and the common representation can be learned; and complementarity and consistency among a plurality of views can be automatically balanced.

Description

technical field [0001] The invention belongs to the technical field of image classification, in particular to the technical field of multi-view image classification, and more particularly to a multi-view image classification method based on graph enhancement. Background technique [0002] With the rapid development of deep learning, various deep models have been proposed. Autoencoder (AE), as one of the most representative deep learning algorithms, has been successfully applied in many fields such as computer vision, speech recognition and natural language processing. An autoencoder is an unsupervised learning method that aims to minimize the reconstruction error between the input and the corresponding reconstructed output. Based on the important role of raw data geometry in feature representation, some manifold learning algorithms are introduced into autoencoders. For example, NLSP-SAE takes into account the non-local and local geometry of the data, ensuring that samples ...

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 Applications(China)
IPC IPC(8): G06V10/764G06V10/80G06V10/82G06N3/04G06N3/08
CPCG06V10/764G06V10/806G06V10/82G06N3/08G06N3/045Y02T10/40
Inventor 苟建平谢楠楠刘金华王智欧卫华陈雯柏
Owner JIANGSU 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