Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Multimodal pattern classification method based on analytical dictionary learning

A technology of pattern classification and dictionary learning, applied in character and pattern recognition, instruments, computer parts, etc., it can solve problems such as the inability to solve multi-modal pattern classification problems, so as to improve judgment ability and robustness, and improve classification. The effect of precision

Inactive Publication Date: 2017-11-24
DALIAN UNIV OF TECH
View PDF0 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] However, current analytical dictionary learning methods are not able to solve multimodal pattern classification problems

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
  • Multimodal pattern classification method based on analytical dictionary learning
  • Multimodal pattern classification method based on analytical dictionary learning
  • Multimodal pattern classification method based on analytical dictionary learning

Examples

Experimental program
Comparison scheme
Effect test

experiment example

[0074] In order to describe the specific implementation of the present invention and verify the effectiveness of the present invention, the method proposed by the present invention is applied to a public object image database, namely NUS-WIDE-OBJECT database. The database contains 30,000 images in 31 categories, fully simulating the actual environment. The extracted features are 64-dimensional color histogram feature, 144-dimensional color correlation map feature, 73-dimensional edge direction histogram feature, 128-dimensional wavelet texture feature, 225-dimensional block color moment feature and 500-dimensional SIFT feature. Characteristics. Randomly select 17927 images from this database as training sample points, and the remaining 12073 images as test sample points.

[0075] First, all sample data are input into the classification model function of the embodiment of the present invention for training, wherein the parameters , , , respectively by cross-validation ...

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 multimodal pattern classification method based on analytical dictionary learning, and can enhance the classification accuracy. Characteristic information and shared class information of different modal data are reinforced, and the multimodal image characteristic information is arranged under a framework performing dictionary and classifier learning simultaneously to perform crucial refining so that the subsequent classification based on the classifier regression result maximum position index is facilitated. Meanwhile, the classification target having flexibility is learnt by applying the interval target strategy so that the crucial capacity and the robustness of the whole model can be enhanced and the classification accuracy can be enhanced.

Description

technical field [0001] The invention relates to a pattern classification technology, in particular to a multi-mode pattern classification method based on analytic dictionary learning which can improve classification accuracy. Background technique [0002] Pattern classification technology is an extremely important theory in the subject of pattern recognition, which has been widely used in various fields, such as biometric authentication, gesture recognition, data mining, information retrieval, signal processing, etc. Conventional pattern classification techniques mainly include two main processes, namely the training process and the testing process. The training process includes extracting the characteristics of the training samples and establishing a classification model; the testing process is to extract the test samples and use the classification model obtained in the training stage to classify and predict the test samples. [0003] With the rapid development of multimed...

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
IPC IPC(8): G06K9/62
CPCG06F18/243G06F18/2451
Inventor 郭艳卿
Owner DALIAN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com