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

Image labeling method based on activation diffusion theory

A technique of diffusion theory and image annotation, which is applied in the fields of electrical digital data processing, special data processing applications, instruments, etc., and can solve problems such as the unsatisfactory performance of automatic image annotation algorithms

Inactive Publication Date: 2013-04-24
NANJING UNIV OF POSTS & TELECOMM
View PDF3 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Due to the great difference between the underlying feature description and semantic concept description of image content, the performance of existing automatic image annotation algorithms is not ideal.

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
  • Image labeling method based on activation diffusion theory
  • Image labeling method based on activation diffusion theory
  • Image labeling method based on activation diffusion theory

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] The following is attached figure 1 The technical scheme of the present invention will be further described in detail with the examples, but the protection scope of the present invention should not be limited to the following examples.

[0055] 1 Input an unlabeled image set.

[0056] 2 Determine candidate labels.

[0057] The process of determining candidate image labels is a process of assigning appropriate labels to unlabeled images. In this process, the influence of visual information and label information should be considered at the same time. Suppose Q is a test image, and S is a training set T. image and the label set of S is:

[0058] A s =(a 1 ,a 2 ,...,a m ) (1)

[0059] Among them, m is the number of tags;

[0060] The joint distribution probability A of the label set of the test image Q is calculated by the following process Q :

[0061] Step 1: The uniform probability of selecting an image S from the training set T is:

[0062] P ...

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 an image labeling method based on an activation diffusion theory. Firstly, an image label is regarded as a node of semantic network, a progressive correlation model is utilized to obtain a joint probability among multiple image labels, candidate image labels are generated, then the activation diffusion theory is utilized to compute correlation value of each candidate image label, effective transmissibility of each candidate image label is refined, and corresponding numerical value is given on each candidate image label. The size of the numerical value depends on the relation between the each candidate image labels and other candidate image labels, and a final label of each image is the candidate image label of the highest correlation value.

Description

technical field [0001] The invention relates to an image tagging method in the field of computer technology, in particular to an image tagging method based on activation diffusion theory. Background technique [0002] As more and more digital images appear on the network, personal computers and digital devices, the desire to effectively organize, manage and utilize such massive information based on image content analysis technology is becoming stronger and stronger. Among them, the research on image annotation is the most important and critical step to achieve content-based image indexing, retrieval and other related applications. Its purpose is to establish an accurate relationship between visual information at the perceptual level and language description at the semantic level Correspondence. [0003] From the perspective of pattern recognition, the image annotation problem can be regarded as a pattern classification problem, which is completed using machine learning tech...

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): G06K9/62G06F17/30
Inventor 朱松豪邹黎明刘佳伟梁志伟
Owner NANJING UNIV OF POSTS & TELECOMM
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