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

Automatic image annotation algorithm

A technology of automatic image labeling and image labeling, which is applied in the field of classification and recognition, image retrieval, and can solve the problems of manpower-consuming unlabeled data structure and full utilization

Active Publication Date: 2014-02-26
CHINA JILIANG UNIV
View PDF3 Cites 32 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problems of manpower consumed in the process of labeling large data images and the full utilization of unlabeled data structures, the present invention provides an automatic image labeling algorithm, including steps:

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
  • Automatic image annotation algorithm
  • Automatic image annotation algorithm
  • Automatic image annotation algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] The present invention will be further described below in conjunction with accompanying drawing.

[0038] Such as figure 1 As shown, the image annotation method based on sparse structural feature selection includes the following steps:

[0039] Step (1) Image dataset feature extraction: The underlying information of the image is obtained by performing feature extraction on the image in the dataset. The selected feature types include: color histogram, block-by-block color moment, edge direction histogram, color correlation map, face feature, wavelet texture and SIFT-based descriptor bag.

[0040] Step (2) Image training set selection: The image automatic labeling algorithm is trained by selecting the most authoritative and standard data set, which must contain multiple features and the most abundant image resources. Select n data from the data set as training samples, some of which have been labeled, and the rest of the samples have not been labeled.

[0041] Step (3) ...

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 automatic image annotation algorithm which includes the steps: (1) extracting features of images in a data set to acquire bottom information of the images; (2) selecting an image training set: training the automatic image annotation algorithm by selecting the most authoritative and the most standard data set with various features and the most abundant image resources, and selecting n data from the data set as training samples; (3) training the image annotation algorithm: selecting features of the obtained samples and optimizing annotation results by bound terms; (4) automatically annotating the images: processing forecast tags by selecting threshold values. Parts of the samples are annotated, and the rest samples are not annotated. The image annotation algorithm based on sparse structure feature selection can be used for automatically annotating the images and is innovative.

Description

technical field [0001] The invention relates to a method for automatic image labeling, which belongs to the field of image retrieval, classification and recognition. Background technique [0002] As digital cameras and other electronic devices become more popular, the number of images is increasing rapidly. Therefore, how to effectively manage and retrieve network multimedia information has become an urgent problem to be solved. In the past few decades, there have been a lot of researches on content-based image retrieval, but limited by the semantic gap, they cannot fully satisfy users' needs. Users are accustomed to using keywords to query, but manual labeling is a very laborious task, which has led to the development of automatic image labeling. Automatic image annotation is to reflect the semantic content and let the computer automatically add images and label images without tags. Its key is to realize the research in the field of semantic retrieval of images. Conten...

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): G06F17/30
CPCG06F16/58G06F16/5838G06F18/213
Inventor 章东平李艳洁杨力芦亚飞
Owner CHINA JILIANG UNIV
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