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

Image marking method based on semi-supervised subject modeling

A topic modeling and image labeling technology, applied in the field of semi-supervised learning, can solve problems such as difficult possibility modeling

Inactive Publication Date: 2014-01-29
ZHEJIANG UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

For image annotation, due to the diversity of image content, it is always difficult to fully likelihood model it

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 marking method based on semi-supervised subject modeling
  • Image marking method based on semi-supervised subject modeling
  • Image marking method based on semi-supervised subject modeling

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0023] Referring to the attached picture:

[0024] A semi-supervised method for image annotation based on topic modeling, the method includes the following steps:

[0025] 1. Obtain images from the Internet, including images with text annotations and unlabeled images;

[0026] 2. Using a model similar to probabilistic latent semantic analysis, the connection between the visual features and text annotations of all images is modeled through latent topics. The modeling process is carried out as follows: for each image i, First use the vector F i Indicates the visual features of the image, the vector W i to represent the image text annotation, where F i ={f 1 ,..., f n}, where f u Indicates the number of times the u-th visual feature word appears in the i-th picture; W i ={w 1 ,…,w n}, where w v Indicates the number of times the v-th text annotation word appears in the i-th picture.

[0027] and assume f i (where i=1,...,n) obey multinomial distribution w i (where i...

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 marking method based on semi-supervised subject modeling. The method comprises the following steps of: firstly, obtaining images from the Internet, including images with text marks and unmarked images; then, modeling the relation between the visual features and text marks of all images through latent subjects by use of a model similar to probabilistic latent semantic analysis; establishing the nearest-neighbor graphs of all images, and adjusting the model according to the manifold structure obtained by modeling the nearest-neighbor graphs; learning the model by an expectation maximization algorithm, and calculating the probability of matching the latent subjects with the images respectively; and finally, calculating the probability of matching each text mark with the unmarked images according to the probability of matching the latent subjects with the images, and selecting the text mark with the highest probability to mark the unmarked images.

Description

technical field [0001] The invention relates to the technical field of semi-supervised learning in machine learning, in particular to an image labeling method based on topic modeling. Background technique [0002] In recent years, due to the increasing popularity of digital cameras, the number of personal digital photos has increased sharply. At the same time, sharing photos on the Internet has also become more and more popular. In order to tap the potential value of large photo collections, users need to be able to effectively retrieve the required information. image. Image annotation, a technique that links the semantic content of text and images, is a good way to reduce the semantic gap and can be used as an intermediate step in image retrieval. It enables users to retrieve images through textual queries, and in terms of semantics, it can provide better results than content-based retrieval. Image annotation has attracted increasing research interest in recent years. ...

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): G06F17/30G06F17/27
Inventor 何晓飞卜佳俊陈纯倪雅博
Owner ZHEJIANG 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