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

Image-based significant object extraction method as well as complementary significance graph learning method and system

An object extraction and salient technology, applied in the field of image processing, can solve the problems of lack of integrity and accuracy of extraction results, and achieve the effect of strengthening generality and robustness, and solving problems of accuracy and robustness.

Inactive Publication Date: 2013-11-06
PEKING UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a method for extracting salient objects in images, a method and system for learning complementary saliency graphs, based on the present invention, can effectively solve the problem of lack of integrity and accuracy in the extraction results of the salient object extraction method in complex scenes

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-based significant object extraction method as well as complementary significance graph learning method and system
  • Image-based significant object extraction method as well as complementary significance graph learning method and system
  • Image-based significant object extraction method as well as complementary significance graph learning method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] In order to make the above objects, features and advantages of the present invention more comprehensible, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0035] The invention solves the problem of extracting salient objects in complex scenes by learning the complementary saliency map and using the complementary saliency as soft prior knowledge. Firstly, the training set is used to learn, and two complementary saliency mapping functions are obtained, and then several coarse saliency maps of the image are mapped and fused based on the mapping function to obtain a complementary saliency map, and finally, according to the soft prior knowledge provided by the complementary saliency, use Binary classifiers classify object and background labels on pixels in an image to extract salient objects. In this way, it effectively solves the lack of accuracy and robustness of general salient object...

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-based significant object extraction method as well as complementary significance graph learning method and system. The image-based significant object extraction method comprises the following steps of: obtaining two complementary significance mapping functions by using an image training set to learn; obtaining a complementary significance graph on the basis of the complementary significance mapping functions by learning; and extracting a significant object according to soft prior knowledge provided by the complementary significance graph. According to the invention, by giving out any image, the significant object in the image can be automatically and accurately extracted and the problems that the extraction result of the general significant object lacks of accuracy and robustness under complex scenes are effectively solved. The invention provides a brand-new idea for the methods.

Description

technical field [0001] The invention relates to the field of image processing, in particular to an image salient object extraction method, a complementary saliency map learning method and a system. Background technique [0002] With the massive increase of digital images in recent years, how to effectively use these images has become a very important issue. Generally speaking, salient objects in an image become the key analysis objects because they can represent the semantics of the entire image. In many image applications (such as automatic image cropping, image compression, advertisement design, content-based image retrieval), whether to correctly extract salient objects and analyze them has become a crucial problem. [0003] For now, the existing research results show satisfactory results in some simple situations (such as simple image background, single salient object, large gap between object visual characteristics and background, etc.), but in complex scenes (such as ...

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): G06K9/46G06K9/70
Inventor 田永鸿余昊男黄铁军高文
Owner PEKING 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