Image matching-based foreground segmentation method and device

A foreground segmentation and image technology, applied in the multimedia field, can solve the problem of low accuracy of automatic foreground segmentation, and achieve the effect of improving efficiency, reducing overall time and improving robustness.

Inactive Publication Date: 2017-10-31
ZTE CORP
View PDF0 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0013] The invention discloses a foreground segmentation method based on image matching, which can solve the problem of low accuracy in the existing automatic foreground segmentation of image matching

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 matching-based foreground segmentation method and device
  • Image matching-based foreground segmentation method and device
  • Image matching-based foreground segmentation method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0066] The inventors of the present application found that the current automatic foreground segmentation scheme mainly focuses on feature extraction of continuous video frames, or static image foreground extraction combined with user intervention and global features. However, the present application proposes to use the dual-image joint automatic foreground extraction method, that is, extract the local features of the picture, obtain the foreground area through feature point matching and cluster analysis, and then use the picture segmentation algorithm to realize the automatic segmentation method for the foreground image. Among them, local features refer to some features that only appear locally, which can appear stably and have good distinguishability. Different from global features such as variance and color, local features can better summarize the information carried by the image, reduce the amount of calculation and improve the anti-interference ability of the algorithm.

...

Embodiment 2

[0125] This embodiment provides a foreground segmentation device based on image matching, which mainly includes the following units.

[0126] The first unit takes the local features of the two input images respectively, and matches the key points according to the extracted local feature information;

[0127] Among them, the first unit takes the local features of two input images including:

[0128] The two images input by the user are processed in gray scale, and the local feature information of the image is extracted using the SURF feature.

[0129] The first unit matches the key points according to the extracted local feature information, including:

[0130] The matching point corresponding to the key point in the first input image in the second input image in the two input images is determined by using the nearest neighbor algorithm.

[0131] The second unit is to filter out the wrong matching points from the matching points of the obtained key points to obtain all the co...

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 matching-based foreground segmentation method and device, and relates to an image matching technology and an image segmentation technology in the field of multimedia. The image matching-based foreground segmentation method is characterized by comprising the following steps of: respectively extracting local features of two input images and matching key points according to the extracted local feature information; screening wrong matching points from matching points of the obtained key points so as to obtain all the correct matching points; obtaining a feature point group on a foreground target from all the correct matching points by using clustering analysis; and obtaining the foreground target in a picture by utilizing a picture segmentation algorithm according to the obtained feature point ground. The invention furthermore discloses the image matching-based foreground segmentation device. By adoption of the image matching-based foreground segmentation method and device, foreground targets in pictures can be objectively obtained, so that the results are more accurate and intuitional, the traditional man-machine interaction method can be replaced, the whole time is decreased and the efficiency is improved.

Description

technical field [0001] The invention relates to image matching technology and image segmentation technology in the multimedia field, in particular to an innovative image local feature extraction scheme. Background technique [0002] Foreground Extraction refers to the extraction of foreground objects of any shape from a static image or a burst of video images. The traditional foreground extraction technology requires the user to mark the foreground pixels or areas. Through the pixel analysis of the area, it can be obtained The approximate outline of the object in the image. [0003] At present, the most commonly used foreground extraction schemes include the following: [0004] 1) Significance detection [0005] By extracting the saliency graph model of the image from global features such as color, brightness, and direction of the image, it can reflect the area in the image that most arouses the user's interest and best expresses the image content. [0006] The problem of...

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): G06T7/00G06K9/62
CPCG06T2207/10016G06F18/23213G06F18/24147G06T7/00
Inventor 邓硕马华东罗圣美傅慧源刘培业
Owner ZTE CORP
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap