Semantic information and edge constraint-based foreground target detection method

A technology of edge constraints and semantic information, applied in the field of computer vision, can solve problems such as the inability to realize image target detection, and achieve the effect of avoiding the input of prior information, complete and continuous contours, and reducing the appearance of cluttered edges

Active Publication Date: 2018-05-15
BEIHANG UNIV
View PDF4 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this technology requires human interaction and cannot achieve automatic image target detection

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
  • Semantic information and edge constraint-based foreground target detection method
  • Semantic information and edge constraint-based foreground target detection method
  • Semantic information and edge constraint-based foreground target detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] In order to better understand the technical solution of the present invention, the specific embodiments of the present invention will be further described below in conjunction with the accompanying drawings:

[0035] 1. Hierarchical Segmentation Based on Semantic Edge Constraints

[0036] Such as figure 1 , 2 As shown, the flow of an image hierarchical segmentation method based on image semantic edge constraints in the present invention is as follows.

[0037] First, the gradient information in different directions is calculated on the Lab color space to obtain the local edge information of the image; then, based on the obtained local edge results, the similarity between the edges is established, and the eigenvector corresponding to the selected minimum eigenvalue is solved. The directional gradient information is calculated on the feature vector space to obtain the salient edge in the image; the local edge and the salient edge are linearly combined to obtain the norm...

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 provides a semantic information and edge constraint-based foreground target detection method, and aims at automatically segmenting foregrounds and backgrounds of images through a segmentation energy function model by means of semantic label information so as to realize detection of foreground targets. A flow chart of the method is shown in an abstract drawing. The method comprises the following five steps of: 1, semantic edge constraint-based image layering and segmentation; 2, establishment of a position model; 3, establishment of an appearance model; 4, establishing of a smoothconstraint; and 5, iterative optimization of a segmentation model. Experiments prove that the method has feasibility, correctness and universality, and can be used in various high-level image analysis and image understanding.

Description

technical field [0001] The invention relates to a foreground object detection method based on semantic information and edge constraints. The segmentation energy function model completes the automatic segmentation of image foreground and background with the help of semantic label information, thereby realizing the detection of foreground objects, which has certain effectiveness and universality. belongs to the field of computer vision. Background technique [0002] The detection of the foreground target can be realized by image segmentation technology. Image segmentation refers to the technology and process of using image features, such as color, texture, etc., to divide the image into regions with various characteristics and extract the target of interest. [0003] For image segmentation, it can be roughly divided into two categories. The first is the superpixel segmentation algorithm that divides the image into some relatively small superpixel regions, and the other is the ...

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): G06T7/00G06T7/194G06T7/136G06T7/13G06T7/90
CPCG06T7/0002G06T7/13G06T7/136G06T7/194G06T7/90G06T2207/10024G06T2207/20081
Inventor 袁丁强晶晶胡晓辉张弘
Owner BEIHANG UNIV
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