Stereoscopic image significance region detection method

A technology of stereoscopic image and area detection, applied in image analysis, image enhancement, image data processing and other directions, can solve the problem of low accuracy

Active Publication Date: 2015-03-25
BEIJING UNIV OF TECH
View PDF3 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] An object of the present invention is to solve the defect that the detection accuracy of the salient region of the stereo image is not high, and provide at least the advantages that will be described later

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
  • Stereoscopic image significance region detection method
  • Stereoscopic image significance region detection method
  • Stereoscopic image significance region detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0044] The present invention will be further described in detail below in conjunction with the accompanying drawings, so that those skilled in the art can implement it with reference to the description.

[0045] The input stereoscopic image database involved in the present invention comes from the establishment of Niu in 2012. The source library is 1000 stereoscopic image pairs in the form of left and right views. However, we found that the principle that Niu thinks that images are available is to use three people to identify the target, If the recognition of the three people is consistent, the image is considered to be available, and the recognized target is obtained as GroundTruth. However, we think that it is not enough to use only 3 people to do the target recognition experiment, and the small sample size leads to increased randomness. Therefore, the present invention screens the 1000 stereo images again, and only keeps the foreground with a single target. images, remove t...

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 a stereoscopic image significance region detection method. The method includes the steps of firstly, obtaining flow information of all pixel points for a left eye view and a right eye view of a stereoscopic image; secondly, matching the flow information to obtain a disparity map; thirdly, selecting one of the left eye view and the right eye view, and dividing the selected eye view into T non-overlapping square image blocks; fourthly, calculating the disparity influence degree of each image block on the disparity map; fifthly, calculating the central polarization characteristic and the space dissimilarity of each image block on the selected left eye view or the selected right eye view, and multiplying the corresponding central polarization characteristic, the corresponding space dissimilarity and the corresponding disparity influence degree to obtain the significance value of each image block; sixthly, obtaining a significance gray-scale map of the stereoscopic image through the significance values of the image blocks. The stereoscopic significance extraction method on the basis of the disparity influence degrees and the space dissimilarity is provided, deep information is obtained through disparity, and the more accurate stereoscopic significance region detection is achieved through the combination of the vision central polarization characteristics and the space dissimilarity.

Description

technical field [0001] The invention relates to a method for detecting a salient area of ​​a stereo image. More specifically, the present invention relates to a method for detecting salient regions of stereoscopic images based on parallax influence degree and spatial dissimilarity degree. Background technique [0002] Visual saliency has attracted extensive attention in the research field because it can simulate the characteristics of human visual attention mechanism, and it has become a research topic in neuroscience, robotics, computer vision and other fields. The recognition of salient regions can be applied to object recognition, image relocation, visual tracking technology, image segmentation, and can also be used to analyze the selection of human viewpoints. At present, the methods for saliency extraction are mostly oriented to a single 2D image. With the development of information technology, more stereoscopic images appear in people's lives, which also poses new ch...

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): G06T7/00G06V10/25
CPCG06T7/11G06T7/593G06T2207/10021G06T2207/20021G06T7/13G06T7/174G06T7/269G06V10/451G06V10/25G06T5/10G06T2207/10012
Inventor 段立娟邱硕马伟苗军李甲
Owner BEIJING UNIV OF TECH
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