Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

A collaborative image segmentation method based on foreground background estimation and hierarchical region association

A technology of foreground background and graded region, which is applied in the field of image collaborative segmentation based on foreground background estimation and graded region association, to achieve the effect of increasing consistency

Inactive Publication Date: 2019-02-26
NINGBO UNIVERSITY OF TECHNOLOGY
View PDF1 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the aforementioned early methods can only extract approximately the same foreground objects from different backgrounds.

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
  • A collaborative image segmentation method based on foreground background estimation and hierarchical region association
  • A collaborative image segmentation method based on foreground background estimation and hierarchical region association
  • A collaborative image segmentation method based on foreground background estimation and hierarchical region association

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] The image cooperative segmentation method based on progressive foreground estimation and hierarchical region association of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0033] Such as figure 1 As shown, the image collaborative segmentation method based on foreground background estimation and hierarchical region association includes steps:

[0034] S1, for image set I={I 1 , I 2 ,...,I N} in each image is subjected to superpixel segmentation operation to obtain superpixel collections of all images at different scales, where image I k The set of superpixels for R k , superpixel ν r(k) contains R k Each superpixel in: N is a positive integer, k∈N;

[0035]S2. For image set I, pre-estimate each superpixel category label X in each image i and pixel class labels X j ,in and node ν r (k) and ν p (k) represent image I respectively k of superpixels and pixels;

[0036] S3, estimat...

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 an image synergistic segmentation method based on foreground background estimation and hierarchical region association. The method comprises the following steps: S1, obtaining the super pixel sets of all images at different scales; S2, estimating each super pixel class label and pixel class label in each image; S3, estimating a foreground Gaussian mixture model and a background Gaussian mixture model of all images; S4, training an appearance model corresponding to each image by using a texture classifier model; 5, according to that foreground Gaussian mixture model, thebackground Gaussian mixture model, the texture classifier model and the consistency of the superpixel label among the images of the image set, estimating the pixel / superpixel labeling of the image; S6, updating the superpixel class label and the pixel class label of all the estimated images; 7, judging whether that update is complete. The above method overcomes the prior constraint of obvious foreground and background difference, and has more robustness.

Description

technical field [0001] The invention relates to the field of image segmentation, in particular to an image collaborative segmentation method based on foreground background estimation and hierarchical area association. Background technique [0002] As a research hotspot in the field of computer vision, image segmentation technology has broad application prospects and has achieved rapid development in the past two decades. Among them, image collaborative segmentation technology is a research branch in this field, and its goal is to distinguish the foreground object in the image from the background area on the basis of two or more images. However, solving this ill-posed problem is quite challenging due to the existence of factors such as the change of shooting angle and pose, and the intra-class diversity of foreground objects. So far, researchers have applied image collaborative segmentation technology to many popular fields such as image retrieval, visual summary generation,...

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): G06T7/11
CPCG06T2207/20081G06T7/11
Inventor 姚拓中安鹏何加铭
Owner NINGBO UNIVERSITY OF TECHNOLOGY
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
Eureka Blog
Learn More
PatSnap group products