Unlock instant, AI-driven research and patent intelligence for your innovation.

A Hypergraph Optimization Method for Salient Object Detection Based on Foreground and Background Seeds

A target detection and graph optimization technology, applied in the field of image processing, can solve problems such as the inability to describe the multi-order relationship of multiple nodes, and achieve the effect of improving performance

Active Publication Date: 2021-04-06
SOUTHEAST UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method can only describe the second-order relationship in the image but cannot describe the multi-order relationship between multiple nodes

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 Hypergraph Optimization Method for Salient Object Detection Based on Foreground and Background Seeds
  • A Hypergraph Optimization Method for Salient Object Detection Based on Foreground and Background Seeds
  • A Hypergraph Optimization Method for Salient Object Detection Based on Foreground and Background Seeds

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] Such as figure 1 As shown, the hypergraph optimization method for salient object detection based on the foreground and background seeds of this embodiment includes the following steps in turn:

[0044] S1: Use the existing SLIC algorithm to over-segment the image into superpixels, and calculate the position and color features of each superpixel;

[0045] The image to be processed is over-segmented into 300 homogeneous superpixels using the SLIC method, and its spatial position feature and CIELab color feature are extracted for each superpixel;

[0046] S2: Define superpixels as the nodes of the hypergraph, and construct a probability hypergraph according to the global position correlation, local position correlation and color correlation between superpixels to describe the input image;

[0047] The superpixels formed by over-segmentation are defined as the nodes of the hypergraph. Based on the local position correlation based on each node v i Construct a hyperedge: 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 hypergraph optimization method for salient target detection based on foreground and background seeds, comprising the following steps: using SLIC algorithm to over-segment an image into superpixels, and calculating the position and color features of each superpixel; Defined as a node of a hypergraph, a probability hypergraph is constructed according to the global position correlation, local position correlation and color correlation between superpixels to describe the input image; based on the image edge superpixels and the constructed probability hypergraph, Foreground seed and background seed information, obtain foreground seed and background seed information; propose a probabilistic hypergraph optimization framework, fuse the constructed probabilistic hypergraph, and detect salient objects in natural scene images. The present invention fully considers the foreground seed and background seed information in the input image, constructs a probability hypergraph capable of describing complex relationships in the image, and improves the performance of salient target detection in complex natural scene images. The detection results obtained by the present invention are consistent with the true values ​​in the database The graphs are more consistent.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a hypergraph optimization method for salient object detection based on foreground and background seeds. Background technique [0002] Since salient object detection can be widely used in computer vision tasks such as image segmentation, image quality assessment, image compression, and object recognition, salient object detection has attracted a large number of scholars to study it in recent years. Since graphs can conveniently describe the information contained in images, some scholars have proposed graph-based salient object detection methods. These methods represent each input image as a graph, and obtain the final salient object detection results by propagating information on the edges of the graph. [0003] These methods generally explicitly use a kind of seed node information, that is, foreground seed or background seed. The purpose of salient object detection is ...

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): G06T7/11G06T7/136G06T7/194G06T7/66G06T7/90
CPCG06T2207/10024G06T7/11G06T7/136G06T7/194G06T7/66G06T7/90
Inventor 张金霞魏海坤谢利萍
Owner SOUTHEAST UNIV