Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

An image saliency detection method based on an improved graph model

A graphical model and salient technology, which is applied in the field of image detection and computer vision, can solve the problems of not evenly highlighting the interior of salient objects, incomplete detection of salient objects, etc., and achieve the effect of improving accuracy

Active Publication Date: 2021-04-30
无锡尚合达智能科技有限公司
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In view of the above problems, the present invention proposes an image saliency detection method based on an improved graph model to solve the problem of incomplete detection of salient objects in complex environments or the inability to uniformly highlight the inside of salient objects, and can completely detect and evenly segment the entire salient object. Improving the performance of image saliency detection algorithms will greatly promote the further research and development of image-related fields

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
  • An image saliency detection method based on an improved graph model
  • An image saliency detection method based on an improved graph model
  • An image saliency detection method based on an improved graph model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0095] This embodiment provides an image saliency detection method based on an improved graph model, see figure 1 , the method includes the following steps:

[0096] Step 1. Segment the input image into N superpixels using Simple Linear Iterative Clustering (SLIC), and the i-th superpixel is denoted by v i Indicates that the jth superpixel is denoted by v j Indicates that i, j ∈ 1, 2, ..., N.

[0097] Step 2. Use the underlying features of the image to calculate the similarity between the superpixels to form a similarity matrix A=[a ij ] N×N , a ij Denotes superpixel v i and v j The degree of similarity, i, j∈1, 2,..., N;

[0098] (2.1) CLELAB color mean value c=(l, a, b) of the pixels included in the superpixel T To represent the color feature of the superpixel, the 59-dimensional vector t formed by the equivalent mode of the local binary pattern (LBP) represents the texture feature of the superpixel; the superpixel v i and v j The color feature distance Dc ij and ...

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 saliency detection method based on an improved graph model, and belongs to the technical fields of computer vision and image detection. This method uses simple linear iterative clustering to divide the image into superpixels, constructs an undirected graph with superpixels as vertices, and uses the underlying image features and prior knowledge to extract high-level features on the basis of improving the graph model, and obtains the significant features based on the underlying features. picture. Then, the foreground and background seed nodes are selected by using the high-level features and the compactness of salient objects, and the saliency maps based on the foreground and background seeds are respectively calculated and fused. Finally, the saliency maps obtained in the two stages are fused to obtain the final saliency map. The invention can completely detect and uniformly highlight salient objects in an image, improve the accuracy of salient object detection in complex environments, meet the design requirements of actual engineering systems, and solve the problem of low accuracy rate of salient object detection in complex environments.

Description

technical field [0001] The invention relates to an image saliency detection method based on an improved graph model, and belongs to the technical fields of computer vision and image detection. Background technique [0002] Saliency detection aims to enable computers to have a human-like visual attention mechanism to find the most interesting and valuable information from complex scenes. The early saliency detection algorithm is aimed at the detection of visual attention, and the purpose is to predict the gaze point of the human eye in the image. Later, many salient region detections aimed at segmenting the entire salient object emerged. Compared with the former, salient region detection has higher application value. Saliency detection models can be divided into bottom-up and top-down categories. The bottom-up model is driven by data, using image color, contrast, etc. to calculate saliency; the top-down model is task-driven, and often needs to train a large number of sampl...

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): G06K9/34G06K9/46G06K9/62
CPCG06V10/267G06V10/462G06F18/23213G06F18/22
Inventor 葛洪伟张莹莹羊洁明江明
Owner 无锡尚合达智能科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com