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

Symmetry property-based method for detecting salient regions of images

A technology for region detection and symmetric graphs, applied in the field of computer vision

Active Publication Date: 2011-10-19
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF1 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the existing problems of the existing Itti model detection salient regions, and propose a method for detecting salient regions of images based on symmetry characteristics

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
  • Symmetry property-based method for detecting salient regions of images
  • Symmetry property-based method for detecting salient regions of images
  • Symmetry property-based method for detecting salient regions of images

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0015] Embodiment 1: Take the symmetric image detection of an artificial image as an example.

[0016] Such as figure 2 As shown in a, first construct an image of a circle containing an inscribed square as the target image, and the image size is 181×181. The process of the specific detection method is as follows: figure 1 As shown, the specific process is as follows:

[0017] S1. Create a multi-scale image for the target image. Here, the Gaussian pyramid model is used to establish a multi-scale image, and the information at different scales of the image is extracted, that is, the target image is down-sampled step by step. After each sampling, the image is Gaussian smoothed to reduce noise interference. A total of five samplings are performed here to obtain five images of different scales. Specifically, Gaussian images at different scales can be obtained by convolving the target image with the Gaussian kernel, and the scale factor in the Gaussian kernel controls the degree...

Embodiment 2

[0026] Embodiment 2: Take salient region detection of a natural image as an example.

[0027] This image, as well as the eye movement diagram, were downloaded from the online database provided by Neil.D.B.Bruce et al. The size of the images is 681×511, and the eye movement images are obtained by 20 subjects viewing the images freely. The flow chart of the specific detection method is as follows: figure 1 As shown, the specific process is the same as the first embodiment, except that in step S2, the radius of the isotropic symmetry operator is changed to 8 pixels.

[0028] image 3 It is a saliency map obtained by using the method of the present invention to actually detect natural images, a saliency map obtained by using the Itti model in the background technology, and a group of effect comparison maps of eye movement fixation maps. Among them: 3a. original image, 3b. human eye gaze map, 3c. saliency map detected by the method of the present invention, 3d. saliency map dete...

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 belongs to the technical field of computer vision, and discloses a symmetry property-based method for detecting salient regions of images, aiming at solving the problem of salient regions during the existing Itti model detection. The method comprises the following steps of processing target images to obtain images with different dimensions, respectively extracting frequency domain symmetric properties and time domain symmetric properties of the images with the different dimensions, and merging the images to obtain a time frequency characteristic image so as to obtain a final saliency map. The method combines the time domain symmetric properties with the frequency domain properties to accomplish the detection of the salient regions of the images. The detection method can more completely detect salient targets by utilizing the function of the symmetry properties, which is played in the process of eye fixation, so that the detected salient regions of the images are more in line with the result of the eye fixation.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and in particular relates to a method for detecting a salient area of ​​an image. Background technique [0002] The human visual system can quickly and efficiently process a large amount of visual information from the outside world. Based on such a feature, more and more researchers are now hoping to simulate the visual selective attention mechanism to complete various image processing tasks, such as image recognition, object tracking and scene analysis. So far, there have been many models based on the visual selective attention mechanism to extract salient regions. Among them, the most typical model is a bottom-up selective attention model proposed by Itti and Koch et al. in 1998——Itti model, For details, please refer to the literature: L.Itti, C.Koch.E.Niebur.A model of saliency-based visual attention for rapid scene analysis. Pattern Analysis and Machine Intelligence, IEEE Transaction...

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
IPC IPC(8): G06T5/00G06K9/46
Inventor 李永杰陈丽霞李朝义
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
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