Global-local-low-rank-based image salient target detection method

A salient and global technology, applied in the fields of computer vision and image processing, which can solve the problems of cluttered salient objects and difficult application of saliency maps.

Inactive Publication Date: 2015-03-11
LANZHOU UNIVERSITY OF TECHNOLOGY
View PDF2 Cites 18 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the traditional saliency detection method needs to combine multiple image features, and often introduces a lot of non-salient information in the image background when the salient target is obtained, so that the obtained salient target is in a relatively messy environment, which is salient Further applications of graphs pose great difficulties

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
  • Global-local-low-rank-based image salient target detection method
  • Global-local-low-rank-based image salient target detection method
  • Global-local-low-rank-based image salient target detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] Attached below Figure 1 to Figure 4 And specific embodiments further illustrate the present invention.

[0047] figure 1 The overall process of image salient object detection method based on global and local low rank is given.

[0048] In this paper, a salient object detection method based on global and local low-rank images is invented. The main steps are as follows:

[0049] Step (1), obtain the initial saliency map: the original image subtracts the mean value of each corresponding channel from the three channels of the CIE Lab color space to obtain the three-channel difference feature matrix, and then calculate the standard deviation and 2-D of the three-channel feature matrix Entropy, select the result with the largest standard deviation and the smallest 2-D entropy and give different weights to fuse to obtain the initial saliency map;

[0050] Step (2), global and local low-rank processing: Here, combined with the global and local perception characteristics of hu...

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 global-local-low-rank-based image salient target detection method. The method comprises the followings steps: step (1), obtaining an initial salient image; step (2), carrying out global and local low-rank processing; and step (3), carrying out global and local salient image combination; to be specific, combining a global salient image and a local salient image, providing different weights based on contributions of the two images and carrying out fusion to obtain a final result. According to the invention, an image initial salient image is obtained by combining a CIE Lab color space contrast feature; and non-salient information of the initial salient image is inhibited in the global and local ways. Because the obtained salient image contains less background non-salient information, the processing result can be applied to more tasks like computer vision and image processing. The less background non-salient information the salient target image contains, the clearer and more reliable the obtained salient target becomes.

Description

technical field [0001] The invention relates to the technical fields of computer vision and image processing, in particular to an image salient target detection method based on global and local low ranks. Background technique [0002] With the development of computer vision, image salient object detection as a basic task in the field of computer vision has become a research hotspot. Detecting salient objects can be applied to image segmentation, object recognition, image scaling, image compression and other fields. A high-quality saliency map can provide better preprocessing for the above work. [0003] There are many methods for salient object detection at this stage, and the accuracy of salient detection has been greatly improved. However, the traditional saliency detection method needs to combine multiple image features, and often introduces a lot of non-salient information in the image background when the salient target is obtained, so that the obtained salient target ...

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): G06T7/00
CPCG06T7/73
Inventor 李策胡治佳肖利梅李铭万腾
Owner LANZHOU 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
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