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

Salient object identification method

A target recognition and remarkable technology, which is applied in the direction of character and pattern recognition, instruments, computer parts, etc., to achieve the effect of reducing interference, improving accuracy and good consistency

Inactive Publication Date: 2017-05-03
SICHUAN YONGLIAN INFORMATION TECH CO LTD
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Aiming at the above-mentioned problems existing in the prior art, the present invention comprehensively considers the two factors of local contrast and global contrast, uses the color distance with the spatial distance as the weight to represent the salience, and utilizes the distribution of colors in the image to represent the salience with the global The angle defines the target as a whole, which solves the problems caused by one-sided consideration, and adopts the SLIC algorithm to abstract the image into multiple superpixels. The number of pixels, in order to achieve the purpose of multi-scale analysis, so as to better handle the situation of different sizes of salient objects

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
  • Salient object identification method
  • Salient object identification method
  • Salient object identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0013] Next, this invention will be specifically described.

[0014] 1. Image abstraction processing

[0015] Based on the SLIC algorithm, the pixels with similar colors and close distances are clustered to form a visually uniform area and generate superpixels. Use K-means clustering to segment the image in the Lab color space to generate compact superpixels with clear boundaries. Therefore, the generation method of the superpixels in the above step 1 is as follows: First: set the abstract image into three scales, including The number of superpixels is N K ={100,400,1000}, and define the color of the i-th superpixel in the k-th scale on the Lab color space as C ik , the spatial position is P ik , where the color of a superpixel is represented by the average color of all pixels inside it, and the spatial position is represented by the center of the superpixel.

[0016] 2. Calculate the saliency of single-scale superpixels

[0017] Because: (1) regions with different colors...

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 a salient object identification method. The method aims to increase an image identification rate. The method comprises the following main steps of 1: segmenting an image in a Lab color space to generate compact superpixels with clear boundary; 2: representing scarcity Rik of the superpixels by taking a spatial distance among the superpixels as a sum of weighted color distances of weight values; 3: selecting a sum of spatial distances between the superpixels and other superpixels similar to the superpixels in color to represent spatial dispersity Dik; 4: calculating a sum of the spatial distance among the superpixels to represent spatial dispersity Sik; 5: normalizing salient values of all dimensions to [0,55], and expressing the normalized values in the form of a grey-scale map; and 6: obtaining salient maps of all dimensions and fusing the salient maps of all the dimensions to obtain a final salient map.

Description

technical field [0001] The method relates to the technical fields of image processing, image recognition and computer vision. Background technique [0002] Salient object recognition is an important application of visual saliency. Visual saliency reflects the ability of objects in a scene to attract human visual attention. This ability can be quantified and estimated by establishing a visual saliency calculation model, and expressed with a grayscale image. Visual saliency is widely used in various fields of visual computing, including salient object recognition, video analysis, remote sensing image processing, image compression, etc. At first, the visual saliency calculation model was established based on features such as color, direction, and brightness; later, the calculation model of the Markov random chain was proposed. These calculation models are all calculation models based on local contrast, which can effectively detect target pixels. The size of the feature diffe...

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): G06K9/62
CPCG06F18/23213G06F18/251
Inventor 姜艾佳胡成华
Owner SICHUAN YONGLIAN INFORMATION TECH CO LTD
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