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

Image foreground and background boundary saliency assessment method and system

A background boundary and evaluation system technology, applied in the field of image processing, can solve problems such as poor results

Active Publication Date: 2019-05-14
GUANGZHOU XINGFU NETWORK TECH
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] Image segmentation is one of the basic operations of computer image processing. At present, some image segmentation algorithms (such as GrabCut) have a good segmentation effect when the contrast between the foreground and background boundaries is large, but the contrast between the foreground and background boundaries is not significant. does not work well for
[0003] However, in many cases, it is necessary to estimate whether the segmentation effect is good before segmenting the image. For example, in the automatic image matting process of ID photos, it is necessary to estimate the segmentation effect so that it can be fed back to the user in time, which is not satisfactory Prompt user to retake when effect

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
  • Image foreground and background boundary saliency assessment method and system
  • Image foreground and background boundary saliency assessment method and system
  • Image foreground and background boundary saliency assessment method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0084] Such as figure 1 As shown, the image foreground and background boundary saliency assessment method of this embodiment includes boundary positioning, boundary contrast calculation and boundary saliency scoring;

[0085] Boundary positioning is the basis. The original image is scaled to an image with a column width of a certain pixel (240 pixels in this example), and some specified 4*4 pixel square areas are used as the background. First, these background areas are used as the seeds of the background queue (Applied to the invention patent application with application number 201510547197.5 and titled "A Camera and Method for Automatically Cutting Portraits", two rectangles that completely contain the foreground can be used, and these two rectangles are determined by the previous period. According to the algorithm, in other general applications, the point determined as the background can be specified, and all 4*4 pixel square areas outside the edge of the image initialize t...

Embodiment 2

[0117] The difference between this embodiment and the above-mentioned embodiment 1 is that in boundary saliency scoring, considering that continuous low-contrast boundary areas have a greater impact on matting than scattered single-contrast boundary blocks, each penalty item marking is For proper correction, the distance factor from other low-contrast borders is added as a deduction factor to the deduction item marking of each low-contrast border, and the deduction formula for a single low-contrast border is modified as follows:

[0118]

[0119] in, Δs=∑(|Δx|+|Δy|), Δs is the measure of the distance from each low-contrast border to other low-contrast borders, |Δx|, |Δy| represent the row distance and column distance from other low-contrast borders respectively; add factor The purpose of is similar to that of Embodiment 1, reducing the penalty at the boundary of higher contrast and increasing the penalty at the boundary of low contrast.

[0120] Therefore, subtract the ...

Embodiment 3

[0125] Such as Figure 9 As shown, the image foreground and background boundary saliency evaluation system of this embodiment includes a boundary positioning module, a boundary contrast calculation module and a boundary saliency scoring module;

[0126] The boundary positioning module includes:

[0127] The scaling unit is used to scale the original image in proportions such as length and width, and scale it to an image whose column width is a certain pixel;

[0128] The creation unit is used to create two two-dimensional matrices of the same size as the scaled image, which are respectively recorded as MIV and MIBQ, and initialize each element in MIV and MIBQ; among them, MIV is used to mark the image to determine the background information, and MIBQ Used to indicate whether the image block is in the background queue Q;

[0129] The setting unit is used to set a number of points determined as the background, a set of square areas of 4*4 pixels starting from these points as 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 method and system for evaluating significance of image foreground and background boundaries. The method comprises boundary positioning, boundary contrast calculation and boundary significance scoring. The corresponding system comprises a boundary positioning module, a boundary contrast calculation module and a boundary significance scoring module. According to the method and system disclosed by the invention, for an image with a pure background, in the case that a plurality of background points are given, foreground and background boundaries can be automatically positioned, a boundary contrast is calculated, and whether the foreground and background boundaries of the given image are significant are evaluated to effectively evaluate whether the background is well distinguished from the foreground, so as to evaluate image division quality in advance. The method and system can be widely used in the technical field of image division, for example, sectional drawing of identification pictures shot by mobile terminals.

Description

technical field [0001] The invention relates to a boundary saliency assessment method and system, in particular to an image foreground and background boundary saliency assessment method and system, belonging to the field of image processing. Background technique [0002] Image segmentation is one of the basic operations of computer image processing. At present, some image segmentation algorithms (such as GrabCut) have a good segmentation effect when the contrast between the foreground and background boundaries is large, but the contrast between the foreground and background boundaries is not significant. The effect is not good in the case. [0003] However, in many cases, it is necessary to estimate whether the segmentation effect is good before segmenting the image. For example, when the ID photo is automatically cut out, it is necessary to estimate the segmentation effect so that it can be fed back to the user in time, which does not satisfy the best cutout Prompt the use...

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): G06T7/00
Inventor 易焕银叶燕妮其他发明人请求不公开姓名
Owner GUANGZHOU XINGFU NETWORK TECH
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