Unlock instant, AI-driven research and patent intelligence for your innovation.

A Saliency Calculation Method of RGB-D Image

A technology of RGB-D and calculation methods, applied in the field of image processing, can solve problems such as limited accuracy, difficult post-processing of images, unsatisfactory complex scene results, etc., and achieve the effect of improving accuracy

Active Publication Date: 2020-04-03
ANHUI UNIVERSITY
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Traditional saliency detection is based on two-dimensional color images, and the accuracy is limited. The detection results for simple scenes are better, but the results for complex scenes are unsatisfactory, and it is difficult to be used for post-processing of images.

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
  • A Saliency Calculation Method of RGB-D Image
  • A Saliency Calculation Method of RGB-D Image
  • A Saliency Calculation Method of RGB-D Image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, so as to define the protection scope of the present invention more clearly.

[0034] Embodiments of the invention include:

[0035] A RGB-D image saliency calculation method, which defines the background of an RGB-D image as the upper, lower, left, right, front, and rear six surfaces in a three-dimensional space, and uses this as prior knowledge to perform RGB-D Image saliency calculation. In this method, an RGB-D image is composed of an RGB color image and a Depth image.

[0036] see figure 1 , after superpixel segmentation of the RGB color image, each superpixel has a depth value according to the depth information provided by the Depth depth map, so that each superpixel will no longer be a superpixel in the RGB color pla...

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 for calculating the saliency of an RGB-D image, which defines the background of an RGB-D image as six surfaces of upper, lower, left, right, front and rear in a three-dimensional space, and takes this as prior knowledge Carry out the RGB-D image saliency calculation process. The method regards the RGB-D image as a three-dimensional image in three-dimensional space, fully utilizes the depth information of the RGB-D image, accurately defines the boundary of the three-dimensional space, and more truly reflects the human visual attention mechanism. detection accuracy. The method of the present invention proves its effectiveness and obvious advantages in effect through image database test comparison.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a RGB-D image saliency calculation method with a three-dimensional space boundary as a priori background. Background technique [0002] Humans obtain information through various channels, such as vision, hearing, touch, smell, etc., and 83% of them are obtained through the visual system. It is conceivable how important vision is to human life. Further research shows that people Often there is a tendency to selectively filter visual information rather than put all the information into the brain for processing. This selective process is called saliency detection in the field of computer graphics. When a computer processes an image, it can use a similar strategy. First, preprocess the image, select the important part that is easy to attract people's attention, and filter out the unimportant part that is not easy to attract people's attention. Post-processing is of great s...

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): G06T7/50
Inventor 刘政怡谢丰宋腾飞郭星李炜
Owner ANHUI UNIVERSITY