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

Color image color consistency description method

A color image, consistent technology, applied in image enhancement, image analysis, image data processing, etc., can solve the problem of different image appearance

Active Publication Date: 2019-08-16
TIANJIN UNIV OF SCI & TECH
View PDF14 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to solve the problem that the appearance of the captured image is different due to the influence of the change of lighting conditions with time and space when the digital acquisition device captures the same scene image, and proposes a color image with consistent colors. method of sexual description

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
  • Color image color consistency description method
  • Color image color consistency description method
  • Color image color consistency description method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0069] Image 6 (a) is a color image captured by a digital acquisition device, the overall tone of the image is reddish, the size of the image is 3328×2496 pixels, and the range of color values ​​of image pixels R, G, and B is 0-255, Ω =255. see Figure 1-4 And steps 1-4, using the R, G, B data of the image and the technical solution provided by the present invention to obtain the L', a', b' data of the image and carry out cluster analysis of k sub-region divisions on the image, preferably , select the "K-means" method for clustering, and obtain the sub-regional division image of the "k sub-regional division" of the image, and its display effect is as follows Image 6 As shown in (b), the k sub-regions are all geometric shapes that do not have a square or a rectangle, and the conditions for dividing the image sub-regions are as follows: the ratio of the pixel area of ​​the smallest sub-region to the total area of ​​the image is 0.55%, within the sub-region Δ PL′a′b′ The ma...

Embodiment 2

[0072] Figure 7 (a) is a color image captured by a digital acquisition device, the R, G, and B color values ​​of the image range from 0 to 1024, Ω=1024, the size of the image is 3328×2496 pixels, and the overall tone of the image Weak, dark. see Figure 1-4 And steps 1-4, using the R, G, B data of the image and the technical solution provided by the present invention to obtain the L', a', b' data of the image and carry out cluster analysis of k sub-region divisions on the image, preferably , select the "K-means" method for clustering, and obtain the sub-regional division image of the "k sub-regional division" of the image, and its display effect is as follows Figure 7 As shown in (b), the k sub-regions are all geometric shapes that do not have a square or a rectangle, and the conditions for dividing the image sub-regions are as follows: the ratio of the pixel area of ​​the smallest sub-region to the total area of ​​the image is 1.5%, within the sub-region Δ PL′a′b′ The m...

Embodiment 3

[0076] Figure 8 (a) is a color image captured by a digital acquisition device, the overall tone of the image is greenish, the size of the image is 3328×2496 pixels, and the range of color values ​​of image pixels R, G, and B is 0-255, Ω =255. see Figure 1-4 And steps 1-4, using the R, G, B data of the image and the technical solution provided by the present invention to obtain the L', a', b' data of the color image, and carry out the cluster analysis of k sub-region divisions to the image, Preferably, the "K-means" method is selected for clustering, and the sub-region division image of "k sub-region division" of the image is obtained, and the display effect diagram is as follows Figure 8 As shown in (b), two of the k sub-regions are square or rectangular geometric shapes, and the image division conditions adopted are: the ratio of the pixel area of ​​the smallest sub-region to the total image area is 0.5%, and the Δ PL′a′b′ The maximum value of is 3.0, and the Δ between...

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 color image color consistency description method. The device can work by using one image; the method comprises the following steps: firstly, dividing an image into a pluralityof sub-regions with irregular geometrical shapes according to the visual similarity of colors and textures; constructing a specific sub-region analysis data set and carrying out linear fitting processing; t obtaining an image sub-region and the external illumination condition of the whole image, then realizing color consistency description of the image by utilizing a specific pixel conversion relationship, obtaining the intrinsic appearance characteristics of image contents, and solving the problem that the appearances of the captured images of the same scene under different illumination conditions are inconsistent. According to the method, manual intervention is not needed, the image content is actively analyzed to obtain the key conversion parameters, the parameters have the characteristics of one-time obtaining and pixel sharing, the processing speed is high, and practicability is high. The method belongs to the fields of artificial intelligence and intrinsic color recognition, andhas good application potential in the fields of image segmentation, printed product quality detection, target recognition and machine vision.

Description

technical field [0001] The invention relates to a method for describing the color consistency of a color image, which belongs to the category of artificial intelligence and intrinsic color recognition, and can be applied to the fields of image segmentation, print quality detection, target recognition, machine vision and the like. Background technique [0002] As an important attribute of color images, color has been widely used in image reproduction, machine detection and artificial intelligence, such as printing quality inspection, image segmentation, object recognition and tracking, etc. However, the color value of an image pixel is not only related to the essential attributes of the external scenery captured by the acquisition device, but also closely related to the external lighting conditions when the image is acquired. Therefore, if the external lighting conditions change, even if the same equipment is used to collect image data of the same external scene, it is diffic...

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): G06T7/90G06K9/62
CPCG06T7/90G06T2207/10024G06F18/23213
Inventor 陈永利田婕慧许浩墨
Owner TIANJIN UNIV OF SCI & 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