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

Coloring method and device based on weight learning

A weight and learning model technology, applied in the direction of texture/color, 2D image generation, instrument, etc., can solve problems such as no clear description, and achieve good coloring effect

Active Publication Date: 2021-01-08
SHANDONG NORMAL UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Many methods have been improved based on this, and some methods define different weight functions, but the weight calculation methods used in these methods are all predefined, and it is not clearly stated which weight can get better results

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
  • Coloring method and device based on weight learning
  • Coloring method and device based on weight learning
  • Coloring method and device based on weight learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] This embodiment discloses a coloring method based on weight learning, including the following steps:

[0043] Step 1: Select multiple grayscale images and corresponding color images, respectively calculate the feature difference between the adjacent pixels of the gray image and the corresponding color images, as the training data set;

[0044] The training set includes two parts, one is the feature difference between pixels in the grayscale image, and the other is the color difference in the color space. On the prepared grayscale image and color image pair, calculate the feature difference F rs and color difference D rs , Frs and D rs Do vectorization processing respectively, F rs Is a two-dimensional vector, representing the two-dimensional vector of brightness and gradient of adjacent pixels, D rs is a one-dimensional vector representing the color difference. L represents the number of relationship pairs between adjacent pixels. Will (F rs ,D rs ) as the train...

Embodiment 2

[0071] The purpose of this embodiment is to provide a computing device.

[0072] A coloring device based on weight learning, comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, the processor implements the following steps when executing the program, including:

[0073] Step 1: Select multiple grayscale images and corresponding color images, respectively calculate the feature difference between the adjacent pixels of the gray image and the corresponding color images, as the training data set;

[0074] Step 2: Based on the training data set, use the random forest algorithm to train the weight learning model;

[0075] Step 3: mark the color on the target grayscale image to be colored;

[0076] Step 4: Extract the feature difference between adjacent pixels from the target grayscale image, and use it as the input of the weight learning model to obtain the optimal weight;

[0077] Step 5: Perform color transfer according to ...

Embodiment 3

[0079] The purpose of this embodiment is to provide a computer-readable storage medium.

[0080] A computer-readable storage medium having stored thereon a computer program for coloring grayscale images, the program performing the following steps when executed by a processor:

[0081] Step 1: Select multiple grayscale images and corresponding color images, respectively calculate the feature difference between the adjacent pixels of the gray image and the corresponding color images, as the training data set;

[0082] Step 2: Based on the training data set, use the random forest algorithm to train the weight learning model;

[0083] Step 3: mark the color on the target grayscale image to be colored;

[0084] Step 4: Extract the feature difference between adjacent pixels from the target grayscale image, and use it as the input of the weight learning model to obtain the optimal weight;

[0085] Step 5: Perform color transfer according to the color mark and the optimal weight, an...

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 coloring method based on weight learning. The method includes the following steps: selecting a plurality of grayscale images and corresponding color images, respectively calculating feature differences between adjacent pixels of the gray images and the corresponding color images, and using the same as a training data set; utilizing a random forest algorithm to train a weight learning model on the basis of the training data set; marking colors on a to-be-colored target grayscale image; extracting feature differences between adjacent pixels from the target grayscale image, and using the same as input of the weight learning model to obtain optimal weights; and carrying out color transfer according to color marks and the optimal weights to acquire a color image corresponding to the target grayscale image. According to the coloring method of the invention, a manner of learning is utilized to obtain the weights, better interrelations between the pixels can be obtained, and a better coloring effect can be obtained.

Description

technical field [0001] The invention relates to a computer-aided image coloring method, in particular to an image coloring method based on weight learning. Background technique [0002] As a carrier of information, images are a true reflection of human visual perception, and color is very important information for people to understand images, and is one of the most important attributes of images. People have experienced the transition from black and white images to color images, but in the early days, the photography technology at that time was limited and could only generate black and white photos and videos, so adding appropriate colors to these old photos and videos made them more ornamental , becomes a very important task. [0003] The term colorization was first coined by Wilson Markle in 1970, and was defined as a process of coloring black and white or monochrome images and videos with the aid of a computer [1] . The emergence of coloring technology can restore, enh...

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): G06T11/00
CPCG06T11/001
Inventor 郑元杰宋双连剑刘弘魏本征
Owner SHANDONG NORMAL UNIV
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