Image colorization method and system and computer readable storage medium

A storage medium and colorization technology, applied in the field of computer vision, can solve the problems of gradient disappearance and overfitting of learning models, and achieve the effect of great practical application value, good coloring effect, and high image definition.

Active Publication Date: 2020-05-12
YUNNAN UNIV
View PDF8 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

To solve the problem of gradient disappearance and overfitting of the learning model in the process of image colorization based on neural network

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 colorization method and system and computer readable storage medium
  • Image colorization method and system and computer readable storage medium
  • Image colorization method and system and computer readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0054] see Figure 4 The architecture of the colorization method, this embodiment discloses an image colorization method based on a deep convolutional autoencoder and multiple layer skipping, which includes the following steps:

[0055] A. Convert the image from RGB color space to YUV color space, and separate the Y channel data.

[0056] B. Copy the Y-channel data, and construct two-channel data together with the Y-channel data. to keep the same size as the UV channel that will be predicted.

[0057] C. Use two-channel data as the input of the deep convolutional self-encoder to predict the UV channel respectively; wherein, the deep convolutional self-encoder is composed of several re-skip layer connections.

[0058] D. Combining the Y channel data with the UV channel data learned in step C to construct a complete YUV color space image.

[0059] E. Converting the YUV color space image into an RGB color space image to obtain the final colorized image.

Embodiment 2

[0061] see Figure 4 The architecture of the colorization method, this embodiment discloses an image colorization method based on a deep convolutional autoencoder and multiple layer skipping, which includes the following steps:

[0062] A. Convert the image from RGB color space to YUV color space, and separate the Y channel as a grayscale image. The method of converting RGB color space to YUV color space is as follows:

[0063] Y=0.299R+0.587G+0.114B;

[0064] U=-0.147R-0.289G+0.436B;

[0065] V=0.615R-0.515G+0.100B.

[0066] Where R is the red channel, G is the green channel, and B is the blue channel. Y, U, and V are the three channels of the YUV color space, respectively.

[0067] B. Copy the Y channel data to obtain the Y' channel, and construct the two-channel data YY' together with the Y channel data. Thus the data size remains the same size as the output.

[0068] By duplicating the Y channel separated in step A, the Y' channel is obtained, and two channels YY' a...

Embodiment 3

[0112] Such as Figure 1~3 As shown, this embodiment discloses an image colorization method based on a deep convolutional autoencoder and multiple layer skipping, including the following steps:

[0113] S101: Acquiring a grayscale image:

[0114] Get the prepared grayscale image and convert the image into Numpy data format.

[0115] S102: convert the image from RGB color space to YUV color space:

[0116] The conversion from RGB color space to YUV color space can be described as:

[0117] Y=0.299R+0.587G+0.114B

[0118] U=-0.147R-0.289G+0.436B

[0119] V=0.615R-0.515G+0.100B

[0120] Where R is the red channel, G is the green channel, and B is the blue channel. Y, U, V are the three channels of the YUV color space.

[0121] S103: Separate the Y channel data:

[0122] First, the YUV data information of the grayscale image is separated into three channels, namely Y, U, and V channels, and the data of the Y channel is taken for operation.

[0123] S104: copy the Y channe...

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 an image colorization method and system and a computer readable storage medium. The method comprises the following steps: A, converting an image to be colorized from an RGB color space to a YUV color space, and separating Y channel data; B, copying the Y channel data, and constructing two-channel data together with the Y channel data; C, using the data of the two channels as the input of a depth convolution auto-encoder to predict UV channels respectively, wherein the depth convolution auto-encoder is formed by connecting a plurality of rejump layers; D, combining the Ychannel data with the UV channel data predicted in the step C to construct a complete YUV color space image; and E, converting the YUV color space image into an RGB color space image to obtain a final colorized image. According to the method, the problems of model gradient disappearance and over-fitting can be better solved, a better coloring effect and better image definition are achieved, artifacts generated by the image can be effectively reduced, and the color saturation can be effectively enhanced.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to an image colorization method based on a deep convolutional self-encoder and multiple layer jumps, a computer-readable storage medium and a corresponding system. Background technique [0002] Image colorization is a technique to assign an appropriate color to each pixel of the target grayscale image to make it look real and natural. Image colorization technology can provide humans with rich target scene information, so this technology has been widely used in various fields, such as color restoration of old photos, color schemes for assisting artists to design sketches, remote sensing images and night vision imaging systems, etc. And in the past few years, image colorization methods have received more and more attention and research from researchers. However, due to the diversity of object colors in the real world and the ambiguity of human perception of color, image colorization is...

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): G06T11/00G06T7/90G06N3/04
CPCG06T11/001G06T7/90G06N3/045
Inventor 邸一得金鑫江倩黄姗姗周维储星姚绍文王云
Owner YUNNAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap