Self-supervised color correction method for multi-device domain AWB enhancement
A color correction, multi-device technology, applied in image enhancement, neural learning methods, texture/color, etc., can solve the problems of different sensors of different cameras, limited application, difficult to promote the performance of a single model, etc., to achieve generalization and accuracy The effect of improving and reducing the consumption of human and material resources
Pending Publication Date: 2022-05-10
BEIJING UNIV OF POSTS & TELECOMM
View PDF0 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] However, learning-based methods cannot directly use data from different camera (Camera) domains. The reason is that the sensors of different cameras are different.
It has been demonstrated that data from different sensors hardly contribute to the performance of a single model, which further limits the practical application of learning-based models
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0041] Specific embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, but are not intended to limit the protection scope of the present invention.
[0042] In the fields of image restoration, photography, etc., due to the influence of light, photos often have color casts. The general method is that professionals use the naked eye to adjust the color of the image to a comfortable state, but this method is unreliable and difficult. Promoted. The purpose of the color constancy technology is to automatically remove the influence of light and correct the color cast image to the condition of no color cast in white light. However, in the prior art, this method has been facing the problem of data scarcity.
[0043] In view of the above problems, an embodiment of the present invention first proposes an image generation method based on an illumination invariant description. Such as figure 1 As shown, it mainly includes th...
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
Login to View More Abstract
The invention provides a self-supervised color correction method for multi-device domain AWB enhancement, aiming at the problem of data shortage in the existing color constancy technology, and the self-supervised color correction method comprises the following steps: generating an illumination device irrelevant image through an illumination invariant description theory; utilizing a Von Kries model and an illumination invariant description theory to generate a self-supervision label correspondingly learned by the illumination equipment irrelevant image; a deep convolutional neural network is used to learn mapping of illumination equipment-independent images to corresponding labels. According to the method, the purpose of enhancing the AWB by multi-device domain data joint training can be achieved, so that the generalization and precision of the AWB technology are effectively improved, and the consumption of manpower and material resources is reduced.
Description
technical field [0001] The invention relates to the technical field of color constancy, in particular to a color cast image correction method based on an illumination invariant description. Background technique [0002] Research on visual cognition and psychology shows that the human system has the function of "Color Constancy", that is, the human visual system can automatically adjust the color of the scene object in the human eye according to the scene seen, so as to eliminate the Due to the influence of the color of the light source on the color of the object in the scene, the color of the object can maintain a certain "constancy" under different light sources. But for a computer system, it is difficult to judge what is an appropriate color, and it is difficult to automatically correct the color cast of an object in a huge lighting space. Therefore, the color constancy technology hopes to automatically restore the uncolored image under white light from the color cast pic...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06T5/00G06T11/00G06N3/04G06N3/08
CPCG06T11/001G06N3/08G06N3/045G06T5/77
Inventor 明安龙唐宇翔李春晓康学净
Owner BEIJING UNIV OF POSTS & TELECOMM