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

A video matting method based on three elements of color

A three-element, matting technology, applied to color TVs, color TV parts, TVs, etc., can solve problems such as lack of universality, inability to deal with background colors, cumbersome matting work, etc., to improve visual effects and restore Color and transparency, effects that enhance processing

Active Publication Date: 2017-02-22
BEIJING DAYANG TECH DEV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

It can be seen from the whole algorithm that the user needs to finely set a lot of parameters to determine the three polyhedrons and overflow color processing to achieve a satisfactory effect, and the work of matting is cumbersome.
[0004] The matting scheme of Vlahos assumes that the red, blue and green components satisfy a certain proportional relationship. On this basis, a series of empirical formulas for determining the alpha value are given after repeated trials. Although this method is simple and easy to implement, it is found in the experiment that it cannot be very Good for reverting the color of translucent areas, and not universal Can't handle the case where the background color is yellow-cyan-magenta

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 video matting method based on three elements of color
  • A video matting method based on three elements of color
  • A video matting method based on three elements of color

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0034] This embodiment is a video keying method based on the three elements of color, and the process of keying is as follows: figure 1 shown.

[0035] The following definitions and symbols are introduced:

[0036] make Indicates the color in the RGB space, and the corresponding color in the HSI space is , Indicates how opaque the color is, ,when , called fully transparent, when , called opaque, when , called translucency. and The opposite is transparency, with express, .

[0037] Select a certain pixel in the current frame image, and call this pixel the current pixel.

[0038] Other definitions are as follows:

[0039] current color : Indicates the color of the current pixel, expressed in RGB space as , expressed in HSI space as: ;

[0040] background color : Indicates the color of the background screen corresponding to the current color, expressed in RGB space as , expressed in HSI space as: ;

[0041] foreground color : Indicates the...

Embodiment 2

[0121] This embodiment is an improvement of the first embodiment, which is the refinement of the calculation of the background pixel point set in the first embodiment. The calculation parameters according to the background pixel point set include the following sub-steps:

[0122] Calculate the average hue of the background pixel point set, and compare it with the hues of 6 background colors, including blue, magenta, red, yellow, green, and cyan. The background color with the smallest difference is determined as the keying type. The upper and lower boundaries of the corresponding background color tone, and the overflow color upper and lower boundaries of the background color tone are determined accordingly.

[0123] Remove the points whose hue in the background pixel point set is not between the upper and lower boundaries of the background color hue.

[0124] Determine other parameters in the background information according to the remaining pixels. The background information c...

Embodiment 3

[0128] This embodiment is an improvement of the above-mentioned embodiment, and is a refinement of the above-mentioned embodiment regarding color overflow processing. The manner of overflow color processing described in the present embodiment comprises:

[0129] Sets the saturation of the foreground color that reflects the background color to zero.

[0130] For the foreground where the hue of the foreground color is near the upper and lower boundaries of the background color hue, the hue of the foreground color is shifted to the overflow boundary of the background color hue.

[0131] Color overflow processing refers to the further processing of the color that the human eye thinks has not been cut out. There are two main reasons for the visual appearance of not being cut out cleanly: the foreground that reflects the background color, the foreground color tone is above the background color tone, The foreground near the lower border. For the first case, it is often because the ...

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 relates to a video image matting method based on three elements of color. The method comprises collection of a video sequence; conversion of color space; acquisition of background information; distinguishing of an opaque area, a full transparent or semitransparent area and an area where background color is reflected in the foreground; calculation of the background color; calculation of opacity; calculation of foreground color; judgment of color bleeding processing; color bleeding; and finishing. According to the method, an image with a better effect can be cut out by respectively processing the opaque area, the full transparent or semitransparent area and the area where the background color is reflected in the foreground. Compared with the prior art, each pixel point in the matted image is accepted or rejected according to the brightness, the hue and the saturability of the pixel point instead of being simply reserved or abandoned, so that the image matting processing effect of a semitransparent foreground object is improved, the visual effect is improved simultaneously through the color bleeding processing, and the colors and the transparency of semitransparent objects such as glass, smoke, fur and shadows can be effectively recovered.

Description

technical field [0001] The present invention relates to a video matting method based on three elements of color, which is a computer image processing method and a digital video image processing method. Background technique [0002] Today's more popular keying systems include Red Giant's Primatte Keyer and Ultimatte's Advantage edge. After Effect, Digital Fusion, Shake and many other mainstream software support PrimatteKeyer. Ultimatte uses Vlahos' matting scheme, [0003] Primatte Keyer works in a three-dimensional RGB color space. The color space is divided into four separate categories by user selection of background pixels. These four categories are identified by three 128-sided polyhedra defined in 3D space. The first type is the area defined by the smallest polyhedron in the center, which contains 100% of all pixels of the background color, that is to say, any color of the small polyhedron will be completely replaced by the pixels of the background image, and color ...

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): H04N5/272
Inventor 姚威赵玉华
Owner BEIJING DAYANG TECH DEV
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