CN101489144A - Automatic color space and scenery conversion method for camera - Google Patents
Automatic color space and scenery conversion method for camera Download PDFInfo
- Publication number
- CN101489144A CN101489144A CNA2008100041222A CN200810004122A CN101489144A CN 101489144 A CN101489144 A CN 101489144A CN A2008100041222 A CNA2008100041222 A CN A2008100041222A CN 200810004122 A CN200810004122 A CN 200810004122A CN 101489144 A CN101489144 A CN 101489144A
- Authority
- CN
- China
- Prior art keywords
- color space
- target image
- raw video
- brightness
- camera
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
A camera automatic color space and situation conversion method comprise at least the following steps: selecting an original image and an object image; executing an automatic color space conversion, respectively converting the color space of the two images to another color space; then executing a color stage match and performing a characteristic grouping and a most similar neighborhood searching; and executing a color information copy, combining the mutual matched relative regions of the object image and the original image, then copying the color and brightness information of the object region to the relative region of the original image, and reaching the objective of situation conversion.
Description
Technical field
The present invention relates to a kind of auto color space and scenery conversion method of camera.
Background technology
General camera provides the exposal model of various different situations conversions, in order to present different visual effects, illustrate as pseudo-classic, black and white, with the exposal model of situation conversion such as embossment.
The exposal model of situation that prior art provides conversion, not only dullness and pattern are limited, do not apply public demand already.Utilize color space conversion, colour temperature to change or the edge special processing in the prior art, reach the purpose of situation conversion.Illustrate, the video conversion of captured by day high colour temperature is become the image of low colour temperature when the dusk.Wherein video conversion only changes the gain (gain) or the color correction matrix (color correction matrix) of each color, and this colour temperature of utilizing changes the video conversion mode that reaches, and is more unnatural, and exposal model is limited.
According to No. 558670 announcements of TaiWan, China patent of invention Announcement Number a kind of " digital camera " with light source setting and color correction functions, utilize colour temperature to change the gain that changes each color, and colour correction, video conversion is provided, resulting image is nature more not, and exposal model is limited.
So, the part improved of the above-mentioned defective of inventor's thoughts, and according to the correlation experience of being engaged in for many years in this respect, the concentrated observation and research, and cooperate the utilization of scientific principle, and propose a kind of reasonable in design and effectively improve the technical solution of the present invention of above-mentioned defective.
Summary of the invention
Therefore purpose of the present invention is exactly auto color space and the scenery conversion method that proposes a kind of camera, and its resulting image is more natural, and has diversified exposal model, reaches the purpose of situation conversion.
According to above-mentioned purpose of the present invention, the invention provides a kind of auto color space and scenery conversion method of camera, comprise at least: a selected raw video and a target image; Then carry out an auto color space conversion, convert a color space of raw video and target image to another color space respectively, and obtain raw video and target image and distribute in the brightness color range of another color space, a plurality of pixel values of brightness color range distribution calculating raw video and target image correspond to the appearance probability in a plurality of brightness color range values of another color space respectively, and each those pixel value has chroma information; Carry out color stage match again, at first raw video and target image are carried out a characteristic grouping respectively in the brightness color range distribution of another color space, carry out similar neighborhood searching again, find out raw video and distribute in the brightness color range of another color space and be matched with target image most and distribute in the brightness color range of another color space; And carry out multimedia message of the same colour breath and duplicate, with target image and raw video in another color space the brightness color range of coupling distribute, find out the pixel value of the target image that corresponds to, duplicate this chroma information and arrive raw video, to reach the purpose that situation is changed.
The present invention provides a kind of auto color space and scenery conversion method of camera in addition, comprises at least: a selected raw video and a target image; Carry out an auto color space conversion, convert a color space of raw video and target image to another color space respectively, obtaining raw video and target image distributes in a brightness color range of another color space, a plurality of pixel values of brightness color range distribution calculating raw video and target image correspond to the appearance probability in a plurality of brightness color range values of another color space respectively, and each those pixel value has chroma information; Then carry out a gamma correction, becoming the brightness color range distribution non-linear conversion of another color space of raw video, the brightness color range of another color space of approximate target image distributes, obtain the color range of raw video after another color space brightness correction and distribute, can obtain the pixel value behind the raw video gamma correction as calculated; Carry out a characteristic grouping, the pixel value of target image is carried out statistical calculation, obtain the characteristic value of the pixel value of target image; Carry out similar neighborhood searching, after the characteristic value normalization (Normalization) with the pixel value of the pixel value of the gamma correction of raw video and target image, search and compare and find out matching value; And carry out multimedia message of the same colour breath and duplicate, with target image and raw video in another color space the brightness color range of coupling distribute, find out the value of the target image that corresponds to, duplicate chroma information and arrive raw video, reach the purpose that situation is changed.
The present invention has following beneficial effect: at first selected raw video and target image, carry out auto color space conversion and color range coupling respectively, or raw video carried out gamma correction, at last with the relative brightness of target image and chroma information reproduction to raw video, reach the purpose of situation conversion.
In order to make narration of the present invention more detailed and complete, in the following summary of the invention, provide many different embodiment or example, can be used for understanding the application of the different characteristic in different embodiment with reference to following description and conjunction with figs..
Description of drawings
Fig. 1 is the process step figure that illustrates according to the method for a preferred embodiment of the present invention.
200-209: the method flow step
Embodiment
Please refer to Fig. 1 and be the process step figure that illustrates according to the method 200 of a preferred embodiment of the auto color space of camera and scenery conversion method.
Selected image, a selected raw video.Illustrate: be used as raw video at the image that summer is captured; Or captured image is used as raw video in the Taibei.
Selected image, a selected target image.Illustrate: with respect to above-mentioned carry summer captured image be used as raw video, and captured in autumn image is used as target image; Or be used as raw video with respect to the above-mentioned mentioned image captured in the Taibei, and captured image is used as target image in Kaohsiung.Moreover, in the present embodiment, this target image further for the image that stores several particular context patterns in advance in camera, and give the title of its situation, can select contextual model decided at the higher level but not officially announced or user contextual model is provided voluntarily.
Carry out the auto color space conversion, for a color space conversion of raw video being become another color space of raw video, and obtain raw video and distribute in a brightness color range of another color space, a plurality of pixel values of brightness color range distribution calculating raw video correspond to the appearance probability in a plurality of brightness color range values of another color space, and each those pixel value has chroma information.Illustrate, the color space of raw video is for comprising at least: RGB or CIE XYZ; Another color space of raw video is for comprising at least: CIE LAB (1 α β) or YUV (YCbCr).
Wherein, in rgb color space, R defines red value, and G defines green value, and B defines blue valve; In CIE XYZ color space, X defines red values, and Y defines green values, and Z defines blue values; In CIE LAB color space, the bright value of L definition, A definition red value (on the occasion of) to green value (negative value), the yellow value of B definition (on the occasion of) to blue valve (negative value); In the YUV color space, the bright value of Y definition, U defines chromatic value, and V defines concentration value.
Convert the pixel value of raw video to brightness color range value respectively, make that the color relevance at another color space reduces at another color space.Illustrate: on the image of rgb color space, the red value in the rgb color space, green value and blue valve each other may relevant property.Such as, at the pixel value of most of image, when wherein blue valve is bigger than normal, red value and green value also can be bigger than normal, proofread and correct a kind of brightness color range value if will in color space, adjust, also will consider the variation of the brightness color range value of other two dimensions, make the color space conversion will become very complicated.So rgb color space is converted to CIE LAB color space, can be so that the color space conversion becomes easy.
Carry out the auto color space conversion, convert another color space of raw video to for color space with target image, obtain raw video and distribute in the brightness color range of another color space, a plurality of pixel values of brightness color range distribution calculating raw video correspond to the appearance probability in a plurality of brightness color range values of another color space.
Carry out gamma correction, in another color space, become the brightness color range distribution non-linear conversion of raw video that the brightness color range of approximate target image distributes, obtain raw video to distribute, calculate the pixel value of the gamma correction of raw video in another brightness color range of another color space.
Carry out characteristic grouping, the pixel value of target image is carried out statistical calculation, obtain the characteristic value of the pixel value of target image.
Characteristic grouping wherein, with pixel value in the target image, be at least about 5 * 5 local size for its pixel value, carry out brightness average (mean), standard deviation (standard deviation), with the statistical calculation of gradient (gradient), resulting statistical value is used as the characteristic value of pixel value in the target image.Because target image is very big in the repeatability (redundancy) that the brightness color range of another color space distributes, utilizing k-mean to hive off, statistical method is simple and easy hives off, and makes follow-up image search space reduce.In addition, (mode of taking binary tree to set up converts the characteristic value of pixel value in the target image to characteristics tree, more makes follow-up image search space significantly reduce for VQ, statistical method vectorquantization) also can to utilize general vector quantization.
Step 207
Carry out similar neighborhood searching, after the characteristic value normalization (Normalization) with the pixel value of the pixel value of the gamma correction of step 205 gained raw video and target image, search and compare and find out matching value.
In addition, if before execution in step 205, the brightness color range distribution of another color space of discovery raw video distributes very approximate with the brightness color range of another color space of target image, with regard to not needing raw video is carried out gamma correction step 205, but accelerate for making similar neighborhood searching step 207 searching efficiency, the brightness color range of another color space of raw video distributed carry out characteristic grouping step 206, obtain the characteristic value of the pixel value of raw video, carry out similar neighborhood searching step 207 again, with the characteristic value of pixel value in resultant raw video and the target image after the characteristic grouping, search and comparison find out pixel value in raw video and the target image between the characteristic value of mating most.Wherein, definition color range coupling (HistogramMatching) comprises in regular turn at least: characteristic grouping step 206 and similar neighborhood searching step 207.
The execution color information is duplicated, with target image and raw video in another color space the brightness color range of coupling distribute, find out the chroma value of the target image that corresponds to, copy to raw video, change back original color space at last, obtain a new image.
Carry out the image that shifts, reach the purpose of situation conversion.
The present invention utilizes the method for auto color space conversion, at first selected raw video and target image, then carry out the auto color space conversion, the color space of image is converted to another color space, then carries out color range coupling, carry out characteristic grouping and similar neighborhood searching, carry out the image that shifts at last, near the color part of the target image of raw video, copy to raw video with, reach the purpose of situation conversion.Wherein characteristic grouping utilize k-mean to hive off statistical method is simple and easy hives off, more comprise the statistical method of general vector quantization.
Though the present invention discloses as above with a preferred embodiment; right its is not in order to qualification the present invention, any those skilled in the art, without departing from the spirit and scope of the present invention; when can being used for a variety of modifications and variations, so protection scope of the present invention is as the criterion when the scope that the right claim is defined.
Claims (10)
1. the auto color space and the scenery conversion method of a camera is characterized in that, comprise at least:
A selected raw video and a target image;
Carry out an auto color space conversion, convert a color space of this raw video and this target image to another color space respectively, obtaining this raw video and this target image distributes in a brightness color range of this another color space, a plurality of pixel values that this raw video and this target image are calculated in this brightness color range distribution correspond to the appearance probability in a plurality of brightness color range values of this another color space respectively, and each those pixel value has chroma information;
Carry out color stage match, at first this raw video and this target image are carried out a characteristic grouping respectively in this brightness color range distribution of this another color space, carry out similar neighborhood searching again, find out this raw video and distribute in this brightness color range of this another color space and be matched with this target image most and distribute in this brightness color range of this another color space; And
Carrying out multimedia message breath of the same colour duplicates, with this target image and this raw video in this another color space this brightness color range of coupling distribute, find out those pixel values of this target image that corresponds to, duplicate this chroma information, to reach the purpose of situation conversion to this raw video.
2. the auto color space and the scenery conversion method of camera as claimed in claim 1 is characterized in that, this color space is for comprising at least: RGB or CIEXYZ.
3. the auto color space and the scenery conversion method of camera as claimed in claim 1 is characterized in that, this another color space is for comprising at least: CIELAB (1 α β) or YUV (YCbCr).
4. the auto color space and the scenery conversion method of camera as claimed in claim 1 is characterized in that, this characteristic grouping comprises the k-mean statistical method of hiving off at least.
5. the auto color space and the scenery conversion method of camera as claimed in claim 1 is characterized in that, this characteristic grouping more comprises the statistical method of vector quantization.
6. the auto color space and the scenery conversion method of a camera is characterized in that, comprise at least:
A selected raw video and a target image;
Carry out an auto color space conversion, convert a color space of this raw video and this target image to another color space respectively, obtaining this raw video and this target image distributes in a brightness color range of this another color space, a plurality of pixel values that this raw video and this target image are calculated in this brightness color range distribution correspond to the appearance probability in a plurality of brightness color range values of this another color space respectively, and each those pixel value has chroma information;
Carry out a gamma correction, becoming this brightness color range distribution non-linear conversion of this another color space of this raw video, this brightness color range of this another color space of approximate this target image distributes, another brightness color range that obtains this another color space of this raw video distributes, and calculates those pixel values of the gamma correction of this raw video;
Carry out a characteristic grouping, those pixel values of this target image are carried out statistical calculation, obtain the characteristic value of those pixel values of this target image;
Carry out similar neighborhood searching, after the characteristic value normalization with those pixel values of those pixel values of the gamma correction of this raw video and this target image, search and compare and find out matching value; And
Carrying out multimedia message breath of the same colour duplicates, with this target image and this raw video in this another color space this brightness color range of coupling distribute, find out those pixel values of this target image that corresponds to, duplicate this chroma information, to reach the purpose of situation conversion to this raw video.
7. the auto color space and the scenery conversion method of camera as claimed in claim 6 is characterized in that, this color space is for comprising at least: RGB or CIEXYZ.
8. the auto color space and the scenery conversion method of camera as claimed in claim 6 is characterized in that, this another color space is for comprising at least: CIELAB (1 α β) or YUV (YCbCr).
9. the auto color space and the scenery conversion method of camera as claimed in claim 6 is characterized in that, this characteristic grouping comprises the k-mean statistical method of hiving off at least.
10. the auto color space and the scenery conversion method of camera as claimed in claim 6 is characterized in that, this characteristic grouping more comprises the statistical method of vector quantization.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008100041222A CN101489144B (en) | 2008-01-17 | 2008-01-17 | Automatic color space and scenery conversion method for camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008100041222A CN101489144B (en) | 2008-01-17 | 2008-01-17 | Automatic color space and scenery conversion method for camera |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101489144A true CN101489144A (en) | 2009-07-22 |
| CN101489144B CN101489144B (en) | 2010-09-29 |
Family
ID=40891784
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008100041222A Expired - Fee Related CN101489144B (en) | 2008-01-17 | 2008-01-17 | Automatic color space and scenery conversion method for camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101489144B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102968877A (en) * | 2012-11-15 | 2013-03-13 | 镇江石鼓文智能化系统开发有限公司 | Flame detection device based on video image analysis |
| CN103258183A (en) * | 2012-11-15 | 2013-08-21 | 镇江石鼓文智能化系统开发有限公司 | Video image preprocessing module based on video image analysis flame |
| CN107341835A (en) * | 2017-07-07 | 2017-11-10 | 武汉斗鱼网络科技有限公司 | Image processing method, device, electronic equipment and computer-readable recording medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0650299B1 (en) * | 1993-10-20 | 1998-07-22 | Laboratoires D'electronique Philips S.A.S. | Method of processing luminance levels in a composite image and image processing system applying this method |
| JP4662356B2 (en) * | 2005-10-21 | 2011-03-30 | キヤノン株式会社 | Imaging apparatus and control method thereof, control program thereof, and storage medium storing control program |
-
2008
- 2008-01-17 CN CN2008100041222A patent/CN101489144B/en not_active Expired - Fee Related
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102968877A (en) * | 2012-11-15 | 2013-03-13 | 镇江石鼓文智能化系统开发有限公司 | Flame detection device based on video image analysis |
| CN103258183A (en) * | 2012-11-15 | 2013-08-21 | 镇江石鼓文智能化系统开发有限公司 | Video image preprocessing module based on video image analysis flame |
| CN107341835A (en) * | 2017-07-07 | 2017-11-10 | 武汉斗鱼网络科技有限公司 | Image processing method, device, electronic equipment and computer-readable recording medium |
| CN107341835B (en) * | 2017-07-07 | 2018-08-03 | 武汉斗鱼网络科技有限公司 | Image processing method, device, electronic equipment and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101489144B (en) | 2010-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110691277B (en) | Video signal processing method and device | |
| JP4708192B2 (en) | Dynamic camera color correction device and video search device using the same | |
| WO2004098167A3 (en) | A system and process for generating high dynamic range video | |
| CN1977542A (en) | Dominant color extraction using perceptual rules to produce ambient light derived from video content | |
| CN101998063B (en) | Foreground image separation method | |
| US20080292184A1 (en) | Apparatus and method for automatically adjusting white balance | |
| JP2009055465A (en) | Image processing device and method | |
| CN103491357A (en) | Auto white balance treatment method of image sensor | |
| CN101355634B (en) | Method and system for regulating image chromaticity | |
| CN101287130A (en) | Apparatus and method for generating wide colour gamut signal in image capturing device | |
| CN101489144B (en) | Automatic color space and scenery conversion method for camera | |
| US20070098256A1 (en) | Motion picture content editing | |
| US7821580B2 (en) | Contour free point operation for video skin tone correction | |
| WO2022218245A1 (en) | Image processing method and apparatus, electronic device, and readable storage medium | |
| Kotera | A scene-referred color transfer for pleasant imaging on display | |
| CN111107330A (en) | Color cast correction method for Lab space | |
| Lam et al. | Automatic white balancing using standard deviation of RGB components | |
| Lam et al. | Automatic white balancing using luminance component and standard deviation of RGB components [image preprocessing] | |
| CN1977529A (en) | Dominant color extraction for ambient light derived from video content mapped through unrendered color space | |
| CN106331663B (en) | A kind of interaction material acquisition system and method for portable equipment | |
| CN102957919A (en) | Digital image brightness enhancement method | |
| TWI380677B (en) | Method of color space transformation to transfer prospect for camera sphere | |
| US20070076014A1 (en) | Universal color decoder an method for decoding input signal for a multiple primary color display system | |
| CN112598570B (en) | Banner background self-adaptive color migration method and system | |
| CN110490945B (en) | Image color adjusting method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100929 Termination date: 20190117 |