CN106131526A - A kind of white balancing treatment method based on rgb space and device - Google Patents

Abstract

A kind of white balancing treatment method based on rgb space of disclosure and device, in the method, in obtaining pending image, each pixel is respectively at R, G, after the original value of tri-Color Channels of B, the reddish blue calculating described pixel is poor, and by the poor and mellow lime pixel Rule of judgment of reddish blue, determine mellow lime pixel, wherein, with the change of current colour temperature, iteration updates mellow lime pixel Rule of judgment, number further according to mellow lime pixel accounts for the ratio of pixel sum in described pending image, it is calculated described pixel respectively at R, G, the yield value of tri-Color Channels of B, according to yield value, obtain the corrected value of pixel, again by each pixel in pending image respectively at R, G, the original value of tri-Color Channels of B adjusts to corrected value, realize the white balance to pending image to process.By scheme disclosed in the present application, it is possible to realize the white balance to image and process, thus reduce the aberration in image, improve picture quality.

Description

A kind of white balancing treatment method based on rgb space and device

Technical field

The present embodiments relate to technical field of image processing, particularly relate to a kind of white balance based on rgb space and process Method and device.

Background technology

Different light sources has different spectral components and distribution, and this phenomenon is referred to as colour temperature in colorimetry.Same White object, there will be aberration under the light source of different-colour irradiates, such as, under the light source of low colour temperature irradiates, and white object Can be the reddest, and under the light source of high color temperature irradiates, white object can be partially blue.

Wherein, the aberration caused by different-colour is corrected, so that white object presents real white, is referred to as For white balance.Such as, human visual system can adjust lightsensitivity automatically according to the colour temperature of current light source, makes human eye perception The object color approximate constant arrived, therefore, when human visual system watches white object under different-colour, all can experience To real white, it is to avoid the impact of aberration, it is achieved white balance.

But, inventor finds in the research process of the application, and image capturing system often cannot realize white balance, leads There is aberration in the image causing to collect, causes the poor image quality collected.

Summary of the invention

For overcoming problem present in correlation technique, the embodiment of the present invention provides at a kind of white balance based on rgb space Reason method and device.

In order to solve above-mentioned technical problem, the embodiment of the invention discloses following technical scheme:

First aspect according to embodiments of the present invention, it is provided that a kind of white balancing treatment method based on rgb space, including:

Obtain in pending image each pixel respectively at the original value of tri-Color Channels of R, G, B, and according to described The reddish blue of pixel described in raw calculation is poor；

Poor and mellow lime for the reddish blue of described pixel pixel Rule of judgment is contrasted, is determined for compliance with described middle spodogram The pixel of vegetarian refreshments Rule of judgment is mellow lime pixel, and accounts in described pending image according to the number of described mellow lime pixel The ratio described pixel of calculating of pixel sum is respectively at the yield value of tri-Color Channels of R, G, B, wherein, and described middle spodogram With the change of current colour temperature, iteration updates vegetarian refreshments Rule of judgment；

According to described pixel respectively at the yield value of tri-Color Channels of R, G, B, and described pixel respectively R, The original value of tri-Color Channels of G, B, calculates described pixel respectively at the corrected value of tri-Color Channels of R, G, B, and by institute State each pixel in pending image to adjust to described corrected value at the original value of tri-Color Channels of R, G, B respectively.

Optionally, also include:

After obtaining original image, described original image is averagely divided into n region, and obtains pixel in each region Point is respectively in the meansigma methods of tri-Color Channels of R, G, B, and wherein, n is default integer；

By each pixel in described each region respectively the value of tri-Color Channels of R, G, B be adjusted to described averagely Value；

It is a pixel by described each area reduction, and the image after reducing is as described pending image.

Optionally, also include:

In obtaining pending image, each pixel is respectively after the original value of tri-Color Channels of R, G, B, and at meter Before calculating the reddish blue difference of described pixel, according to the original value of described tri-Color Channels of R, G, B, search overexposure therein and/ Or the darkest pixel, and delete described overexposure and/or the darkest pixel.

Optionally, gray pixel during described mellow lime pixel Rule of judgment includes the first mellow lime pixel Rule of judgment and second Point Rule of judgment；

Described first mellow lime pixel Rule of judgment is:

$\left\{\begin{array}{c}sum\_min\_I<r+b<sum\_max\_I\\ r>r\_\min \_I\\ b>b\_\min \_I\end{array}\right.;$

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, sum_min_I, sum_max_I, r_ Min_I and b_min_I is respectively the first judgement parameter, and described first judges that parameter is during described pixel is under certain color temperature The judgement parameter met is needed during spodogram vegetarian refreshments；

Described second mellow lime pixel Rule of judgment is:

$\left\{\begin{array}{c}sum\_min\_II<r+b<sum\_max\_II\\ r>r\_\min \_II\\ b>b\_\min \_II\end{array}\right.;$

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, sum_min_II, sum_max_II, r_ Min_II and b_min_II is respectively the second judgement parameter, and described second judges that parameter is that described pixel is under current colour temperature The judgement parameter met is needed during mellow lime pixel.

Optionally, poor and mellow lime for the reddish blue of described pixel pixel Rule of judgment is contrasted, is determined for compliance with institute The pixel stating mellow lime pixel Rule of judgment is mellow lime pixel, and according to the number of described mellow lime pixel account for described in wait to locate In reason image, pixel described in the ratio-dependent of pixel sum is respectively at the yield value of tri-Color Channels of R, G, B, including:

41) poor and described for the reddish blue of described pixel the first mellow lime pixel Rule of judgment is contrasted, determine full The pixel of the described first mellow lime pixel Rule of judgment of foot is first object pixel, and calculates described first object pixel Number account for the first ratio of the sum of pixel in described pending image；

42) judge that described first ratio, whether less than the first preset ratio, compares if described first ratio is preset less than first Example, it is determined that described first object pixel is described mellow lime pixel, and determine that described pixel is respectively in tri-face of R, G, B The yield value of chrominance channel keeps constant, or, if described first ratio is less than the first preset ratio, keep described second to judge ginseng Number constant, and perform step 43) operation；

If described first ratio is not less than described first preset ratio, then updates described second according to equation below and judge ginseng Number, and in the updated perform step 43) operation:

$\left\{\begin{array}{c}sum\_max\_II=r_{mean}+b_{mean}+q\\ sum\_min\_II=r_{mean}+b_{mean}-q\\ r\_\min \_II=r_{mean}-q\\ b\_\min \_II=b_{mean}-q\end{array}\right.;$

Wherein,

Rsum_I be the cumulative of red color r of the pixel meeting the first mellow lime pixel Rule of judgment and, Bsum_I For meet the cumulative of blue color difference b of the pixel of the first mellow lime pixel Rule of judgment and, nNum_I is mellow lime for meeting first The number of the pixel of pixel Rule of judgment, r_meanFor meeting the red color of the pixel of the first mellow lime pixel Rule of judgment The meansigma methods of difference r, b_meanFor meeting the meansigma methods of blue color difference b of the pixel of the first mellow lime pixel Rule of judgment, q is pre- If parametric variable；

43) poor and described for the reddish blue of described pixel the second mellow lime pixel Rule of judgment is contrasted, determine full The pixel of the described second mellow lime pixel Rule of judgment of foot is described mellow lime pixel, and calculates the individual of described mellow lime pixel Number accounts for the second ratio of pixel sum in described pending image；

44) judge that described second ratio, whether less than the second preset ratio, compares if described second ratio is preset less than second Example, it is determined that described pixel keeps constant, if described second ratio is the least at the yield value of tri-Color Channels of R, G, B respectively In the second preset ratio, then calculate described pixel respectively at the yield value of tri-Color Channels of R, G, B according to equation below:

ORgain=Gsum_II/Rsum_II

OGgain=1；

OBgain=Gsum_II/Bsum_II

Wherein,

ORgain is the described pixel yield value at R Color Channel, and OGgain is that described pixel is at G Color Channel Yield value, OBgain is the described pixel yield value at B color channel, and Rsum_II judges for meeting the second mellow lime pixel The original R of the red color channel value of the pixel of condition cumulative and, Gsum_II judges bar for meeting the second mellow lime pixel The original value G of the green color channel of the pixel of part cumulative and, Bsum_II is for meeting the second mellow lime pixel Rule of judgment Pixel blue color channels original value B cumulative and.

Optionally, also include: calculate first and judge parameter；

Described calculating first judges that parameter includes:

Obtain the color chart under different-colour, and the reddish blue calculating mellow lime pixel in each color chart is poor；

Reddish blue according to the mellow lime pixel in each color chart described is poor, obtains the reddish blue of described mellow lime pixel Difference Butut；

Obtain the common set of described reddish blue difference Butut, and it is the fullest to calculate each pixel in described common set During the described first mellow lime pixel Rule of judgment of foot, described first judges the numerical value of parameter.

Optionally, described pixel is calculated respectively at the corrected value of tri-Color Channels of R, G, B by below equation:

OR=R*ORgain

OG=G*OGgain；

OB=B*OBgain

Wherein, ORgain is the described pixel yield value at R Color Channel, and OGgain is that described pixel is in G color The yield value of passage, OBgain is the described pixel yield value at B color channel, and R is that described pixel is at R Color Channel Original value, G is the described pixel original value at G Color Channel, and B is the described pixel original value at B color channel, and oR is Described pixel is at the corrected value of R Color Channel, and oG is the described pixel corrected value at G Color Channel, and oB is described pixel Point is at the corrected value of B color channel.

Second aspect according to embodiments of the present invention, it is provided that a kind of white balance processing equipment based on rgb space, including:

Colorimetry module, for obtaining in pending image each pixel respectively at tri-Color Channels of R, G, B Original value, and the reddish blue according to pixel described in described raw calculation is poor；

Gain value computation module is right for being carried out by the poor and mellow lime pixel Rule of judgment of reddish blue of described pixel Ratio, the pixel being determined for compliance with described mellow lime pixel Rule of judgment is mellow lime pixel, and according to described mellow lime pixel Number accounts for the ratio of pixel sum in described pending image and calculates described pixel respectively at tri-Color Channels of R, G, B Yield value, wherein, with the change of current colour temperature, iteration updates described mellow lime pixel Rule of judgment；

Adjusting module, is used for according to described pixel respectively at the yield value of tri-Color Channels of R, G, B, and described picture Vegetarian refreshments, respectively at the original value of tri-Color Channels of R, G, B, calculates described pixel respectively in the school of tri-Color Channels of R, G, B On the occasion of, and each pixel in described pending image is adjusted to described at the original value of tri-Color Channels of R, G, B respectively Corrected value.

Optionally, also include: pending image collection module,

Described pending image collection module is for, after obtaining original image, being averagely divided into n by described original image Individual region, and obtain in each region that pixel is respectively in the meansigma methods of tri-Color Channels of R, G, B, wherein, n is default whole Number；

By each pixel in described each region respectively the value of tri-Color Channels of R, G, B be adjusted to described averagely Value；

It is a pixel by described each area reduction, and the image after reducing is as described pending image.

Optionally, described Colorimetry module is additionally operable to, and in obtaining pending image, each pixel is respectively at R, G, B After the original value of three Color Channels, and before the reddish blue difference calculating described pixel, lead to according to tri-colors of described R, G, B The original value in road, searches overexposure therein and/or the darkest pixel, and deletes described overexposure and/or the darkest pixel.

Optionally, gray pixel during described mellow lime pixel Rule of judgment includes the first mellow lime pixel Rule of judgment and second Point Rule of judgment；

Described first mellow lime pixel Rule of judgment is:

$\left\{\begin{array}{c}sum\_min\_I<r+b<sum\_max\_I\\ r>r\_\min \_I\\ b>b\_\min \_I\end{array}\right.;$

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, sum_min_I, sum_max_I, r_ Min_I and b_min_I is respectively the first judgement parameter, and described first judges that parameter is during described pixel is under certain color temperature The judgement parameter met is needed during spodogram vegetarian refreshments；

Described second mellow lime pixel Rule of judgment is:

$\left\{\begin{array}{c}sum\_min\_II<r+b<sum\_max\_II\\ r>r\_\min \_II\\ b>b\_\min \_II\end{array}\right.;$

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, sum_min_II, sum_max_II, r_ Min_II and b_min_II is respectively the second judgement parameter, and described second judges that parameter is that described pixel is under current colour temperature The judgement parameter met is needed during mellow lime pixel.

Optionally, described gain value computation module includes:

First ratio computing unit, for judging bar by poor for the reddish blue of described pixel with described first mellow lime pixel Part contrasts, and determines that the pixel meeting described first mellow lime pixel Rule of judgment is first object pixel, and calculates The number of described first object pixel accounts for the first ratio of pixel sum in described pending image；

Contrast processing unit, is used for judging whether described first ratio is less than the first preset ratio, if described first ratio Less than the first preset ratio, it is determined that described first object pixel is described mellow lime pixel, and determine that described pixel divides The not yield value at tri-Color Channels of R, G, B keeps constant, or, if described first ratio is less than the first preset ratio, protect Hold described second and judge parameter constant, and performed corresponding operating by the second contrast unit；

If described first ratio is not less than described first preset ratio, then updates described second according to equation below and judge ginseng Number, and in the updated by the second contrast unit execution corresponding operating:

$\left\{\begin{array}{c}sum\_max\_II=r_{mean}+b_{mean}+q\\ sum\_min\_II=r_{mean}+b_{mean}-q\\ r\_\min \_II=r_{mean}-q\\ b\_\min \_II=b_{mean}-q\end{array}\right.;$

Wherein,

Rsum_I be the cumulative of red color r of the pixel meeting the first mellow lime pixel Rule of judgment and, Bsum_I For meet the cumulative of blue color difference b of the pixel of the first mellow lime pixel Rule of judgment and, nNum_I is mellow lime for meeting first The number of the pixel of pixel Rule of judgment, r_meanFor meeting the red color of the pixel of the first mellow lime pixel Rule of judgment The meansigma methods of difference r, b_meanFor meeting the meansigma methods of blue color difference b of the pixel of the first mellow lime pixel Rule of judgment, q is pre- If parametric variable；

Second ratio computing unit, for judging bar by poor for the reddish blue of described pixel with described second mellow lime pixel Part contrasts, and determines that the pixel meeting described second mellow lime pixel Rule of judgment is described mellow lime pixel, and calculates The number of described mellow lime pixel accounts for the second ratio of pixel sum in described pending image；

Yield value computing unit, is used for judging whether described second ratio is less than the second preset ratio, if described second ratio Example is less than the second preset ratio, it is determined that described pixel keeps constant at the yield value of tri-Color Channels of R, G, B respectively, if Described second ratio is not less than the second preset ratio, then calculate described pixel respectively in tri-face of R, G, B according to equation below The yield value of chrominance channel:

ORgain=Gsum_II/Rsum_II

OGgain=1；

OBgain=Gsum_II/Bsum_II

Wherein,

ORgain is the described pixel yield value at R Color Channel, and OGgain is that described pixel is at G Color Channel Yield value, OBgain is the described pixel yield value at B color channel, and Rsum_II judges for meeting the second mellow lime pixel The original value R of the red color passage of the pixel of condition cumulative and, Gsum_II judges bar for meeting the second mellow lime pixel The original value G of the green color channel of the pixel of part cumulative and, Bsum_II is for meeting the second mellow lime pixel Rule of judgment Pixel blue color channels original value B cumulative and.

Optionally, also include: first judges parameter calculating module, judge parameter for calculating first；

Described first judges that parameter calculating module includes:

Color chart acquiring unit, for obtaining the color chart under different-colour, and calculates the middle spodogram in each color chart The reddish blue of vegetarian refreshments is poor；

Scattergram acquiring unit, poor for the reddish blue according to the mellow lime pixel in each color chart described, obtain institute State the reddish blue difference Butut of mellow lime pixel；

First judges parameter calculation unit, for obtaining the common set of described reddish blue difference Butut, and calculates described When each pixel in common set is satisfied by described first mellow lime pixel Rule of judgment, described first judges the number of parameter Value.

Optionally, described adjusting module calculates described pixel respectively at tri-Color Channels of R, G, B by below equation Corrected value:

OR=R*ORgain

OG=G*OGgain；

OB=B*OBgain

Wherein, ORgain is the described pixel yield value at R Color Channel, and OGgain is that described pixel is in G color The yield value of passage, OBgain is the described pixel yield value at B color channel, and R is that described pixel is at R Color Channel Original value, G is the described pixel original value at G Color Channel, and B is the described pixel original value at B color channel, and oR is Described pixel is at the corrected value of R Color Channel, and oG is the described pixel corrected value at G Color Channel, and oB is described pixel Point is at the corrected value of B color channel.

The technical scheme that the embodiment of the present invention provides can include following beneficial effect:

By scheme disclosed in the present application, it is possible to realize the white balance to image and process, thus reduce the aberration in image, Improve picture quality.

It should be appreciated that it is only exemplary and explanatory, not that above general description and details hereinafter describe The present invention can be limited.

Accompanying drawing explanation

Accompanying drawing herein is merged in description and constitutes the part of this specification, it is shown that meet the enforcement of the present invention Example, and for explaining the principle of the present invention together with description.

Fig. 1 is according to a kind of based on rgb space the white balancing treatment method shown in the present invention one exemplary embodiment Workflow schematic diagram；

Fig. 2 is according in a kind of based on rgb space the white balancing treatment method shown in the present invention one exemplary embodiment, Calculate the workflow schematic diagram of yield value；

Fig. 3 is according in a kind of based on rgb space the white balancing treatment method shown in the present invention one exemplary embodiment Reddish blue difference Butut；

Fig. 4 is according to a kind of based on rgb space the white balance processing equipment shown in the present invention one exemplary embodiment Structural representation.

Detailed description of the invention

Here will illustrate exemplary embodiment in detail, its example represents in the accompanying drawings.Explained below relates to During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represents same or analogous key element.Following exemplary embodiment Described in embodiment do not represent all embodiments consistent with the present invention.On the contrary, they are only with the most appended The example of the apparatus and method that some aspects that described in detail in claims, the present invention are consistent.

The embodiment of the present invention discloses a kind of white balancing treatment method based on rgb space and device, to solve prior art There is aberration in the image that middle image capturing system collects, the problem causing the poor image quality collected.

The open white balancing treatment method based on rgb space of the embodiment of the present invention one.Workflow shown in Figure 1 is shown Being intended to, described white balancing treatment method based on rgb space comprises the following steps:

Step S11, obtain in pending image each pixel respectively at the original value of tri-Color Channels of R, G, B, and Reddish blue according to pixel described in described raw calculation is poor.

When the reddish blue difference according to pixel described in described raw calculation, it then follows formula below:

R=R/ (R+G+B) r ∈ [0,1]；

B=B/ (R+G+B) b ∈ [0,1].

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, and R, G and B then divide for described pixel Not at the original value of tri-Color Channels of R, G, B.

Step S12, poor and mellow lime for the reddish blue of described pixel pixel Rule of judgment is contrasted, be determined for compliance with institute The pixel stating mellow lime pixel Rule of judgment is mellow lime pixel, and according to the number of described mellow lime pixel account for described in wait to locate In reason image, the ratio described pixel of calculating of pixel sum is respectively at the yield value of tri-Color Channels of R, G, B, wherein, and institute Iteration updates with the change of current colour temperature to state mellow lime pixel Rule of judgment.

In this application, described mellow lime pixel refers to the pixel in addition to color pixel cell, including white pixel Point, black pixel point and gray pixels point.Wherein, described mellow lime pixel generally meets the condition of R=G=B.It addition, at this Application is provided with mellow lime pixel Rule of judgment, and described mellow lime pixel Rule of judgment changes and iteration with current colour temperature Update such that it is able to ensure and meet the pixel of described mellow lime pixel Rule of judgment under current colour temperature, for mellow lime pixel.

Step S13, according to described pixel respectively at the yield value of tri-Color Channels of R, G, B, and described pixel Respectively at the original value of tri-Color Channels of R, G, B, calculate described pixel respectively in the correction of tri-Color Channels of R, G, B Value, and each pixel in described pending image is adjusted to described school at the original value of tri-Color Channels of R, G, B respectively On the occasion of.

Step S11 of the application to step S13 discloses a kind of white balancing treatment method based on rgb space, in the method, In obtaining pending image, each pixel is respectively after the original value of tri-Color Channels of R, G, B, calculates described pixel Reddish blue poor, and by the poor and mellow lime pixel Rule of judgment of described reddish blue, determine mellow lime pixel, further according to described The number of mellow lime pixel accounts for the ratio of the sum of pixel in described pending image, be calculated described pixel respectively R, The yield value of tri-Color Channels of G, B, according to described yield value, obtains the corrected value of pixel, then by described pending image In each pixel adjust to described corrected value at the original value of tri-Color Channels of R, G, B respectively, it is achieved to pending image White balance process.

Pass through disclosed method, it is possible to realize the white balance to image and process, thus reduce the aberration in image, Improve picture quality.

In step s 11, described pending image can be the original image that image capturing system collects.It addition, be Improve the quality that image is carried out white balance process, it is also possible to according to the meansigma methods of pixel, original image is reduced Process, and using reducing the image after process as pending image, in this case, disclosed in the present application based on rgb space White balancing treatment method, further comprising the steps of:

First, after obtaining original image, described original image is averagely divided into n region, and obtains each region Middle pixel is respectively in the meansigma methods of tri-Color Channels of R, G, B, and wherein, n is default integer.

Then, each pixel in described each region is adjusted to described in the value of tri-Color Channels of R, G, B respectively Meansigma methods.

In this step, first obtain each pixel respectively at the original value of tri-Color Channels of R, G, B, count the most respectively Calculate in each region, each pixel respectively in the meansigma methods of tri-Color Channels of R, G, B, then by each pixel respectively R, The value of tri-Color Channels of G, B is adjusted to described meansigma methods respectively.

Finally, it is a pixel by described each area reduction, and the image after reducing is as described pending figure Picture, in order to perform the operation of step S11.

By above-mentioned operation, it is possible to described original image is reduced into original 1/n.Such as, can be by described original graph As being averagely divided into 16 regions, say, that n is equal to 16, in this case, it is possible to be reduced into originally by described original image 1/16.

Original image is carried out above-mentioned reduce process after, it is possible to remove the impact of noise, this situation in pixel Under, the image after reducing is as pending image, it is possible to increase pending image carries out the degree of accuracy of white balance process, enters one Step improves picture quality.Further, after carrying out reducing process, in pending image, each pixel is respectively in tri-face of R, G, B The value of chrominance channel, is each pixel respectively at the original value of tri-Color Channels of R, G, B.

Further, white balancing treatment method based on rgb space disclosed in the present application also includes:

In obtaining pending image, each pixel is respectively after the original value of tri-Color Channels of R, G, B, and at meter Before calculating the reddish blue difference of described pixel, according to the original value of described tri-Color Channels of R, G, B, search overexposure therein and/ Or the darkest pixel, and delete described overexposure and/or the darkest pixel.

It addition, after deleting described overexposure and/or the darkest pixel, the application calculates the red indigo plant of remaining pixel Aberration, and the reddish blue according to remaining described pixel is poor.In this case, owing to deleting overexposure and/or the darkest picture Vegetarian refreshments, eliminates overexposure and/or the interference of the darkest pixel such that it is able to improves pending image and carries out white balance process Degree of accuracy, improves picture quality further.

In this application, overexposure and the darkest pixel can be determined by below equation:

$\left\{\begin{array}{c}R<R\_\max \\ G<G\_\max \\ B<B\_\max \\ Y\&GreaterEqual;Y\_\min \end{array}\right..$

Wherein, Y=R+G+B, R are the described pixel original value at R Color Channel, and G is that described pixel leads in G color The original value in road, B is the described pixel original value at B color channel, and R_max, G_max and B_max respectively preset The fixed value of tri-Color Channels of R, G, B, such as, R_max, G_max and B_max can be set and be 255, Y_min is in advance The fixed value set, such as, can set Y_min as 30.If a certain pixel meets above-mentioned formula, then this pixel is described not Overexposure or the darkest pixel, and if a certain pixel is unsatisfactory for above-mentioned formula, then illustrate that this pixel is overexposure or the darkest Pixel, can delete.

Further, in step s 12, disclose poor for the reddish blue of described pixel and mellow lime pixel Rule of judgment Carry out the operation contrasted.Wherein, during described mellow lime pixel Rule of judgment includes the first mellow lime pixel Rule of judgment and second Spodogram vegetarian refreshments Rule of judgment.

Described first mellow lime pixel Rule of judgment is:

$\left\{\begin{array}{c}sum\_min\_I<r+b<sum\_max\_I\\ r>r\_\min \_I\\ b>b\_\min \_I\end{array}\right.;$

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, sum_min_I, sum_max_I, r_ Min_I and b_min_I is respectively the first judgement parameter, and described first judges that parameter is during described pixel is under certain color temperature The judgement parameter met is needed during spodogram vegetarian refreshments.If judging it is to say, a certain pixel meets described first mellow lime pixel Condition, then it represents that this pixel is mellow lime pixel under certain color temperature.Wherein, described first judges that parameter is fixed value, no Change with the change of environment colour temperature.

Described second mellow lime pixel Rule of judgment is:

$\left\{\begin{array}{c}sum\_min\_II<r+b<sum\_max\_II\\ r>r\_\min \_II\\ b>b\_\min \_II\end{array}\right.;$

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, sum_min_II, sum_max_II, r_ Min_II and b_min_II is respectively the second judgement parameter, and described second judges that parameter is that described pixel is under current colour temperature The judgement parameter met is needed during mellow lime pixel.Judge it is to say, a certain pixel meets described second mellow lime pixel Condition, then it represents that this pixel is mellow lime pixel under the colour temperature of current environment, wherein, described second judges that parameter is with currently The change of the colour temperature of environment and change, in this application, can sentence by the first mellow lime pixel Rule of judgment comprises first Disconnected parameter calculates second and judges parameter, and judges to described second that in the processing procedure of image parameter is iterated updating.

In step S12 of the application, the poor and mellow lime pixel Rule of judgment of reddish blue of described pixel is entered by disclosure Row contrast, the pixel being determined for compliance with described mellow lime pixel Rule of judgment is mellow lime pixel, and according to described middle gray pixel The number of point accounts for pixel described in the ratio-dependent of pixel sum in described pending image and leads in tri-colors of R, G, B respectively The operation of the yield value in road.Workflow schematic diagram shown in Figure 2, this operation comprises the following steps:

Step S21, poor and described for the reddish blue of described pixel the first mellow lime pixel Rule of judgment is contrasted, really Surely the pixel meeting described first mellow lime pixel Rule of judgment is first object pixel, and calculates described first object picture The number of vegetarian refreshments accounts for the first ratio of pixel sum in described pending image.

Step S22, judge that described first ratio, whether less than the first preset ratio, if so, performs the operation of step S23, If it is not, perform the operation of step S24.

Wherein, described first preset ratio is arranged by staff, generally may be configured as 5%, it is of course also possible to by first Preset ratio is set to other numerical value, and this is not limited by the application.

If described first ratio of step S23 is less than the first preset ratio, keeps described second to judge parameter constant, and hold The operation of row step S25.

If described first ratio is less than the first preset ratio, then illustrate that described second judges that parameter is without being iterated more Newly, in this case, described second is kept to judge parameter constant.

Or, if described first ratio is less than the first preset ratio, it is determined that described first object pixel be described in Spodogram vegetarian refreshments, and determine that described pixel keeps constant at the yield value of tri-Color Channels of R, G, B respectively, and terminate this behaviour Make.It is to say, if described first ratio is less than the first preset ratio, in pending image, each pixel is R, G, B tri- The yield value of Color Channel is without updating.

Generally when image being carried out white balance process first by disclosed method, if described first ratio is little In the first preset ratio, often determine that described pixel keeps constant, no longer at the yield value of tri-Color Channels of R, G, B respectively It is contrasted with the second Rule of judgment.In this case, described pixel is respectively in the gain of tri-Color Channels of R, G, B The yield value that value uses when can gather image according to the last time determines.Such as, if during the last collection image, image acquisition System is 1 at the yield value of tri-Color Channels of R, G, B, then can determine that when this white balance processes, pixel respectively R, The yield value of tri-Color Channels of G, B is 1.

If described first ratio of step S24 is not less than described first preset ratio, then update described the according to equation below Two judge parameter, and perform the operation of step S25 in the updated:

$\left\{\begin{array}{c}sum\_max\_II=r_{mean}+b_{mean}+q\\ sum\_min\_II=r_{mean}+b_{mean}-q\\ r\_\min \_II=r_{mean}-q\\ b\_\min \_II=b_{mean}-q\end{array}\right.;$

Wherein,

Rsum_I+=r

Bsum_I+=b.

nNum_I++

In above-mentioned formula, Rsum_I is red color r of the pixel meeting the first mellow lime pixel Rule of judgment Cumulative and, Bsum_I be the cumulative of blue color difference b of the pixel meeting the first mellow lime pixel Rule of judgment and, nNum_I is Meet the number of the pixel of the first mellow lime pixel Rule of judgment, r_meanFor meeting the picture of the first mellow lime pixel Rule of judgment The meansigma methods of red color r of vegetarian refreshments, b_meanFor meeting blue color difference b of the pixel of the first mellow lime pixel Rule of judgment Meansigma methods, q is default parametric variable.Wherein, the numerical value of q can be set according to the process demand of image by staff Putting, such as, can be 51.2 by arranging q, it is of course also possible to q is set to other numerical value, this not be limited by the application.

In this step, however, it is determined that described first ratio is not less than described first preset ratio, then need to obtain to calculate Rsum_I, Bsum_I and nNum_I, and according to Rsum_I, Bsum_I and nNum_I, it is calculated r_meanAnd b_mean, further according to upper State formula, calculate described second and judge parameter, thus realize judging described second that the iteration of parameter updates.

Step S25, poor and described for the reddish blue of described pixel the second mellow lime pixel Rule of judgment is contrasted, really Surely the pixel meeting described second mellow lime pixel Rule of judgment is described mellow lime pixel, and calculates described mellow lime pixel Number account for the second ratio of the sum of pixel in described pending image.

After renewal described second judges parameter, described second mellow lime pixel Rule of judgment updates the most therewith.Should In step, poor and described for the reddish blue of described pixel the second mellow lime pixel Rule of judgment is contrasted, wherein, if a certain Pixel meets described second mellow lime pixel Rule of judgment, then it represents that this pixel is middle spodogram under the colour temperature of current environment Vegetarian refreshments.

Step S26, judge that described second ratio, whether less than the second preset ratio, if so, performs the operation of step S27, If it is not, perform the operation of step S28.

Wherein, described second preset ratio is arranged by staff, generally may be configured as 5%, it is of course also possible to by second Preset ratio is set to other numerical value, and this is not limited by the application.

If described second ratio of step S27 is less than the second preset ratio, it is determined that described pixel is respectively at R, G, B tri- The yield value of individual Color Channel keeps constant, and performs the operation of step S29.

If it is to say, described second ratio is less than the second preset ratio, it is determined that described pixel is respectively at R, G, B tri- The yield value of individual Color Channel keeps constant, it then follows equation below:

ORgain=ORgain

OGgain=OGgain.

OBgain=OBgain

Wherein, ORgain is the described pixel yield value at R Color Channel, and OGgain is that described pixel is in G color The yield value of passage, OBgain is the described pixel yield value at B color channel.In this case, described pixel is respectively Yield value is used to determine when the yield value of tri-Color Channels of R, G, B can gather image according to the last time.Such as, if recently When once gathering image, image capturing system is 1 at the yield value of tri-Color Channels of R, G, B, then can determine that this is put down in vain When weighing apparatus processes, pixel is 1 at the yield value of tri-Color Channels of R, G, B respectively.

If described second ratio of step S28 is not less than the second preset ratio, then calculate described pixel according to equation below Respectively at the yield value of tri-Color Channels of R, G, B:

ORgain=Gsum_II/Rsum_II

OGgain=1；

OBgain=Gsum_II/Bsum_II

In above-mentioned formula,

Wherein, ORgain is the described pixel yield value at R Color Channel, and OGgain is that described pixel is in G color The yield value of passage, OBgain is the described pixel yield value at B color channel, and Rsum_II is for meeting gray pixel in second The original value R of the red color passage of the pixel of some Rule of judgment cumulative and, Gsum_II is for meeting the second mellow lime pixel The original value G of the green color channel of the pixel of Rule of judgment cumulative and, Bsum_II sentences for meeting the second mellow lime pixel The original value B of the blue color channels of the pixel of broken strip part cumulative and.

By above-mentioned formula, pixel can be calculated respectively at the yield value of tri-Color Channels of R, G, B.

In the application, described first judges that parameter is fixed value, does not changes with the change of environment colour temperature, and meets in first The pixel of spodogram vegetarian refreshments Rule of judgment is mellow lime pixel under certain color temperature, but under current colour temperature, meets described The pixel of one mellow lime pixel Rule of judgment is also possible to the most mellow lime pixel, therefore, can pass through the second mellow lime pixel Rule of judgment judges further, if this pixel also meets described second mellow lime pixel Rule of judgment, then it represents that this picture Vegetarian refreshments is mellow lime pixel under the colour temperature of current environment.Wherein, described second judges the parameter change with the colour temperature of current environment Change and change, in this application, can judge that parameter calculates second by the first mellow lime pixel Rule of judgment comprises first Judge parameter, and in the processing procedure of image, judge to described second that parameter is iterated updating.

During the calculating of above-mentioned yield value, when determining the first ratio not less than the first preset ratio, according to first Judge that parameter updates second and judges parameter such that it is able to make the second judgement parameter update with the continuous iteration of current environment colour temperature, i.e. The distribution of the mellow lime pixel under current colour temperature is reduced further such that it is able to calculate and be more applicable for current colour temperature RGB triple channel gain, thus the RGB image after drawing color correction more accurately.

Further, in step S21 to step S28, apply the first judgement parameter.Accordingly, in disclosure White balancing treatment method based on rgb space in, also include: calculate and first judge parameter.

Wherein, described calculating first judges that parameter comprises the following steps:

First, obtain the color chart under different-colour, and the reddish blue calculating mellow lime pixel in each color chart is poor.

In this step, image capturing system, under different-colour environment, gathers color chart picture such that it is able to obtain the most homochromy Color chart under Wen.It addition, for the computational accuracy improving the first judgement parameter, image capturing system is under gathering different-colour Color chart time, the yield value of each passage can be adjusted to 1, and, adjust time of exposure, make each mellow lime pixel in color chart In the value of tri-Color Channels of R, G, B close to 255, but unsaturated, thus avoid color chart occurs less exposure and/or the darkest Pixel, improves the picture quality of the color chart collected.

Further, before the reddish blue difference of the mellow lime pixel in calculating each color chart, it is also possible to search colour atla The exposure occurred in figure and/or the darkest pixel deleting, in order to the reddish blue calculating remaining mellow lime pixel is poor, it is thus possible to Enough further minimizing overexposure and/or the interference of the darkest pixel.

In this step, overexposure and the darkest pixel can be determined by below equation:

$\left\{\begin{array}{c}R<R\_\max \\ G<G\_\max \\ B<B\_\max \\ Y\&GreaterEqual;Y\_\min \end{array}\right..$

Wherein, Y=R+G+B, R are the described pixel original value at R Color Channel, and G is that described pixel leads in G color The original value in road, B is the described pixel original value at B color channel, and R_max, G_max and B_max respectively preset The fixed value of tri-Color Channels of R, G, B, such as, R_max, G_max and B_max can be set and be 255, Y_min is in advance The fixed value set, such as, can set Y_min as 30.If a certain pixel meets above-mentioned formula, then this pixel is described not Overexposure or the darkest pixel, if a certain pixel is unsatisfactory for above-mentioned formula, then illustrate that this pixel is overexposure or the darkest picture Vegetarian refreshments, can delete.

Then, the reddish blue according to the mellow lime pixel in each color chart described is poor, obtains described mellow lime pixel Reddish blue difference Butut.

Wherein, the mellow lime pixel in each color chart described can be determined by staff.It addition, mellow lime pixel Reddish blue difference can be calculated by below equation:

$r=\frac{R}{R+G+B},b=\frac{B}{R+G+B}.$

Wherein, r is red color, and b is blue color difference.

Finally, obtain the common set of described reddish blue difference Butut, and calculate each pixel in described common set When point is satisfied by described first mellow lime pixel Rule of judgment, described first judges the numerical value of parameter.

Reddish blue difference Butut shown in Figure 3, wherein, in this reddish blue difference Butut, abscissa is blue color Difference, ordinate axle is red color, and each point in figure represents the mellow lime pixel in each color chart.It addition, in this figure Article four, the region that straight line is divided is the common set of described reddish blue difference Butut, and this common set is that described reddish blue is poor The close quarters of the mellow lime pixel distribution in scattergram.

Each pixel in described common set is mellow lime pixel, therefore, each picture in described common set Vegetarian refreshments is satisfied by described first mellow lime pixel Rule of judgment, say, that each pixel in described common set all accords with Conjunction equation below:

The described first numerical value judging parameter can be calculated accordingly.

Wherein, if described reddish blue difference Butut and described common set are as it is shown on figure 3, pass through and calculate, it may be determined that institute State the first judgement parameter as follows:

$\left\{\begin{array}{c}sum\_\max \_I=0.64\\ sum\_\min \_I=0.4\\ r\_\min \_I=0.1\\ b\_\min \_I=0.1\end{array}\right..$

Further, step S13 of the application discloses according to described pixel respectively at tri-Color Channels of R, G, B Yield value, and described pixel is respectively at the original value of tri-Color Channels of R, G, B, calculate described pixel respectively R, The operation of the corrected value of tri-Color Channels of G, B.Concrete, described pixel can be calculated respectively at R, G, B by below equation The corrected value of three Color Channels:

OR=R*ORgain

OG=G*OGgain；

OB=B*OBgain

Wherein, ORgain is the described pixel yield value at R Color Channel, and OGgain is that described pixel is in G color The yield value of passage, OBgain is the described pixel yield value at B color channel, and R is that described pixel is at R Color Channel Original value, G is the described pixel original value at G Color Channel, and B is the described pixel original value at B color channel, and oR is Described pixel is at the corrected value of R Color Channel, and oG is the described pixel corrected value at G Color Channel, and oB is described pixel Point is at the corrected value of B color channel.

According to above-mentioned formula, pixel can be calculated respectively at the corrected value of tri-Color Channels of R, G, B.Then, Each pixel in described pending image is adjusted to described corrected value at the original value of tri-Color Channels of R, G, B respectively, I.e. can get the image after white balance processes.

Accordingly, the second embodiment of the present invention discloses a kind of white balance processing equipment based on rgb space.See Fig. 4 Shown structural representation, described white balance processing equipment based on rgb space includes: Colorimetry module 100, yield value meter Calculate module 200 and adjusting module 300.

Wherein, described Colorimetry module 100, it is used for obtaining in pending image each pixel respectively at R, G, B tri- The original value of individual Color Channel, and the reddish blue according to pixel described in described raw calculation is poor；

Described gain value computation module 200, for by poor for the reddish blue of described pixel and mellow lime pixel Rule of judgment Contrasting, the pixel being determined for compliance with described mellow lime pixel Rule of judgment is mellow lime pixel, and according to described middle spodogram The number of vegetarian refreshments accounts for the ratio of pixel sum in described pending image and calculates described pixel respectively in tri-colors of R, G, B The yield value of passage, wherein, with the change of current colour temperature, iteration updates described mellow lime pixel Rule of judgment；

Described adjusting module 300, for according to described pixel respectively at the yield value of tri-Color Channels of R, G, B, with And described pixel is respectively at the original value of tri-Color Channels of R, G, B, calculate described pixel respectively in tri-colors of R, G, B The corrected value of passage, and each pixel in described pending image is adjusted at the original value of tri-Color Channels of R, G, B respectively Whole to described corrected value.

In the application, it is poor that described Colorimetry module 100 can calculate the reddish blue of described pixel by equation below:

R=R/ (R+G+B) r ∈ [0,1]；

B=B/ (R+G+B) b ∈ [0,1].

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, and R, G and B then divide for described pixel Not at the original value of tri-Color Channels of R, G, B.

It addition, in this application, described mellow lime pixel refers to the pixel in addition to color pixel cell, including white Pixel, black pixel point and gray pixels point.Wherein, described mellow lime pixel generally meets the condition of R=G=B.It addition, It is provided with mellow lime pixel Rule of judgment in this application, and described mellow lime pixel Rule of judgment is with the change of current colour temperature Iteration updates such that it is able to ensures and meets the pixel of described mellow lime pixel Rule of judgment under current colour temperature, for middle spodogram Vegetarian refreshments.

Second embodiment of the application discloses a kind of white balance processing equipment based on rgb space, by this device, it is possible to Realize the white balance to image to process, thus reduce the aberration in image, improve picture quality.

Further, described white balance processing equipment based on rgb space also includes: pending image collection module.

Described pending image collection module is for, after obtaining original image, being averagely divided into n by described original image Individual region, and obtain in each region that pixel is respectively in the meansigma methods of tri-Color Channels of R, G, B, wherein, n is default whole Number；Each pixel in described each region is adjusted to described meansigma methods in the value of tri-Color Channels of R, G, B respectively；By institute Stating each area reduction is a pixel, and the image after reducing is as described pending image.

By described pending image collection module, it is possible to described original image is reduced into original 1/n.Such as, may be used Described original image is averagely divided into 16 regions, say, that n is equal to 16, in this case, it is possible to by described original graph As being reduced into original 1/16.

After original image being reduced process according to described pending image collection module, it is possible to remove in pixel and make an uproar The impact of point, in this case, the image after reducing is as pending image, it is possible to increase pending image carries out white balance The degree of accuracy processed, improves picture quality further.

Further, in described white balance processing equipment based on rgb space, described Colorimetry module 100 is also used In, in obtaining pending image, each pixel is respectively after the original value of tri-Color Channels of R, G, B, and described calculating Before the reddish blue difference of pixel, according to the original value of described tri-Color Channels of R, G, B, search overexposure therein and/or the darkest Pixel, and delete described overexposure and/or the darkest pixel.

It addition, after deleting described overexposure and/or the darkest pixel, described Colorimetry module 100 is according to residue The reddish blue of pixel poor, and the reddish blue according to remaining described pixel is poor.Owing to deleting overexposure and/or the darkest Pixel, eliminates overexposure and/or the interference of the darkest pixel such that it is able to improves pending image and carries out white balance process Degree of accuracy, improve picture quality further.

In this application, overexposure and the darkest pixel can be determined by below equation:

$\left\{\begin{array}{c}R<R\_\max \\ G<G\_\max \\ B<B\_\max \\ Y\&GreaterEqual;Y\_\min \end{array}\right..$

Wherein, Y=R+G+B, R are the described pixel original value at R Color Channel, and G is that described pixel leads in G color The original value in road, B is the described pixel original value at B color channel, and R_max, G_max and B_max respectively preset The fixed value of tri-Color Channels of R, G, B, such as, R_max, G_max and B_max can be set and be 255, Y_min is in advance The fixed value set, such as, can set Y_min as 30.If a certain pixel meets above-mentioned formula, then this pixel is described not Overexposure or the darkest pixel, and if a certain pixel is unsatisfactory for above-mentioned formula, then illustrate that this pixel is overexposure or the darkest Pixel, can delete.

It addition, in this application, described mellow lime pixel Rule of judgment includes the first mellow lime pixel Rule of judgment and Two mellow lime pixel Rule of judgment；

Described first mellow lime pixel Rule of judgment is:

$\left\{\begin{array}{c}sum\_min\_I<r+b<sum\_max\_I\\ r>r\_\min \_I\\ b>b\_\min \_I\end{array}\right.;$

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, sum_min_I, sum_max_I, r_ Min_I and b_min_I is respectively the first judgement parameter, and described first judges that parameter is during described pixel is under certain color temperature The judgement parameter met is needed during spodogram vegetarian refreshments.If judging it is to say, a certain pixel meets described first mellow lime pixel Condition, then it represents that this pixel is mellow lime pixel under certain color temperature.Wherein, described first judges that parameter is fixed value, no Change with the change of environment colour temperature.

Described second mellow lime pixel Rule of judgment is:

$\left\{\begin{array}{c}sum\_min\_II<r+b<sum\_max\_II\\ r>r\_\min \_II\\ b>b\_\min \_II\end{array}\right.;$

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, sum_min_II, sum_max_II, r_ Min_II and b_min_II is respectively the second judgement parameter, and described second judges that parameter is that described pixel is under current colour temperature The judgement parameter met is needed during mellow lime pixel.Judge it is to say, a certain pixel meets described second mellow lime pixel Condition, then it represents that this pixel is mellow lime pixel under the colour temperature of current environment, wherein, described second judges that parameter is with currently The change of the colour temperature of environment and change, in this application, can sentence by the first mellow lime pixel Rule of judgment comprises first Disconnected parameter calculates second and judges parameter, and judges to described second that in the processing procedure of image parameter is iterated updating.

Further, in white balance processing equipment based on rgb space disclosed in the present application, described yield value calculates mould Block includes:

First ratio computing unit, for judging bar by poor for the reddish blue of described pixel with described first mellow lime pixel Part contrasts, and determines that the pixel meeting described first mellow lime pixel Rule of judgment is first object pixel, and calculates The number of described first object pixel accounts for the first ratio of pixel sum in described pending image；

Contrast processing unit, is used for judging whether described first ratio is less than the first preset ratio, if described first ratio Less than the first preset ratio, it is determined that described first object pixel is described mellow lime pixel, and determine that described pixel divides The not yield value at tri-Color Channels of R, G, B keeps constant, or, if described first ratio is less than the first preset ratio, protect Hold described second and judge parameter constant, and performed corresponding operating by the second contrast unit；

If described first ratio is not less than described first preset ratio, then updates described second according to equation below and judge ginseng Number, and in the updated by the second contrast unit execution corresponding operating:

$\left\{\begin{array}{c}sum\_max\_II=r_{mean}+b_{mean}+q\\ sum\_min\_II=r_{mean}+b_{mean}-q\\ r\_\min \_II=r_{mean}-q\\ b\_\min \_II=b_{mean}-q\end{array}\right.;$

Wherein,

Rsum_I be the cumulative of red color r of the pixel meeting the first mellow lime pixel Rule of judgment and, Bsum_I For meet the cumulative of blue color difference b of the pixel of the first mellow lime pixel Rule of judgment and, nNum_I is mellow lime for meeting first The number of the pixel of pixel Rule of judgment, r_meanFor meeting the red color of the pixel of the first mellow lime pixel Rule of judgment The meansigma methods of difference r, b_meanFor meeting the meansigma methods of blue color difference b of the pixel of the first mellow lime pixel Rule of judgment, q is pre- If parametric variable；

Second ratio computing unit, for judging bar by poor for the reddish blue of described pixel with described second mellow lime pixel Part contrasts, and determines that the pixel meeting described second mellow lime pixel Rule of judgment is described mellow lime pixel, and calculates The number of described mellow lime pixel accounts for the second ratio of pixel sum in described pending image；

Yield value computing unit, is used for judging whether described second ratio is less than the second preset ratio, if described second ratio Example is less than the second preset ratio, it is determined that described pixel keeps constant at the yield value of tri-Color Channels of R, G, B respectively, if Described second ratio is not less than the second preset ratio, then calculate described pixel respectively in tri-face of R, G, B according to equation below The yield value of chrominance channel:

ORgain=Gsum_II/Rsum_II

OGgain=1；

OBgain=Gsum_II/Bsum_II

Wherein,

ORgain is the described pixel yield value at R Color Channel, and OGgain is that described pixel is at G Color Channel Yield value, OBgain is the described pixel yield value at B color channel, and Rsum_II judges for meeting the second mellow lime pixel The original value R of the red color passage of the pixel of condition cumulative and, Gsum_II judges bar for meeting the second mellow lime pixel The original value G of the green color channel of the pixel of part cumulative and, Bsum_II is for meeting the second mellow lime pixel Rule of judgment Pixel blue color channels original value B cumulative and.

Wherein, described first preset ratio is arranged by staff, generally may be configured as 5%, it is of course also possible to by first Preset ratio is set to other numerical value, and this is not limited by the application.It addition, described second preset ratio is also set by staff Putting, generally may be configured as 5%, it is of course also possible to the second preset ratio is set to other numerical value, this is not limited by the application Fixed.

In the application, described first judges that parameter is fixed value, does not changes with the change of environment colour temperature, and meets in first The pixel of spodogram vegetarian refreshments Rule of judgment is mellow lime pixel under certain color temperature, but under current colour temperature, meets described The pixel of one mellow lime pixel Rule of judgment is also possible to the most mellow lime pixel, therefore, can pass through the second mellow lime pixel Rule of judgment judges further, if this pixel also meets described second mellow lime pixel Rule of judgment, then it represents that this picture Vegetarian refreshments is mellow lime pixel under the colour temperature of current environment, and wherein, described second judges the parameter change with the colour temperature of current environment Change and change, in this application, can judge that parameter calculates second by the first mellow lime pixel Rule of judgment comprises first Judge parameter, and in the processing procedure of image, judge to described second that parameter is iterated updating.

During the calculating of above-mentioned yield value, when determining the first ratio not less than the first preset ratio, according to first Judge that parameter updates second and judges parameter such that it is able to make the second judgement parameter update with the continuous iteration of current environment colour temperature, i.e. The distribution of the mellow lime pixel under current colour temperature is reduced further such that it is able to calculate and be more applicable for current colour temperature RGB triple channel gain, thus the RGB image after drawing color correction more accurately.

Further, described white balance processing equipment based on rgb space also includes: first judges parameter calculating module, Parameter is judged for calculating first.

Wherein, described first judges that parameter calculating module includes:

Color chart acquiring unit, for obtaining the color chart under different-colour, and calculates the middle spodogram in each color chart The reddish blue of vegetarian refreshments is poor；

Scattergram acquiring unit, poor for the reddish blue according to the mellow lime pixel in each color chart described, obtain institute State the reddish blue difference Butut of mellow lime pixel；

First judges parameter calculation unit, for obtaining the common set of described reddish blue difference Butut, and calculates described When each pixel in common set is satisfied by described first mellow lime pixel Rule of judgment, described first judges the number of parameter Value.

Reddish blue difference Butut shown in Figure 3, wherein, in this reddish blue difference Butut, abscissa is blue color Difference, ordinate axle is red color, and each point in figure represents the mellow lime pixel in each color chart.It addition, in this figure Article four, the region that straight line is divided is the common set of described reddish blue difference Butut, and this common set is that described reddish blue is poor The close quarters of the mellow lime pixel distribution in scattergram.

Each pixel in described common set is mellow lime pixel, therefore, each picture in described common set Vegetarian refreshments is satisfied by described first mellow lime pixel Rule of judgment, say, that each pixel in described common set all accords with Conjunction equation below:

The described first numerical value judging parameter can be calculated accordingly.

Wherein, if described reddish blue difference Butut and described common set are as it is shown on figure 3, pass through and calculate, it may be determined that institute State the first judgement parameter as follows:

$\left\{\begin{array}{c}sum\_\max \_I=0.64\\ sum\_\min \_I=0.4\\ r\_\min \_I=0.1\\ b\_\min \_I=0.1\end{array}\right..$

Further, described adjusting module 300 calculates described pixel respectively in tri-colors of R, G, B by below equation The corrected value of passage:

OR=R*ORgain

OG=G*OGgain；

OB=B*OBgain

Wherein, ORgain is the described pixel yield value at R Color Channel, and OGgain is that described pixel is in G color The yield value of passage, OBgain is the described pixel yield value at B color channel, and R is that described pixel is at R Color Channel Original value, G is the described pixel original value at G Color Channel, and B is the described pixel original value at B color channel, and oR is Described pixel is at the corrected value of R Color Channel, and oG is the described pixel corrected value at G Color Channel, and oB is described pixel Point is at the corrected value of B color channel.

According to above-mentioned formula, described adjusting module 300 can be calculated pixel respectively at tri-Color Channels of R, G, B Corrected value.Then, each pixel in described pending image is adjusted at the original value of tri-Color Channels of R, G, B respectively To described corrected value, i.e. can get the image after white balance processes.

About the device in above-described embodiment, wherein modules performs the concrete mode of operation in relevant the method Embodiment in be described in detail, explanation will be not set forth in detail herein.

Those skilled in the art, after considering description and putting into practice invention disclosed, will readily occur to other reality of the present invention Execute scheme.The application is intended to any modification, purposes or the adaptations of the present invention, these modification, purposes or suitable The change of answering property is followed the general principle of the present invention and includes the undocumented common knowledge or used in the art of the present invention Use technological means.Description and embodiments is considered only as exemplary, and true scope and spirit of the invention are by following right Requirement is pointed out.

It should be appreciated that the invention is not limited in precision architecture described above and illustrated in the accompanying drawings, and And various modifications and changes can carried out without departing from the scope.The scope of the present invention is only limited by appended claim.

Claims (14)

1. a white balancing treatment method based on rgb space, it is characterised in that including:

Obtain in pending image each pixel respectively at the original value of tri-Color Channels of R, G, B, and according to described original The reddish blue that value calculates described pixel is poor；

Poor and mellow lime for the reddish blue of described pixel pixel Rule of judgment is contrasted, is determined for compliance with described mellow lime pixel The pixel of Rule of judgment is mellow lime pixel, and accounts for pixel in described pending image according to the number of described mellow lime pixel The ratio described pixel of calculating of some sum is respectively at the yield value of tri-Color Channels of R, G, B, wherein, and described mellow lime pixel With the change of current colour temperature, iteration updates Rule of judgment；

According to described pixel respectively at the yield value of tri-Color Channels of R, G, B, and described pixel is respectively at R, G, B tri- The original value of individual Color Channel, calculates described pixel respectively at the corrected value of tri-Color Channels of R, G, B, and described waiting is located In reason image, each pixel adjusts to described corrected value at the original value of tri-Color Channels of R, G, B respectively.

White balancing treatment method based on rgb space the most according to claim 1, it is characterised in that also include:

Obtaining after original image, described original image is averagely being divided into n region, and obtaining pixel in each region and divide Not in the meansigma methods of tri-Color Channels of R, G, B, wherein, n is default integer；

Each pixel in described each region is adjusted to described meansigma methods in the value of tri-Color Channels of R, G, B respectively；

It is a pixel by described each area reduction, and the image after reducing is as described pending image.

White balancing treatment method based on rgb space the most according to claim 1, it is characterised in that also include:

In obtaining pending image, each pixel is respectively after the original value of tri-Color Channels of R, G, B, and is calculating institute Before stating the reddish blue difference of pixel, according to the original value of described tri-Color Channels of R, G, B, search overexposure therein and/or mistake Dark pixel, and delete described overexposure and/or the darkest pixel.

White balancing treatment method based on rgb space the most according to claim 1, it is characterised in that

Described mellow lime pixel Rule of judgment includes the first mellow lime pixel Rule of judgment and the second mellow lime pixel Rule of judgment；

Described first mellow lime pixel Rule of judgment is:

$\left\{\begin{array}{c}sum\_\min \_I<r+b<sum\_\max \_I\\ r>r\_\min \_I\\ b>b\_\min \_I\end{array}\right.;$

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, sum_min_I, sum_max_I, r_min_I Be respectively the first judgement parameter with b_min_I, described first judge parameter be described pixel under certain color temperature as middle spodogram The judgement parameter met is needed during vegetarian refreshments；

Described second mellow lime pixel Rule of judgment is:

$\left\{\begin{array}{c}sum\_\min \_II<r+b<sum\_\max \_II\\ r>r\_\min \_II\\ b>b\_\min \_II\end{array}\right.;$

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, sum_min_II, sum_max_II, r_min_ II and b_min_II is respectively the second judgement parameter, described second judge parameter be described pixel under current colour temperature as mellow lime The judgement parameter met is needed during pixel.

White balancing treatment method based on rgb space the most according to claim 4, it is characterised in that by described pixel The poor and mellow lime pixel Rule of judgment of reddish blue contrast, be determined for compliance with the pixel of described mellow lime pixel Rule of judgment For mellow lime pixel, and account for the ratio of pixel sum in described pending image according to the number of described mellow lime pixel and calculate Described pixel respectively at the yield value of tri-Color Channels of R, G, B, including:

41) poor and described for the reddish blue of described pixel the first mellow lime pixel Rule of judgment is contrasted, determine and meet institute The pixel stating the first mellow lime pixel Rule of judgment is first object pixel, and calculates the individual of described first object pixel Number accounts for the first ratio of pixel sum in described pending image；

42) judge that whether described first ratio is less than the first preset ratio；

If described first ratio is less than the first preset ratio, it is determined that described first object pixel is described mellow lime pixel, And determine that described pixel keeps constant at the yield value of tri-Color Channels of R, G, B respectively, or, if described first ratio is little In the first preset ratio, keep described second to judge parameter constant, and perform step 43) operation；

If described first ratio is not less than described first preset ratio, then updates described second according to equation below and judge parameter, And in the updated perform step 43) operation:

$\left\{\begin{array}{c}sum\_max\_II=r_{mean}+b_{mean}+q\\ sum\_min\_II=r_{mean}+b_{mean}-q\\ r\_\min \_II=r_{mean}-q\\ b\_\min \_II=b_{mean}-q\end{array}\right.;$

Wherein,

Rsum_I be the cumulative of red color r of the pixel meeting the first mellow lime pixel Rule of judgment and, Bsum_I is full Blue color difference b of the pixel of foot the first mellow lime pixel Rule of judgment cumulative and, nNum_I is for meeting gray pixel in first The number of the pixel of some Rule of judgment, r_meanFor meeting red color r of the pixel of the first mellow lime pixel Rule of judgment Meansigma methods, b_meanFor meeting the meansigma methods of blue color difference b of the pixel of the first mellow lime pixel Rule of judgment, q is default Parametric variable；

43) poor and described for the reddish blue of described pixel the second mellow lime pixel Rule of judgment is contrasted, determine and meet institute The pixel stating the second mellow lime pixel Rule of judgment is described mellow lime pixel, and the number calculating described mellow lime pixel accounts for Second ratio of pixel sum in described pending image；

44) judge whether described second ratio is less than the second preset ratio, if described second ratio is less than the second preset ratio, then Determine that described pixel keeps constant at the yield value of tri-Color Channels of R, G, B respectively, if described second ratio is not less than the Two preset ratio, then calculate described pixel respectively at the yield value of tri-Color Channels of R, G, B according to equation below:

ORgain=Gsum_II/Rsum_II

OGgain=1；

OBgain=Gsum_II/Bsum_II

Wherein,

ORgain is the described pixel yield value at R Color Channel, and OGgain is the gain at G Color Channel of the described pixel Value, OBgain is the described pixel yield value at B color channel, and Rsum_II is for meeting the second mellow lime pixel Rule of judgment Pixel red color passage original value R cumulative and, Gsum_II is to meet the second mellow lime pixel Rule of judgment The original value G of the green color channel of pixel cumulative and, Bsum_II is the picture meeting the second mellow lime pixel Rule of judgment The original value B of the blue color channels of vegetarian refreshments cumulative and.

White balancing treatment method based on rgb space the most according to claim 4, it is characterised in that also include: calculate the One judges parameter；

Described calculating first judges that parameter includes:

Obtain the color chart under different-colour, and the reddish blue calculating mellow lime pixel in each color chart is poor；

Reddish blue according to the mellow lime pixel in each color chart described is poor, obtains the reddish blue difference of described mellow lime pixel Butut；

Obtain the common set of described reddish blue difference Butut, and each pixel calculated in described common set is satisfied by institute When stating the first mellow lime pixel Rule of judgment, described first judges the numerical value of parameter.

White balancing treatment method based on rgb space the most according to claim 1, it is characterised in that pass through below equation Calculate described pixel respectively at the corrected value of tri-Color Channels of R, G, B:

OR=R*ORgain

OG=G*OGgain；

OB=B*OBgain

Wherein, ORgain is the described pixel yield value at R Color Channel, and OGgain is that described pixel is at G Color Channel Yield value, OBgain is the described pixel yield value at B color channel, and R is original at R Color Channel of described pixel Value, G is the described pixel original value at G Color Channel, and B is the described pixel original value at B color channel, and oR is described Pixel is at the corrected value of R Color Channel, and oG is the described pixel corrected value at G Color Channel, and oB is that described pixel is at B The corrected value of Color Channel.

8. a white balance processing equipment based on rgb space, it is characterised in that including:

Colorimetry module, for obtaining original at tri-Color Channels of R, G, B respectively of each pixel in pending image Value, and the reddish blue according to pixel described in described raw calculation is poor；

Gain value computation module, for contrasting the poor and mellow lime pixel Rule of judgment of reddish blue of described pixel, really Surely the pixel meeting described mellow lime pixel Rule of judgment is mellow lime pixel, and accounts for according to the number of described mellow lime pixel In described pending image, the ratio of pixel sum calculates described pixel respectively in the gain of tri-Color Channels of R, G, B Value, wherein, with the change of current colour temperature, iteration updates described mellow lime pixel Rule of judgment；

Adjusting module, is used for according to described pixel respectively at the yield value of tri-Color Channels of R, G, B, and described pixel Respectively at the original value of tri-Color Channels of R, G, B, calculate described pixel respectively in the correction of tri-Color Channels of R, G, B Value, and each pixel in described pending image is adjusted to described school at the original value of tri-Color Channels of R, G, B respectively On the occasion of.

White balance processing equipment based on rgb space the most according to claim 8, it is characterised in that also include: pending Image collection module,

Described pending image collection module is for, after obtaining original image, being averagely divided into n district by described original image Territory, and obtain in each region that pixel is respectively in the meansigma methods of tri-Color Channels of R, G, B, wherein, n is default integer；

Each pixel in described each region is adjusted to described meansigma methods in the value of tri-Color Channels of R, G, B respectively；

It is a pixel by described each area reduction, and the image after reducing is as described pending image.

White balance processing equipment based on rgb space the most according to claim 8, it is characterised in that

Described Colorimetry module is additionally operable to, and in obtaining pending image, each pixel leads in tri-colors of R, G, B respectively After the original value in road, and before the reddish blue difference calculating described pixel, original according to described tri-Color Channels of R, G, B Value, searches overexposure therein and/or the darkest pixel, and deletes described overexposure and/or the darkest pixel.

11. white balance processing equipments based on rgb space according to claim 8, it is characterised in that

Described mellow lime pixel Rule of judgment includes the first mellow lime pixel Rule of judgment and the second mellow lime pixel Rule of judgment；

Described first mellow lime pixel Rule of judgment is:

$\left\{\begin{array}{c}sum\_\min \_I<r+b<sum\_\max \_I\\ r>r\_\min \_I\\ b>b\_\min \_I\end{array}\right.;$

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, sum_min_I, sum_max_I, r_min_I Be respectively the first judgement parameter with b_min_I, described first judge parameter be described pixel under certain color temperature as middle spodogram The judgement parameter met is needed during vegetarian refreshments；

Described second mellow lime pixel Rule of judgment is:

$\left\{\begin{array}{c}sum\_\min \_II<r+b<sum\_\max \_II\\ r>r\_\min \_II\\ b>b\_\min \_II\end{array}\right.;$

Wherein, r is the red color of pixel, and b is the blue color difference of pixel, sum_min_II, sum_max_II, r_min_ II and b_min_II is respectively the second judgement parameter, described second judge parameter be described pixel under current colour temperature as mellow lime The judgement parameter met is needed during pixel.

12. white balance processing equipments based on rgb space according to claim 11, it is characterised in that described yield value Computing module includes:

First ratio computing unit, for entering the poor and described first mellow lime pixel Rule of judgment of reddish blue of described pixel Row contrast, determines that the pixel meeting described first mellow lime pixel Rule of judgment is first object pixel, and calculates described The number of first object pixel accounts for the first ratio of pixel sum in described pending image；

Contrast processing unit, is used for judging whether described first ratio is less than the first preset ratio, if described first ratio is less than First preset ratio, it is determined that described first object pixel is described mellow lime pixel, and determine that described pixel exists respectively The yield value of tri-Color Channels of R, G, B keeps constant, or, if described first ratio is less than the first preset ratio, keep institute State the second judgement parameter constant, and performed corresponding operating by the second contrast unit；

If described first ratio is not less than described first preset ratio, then updates described second according to equation below and judge parameter, And in the updated by the second contrast unit execution corresponding operating:

$\left\{\begin{array}{c}sum\_max\_II=r_{mean}+b_{mean}+q\\ sum\_min\_II=r_{mean}+b_{mean}-q\\ r\_\min \_II=r_{mean}-q\\ b\_\min \_II=b_{mean}-q\end{array}\right.;$

Wherein,

Rsum_I be the cumulative of red color r of the pixel meeting the first mellow lime pixel Rule of judgment and, Bsum_I is full Blue color difference b of the pixel of foot the first mellow lime pixel Rule of judgment cumulative and, nNum_I is for meeting gray pixel in first The number of the pixel of some Rule of judgment, r_meanFor meeting red color r of the pixel of the first mellow lime pixel Rule of judgment Meansigma methods, b_meanFor meeting the meansigma methods of blue color difference b of the pixel of the first mellow lime pixel Rule of judgment, q is default Parametric variable；

Second ratio computing unit, for entering the poor and described second mellow lime pixel Rule of judgment of reddish blue of described pixel Row contrast, determines that the pixel meeting described second mellow lime pixel Rule of judgment is described mellow lime pixel, and calculates described The number of mellow lime pixel accounts for the second ratio of pixel sum in described pending image；

Yield value computing unit, is used for judging whether described second ratio is less than the second preset ratio, if described second ratio is little In the second preset ratio, it is determined that described pixel keeps constant at the yield value of tri-Color Channels of R, G, B respectively, if described Second ratio is not less than the second preset ratio, then calculate described pixel according to equation below and lead in tri-colors of R, G, B respectively The yield value in road:

ORgain=Gsum_II/Rsum_II

OGgain=1；

OBgain=Gsum_II/Bsum_II

Wherein,

ORgain is the described pixel yield value at R Color Channel, and OGgain is the gain at G Color Channel of the described pixel Value, OBgain is the described pixel yield value at B color channel, and Rsum_II is for meeting the second mellow lime pixel Rule of judgment Pixel red color passage original value R cumulative and, Gsum_II is to meet the second mellow lime pixel Rule of judgment The original value G of the green color channel of pixel cumulative and, Bsum_II is the picture meeting the second mellow lime pixel Rule of judgment The original value B of the blue color channels of vegetarian refreshments cumulative and.

13. white balance processing equipments based on rgb space according to claim 11, it is characterised in that also include: first Judge parameter calculating module, judge parameter for calculating first；

Described first judges that parameter calculating module includes:

Color chart acquiring unit, for obtaining the color chart under different-colour, and calculates the mellow lime pixel in each color chart Reddish blue poor；

Scattergram acquiring unit, for poor according to the reddish blue of the mellow lime pixel in each color chart described, obtain described in The reddish blue difference Butut of spodogram vegetarian refreshments；

First judges parameter calculation unit, for obtaining the common set of described reddish blue difference Butut, and calculates described public When each pixel in set is satisfied by described first mellow lime pixel Rule of judgment, described first judges the numerical value of parameter.

14. white balance processing equipments based on rgb space according to claim 8, it is characterised in that described adjusting module Described pixel is calculated respectively at the corrected value of tri-Color Channels of R, G, B by below equation:

OR=R*ORgain

OG=G*OGgain；

OB=B*OBgain

Wherein, ORgain is the described pixel yield value at R Color Channel, and OGgain is that described pixel is at G Color Channel Yield value, OBgain is the described pixel yield value at B color channel, and R is original at R Color Channel of described pixel Value, G is the described pixel original value at G Color Channel, and B is the described pixel original value at B color channel, and oR is described Pixel is at the corrected value of R Color Channel, and oG is the described pixel corrected value at G Color Channel, and oB is that described pixel is at B The corrected value of Color Channel.