CN103473747A - Method and device for transmitting colors - Google Patents

Abstract

The invention discloses a method and device for transmitting colors. The method is carried out based on L alpha beta space, and comprises the steps that one or more pairs of corresponding areas are selected in an original image and a target image according to a color transmitting task; a CTWC is set for each pixel in the original image in each pair of corresponding areas, and the CTWCs of the pixels are inversely proportional to the color statistical distance between the pixels and the currently-selected area of the original image; according to the CTWCs of the pixels of each pair of corresponding areas, color transmission linear conversion is carried out on the original image, and the converted image is obtained. The method and device can achieve local color transmission, and meet the diversified demands of users.

Description

The method and apparatus that color transmits

Technical field

The present invention relates to image processing field, the method and apparatus transmitted in particular to color.

Background technology

The color transmission is that a kind of ad hoc approach that utilizes carries out the method for image color interpolation or change automatically, specify former figure and target figure, utilize the color transmission method that the color information in target figure is delivered in former figure, the color of former figure is changed and there is the color characteristic similar to target figure.

Color transmission of the prior art is normally for overall color transmission, and overall color transmission method is based on the criterion that the image that seems similar has similar statistical information.It carries out linear transformation to single order and the second-order statistics feature of original image and target image, makes original image have the statistical nature similar with target image, and the original image after the conversion has visually just reached the effect similar to the target image color like this.

Yet the mode of this overall color transmission is that the color of view picture original image is all converted, and can't meet the demand of user individual.

For the mode of above-mentioned overall color transmission, can't meet the problem of users ' individualized requirement, effective solution is not yet proposed at present.

Summary of the invention

The method and apparatus that provides a kind of color to transmit is provided the purpose of the embodiment of the present invention, in order to address the above problem.

According to the one side of the embodiment of the present invention, a kind of method that provides color to transmit, the method is carried out based on L α β color space, comprising: select one or more pairs of corresponding regions in former figure and target figure according to the color transmission tasks; For every pair of corresponding region, be that each pixel in former figure arranges CTWC, wherein, the CTWC of pixel and pixel are inversely proportional to the Color Statistical distance of the current selected areas of former figure; According to the every pair of corresponding region, the CTWC of corresponding pixel, carry out the linear transformation of color transmission, the image after being converted to former figure.

According to the embodiment of the present invention on the other hand, the device that provides a kind of color to transmit comprises: module is selected in zone, for according to the color transmission tasks, at former figure and target figure, selecting one or more pairs of corresponding regions; Color transmits weight coefficient module is set, be used for based on L α β color space, every pair of corresponding region selecting module to select for zone is that each pixel in former figure arranges CTWC, and wherein, the CTWC of this pixel and pixel are inversely proportional to the Color Statistical distance of the current selected areas of former figure; The color linear transform module, for every pair of corresponding region that the module setting is set according to color transmission weight coefficient, the CTWC of corresponding pixel, carry out the linear transformation of color transmission, the image after being converted to former figure.

The embodiment of the present invention adopts the stronger L α β color space of Color Channel independence, and the CTWC of pixel and this pixel are set are inversely proportional to the Color Statistical distance of the current selected areas of former figure, then the CTWC based on each pixel carries out the color transmission, can reach local color transmission effect preferably, adapt to user's diversified demand.

The accompanying drawing explanation

The accompanying drawing that forms the application's a part is used to provide a further understanding of the present invention, and schematic description and description of the present invention the present invention does not form inappropriate limitation of the present invention for explaining.In the accompanying drawings:

Fig. 1 is the method flow diagram that the color that provides of the embodiment of the present invention transmits;

Fig. 2 is that the local color that the embodiment of the present invention provides is transmitted the schematic diagram of corresponding region;

Fig. 3 is the apparatus structure block diagram that the color that provides of the embodiment of the present invention transmits.

Embodiment

Hereinafter with reference to accompanying drawing, also describe the present invention in detail in conjunction with the embodiments.It should be noted that, in the situation that do not conflict, embodiment and the feature in embodiment in the application can combine mutually.

Consider that the user has the demand of carrying out the transmission of image local color sometimes, the method and apparatus that the embodiment of the present invention provides a kind of color to transmit, mainly be based on CTWC(Color Transfer Weight Coefficients, color transmits weight coefficient) the local color transmission carried out.

Simultaneously, because L α β color space relatively meets the human visual perception system, and while applying in natural scene by it, L α β significant spatial has reduced the correlativity between each Color Channel, make between each passage to have certain mutual independence, the impact that this variation that can reduce to greatest extent a passage causes to two other passage, thus can carry out different computings at different Color Channels, and the problem that not there will be passage to intersect.In order to reach local color transmission effect preferably, the execution space that the embodiment of the present invention has selected L α β color space to transmit as local color.Wherein, the L passage means the achromaticity passage, i.e. luminance channel, and α means colored Huang-blue channel, β means colored red-green passage.

The method flow diagram that color shown in Figure 1 transmits, the method is carried out based on L α β color space, comprises the following steps:

Step S102 selects one or more pairs of corresponding regions in former figure and target figure according to the color transmission tasks;

Step S104, be that each pixel in former figure arranges CTWC for every pair of corresponding region, and wherein, the CTWC of this pixel and this pixel are inversely proportional to the Color Statistical distance of the current selected areas of former figure;

Step S106, according to the every pair of above-mentioned corresponding region, the CTWC of corresponding pixel, carry out the linear transformation of color transmission, the image after being converted to former figure.

The method of the present embodiment adopts the stronger L α β color space of Color Channel independence, and the CTWC of pixel and this pixel are set are inversely proportional to the Color Statistical distance of the current selected areas of former figure, then the CTWC based on each pixel carries out the color transmission, can reach local color transmission effect preferably, adapt to user's diversified demand.

The needs that transmit according to color can identify out with rectangle frame by the zone of choosing respectively in former figure and target figure, and the pixel color scope in rectangle frame has determined between former figure and target figure to need the color gamut transmitted.When chosen area, can select according to actual needs multipair corresponding region, corresponding one by one with the zone of selecting in target figure for the zone that makes to select in former figure, the above-mentioned step of one or more pairs of corresponding regions of selecting in former figure and target figure according to the color transmission tasks of the present embodiment can comprise: according to the color transmission tasks, select corresponding region one by one in former figure and target figure; For the corresponding region of selecting in former figure and target figure arranges identical sign.By this simple mode, can determine that two identical zones of sign are that color transmits corresponding zone.

Local color is as shown in Figure 2 transmitted the schematic diagram of corresponding region, and the figure in left frame is former figure, and the figure in right frame is target figure.Wherein,

with

be k to the corresponding region in former figure and target figure,

with

be that k is to the corresponding region in former figure and target figure, c _s(i, j) is the value of certain Color Channel of pixel (i, j) in former figure.The needs that transmit according to color are selected corresponding region respectively in former figure and target figure, and the zone of choosing identifies out with rectangle frame, and the pixel color scope in rectangle frame has determined between former figure and target figure to need the color gamut transmitted.While selecting multipair corresponding region, the zone in former figure and target figure is corresponding one by one.

When the CTWC of each pixel is set, can utilize low pass Butterworth(Bart to irrigate hereby) the filter spectrum construction of function, this CTWC weight coefficient means the influence degree that the pixel in former figure is transmitted by the selected areas color, and weight coefficient and pixel are inversely proportional to the Color Statistical distance of selected areas.Calculate single order and the second-order statistics feature of original image and target image in each corresponding region, i.e. average and standard deviation determine the degree of linear transformation when carrying out linear transformation according to CTWC.If select the average of each Color Channel in zone to be for k in former figure

in former figure, select the value of each Color Channel of other pixels (i, j) outside zone to be for k

the Color Statistical distance between pixel (i, j) and selection zone can be calculated with following formula;

$x_{\mathrm{ij}}^k=\sqrt{{(l_{\mathrm{ij}}^k-{\mathrm{\μ}}_l^k)}^2+{({\mathrm{\α}}_{\mathrm{ij}}^k-{\mathrm{\μ}}_{\mathrm{\α}}^k)}^2+{({\mathrm{\β}}_{\mathrm{ij}}^k-{\mathrm{\μ}}_{\mathrm{\β}}^k)}^2};$

For calculating CTWC, need to build a weight function w (x), it and Color Statistical distance

be inversely proportional to,

larger, mean pixel (i, j) and select the field color gap larger, the impact that transmitted by color is less; If

less, mean pixel (i, j) and select the field color gap less, the impact that transmitted by color is larger.The present embodiment calculates CTWC with reference to low pass Butterworth wave filter.Wherein, the characteristics of Buterrworth wave filter are that the frequency response curve in passband is smooth to greatest extent, do not rise and fall, and at suppressed frequency band, drop to gradually zero.On the Bode diagram of the logarithm of amplitude and angular frequency, from a certain boundary angle frequency, amplitude the increase of angular frequency and is gradually reduced, the trend minus infinity.Its low-pass filter can mean with square formula to frequency of following amplitude:

${\left|H\left(w\right)\right|}^2=\frac{1}{1+{\left(\frac{w}{w_c}\right)}^{2n}}=\frac{1}{1+{\mathrm{\ϵ}}^2{\left(\frac{w}{w_{\mathrm{\ρ}}}\right)}^{2n}}$

Wherein, the exponent number of n=wave filter, w _c=cutoff frequency, frequency when its amplitude drops to-3 decibels, w _p=passband marginal frequency.

The frequency w of wave filter corresponds to the Color Statistical distance

cutoff frequency w _ccorrespond to color cut-off statistical distance x _c.Like this, the CTWC weight function of Butterworht filter form can mean with following formula.

$w\left(x_{\mathrm{ij}}^k\right)=\sqrt{\frac{1}{1+{(x_{\mathrm{ij}}^k/x_c)}^{2N}}}$

Wherein, x _cfor the color cut-off statistical distance of setting, the color filtering exponent number of N for setting; From above formula, color cut-off statistical distance x _ccan be used to control with the exponent number N of wave filter the effect that local color is transmitted.

In k selection zone, the CTWC that the pixel in it can be set is 1.

The set-up mode of above-mentioned CTWC can reach local color transmission effect preferably, less on the impact in other territory district.

When the corresponding region of selecting is a pair of, can adopt formula in former figure and target figure $C_s^{\mathrm{new}}(i,j)=C_s(i,j)+w\left(x_{\mathrm{ij}}\right)\·({\mathrm{\μ}}_t+\frac{{\mathrm{\σ}}_t}{{\mathrm{\σ}}_s}(C_s(i,j)-{\mathrm{\μ}}_s)-C_s(i,j))$ The Color Channel of former figure is carried out to the linear transformation of color transmission; Wherein,

the numerical value of corresponding color passage after pixel (i, the j) conversion in former figure; C _s(i, j) is the numerical value of pixel (i, j) corresponding color passage, w (x _ij) be the CTWC of pixel (i, j), μ _s, μ _trespectively the average of the corresponding region corresponding color passage of former figure and target figure, σ _s, σ _tto be respectively the standard deviation of the corresponding region corresponding color passage of former figure and target figure.

The corresponding region of selecting in former figure and target figure is M(M>1) to the time, current task is multizone local color transmission tasks, can adopt Image Iterative fusion method or weighted mean fusion method to realize the transmission of multizone local color.

Above-mentioned Image Iterative fusion method refers to: according to the order of chosen area, take turns doing the local color transmission, and after once transmit before being delivered in once on the basis of result and complete, the final like this figure as a result obtained is the result that all corresponding regions iteration color transmits.Be implemented as follows:

According to selecting regional order, adopt following formula to carry out the local color transmission for every pair of corresponding region successively:

$\begin{array}{c}{C}_s^k(i,j)=C_s^{k-1}(i,j)+w\left(x_{\mathrm{ij}}^k\right)\·({\mathrm{\μ}}_t^k+\frac{{\mathrm{\σ}}_t^k}{{\mathrm{\σ}}_s^k}(C_s^{k-1}(i,j)-{\mathrm{\μ}}_s^k)-C_s^{k-1}(i,j))\\ (k=\mathrm{1,2,3},\·\·\·,M)\end{array}$

Wherein, M>1,

the initial value of a Color Channel of pixel (i, j) in former figure,

the figure as a result obtained after the color transmission of k corresponding region corresponding color passage;

the CTWC of pixel (i, j) for k corresponding region,

respectively the average of k corresponding region corresponding color passage in former figure and target figure,

it is the standard deviation of k corresponding region corresponding color passage in former figure and target figure.

Above-mentioned weighted mean fusion method refers to: the local color transmission is first carried out separately in every a pair of corresponding region, and then the CTWC separately of utilization as a result that every a pair of corresponding region is obtained is weighted on average, obtains final figure as a result.The basic step of the method is as follows:

1) adopt following formula to carry out separately respectively the local color transmission to every pair of corresponding region:

$\begin{array}{c}{C}_s^k(i,j)=C_s(i,j)+w\left(x_{\mathrm{ij}}^k\right)\·({\mathrm{\μ}}_t^k+\frac{{\mathrm{\σ}}_t^k}{{\mathrm{\σ}}_s^k}(C_s(i,j)-{\mathrm{\μ}}_s^k)-C_s(i,j))\\ (k=\mathrm{1,2},\·\·\·,M)\end{array}$

Wherein, M>1,

numerical value after to be pixel (i, j) in former figure transmit for the color of a Color Channel of k corresponding region; C _s(i, j) is the numerical value of pixel (i, j) corresponding color passage,

the CTWC of pixel (i, j) for k corresponding region, respectively the average of k corresponding region corresponding color passage in former figure and target figure,

it is the standard deviation of k corresponding region corresponding color passage in former figure and target figure;

In this step, for each pixel (i, j) in former figure, can calculate respectively in the manner described above the Color Statistical distance of itself and each corresponding region

and then obtain this zone to this pixel $\mathrm{CTWCw}\left(x_{\mathrm{ij}}^k\right)(k=\mathrm{1,2},\·\·\·,M);$

2) for every pair of corresponding region, the CTWC weighted mean value of calculating pixel point (i, j)

$p\left(x_{\mathrm{ij}}^k\right)=\frac{w\left(x_{\mathrm{ij}}^k\right)}{\underset{k=1}{\overset{M}{\mathrm{\Σ}}}w\left(x_{\mathrm{ij}}^k\right)};$

This step is equivalent to the local color transmission is carried out separately in every pair of corresponding region;

3) calculate above-mentioned pixel (i, j) and carry out the result after the color transmission for every pair of corresponding region $C_s^{\mathrm{new}}(i,j)=\underset{k=1}{\overset{M}{\mathrm{\Σ}}}p\left(x_{\mathrm{ij}}^k\right)\·C_s^k(i,j).$

Each pixel is all processed in the manner described above, obtains final color and transmits figure as a result.

Said method transmits corresponding region according to the color of selecting in original image and target image, and the pixel structure CTWC to former figure, determine the influence degree that former figure is transmitted by target figure color, thereby reach the purpose that local color is transmitted.In addition, realize multizone local color transmission image co-registration by using iterative image fusion method formula or weighted mean fusion method, thereby realize the transmission to the multizone local color.

Corresponding to said method, the device that the embodiment of the present invention also provides a kind of color to transmit, the apparatus structure block diagram that color shown in Figure 3 transmits, this device comprises with lower module:

Module 32 is selected in zone, for according to the color transmission tasks, at former figure and target figure, selecting one or more pairs of corresponding regions;

Color transmits weight coefficient module 34 is set, be used for based on L α β color space, every pair of corresponding region selecting module 32 to select for zone is that each pixel in former figure arranges color transmission weight coefficient CTWC, wherein, the CTWC of pixel and pixel are inversely proportional to the Color Statistical distance of the current selected areas of former figure;

Color linear transform module 36, for every pair of corresponding region that module 34 settings are set according to color transmission weight coefficient, the CTWC of corresponding pixel, carry out the linear transformation of color transmission, the image after being converted to former figure.

The device of the present embodiment adopts the stronger L α β color space of Color Channel independence, and the CTWC of pixel and this pixel are set are inversely proportional to the Color Statistical distance of the current selected areas of former figure, then the CTWC based on each pixel carries out the color transmission, can reach local color transmission effect preferably, adapt to user's diversified demand.

Preferably, above-mentioned color transmission weight coefficient arranges module 34 and comprises: the Color Statistical metrics calculation unit, select other pixel (i, j) and k the Color Statistical distance of selecting between zone outside zone for calculating k, former figure wherein, select the average of each Color Channel in regional L α β color space to be for k

in the L α β color space of pixel (i, j), the value of each Color Channel is

the first setting unit for the CTWC that pixel (i, j) is set is:

wherein, x _cfor the color cut-off statistical distance of setting, the color filtering exponent number of N for setting; The second setting unit is 1 for the CTWC that k the pixel in the selection zone is set.

For selecting the situation of a pair of corresponding region in former figure and target figure, above-mentioned color linear transform module 36 can comprise: the first linear transform unit, while being a pair of for the corresponding region selected as former figure and target figure, adopt following formula the Color Channel of former figure to be carried out to the linear transformation of color transmission:

$C_s^{\mathrm{new}}(i,j)=C_s(i,j)+w\left(x_{\mathrm{ij}}\right)\·({\mathrm{\μ}}_t+\frac{{\mathrm{\σ}}_t}{{\mathrm{\σ}}_s}(C_s(i,j)-{\mathrm{\μ}}_s)-C_s(i,j))$

Wherein,

the numerical value of corresponding color passage after pixel (i, the j) conversion in former figure; C _s(i, j) is the numerical value of pixel (i, j) corresponding color passage, w (x _ij) be the CTWC of pixel (i, j), μ _s, μ _trespectively the average of the corresponding region corresponding color passage of former figure and target figure, σ _s, σ _tto be respectively the standard deviation of the corresponding region corresponding color passage of former figure and target figure.

For selecting the situation of multipair corresponding region in former figure and target figure, above-mentioned color linear transform module 36 can adopt following two kinds of specific implementations, mode one: color linear transform module 36 comprises: the iteration integrated unit, for the corresponding region of selecting as former figure and target figure be M to the time, according to selecting regional order, adopt following formula to carry out the local color transmission for every pair of corresponding region successively:

$\begin{array}{c}{C}_s^k(i,j)=C_s^{k-1}(i,j)+w\left(x_{\mathrm{ij}}^k\right)\·({\mathrm{\μ}}_t^k+\frac{{\mathrm{\σ}}_t^k}{{\mathrm{\σ}}_s^k}(C_s^{k-1}(i,j)-{\mathrm{\μ}}_s^k)-C_s^{k-1}(i,j))\\ (k=\mathrm{1,2,3},\·\·\·,M)\end{array}$

Wherein, M>1,

the initial value of a Color Channel of pixel (i, j) in former figure,

the figure as a result obtained after the color transmission of k corresponding region corresponding color passage;

the CTWC of pixel (i, j) for k corresponding region,

respectively the average of k corresponding region corresponding color passage in former figure and target figure,

it is the standard deviation of k corresponding region corresponding color passage in former figure and target figure;

Mode two, color linear transform module 36 comprises with lower unit:

The local color transfer unit, for the corresponding region of selecting as former figure and target figure be M to the time, adopt following formula to carry out separately respectively the local color transmission to every pair of corresponding region: $\begin{array}{c}{C}_s^k(i,j)=C_s(i,j)+w\left(x_{\mathrm{ij}}^k\right)\·({\mathrm{\μ}}_t^k+\frac{{\mathrm{\σ}}_t^k}{{\mathrm{\σ}}_s^k}(C_s(i,j)-{\mathrm{\μ}}_s^k)-C_s(i,j))\\ (k=\mathrm{1,2},\·\·\·,M)\end{array};$

Wherein, M>1, numerical value after to be pixel (i, j) in former figure transmit for the color of a Color Channel of k corresponding region; C _s(i, j) is the numerical value of pixel (i, j) corresponding color passage,

the CTWC of pixel (i, j) for k corresponding region,

respectively the average of k corresponding region corresponding color passage in former figure and target figure,

it is the standard deviation of k corresponding region corresponding color passage in former figure and target figure;

The weighted mean value computing unit, for for every pair of corresponding region, the CTWC weighted mean value of calculating pixel point (i, j)

$p\left(x_{\mathrm{ij}}^k\right)=\frac{w\left(x_{\mathrm{ij}}^k\right)}{\underset{k=1}{\overset{M}{\mathrm{\Σ}}}w\left(x_{\mathrm{ij}}^k\right)};$

The weighted mean integrated unit, carry out the result after the color transmission for calculating pixel point (i, j) for every pair of corresponding region $C_s^{\mathrm{new}}(i,j)=\underset{k=1}{\overset{M}{\mathrm{\Σ}}}p\left(x_{\mathrm{ij}}^k\right)\·C_s^k(i,j).$

Above-described embodiment has been realized the local color transmission to image on the basis of transmitting weight coefficient based on color, and its directive property with delivery areas is higher high with dirigibility.In addition, after using iterative image amalgamation mode or weighted mean amalgamation mode, can realize preferably the local color transmission of multizone.

Obviously, those skilled in the art should be understood that, above-mentioned each module of the present invention or each step can realize with general calculation element, they can concentrate on single calculation element, perhaps be distributed on the network that a plurality of calculation elements form, alternatively, they can be realized with the executable program code of calculation element, thereby, they can be stored in memory storage and be carried out by calculation element, and in some cases, can carry out step shown or that describe with the order be different from herein, perhaps they are made into respectively to each integrated circuit modules, perhaps a plurality of modules in them or step being made into to the single integrated circuit module realizes.Like this, the present invention is not restricted to any specific hardware and software combination.

The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any modification of doing, be equal to replacement, improvement etc., within all should being included in protection scope of the present invention.

Claims (10)

1. the method that color transmits, is characterized in that, described method is carried out based on L α β color space, comprising:

Select one or more pairs of corresponding regions in former figure and target figure according to the color transmission tasks;

For every pair of described corresponding region, be that each pixel in described former figure arranges color and transmits weight coefficient CTWC, wherein, the CTWC of described pixel and described pixel to the Color Statistical of the current selected areas of described former figure apart from being inversely proportional to;

According to the every pair of described corresponding region, the CTWC of corresponding pixel, carry out the linear transformation of color transmission, the image after being converted to described former figure.

2. method according to claim 1, is characterized in that, according to the color transmission tasks, in former figure and target figure, selects one or more pairs of corresponding regions to comprise:

Select corresponding region one by one in former figure and target figure according to the color transmission tasks;

For the corresponding region of selecting in described former figure and described target figure arranges identical sign.

3. method according to claim 1, is characterized in that, for every pair of described corresponding region, is that each pixel in described former figure arranges CTWC and comprises:

Calculate k other pixel (i, j) and described k the Color Statistical distance of selecting between zone of selecting outside zone in described former figure $s_{\mathrm{ij}}^k=\sqrt{{(l_{\mathrm{ij}}^k-{\mathrm{\μ}}_l^k)}^2+{({\mathrm{\α}}_{\mathrm{ij}}^k-{\mathrm{\μ}}_{\mathrm{\α}}^k)}^2+{({\mathrm{\β}}_{\mathrm{ij}}^k-{\mathrm{\μ}}_{\mathrm{\β}}^k)}^2};$ Wherein, select the average of each Color Channel in zone to be for described k

the value of each Color Channel of described pixel (i, j) is

The CTWC that described pixel (i, j) is set is:

wherein, x _cfor the color cut-off statistical distance of setting, the color filtering exponent number of N for setting;

The CTWC that pixel in described k selection zone is set is 1.

4. method according to claim 1, is characterized in that, the CTWC of corresponding pixel according to the every pair of described corresponding region, and the linear transformation that described former figure is carried out to the color transmission comprises:

When the corresponding region of selecting is a pair of, adopt formula in described former figure and described target figure $C_s^{\mathrm{new}}(i,j)=C_s(i,j)+w\left(x_{\mathrm{ij}}\right)\·({\mathrm{\μ}}_t+\frac{{\mathrm{\σ}}_t}{{\mathrm{\σ}}_s}(C_s(i,j)-{\mathrm{\μ}}_s)-C_s(i,j))$ The Color Channel of described former figure is carried out to the linear transformation of color transmission;

Wherein,

the numerical value of corresponding described Color Channel after pixel (i, the j) conversion in described former figure; C _s(i, j) is the numerical value of the corresponding described Color Channel of described pixel (i, j), w (x _ij) be the CTWC of described pixel (i, j), μ _s, μ _trespectively the average of the corresponding described Color Channel in described corresponding region of described former figure and described target figure, σ _s, σ _tto be respectively the standard deviation of the corresponding described Color Channel in described corresponding region of described former figure and described target figure.

5. method according to claim 1, is characterized in that, the CTWC of corresponding pixel according to the every pair of described corresponding region, and the linear transformation that described former figure is carried out to the color transmission comprises:

The corresponding region of selecting in described former figure and described target figure be M to the time, according to selecting regional order, adopt following formula to carry out the local color transmission for every pair of corresponding region successively:

$\begin{array}{c}{C}_s^k(i,j)=C_s^{k-1}(i,j)+w\left(x_{\mathrm{ij}}^k\right)\·({\mathrm{\μ}}_t^k+\frac{{\mathrm{\σ}}_t^k}{{\mathrm{\σ}}_s^k}(C_s^{k-1}(i,j)-{\mathrm{\μ}}_s^k)-C_s^{k-1}(i,j))\\ (k=\mathrm{1,2,3},\·\·\·,M)\end{array}$

Wherein, M>1,

the initial value of a Color Channel of pixel (i, j) in described former figure,

the figure as a result obtained after the color transmission of the corresponding described Color Channel in k corresponding region;

the CTWC of described pixel (i, j) for k corresponding region,

respectively the average of k the corresponding described Color Channel in corresponding region in described former figure and described target figure, it is the standard deviation of k the corresponding described Color Channel in corresponding region in described former figure and described target figure.

6. method according to claim 1, is characterized in that, the CTWC of corresponding pixel according to the every pair of described corresponding region, and the linear transformation that described former figure is carried out to the color transmission comprises:

The corresponding region of selecting in described former figure and described target figure be M to the time, adopt following formula to carry out separately respectively the local color transmission to every pair of described corresponding region: $\begin{array}{c}{C}_s^k(i,j)=C_s(i,j)+w\left(x_{\mathrm{ij}}^k\right)\·({\mathrm{\μ}}_t^k+\frac{{\mathrm{\σ}}_t^k}{{\mathrm{\σ}}_s^k}(C_s(i,j)-{\mathrm{\μ}}_s^k)-C_s(i,j))\\ (k=\mathrm{1,2},\·\·\·,M)\end{array}$

Wherein, M>1, numerical value after to be pixel (i, j) in described former figure transmit for the color of a Color Channel of k corresponding region; C _s(i, j) is the numerical value of the corresponding described Color Channel of described pixel (i, j), the CTWC of described pixel (i, j) for k corresponding region,

respectively the average of k the corresponding described Color Channel in corresponding region in described former figure and described target figure,

it is the standard deviation of k the corresponding described Color Channel in corresponding region in described former figure and described target figure;

For every pair of described corresponding region, calculate the CTWC weighted mean value of described pixel (i, j) $p\left(x_{\mathrm{ij}}^k\right):$

$p\left(x_{\mathrm{ij}}^k\right)=\frac{w\left(x_{\mathrm{ij}}^k\right)}{\underset{k=1}{\overset{M}{\mathrm{\Σ}}}w\left(x_{\mathrm{ij}}^k\right)}$

Calculate described pixel (i, j) and carry out the result after the color transmission for every pair of described corresponding region $C_s^{\mathrm{new}}(i,j)=\underset{k=1}{\overset{M}{\mathrm{\Σ}}}p\left(x_{\mathrm{ij}}^k\right)\·C_s^k(i,j).$

7. the device that color transmits, is characterized in that, comprising:

Module is selected in zone, for according to the color transmission tasks, at former figure and target figure, selecting one or more pairs of corresponding regions;

Color transmits weight coefficient module is set, be used for based on L α β color space, every pair of described corresponding region selecting module to select for described zone is that each pixel in described former figure arranges color transmission weight coefficient CTWC, wherein, the CTWC of described pixel and described pixel are inversely proportional to the Color Statistical distance of the current selected areas of described former figure;

The color linear transform module, for the every pair of described corresponding region that the module setting is set according to described color transmission weight coefficient, the CTWC of corresponding pixel, carry out the linear transformation of color transmission, the image after being converted to described former figure.

8. device according to claim 7, is characterized in that, described color transmission weight coefficient arranges module and comprises:

The Color Statistical metrics calculation unit, select other pixel (i, j) and described k the Color Statistical distance of selecting between zone outside zone for calculating k, described former figure

wherein, select the average of each Color Channel in regional L α β color space to be for described k

in the L α β color space of described pixel (i, j), the value of each Color Channel is

The first setting unit for the CTWC that described pixel (i, j) is set is: wherein, x _cfor the color cut-off statistical distance of setting, the color filtering exponent number of N for setting;

The second setting unit is 1 for the CTWC that the pixel in described k selection zone is set.

9. device according to claim 7, is characterized in that, described color linear transform module comprises:

The first linear transform unit, while being a pair of for the corresponding region selected as described former figure and described target figure, adopts following formula the Color Channel of described former figure to be carried out to the linear transformation of color transmission:

$C_s^{\mathrm{new}}(i,j)=C_s(i,j)+w\left(x_{\mathrm{ij}}\right)\·({\mathrm{\μ}}_t+\frac{{\mathrm{\σ}}_t}{{\mathrm{\σ}}_s}(C_s(i,j)-{\mathrm{\μ}}_s)-C_s(i,j))$

Wherein, the numerical value of corresponding described Color Channel after pixel (i, the j) conversion in described former figure; C _s(i, j) is the numerical value of the corresponding described Color Channel of described pixel (i, j), w (x _ij) be the CTWC of described pixel (i, j), μ _s, μ _trespectively the average of the corresponding described Color Channel in described corresponding region of described former figure and described target figure, σ _s, σ _tto be respectively the standard deviation of the corresponding described Color Channel in described corresponding region of described former figure and described target figure.

10. device according to claim 7, is characterized in that,

Described color linear transform module comprises: the iteration integrated unit, for the corresponding region of selecting as described former figure and described target figure be M to the time, according to selecting regional order, adopt following formula to carry out the local color transmission for every pair of corresponding region successively:

$\begin{array}{c}{C}_s^k(i,j)=C_s^{k-1}(i,j)+w\left(x_{\mathrm{ij}}^k\right)\·({\mathrm{\μ}}_t^k+\frac{{\mathrm{\σ}}_t^k}{{\mathrm{\σ}}_s^k}(C_s^{k-1}(i,j)-{\mathrm{\μ}}_s^k)-C_s^{k-1}(i,j))\\ (k=\mathrm{1,2,3},\·\·\·,M)\end{array}$

Wherein, M>1,

the initial value of a Color Channel of pixel (i, j) in described former figure,

the figure as a result obtained after the color transmission of the corresponding described Color Channel in k corresponding region;

the CTWC of described pixel (i, j) for k corresponding region, respectively the average of k the corresponding described Color Channel in corresponding region in described former figure and described target figure,

it is the standard deviation of k the corresponding described Color Channel in corresponding region in described former figure and described target figure;

Perhaps, described color linear transform module comprises:

The local color transfer unit, for the corresponding region of selecting as described former figure and described target figure be M to the time, adopt following formula to carry out separately respectively the local color transmission to every pair of described corresponding region: $\begin{array}{c}{C}_s^k(i,j)=C_s(i,j)+w\left(x_{\mathrm{ij}}^k\right)\·({\mathrm{\μ}}_t^k+\frac{{\mathrm{\σ}}_t^k}{{\mathrm{\σ}}_s^k}(C_s(i,j)-{\mathrm{\μ}}_s^k)-C_s(i,j))\\ (k=\mathrm{1,2},\·\·\·,M)\end{array};$

Wherein, M>1,

numerical value after to be pixel (i, j) in described former figure transmit for the color of a Color Channel of k corresponding region; C _s(i, j) is the numerical value of the corresponding described Color Channel of described pixel (i, j),

the CTWC of described pixel (i, j) for k corresponding region, respectively the average of k the corresponding described Color Channel in corresponding region in described former figure and described target figure,

it is the standard deviation of k the corresponding described Color Channel in corresponding region in described former figure and described target figure;

The weighted mean value computing unit, for for every pair of described corresponding region, calculate the CTWC weighted mean value of described pixel (i, j)

$p\left(x_{\mathrm{ij}}^k\right)=\frac{w\left(x_{\mathrm{ij}}^k\right)}{\underset{k=1}{\overset{M}{\mathrm{\Σ}}}w\left(x_{\mathrm{ij}}^k\right)};$

The weighted mean integrated unit, carry out the result after the color transmission for calculating described pixel (i, j) for every pair of described corresponding region

Images