CN115665395B - Image white balance method suitable for monotonous scene - Google Patents
Image white balance method suitable for monotonous scene Download PDFInfo
- Publication number
- CN115665395B CN115665395B CN202211593680.3A CN202211593680A CN115665395B CN 115665395 B CN115665395 B CN 115665395B CN 202211593680 A CN202211593680 A CN 202211593680A CN 115665395 B CN115665395 B CN 115665395B
- Authority
- CN
- China
- Prior art keywords
- channel
- image
- color
- value
- pixels
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012937 correction Methods 0.000 claims abstract description 13
- 238000012216 screening Methods 0.000 claims abstract description 10
- 230000011218 segmentation Effects 0.000 claims abstract description 4
- 239000011111 cardboard Substances 0.000 claims description 6
- 238000005286 illumination Methods 0.000 claims description 3
- 238000011088 calibration curve Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000001028 reflection method Methods 0.000 description 5
- 239000003086 colorant Substances 0.000 description 2
- 238000005034 decoration Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
Images
Landscapes
- Processing Of Color Television Signals (AREA)
Abstract
The invention discloses an image white balance method suitable for a monotonous scene, which comprises the following steps: judging whether the integral color condition of the image to be processed belongs to a large-area single-color condition or not according to a space segmentation method; screening out pixels with the pixel similarity of more than 90% of the whole image by a spatial distance method, and then respectively calculating R, G, B three-channel mean values of the screened pixels; respectively calculating the color difference value of the image R, B channel by using a ratio difference method according to the three-channel mean value; according to the color difference value of an image R, B channel and a R, B channel gain curve corresponding to each standard single color, determining gain values Rgain and Bgain of an R channel and a B channel, and multiplying the gain values Rgain and Bgain by the R channel and the B channel of the original image respectively to finish white balance correction of the image. The invention can well restore the color aiming at the condition of integral color temperature deviation of the image in a monotonous scene.
Description
Technical Field
The present invention relates to an image processing method, and in particular, to an image white balance processing method.
Background
In modern society, image processing is closely related to life of people, the color of an image is not separable from the quality of the image, human eyes have constancy to the perception of color, and objects with color cast under various light sources can be restored to the color of the object, but for the current image sensor, the function is not provided, so that the image has color difference under different light sources, and White Balance (WB) can restore the color cast image to the original color of the image.
The current algorithms for white balancing images mainly comprise a gray world method, a complete reflection method and the like. However, for images with monotonous scenes and not rich colors, the common algorithm has no good effect.
(1) The gray world method: the method is based on an assumed situation, and the average values of RGB components of an image tend to be equal on the assumption that the color change of the image is rich enough, but for a monotonous scene, the gray world method is not applicable due to insufficient color, namely, the gray world method has no good effect on color restoration of the monotonous scene.
(2) Complete reflection method: it is also based on a hypothesis: the brightest pixels in an image correspond to points on the shiny or specular surface of an object, which conveys much information about the lighting conditions of the scene. If there is a pure white portion of the scene, then the light source information can be extracted directly from these pixels. Since the mirror or glossy surface does not absorb light by itself, the reflected color is the true color of the light source, since the reflectance function of the mirror or glossy surface remains constant over a long range of wavelengths. The perfect reflection method is to use this characteristic to adjust the image. When the algorithm is executed, the highest luminance pixel in the image is detected and used as the reference white point. Methods based on this idea are all called perfect reflection methods, also called mirror methods. The colloquial means that the brightest point in the whole image is white or reflected by a mirror surface, and the brightest point is the attribute of the light source, but the point is a white point, and the gain value can be calculated on the basis of the white point so as to be corrected. The total reflection method depends on the existence of highlight white blocks in a scene, and the color restoration has no good effect on a monotonous scene without the white blocks.
Disclosure of Invention
The purpose of the invention is as follows: aiming at the integral color temperature deviation of an image in a monotonous scene, the image white balance method suitable for the monotonous scene is provided, and the color can be well restored.
The technical scheme is as follows: an image white balance method suitable for a monotonous scene comprises the following steps:
step 1: judging whether the integral color condition of the image to be processed belongs to a large-area single-color condition or not according to a space segmentation method;
and 2, step: screening out pixels with the pixel similarity of more than 90% of the whole image by a spatial distance method, and then respectively calculating R, G, B three-channel mean values of the screened pixels;
and 3, step 3: according to the step 1, selecting a standard single color corresponding to the image as a reference, and then respectively calculating color difference values Rc and Bc of the R, B channel of the image by using a ratio difference method according to the three-channel mean value obtained in the step 2;
and 4, step 4: determining gain values Rgain and Bgain of an R channel and a B channel according to color difference values Rc and Bc of an image R, B channel and a R, B channel gain curve corresponding to each standard monochrome, and multiplying the gain values Rgain and Bgain of the R channel and the B channel of the original image respectively to finish white balance correction of the image.
Further, the step 1 comprises the following specific steps: taking an R channel as an x axis, a G channel as a y axis and a B channel as a z axis to form a three-dimensional space coordinate system, wherein the value range of each channel is 0-255;
if the R, B, G three-channel values of more than 80% of pixels of the image are all within 0-127, the image is considered to belong to the large-area black condition;
if the R channel value of more than 80% of pixels of the image is within 127-255 and the B, G channel value is within 0-127, the image is considered to be in a large-area red condition;
if the G channel value of more than 80% of pixels of the image is within 127-255 and the R, B channel value is within 0-127, the image is considered to be in the green condition of a large area;
if the R, G channel value of more than 80% of the pixels of the image is within 127-255 and the B channel value is within 0-127, the image is considered to belong to the yellow condition of a large area;
if the B channel value of more than 80% of pixels of the image is within 127-255 and the R, G channel value is within 0-127, the image is considered to belong to the large-area blue condition;
if the R, B channel value of more than 80% of the pixels of the image is within 127-255 and the G channel value is within 0-127, the image is considered to be in the large-area purple condition;
if the B, G channel value of more than 80% of the pixels of the image is within 127-255 and the R channel value is within 0-127, the image is considered to belong to the large-area cyan condition;
if the three channel values of R, G, B of more than 80% of the pixels of the image are all in 127-255, the image is considered to be in a large-area white condition.
Furthermore, the method for screening out the pixels with the similarity of more than 90 percent of the pixels of the whole image by the spatial distance method comprises the following steps:
if the image belongs to a certain monochromatic condition of a large area, setting a spherical area in the three-dimensional space coordinate system by taking a R, G, B three-channel value of a standard monochromatic corresponding to the monochromatic as a spherical center and taking a spatial distance L as a radius, screening out pixels falling in the spherical area in the image, and calculating the similarity P of the pixels falling in the spherical area according to the following formula:
P = (Pr + Pg + Pb)/3
in the formula, pr, pg, pb are the three-channel similarity of the pixel R, G, B in the spherical area respectively;
wherein Rm, gm, bm are the three-channel mean values of the pixel R, G, B falling in the spherical region, ri, gi, bi are the third channel mean values falling in the spherical regioniR, G, B three channel values of each pixel, n being the number of pixels falling in the spherical area;
the value range of L is 0-127, the similarity P is reduced along with the increase of the spatial distance L, and pixels with the pixel similarity P of more than 90% of the whole image are screened out by adjusting the value of the spatial distance L.
Further, the color difference values Rc and Bc of the R, B channel of the image are respectively calculated according to the following formula:
Rc=Rb-Rm*Gb/Gm
Bc=Bb-Bm*Gb/Gm
in the formula, rb, gb and Bb are R, G, B three channel values of the standard monochrome respectively.
Further, the method for acquiring the R, B channel gain curve corresponding to each standard monochrome comprises the following steps:
step A: for each standard monochromatic color, a standard color temperature correction curve is obtained under a standard light source:
shooting standard monochrome cardboards in a color temperature box with uniform illumination according to the sequence of color temperature from low to high, and calculating the mean values Rm ', gm ' and Bm ' of three channels of the image R, G, B shot at each color temperature;
according to three channel values Rb, gb and Bb of R, G, B of standard monochrome, images shot under different color temperatures are respectively corrected according to the following formula:
Rm'' = Gb/Gm' '* Rm'
Gm'' = Gb
Bm ''= Gb/Gm' * Bm'
wherein Rm ', gm ' and Bm ' are the mean values of the three channels R, G, B of the corrected image;
then, respectively calculating gain values Rgain and Bgain corresponding to the R channel and the B channel according to the following formula:
Rgain = Rb/Rm''
Bgain = Bb/Bm''
obtaining a plurality of groups of Rgain and Bgain according to images shot under different color temperatures, and drawing a standard color temperature correction curve corresponding to the standard single color by using Rgian and Bgain;
and B: the relation curve of the color difference value and the color temperature of the R, B channel is obtained by using a ratio difference value method:
for each standard monochrome, the color difference values Rc 'and Bc' of the image R, B channels obtained by shooting the standard monochrome cardboard at different color temperatures are respectively calculated according to the following formula:
Rc'=Rb-Rm'*Gb/Gm'
Bc'=Bb-Bm'*Gb/Gm'
obtaining a plurality of groups of Rc 'and Bc' according to images obtained by shooting under different color temperatures, and obtaining a relation curve between the color difference value and the color temperature of a R, B channel corresponding to the standard monochrome;
and C: and obtaining the R, B channel gain curve corresponding to each standard monochrome, namely a relation curve of the color difference value Rc 'and the gain value Rgain and a relation curve of the color difference value Bc' and the gain value Bgain according to the standard color temperature correction curve and the R, B channel color difference value and color temperature relation curve.
Has the advantages that: the method of the invention performs the processing aiming at the large-area monochromatic scene which is difficult to process by the automatic white balance algorithm of the mainstream image, firstly judges the image, and performs the subsequent processing if the image is in the large-area monochromatic scene, and the method can well restore the color in the monotonic scene; if the image is not in a large-area monochromatic scene, the image can be processed by using a mainstream automatic white balance algorithm, so that the method has strong adaptability and can be matched with the mainstream automatic white balance algorithm for use.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a schematic diagram of a three-dimensional space coordinate system and region division;
FIG. 3 is a graph of spatial distance L versus similarity P;
FIG. 4 is a standard color temperature calibration curve corresponding to a standard yellow color, wherein (a) is a color temperature-R channel calibration curve and (B) is a color temperature-B channel calibration curve;
FIG. 5 shows the R, B channel gain curves corresponding to standard yellow, where (a) is the R channel gain curve and (B) is the B channel gain curve.
Detailed Description
The invention is further explained below with reference to the drawings.
As shown in fig. 1, an image white balance method suitable for a monotonous scene is used for correcting the overall color temperature of a large-area monochromatic image, and includes the following steps:
step 1: and judging whether the overall color condition of the image to be processed belongs to a large-area single-color condition or not according to a space segmentation method.
Step 2: the pixels with the pixel similarity of more than 90% of the whole image are screened out by a spatial distance method, and then R, G, B three-channel mean values of the screened pixels are respectively calculated.
And 3, step 3: and (3) selecting a standard single color corresponding to the image as a reference according to the step 1, and then respectively calculating the color difference values Rc and Bc of the R, B channel of the image by using a ratio difference method according to the three-channel mean value obtained in the step 2.
And 4, step 4: determining gain values Rgain and Bgain of an R channel and a B channel according to color difference values Rc and Bc of an image R, B channel and a R, B channel gain curve corresponding to each standard monochrome, and multiplying the gain values Rgain and Bgain of the R channel and the B channel of the original image respectively to finish white balance correction of the image.
The step 1 comprises the following specific steps: a three-dimensional space coordinate system is formed by taking an R channel as an x axis, a G channel as a y axis and a B channel as a z axis, the value range of each channel is 0-255, and as shown in figure 2, the space coordinate system is divided into 8 areas. If the R, B, G three-channel values of more than 80% of pixels of the image are all within 0-127, the image is considered to belong to the large-area black condition; if the R channel value of more than 80% of pixels of the image is within 127-255 and the B, G channel value is within 0-127, the image is considered to be in a large-area red condition; if the G channel value of more than 80% of pixels of the image is within 127-255 and the R, B channel value is within 0-127, the image is considered to belong to the green condition of a large area; if the R, G channel value of more than 80% of the pixels of the image is within 127-255 and the B channel value is within 0-127, the image is considered to be in the large-area yellow condition; if the B channel value of more than 80% of pixels of the image is within 127-255 and the R, G channel value is within 0-127, the image is considered to belong to the blue condition of a large area; if the R, B channel value of more than 80% of the pixels of the image is within 127-255 and the G channel value is within 0-127, the image is considered to be in the large-area purple condition; if the B, G channel value of more than 80% of the pixels of the image is within 127-255 and the R channel value is within 0-127, the image is considered to belong to the large-area cyan condition; if the three channel values of R, G, B of more than 80% of the pixels of the image are all in 127-255, the image is considered to be in a large-area white condition.
In the step 2, the method for screening out the pixels with the similarity of more than 90 percent of the whole image pixels by a spatial distance method comprises the following steps:
if the image belongs to a certain monochromatic condition of a large area, setting a spherical area in a three-dimensional space coordinate system by taking a R, G, B three-channel value of a standard monochromatic corresponding to the monochromatic as a spherical center and taking a spatial distance L as a radius, screening out pixels falling in the spherical area in the image, and then calculating the similarity P of the pixels falling in the spherical area according to the following formula:
P = (Pr + Pg + Pb)/3
where Pr, pg, pb are the three-channel similarity of the pixel R, G, B in the spherical region, respectively:
wherein Rm, gm, bm are the three-channel mean values of the pixel R, G, B in the spherical region, ri, gi, bi are the third channel mean values in the spherical regioniR, G, B three channel values for each pixel, n is the number of pixels that fall within the spherical region.
The value range of L is 0-127, the value range of P is 0~1, and if the image belongs to a large-area yellow situation, three channel values (231, 199 and 31) of standard color yellow R, G, B are used as the sphere center in the color space, and image pixel similarity screening is carried out according to the method. A two-dimensional coordinate system is created, the spatial distance L is taken as the horizontal axis, and the similarity P is taken as the vertical axis, so as to obtain a relation curve between the spatial distance L and the similarity P as shown in fig. 3. The similarity P is reduced along with the increase of the spatial distance L, and pixels with the similarity P of more than 90% of the pixels of the whole image are screened out by adjusting the value of the spatial distance L.
In step 3, the color difference values Rc and Bc of the R, B channel of the image are respectively calculated according to the following formula:
Rc=Rb-Rm*Gb/Gm
Bc=Bb-Bm*Gb/Gm
in the formula, rb, gb and Bb are R, G, B three-channel values of standard monochrome respectively.
In step 4, the method for acquiring the R, B channel gain curves corresponding to the standard single colors comprises the following steps:
step A: for each standard single color, respectively acquiring standard color temperature correction curves of an R channel and a B channel under a standard light source:
the standard monochrome cardboards are sequentially shot in a color temperature box with uniform illumination according to the sequence of color temperature from low to high, and the mean values Rm ', gm ' and Bm ' of three channels of the image R, G, B shot at each color temperature are calculated.
The white balance processing mainly has the effects of eliminating the color temperature on the image, so that the color of the image accords with the habit of observation of human eyes, and the effects of the color temperature on the image are as follows: the low color temperature makes the image redder, and the high color temperature makes the image bluer. Therefore, the white balance processing of the method is mainly to multiply the R, B channel of the original image by an appropriate gain coefficient, so that the image color can be restored. Since the color temperature mainly affects the R, B channel, the G channel is used as a reference in calculating the correlation gain.
According to R, G, B three-channel values Rb, gb and Bb of a standard monochrome, firstly correcting a G channel, keeping the three-channel mean value proportional relation of an image R, G, B unchanged, and respectively correcting images obtained by shooting under different color temperatures according to the following formula:
Rm'' = Gb/Gm' * Rm'
Gm'' = Gb
Bm ''= Gb/Gm' * Bm'
wherein Rm ', gm ', bm ' are the mean values of the three channels R, G, B of the corrected image, respectively. Then, respectively calculating gain values Rgain and Bgain corresponding to the R channel and the B channel according to the following formula:
Rgain = Rb/Rm''
Bgain = Bb/Bm''
and obtaining a plurality of groups of Rgain and Bgain according to images shot under different color temperatures, and drawing a standard color temperature correction curve corresponding to the standard single color by using Rgian and Bgain.
Taking a standard yellow color as an example, the calibration curve at the standard color temperature corresponding to the yellow color obtained by the above method is shown in fig. 4, in which (a) is a color temperature-R channel calibration curve and (B) is a color temperature-B channel calibration curve.
And B: the relation curve of the color difference value and the color temperature of the R, B channel is obtained by using a ratio difference value method:
for each standard monochrome, taking out the Gb in the three-channel value of R, G, B of the standard monochrome without moving, taking out the mean value Rm ', gm ' and Bm ' of the three-channel of R, G, B of the image shot at each color temperature, multiplying the Gm ' by a coefficient v, adjusting the mean value to be consistent with the Gb value of the standard monochrome, multiplying the Rm ' by the coefficient v to obtain an Rt value, and subtracting the Rt value from the Rb value to obtain the Rc value; and multiplying Bm' by a coefficient v to obtain a Bt value, and subtracting the Bt value from the Bb value to obtain a Bc value, wherein the corresponding process mathematical expression is as follows:
v=Gb/Gm'；Rt=Rm'*v；Rc=Rb-Rt；
namely Rc ' = Rb-Rm '. Gb/Gm '
v=Gb/Gm'；Bt=Bm'*v；Bc=Rb-Bt；
Namely Bc ' = Bb-Bm '. Gb/Gm '
And respectively calculating the color difference values Rc 'and Bc' of R, B channels of the images obtained by shooting the standard monochrome cardboards under different color temperatures according to the formula, and obtaining a plurality of groups of Rc 'and Bc' according to the images obtained by shooting under different color temperatures to obtain a relationship curve between the color difference values and the color temperatures of R, B channels corresponding to the standard monochrome.
And C: and D, obtaining a R, B channel gain curve corresponding to each standard single color, namely a relationship curve of the color difference value Rc 'and the gain value Rgain and a relationship curve of the color difference value Bc' and the gain value Bgain according to the standard color temperature correction curves of the R channel and the B channel obtained in the step A and the R, B channel color difference value and color temperature relationship curve obtained in the step B. Taking the standard yellow color as an example, the R, B channel gain curve corresponding to the standard yellow obtained according to the above method is shown in fig. 5, where (a) is an R channel gain curve and (B) is a B channel gain curve.
The principle that the color difference values Rc and Bc of the R, B channel of the image to be processed are obtained in the step 3 is the same as that in the step B, and the difference is that the average values Rm, gm and Bm of the R, G, B channel with the pixel similarity P of the image to be processed being more than 90% are adopted in calculation; in step 4, rgain and Bgain values corresponding to the Rc and Bc values can be obtained according to a R, B channel gain curve corresponding to a standard single color confirmed in advance.
The 8 standard monochrome images and R, G, B three channel values used in the invention are respectively as follows: red (175,54,60), green (70,148,73), blue (56,61,150), black (52,52,52), white (243,243,242), violet (187,86,149), yellow (231,199,31), cyan (8,133,161).
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (2)
1. An image white balance method suitable for a monotonous scene is characterized by comprising the following steps:
step 1: judging whether the integral color condition of the image to be processed belongs to a large-area monochromatic condition or not according to a space segmentation method;
and 2, step: screening out pixels with the pixel similarity of more than 90% of the whole image by a spatial distance method, and then respectively calculating R, G, B three-channel mean values of the screened pixels;
and step 3: according to the step 1, selecting a standard single color corresponding to the image as a reference, and then respectively calculating color difference values Rc and Bc of the R, B channel of the image by using a ratio difference method according to the three-channel mean value obtained in the step 2;
and 4, step 4: determining gain values Rgain and Bgain of an R channel and a B channel according to color difference values Rc and Bc of an image R, B channel and a R, B channel gain curve corresponding to each standard monochrome, and multiplying the gain values Rgain and Bgain of the R channel and the B channel of the original image respectively to finish white balance correction of the image;
in the step 2, the method for screening out the pixels with the similarity of more than 90 percent of the whole image pixels by using the spatial distance method comprises the following steps:
if the image belongs to a large-area single color, setting a spherical area in a three-dimensional space coordinate system by taking a R, G, B three-channel value of a standard single color corresponding to the single color as a spherical center and a spatial distance L as a radius, screening out pixels falling in the spherical area in the image, and calculating the similarity P of the pixels falling in the spherical area according to the following formula:
P = (Pr + Pg + Pb)/3
in the formula, pr, pg, pb are the three-channel similarity of the pixel R, G, B in the spherical area respectively;
wherein Rm, gm, bm are the three-channel mean values of the pixel R, G, B falling in the spherical region, ri, gi, bi are the third channel mean values falling in the spherical regioniR, G, B triple channel values of each pixel, wherein n is the number of pixels falling in the spherical area;
the value range of L is 0 to 127, the similarity P is reduced along with the increase of the spatial distance L, and pixels with the pixel similarity P of more than 90 percent of the whole image are screened out by adjusting the value of the spatial distance L;
in the step 3, the color difference values Rc and Bc of the R, B channel of the image are respectively calculated according to the following formula:
Rc=Rb-Rm*Gb/Gm
Bc=Bb-Bm*Gb/Gm
in the formula, rb, gb and Bb are R, G, B three channel values of the standard monochrome respectively;
the R, B channel gain curve acquisition method corresponding to each standard monochrome comprises the following steps:
step A: for each standard monochromatic color, a standard color temperature correction curve is obtained under a standard light source:
shooting standard monochrome cardboards in a color temperature box with uniform illumination according to the sequence of color temperature from low to high, and calculating the mean values Rm ', gm ' and Bm ' of three channels of the image R, G, B shot at each color temperature;
according to three channel values Rb, gb and Bb of R, G, B of standard monochrome, images shot under different color temperatures are respectively corrected according to the following formula:
Rm'' = Gb/Gm' * Rm'
Gm'' = Gb
Bm ''= Gb/Gm' * Bm'
wherein Rm ', gm ', bm ' are the mean values of three channels of the corrected image R, G, B;
then, respectively calculating gain values Rgain and Bgain corresponding to the R channel and the B channel according to the following formula:
Rgain = Rb/Rm''
Bgain = Bb/Bm''
obtaining a plurality of groups of Rgain and Bgain according to images shot under different color temperatures, and drawing a standard color temperature correction curve corresponding to the standard single color by using Rgian and Bgain;
and B: the relation curve of the color difference value and the color temperature of the R, B channel is obtained by using a ratio difference value method:
for each standard monochrome, the color difference values Rc 'and Bc' of the R, B channels of the images obtained by shooting the standard monochrome cardboard at different color temperatures are respectively calculated according to the following formula:
Rc'=Rb-Rm'*Gb/Gm'
Bc'=Bb-Bm'*Gb/Gm'
obtaining a plurality of groups of Rc 'and Bc' according to images obtained by shooting under different color temperatures, and obtaining a relation curve between the color difference value and the color temperature of a R, B channel corresponding to the standard monochrome;
step C: and obtaining the R, B channel gain curve corresponding to each standard monochrome, namely a relation curve of the color difference value Rc 'and the gain value Rgain and a relation curve of the color difference value Bc' and the gain value Bgain according to the standard color temperature correction curve and the R, B channel color difference value and color temperature relation curve.
2. The image white balance method suitable for the monotonous scene according to claim 1, wherein the step 1 comprises the following specific steps: taking an R channel as an x axis, a G channel as a y axis and a B channel as a z axis to form a three-dimensional space coordinate system, wherein the value range of each channel is 0-255;
if the R, B, G three-channel values of more than 80% of pixels of the image are all within 0-127, the image is considered to belong to the large-area black condition;
if the R channel value of more than 80% of pixels of the image is within 127-255 and the B, G channel value is within 0-127, the image is considered to be in a large-area red condition;
if the G channel value of more than 80% of pixels of the image is within 127-255 and the R, B channel value is within 0-127, the image is considered to be in the green condition of a large area;
if the R, G channel value of more than 80% of the pixels of the image is within 127-255 and the B channel value is within 0-127, the image is considered to belong to the yellow condition of a large area;
if the B channel value of more than 80% of pixels of the image is within 127-255 and the R, G channel value is within 0-127, the image is considered to belong to the blue condition of a large area;
if the R, B channel value of more than 80% of the pixels of the image is within 127-255 and the G channel value is within 0-127, the image is considered to be in the large-area purple condition;
if the B, G channel value of more than 80% of the pixels of the image is within 127-255 and the R channel value is within 0-127, the image is considered to belong to the large-area cyan condition;
if the three channel values of R, G, B of more than 80% of the pixels of the image are all in 127-255, the image is considered to be in a large-area white condition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211593680.3A CN115665395B (en) | 2022-12-13 | 2022-12-13 | Image white balance method suitable for monotonous scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211593680.3A CN115665395B (en) | 2022-12-13 | 2022-12-13 | Image white balance method suitable for monotonous scene |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115665395A CN115665395A (en) | 2023-01-31 |
| CN115665395B true CN115665395B (en) | 2023-03-10 |
Family
ID=85018407
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211593680.3A Active CN115665395B (en) | 2022-12-13 | 2022-12-13 | Image white balance method suitable for monotonous scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115665395B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118175274A (en) * | 2024-05-15 | 2024-06-11 | 福思(杭州)智能科技有限公司 | Image processing method, device, storage medium and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110213556A (en) * | 2019-06-04 | 2019-09-06 | 福州瑞芯微电子股份有限公司 | Auto white balance method and system, storage medium and terminal under simple color scene |
| CN112788324A (en) * | 2021-02-26 | 2021-05-11 | 广东以诺通讯有限公司 | White balance method, system and terminal for pure color scene |
| CN114841904A (en) * | 2022-03-03 | 2022-08-02 | 浙江大华技术股份有限公司 | Image fusion method, electronic equipment and storage device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI523542B (en) * | 2014-02-24 | 2016-02-21 | 宏碁股份有限公司 | White balance compensation method and electronic apparatus using the same |
-
2022
- 2022-12-13 CN CN202211593680.3A patent/CN115665395B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110213556A (en) * | 2019-06-04 | 2019-09-06 | 福州瑞芯微电子股份有限公司 | Auto white balance method and system, storage medium and terminal under simple color scene |
| CN112788324A (en) * | 2021-02-26 | 2021-05-11 | 广东以诺通讯有限公司 | White balance method, system and terminal for pure color scene |
| CN114841904A (en) * | 2022-03-03 | 2022-08-02 | 浙江大华技术股份有限公司 | Image fusion method, electronic equipment and storage device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115665395A (en) | 2023-01-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9264689B2 (en) | Systems and methods for color compensation in multi-view video | |
| EP2426928B1 (en) | Image processing apparatus, image processing method and program | |
| US9723285B2 (en) | Multi-area white-balance control device, multi-area white-balance control method, multi-area white-balance control program, computer in which multi-area white-balance control program is recorded, multi-area white-balance image-processing device, multi-area white-balance image-processing method, multi-area white-balance image-processing program, computer in which multi-area white-balance image-processing program is recorded, and image-capture apparatus | |
| US7057768B2 (en) | Automatic color balance | |
| TWI293742B (en) | ||
| US9342872B2 (en) | Color correction parameter computation method, color correction parameter computation device, and image output system | |
| CN1941923B (en) | Automatic white balance method for color digital image | |
| EP1592227A2 (en) | Apparatus and method for determining an image processing parameter from image data at a user-selected image position | |
| US20100067030A1 (en) | Image color adjustment | |
| CN110381303B (en) | Portrait automatic exposure white balance correction method and system based on skin color statistics | |
| US7495696B2 (en) | White balance method | |
| JPH0898204A (en) | White balance adjustment extent arithmetic unit | |
| JP2004509486A (en) | Color cast detection and removal method | |
| WO2007007788A1 (en) | Color correction method and device | |
| CN115665395B (en) | Image white balance method suitable for monotonous scene | |
| JP2001148863A (en) | White balance adjustment method and adjustment device | |
| JPH0862741A (en) | Gradation correcting device | |
| US10224004B2 (en) | Automatic white balance based on surface reflection decomposition | |
| US20060177128A1 (en) | White balance with zone weighting | |
| CN113556526A (en) | RGBW filter array-based color enhancement method for color night vision equipment | |
| CN107534758A (en) | Image processing apparatus, image processing method and image processing program | |
| CN111107330A (en) | Color cast correction method for Lab space | |
| Akazawa et al. | N-white balancing: White balancing for multiple illuminants including non-uniform illumination | |
| Akazawa et al. | Multi-color balance for color constancy | |
| CN111010557B (en) | White point range determining method, white balance correcting method and device, and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240207 Address after: No. 88, Wenchang East Road, Yangzhou, Jiangsu 225000 Patentee after: Jiangsu Daoyuan Technology Group Co.,Ltd. Country or region after: China Address before: 211135 enlightenment star Nanjing maker space G41, second floor, No. 188, Qidi street, Qilin science and Technology Innovation Park, Qixia District, Nanjing, Jiangsu Province Patentee before: Jiangsu Peregrine Microelectronics Co.,Ltd. Country or region before: China |