TW201408038A - Color calibration method and image processing device using the same - Google Patents

Abstract

A color calibration method is provided. The method includes: obtaining a first image of a calibration card having color blocks, wherein each of the color block corresponds to a standard value; obtaining first sensing values according to the first image, wherein each of the first sensing values corresponds to one of the standard values; generating a calibration table according to the first sensing values and the standard values; and determining if a recursive condition is satisfied. The method also includes: if the recursive condition is not satisfied, obtaining a second image of the calibration card, obtaining second sensing values according to the second image, adjusting the second sensing values according to the calibration table to generate third sensing values, and updating the calibration table according to the third sensing values and the standard values until the recursive condition is satisfied. Therefore, the method generates the calibration table automatically.

Description

Color correction method and image processing device

The present invention relates to a color correction method and an image processing apparatus using the same.

In general, when a scanner captures an image of an object, the actual brightness of the object does not necessarily equal the brightness of the captured image. There is a nonlinear distortion between the brightness of the object and the brightness of the image. This distortion can be expressed by the following equation (1).

Where X is the brightness value of the object, Y is the brightness value of the image, and γ is a parameter. Because the above nonlinear relationship exists when capturing an image, the scanner removes the luminance or chrominance of the image after acquiring the image to remove the gamma correction. Distortion. The gamma correction compensates for errors in the image in a nonlinear manner, so that the brightness value of the object is linear with the brightness value corrected by the image. The gamma correction can be expressed by the following equation (2).

Where Y' is the gamma corrected brightness value. γ is a parameter used for gamma correction. Therefore, after the operation of equation (2), a relationship between Y' and X becomes linear.

In order to achieve the above gamma correction effect, common methods are analogous and digital. Analogy is the use of nonlinear electronic circuits to generate near The result of the operation of equation (2) above.

The digital method is to directly calculate the exponential operation of equation (2) above. However, on some devices it is not possible to do exponential operations on the fly, so the pre-computed inputs and outputs are stored in a lookup table. When gamma correction is to be performed, the adjusted brightness value can be obtained by querying the lookup table.

It is very efficient to use only one gamma on different scanners. However, a value of γ does not necessarily apply to all scanners. It takes a lot of time to generate a corresponding γ manually for each scanner. Therefore, how to automatically generate gamma is of interest to researchers in this field.

Embodiments of the present invention provide a color correction method and an image processing apparatus that can automatically generate a correction table.

An embodiment of the invention provides a color correction method for an image processing apparatus. The color correction method includes: obtaining a first image of the guide table, the guide table includes a plurality of color blocks, and each color block corresponds to a standard value; and obtaining a plurality of first measurement values according to the first image, wherein each of the first images The measured value corresponds to one of the above standard values; a correction table is generated according to the first measured value and the standard value; and whether the recursive condition is satisfied is satisfied. The color correction method further includes: if the recursive condition is not satisfied, obtaining the second image of the guide table, obtaining a plurality of second measurement values according to the second image, and adjusting the second measurement value according to the calibration table to generate a plurality of third Measured values and based on the third measured value and standard The value updates the correction table until the recursion condition is met.

In an embodiment, the correction table includes a plurality of first output values, each of the first output values corresponding to a first index. The step of generating a calibration table according to the first measured value and the standard value includes: setting a first index of the portion into a first measured value; setting a first output value corresponding to the first measured value as a standard value; and generating The first output value that does not correspond to the first measurement value.

In an embodiment, the step of determining whether the recursive condition is satisfied comprises: determining whether a number of recursive times meets a recursive threshold; and if the number of recursive times meets a recursive threshold, determining that the recursive condition is satisfied. The step of updating the correction table according to the third measured value and the standard value further includes: updating the number of recursive times.

In an embodiment, the step of determining whether the recursive condition is satisfied comprises: calculating an error value according to the first measured value and the standard value, or calculating the error value according to the third measured value and the standard value; determining the error value Whether it is less than an error threshold; and if the error value is less than the error threshold, it is judged that the recursion condition has been satisfied.

In an embodiment, the first measured value includes a minimum measured value and a maximum measured value. The step of generating the first output value that does not correspond to the first measurement value includes: setting a first output value corresponding to the first index smaller than the minimum measurement value as a first preset value; setting a number greater than the maximum measurement value The first output value corresponding to an index is a second preset value; and the remaining first output value is generated by an interpolation algorithm.

In an embodiment, the first index includes a second index, and the second index corresponds to a second output value of the first output value. on The step of updating the correction table according to the third measured value and the standard value includes: establishing a temporary correction table, the temporary correction table including a plurality of third output values, and each of the third output values corresponding to a third index; setting part The third index of the share is the third measured value; setting a third output value corresponding to the third measured value as a standard value; generating a third output value not corresponding to the third measured value; and outputting the second output The value is one of the third indices to obtain a fourth output value of the third output value; and the second output value is set to the fourth output value. Therefore, the updated correction table can be generated by the second index and the second output value.

In an embodiment, the step of calculating an error value according to the first measured value and the standard value comprises: calculating a sum of squared residuals between the first measured value and the standard value to obtain the error value. The step of calculating the error value according to the third measured value and the standard value includes calculating a sum of squared residuals between the third measured value and the standard value to obtain the error value.

In another aspect, an embodiment of the present invention provides an image processing apparatus, including an image capturing unit and a processor. The image capturing unit is used to obtain a first image of the guide. This guide table includes a plurality of color patches, and each color patch corresponds to a standard value. The processor is coupled to the image capturing unit for obtaining a plurality of first measurement values according to the first image, wherein each of the first measurement values corresponds to one of the standard values. The processor is configured to generate a correction table according to the first measured value and the standard value, and determine whether a recursive condition is satisfied. If the recursive condition is not satisfied, the processor is further configured to obtain a second image of the guide table, obtain a plurality of second measurement values according to the second image, and adjust the second measurement value according to the calibration table to generate a plurality of third measurement values. Value, and according to The third measurement value and the standard value update the correction table until the recursion condition is satisfied.

In an embodiment, the correction table includes a plurality of first output values, and each of the first output values corresponds to a first index. The processor is further configured to set a first index of the portion to a first measured value, set a first output value corresponding to the first measured value to a standard value, and generate a first output value that does not correspond to the first measured value. .

In an embodiment, the processor is further configured to determine whether a number of recursive times meets a recursive threshold. If the number of recursions meets the recursive threshold, the processor is also used to determine that the recursive condition has been met. The processor is further configured to update the number of recursive times when the calibration table is updated according to the third measured value and the standard value.

In an embodiment, the processor is further configured to calculate an error value according to the first measured value and the standard value, or calculate the error value according to the third measured value and the standard value. The processor is further configured to determine whether the error value is less than an error threshold. If the error value is less than the error threshold, the processor is further configured to determine that the recursive condition has been met.

In an embodiment, the first measured value includes a minimum measured value and a maximum measured value. The processor is further configured to set a first output value corresponding to the first index that is less than the minimum measured value as a first preset value, and set a first output value corresponding to the first index that is greater than the maximum measured value as a first Two predicted values, and the remaining first output values are generated by an interpolation algorithm.

In an embodiment, the first index includes a second index, and the second index corresponds to a second output value of the first output value. The processor is further configured to establish a temporary correction table, the temporary correction table includes a plurality of third output values, and each of the third output values corresponds to a third index. The processor is further configured to set a third index of the part to a third measurement value, set a third output value corresponding to the third measurement value as a standard value, and generate a third output value that does not correspond to the third measurement value, The second output value is taken as one of the third indexes to obtain a fourth output value of the third output value, and the second output value is set to the fourth output value. Therefore, the updated correction table can be generated by the second index and the second output value.

In an embodiment, the processor is further configured to calculate a sum of squared residuals between the first measured value and the standard value to obtain the error value. Alternatively, the processor calculates a sum of squared residuals between the third measured value and the standard value to obtain the error value.

Based on the above, the color correction method and the image processing apparatus proposed by the embodiments of the present invention can automatically generate a correction table and update the correction table in a recursive manner.

The above described features and advantages of the present invention will be more apparent from the following description.

[First Embodiment]

1 is a block diagram illustrating an image processing apparatus in accordance with an embodiment.

Referring to FIG. 1 , the image processing apparatus 100 includes an image capturing unit 110 , a processor 120 , and a memory 130 . For example, the image processing apparatus 100 can be implemented as a scanner, a digital camera, a digital camera, a computer, or a server.

The image capturing unit 110 is configured to acquire one or more images. For example The image capturing unit 110 includes a lens and a plurality of photosensitive elements. The charge on one photosensitive element forms a color value, and one or more color values can form one pixel. However, in another embodiment, the image capturing unit 110 can be implemented as a communication interface for acquiring images from other devices, and the present invention is not limited thereto.

The processor 120 is configured to execute one or more pieces of code to correct color values in an image. For example, the processor 120 can be a central processing unit (CPU) or an embedded microcontroller.

The memory 130 is used to store code or images. In the present embodiment, the color correction system 132 is stored in the memory 130. The color correction system 132 includes a plurality of modules, each of which provides a particular function. The processor 120 executes the various modules in the color correction system 132.

2 is a schematic diagram illustrating a color correction system in accordance with an embodiment.

Referring to FIG. 2 , the color correction system 132 includes a capture module 210 , a calibration module 220 , and a recursive module 230 .

The processor 120 executes the capture module 210 to control the image capture unit 110 and obtain one or more images. The processor 120 executes the correction module 220 to generate a correction table based on the acquired image. The processor 120 executes the recursive module to determine if a recursive condition is satisfied. If the recursive condition is not met, the processor 120 continues to execute the capture module 210 to obtain another image, and the correction module 220 is executed according to the other image to update the calibration table. Here, when the operations of the capture module 210, the correction module 220, and the recursive module 230 are described, it is indicated that the processor 120 executes these modules. The operation is not repeated here.

3A and 3B are schematic diagrams illustrating an example of a guide table according to an embodiment.

Referring to FIG. 3A and FIG. 3B, the guide table 300 includes color blocks 310(1) to 310(A). The colors in one patch are the same, and one patch corresponds to a standard value. For example, color block 310(1) corresponds to a standard value of "0", color block 310(2) corresponds to a standard value of "11", and so on. The guide table 300 is used to provide standard color blocks 310(1)-310(A), so that the image processing apparatus 100 can generate a correction table according to the color blocks 310(1)-310(A) and the corresponding standard values. .

In this embodiment, each standard value is a numerical value indicating brightness. However, in other embodiments, each standard value may also be a value representing chroma (eg, red, green, blue). Alternatively, each block may correspond to other standard values, and the present invention is not limited thereto.

In order to generate the calibration table, the capture module 210 obtains an image of the guide 300 and obtains a plurality of measurements (also referred to as first measurements) based on the image. For example, the capture module 210 obtains color values of a plurality of pixels in the region of the corresponding color block 310(1) in the image. The capture module 210 calculates an average of these color values as the measured value "2" of the corresponding color block 310(1). Similarly, the capture module 210 also obtains color values of a plurality of pixels in the region of the corresponding color block 310(2) in the image, and calculates an average of the color values as a measure of the corresponding color block 310(2). The value is "4", and so on.

However, due to some hardware errors or noise, the measured values corresponding to the same color block are not necessarily the same as the standard values. For example, measurement The value "2" and the standard value "0" are all corresponding to the color block 310 (1), but the error between them is 2; the measured value "4" and the standard value "11" correspond to the color block 310 (2) ), but the error is 7. The calibration module 220 generates a calibration table based on the measured values and standard values shown in FIG. 3B.

4 and 5 are schematic diagrams illustrating the creation of a correction table in accordance with an embodiment.

Referring to FIG. 4, the calibration table 400 includes output values 410(0)-410(255), and the output values 410(0)-410(255) correspond to indices 420(0)-420(255), respectively. The operation of the correction table 400 is such that when a value is input, the correction module 220 will use this value as an index of the correction table 400 and output a corresponding output value. For example, if the value entered is "2", then this value is treated as index 420(2) and the corresponding output value 410(2) is output.

When the lookup table 400 is created, the calibration module sets the partial index 420(0)~420(255) to the measured value shown in FIG. 3B, and sets the output value corresponding to the measured values to the above-mentioned standard value. . For example, the standard value “0” corresponds to the measured value “2”, so the correction module 220 sets the index 420(2) to the measured value “2” and sets the output value 410(2) to the standard value. "0". The standard value "11" corresponds to the amount side value "4", so the correction module 220 sets the index 420(4) to the measured value "4", and the output value 410(4) corresponding to the measured value is Set to the standard value "11". The standard value "22" corresponds to the measured value "5", so the correction module 220 sets the index 420(5) to the measured value "5" and sets the output value 410(5) to the standard value "22". In this embodiment, the number of standard values is 24, so there are only 24 output values (for example, the output value 410(2), 410(4), 410(5), and 410(204)) will be set to the corresponding standard values. The correction module 220 also produces an output value that does not correspond to the measured value. Specifically, the calibration module 220 first obtains the smallest one of the measured values (also known as the minimum measured value, that is, the measured value 2"), and obtains the largest one (also known as the maximum measured value, ie, Measurement value "204"). The correction module 220 sets the output values 410(0) and 410(1) corresponding to the indexes 420(0)~420(1) that are smaller than the minimum measurement value to be a first preset value. The correction module 220 also sets the output values 410(205)~410(255) corresponding to the indexes 420(205)~420(255) that are greater than the maximum measurement value to a second preset value. Moreover, the correction module 220 generates the remaining output values (eg, the output value 410(3)) with an interpolation algorithm.

Referring to FIG. 5, in this embodiment, the correction module 220 sets the first preset value to 0, and sets the second preset value to 255. In other words, the output values 410(0) and 410(1) will be set to "0"; and the output values 410(205)~410(255) will be set to "255". The correction module 220 performs an interpolation algorithm based on the output values 410(2) and 410(4) to generate an output value 410(3). For example, correction module 220 calculates an average of output values 410(2) and 410(4) to produce output value 410(3).

However, in other embodiments, the correction module 220 can also perform an interpolation algorithm using a low pass filter. Alternatively, the correction module 220 may generate an exponential function or a polynomial function based on the output values 410(2), 410(4), 410(5), 410(204), and the like. The correction module 220 can calculate the output value 410(3) according to the generated function, and the present invention is not limited thereto. In addition, when a pixel's color value is more bits (ie, greater than 8 bits) The correction table 400 will include more indices and output values when represented. Moreover, the first preset value and the second preset value may be set to different values, and the present invention is not limited thereto.

After calculating the correction table 400, the recursive module 230 determines whether a recursive condition has been met. If the recursion condition has been met, the correction module 220 outputs a correction table 400. Thereafter, after the capture module 210 obtains an image, the calibration module 220 can adjust the color value in the image according to the calibration table 400. However, when the recursive condition is not satisfied, the correction module 220 will continuously update the correction table 400 until the recursive condition has been met.

In the process of updating the calibration table 400, the capture module 210 obtains another image (also referred to as a second image) of the guide table 300 and obtains a plurality of measured values (also referred to as second measured values) according to the image. . The calibration module 220 adjusts the second measurements according to the current calibration table 400 to generate a plurality of measurements (also referred to as third measurements). The calibration module 220 updates the calibration table 400 based on the standard values and the third measurements until the recursive module 230 determines that the recursive condition has been met.

6 through 9 are schematic diagrams illustrating an update correction table in accordance with an embodiment.

Referring to FIG. 6 , corresponding to the color blocks 310(1) to 310(3) and 310(A), the second measurement values obtained by the capture module 210 are “0”, “3”, “5”, respectively. ..."204". The calibration module 220 adjusts the second measurements according to the calibration table 400 to generate third measurements "0", "6", "22" ... "255". For example, the correction module 220 will be "3" As an index of the correction table 400, an output value "6" is obtained; the correction module 220 uses "5" as an index of the correction table 400, and an output value "22" can be obtained. It should be noted that the capture module 210 does not necessarily have the same measurement value for the same color patch twice. For example, for the color block 310(2), the first measurement obtained by the capture module 210 is "4" (as shown in FIG. 3B), but the measured value obtained for the second time is "3". .

The correction module 220 creates a temporary correction table in accordance with the manner in which the calibration table 400 is established as described above. Specifically, the calibration module 220 sets the index of the portion of the temporary correction table to the third measurement value, sets the output value corresponding to the third measurement values as a standard value, and generates a third measurement that does not correspond to the third measurement. The output value of the value.

Referring to FIG. 7, the temporary correction table 700 includes output values 710(0)~710(255), and the output values 710(0)~710(255) correspond to indexes 720(0)~720(255), respectively. The correction module 220 sets the indexes 720(0), 720(6), 720(22), and 720(255) to the third measurement values "0", "6", "22", and "255", respectively. The output values 710 (0), 710 (6), 710 (22), and 710 (255) are the corresponding standard values "0", "11", "22", and "255".

Referring to FIG. 8 , the calibration module 220 calculates output values 710(1) to 710(5) and 710(11) that are not corresponding to the third measurement value according to an interpolation algorithm. For example, the correction module 220 produces an output value 710(11) with a value of "14" in a linear interpolation manner.

The correction module 220 combines the temporary correction table 700 with the correction table 400, thereby updating the correction table 400. Specifically, if the correction table 400 includes a second index, and the second index corresponds to one of the correction tables 400 Second output value. The correction module 220 will use this second output value as an index of the temporary correction table 700 to obtain a fourth output value in the temporary correction table. The correction module 220 replaces the second output value with the fourth output value. For example, referring to FIG. 4 and FIG. 8, index 420(4) corresponds to output value 410(4), which has a value of "11". The correction module 220 will use "11" as the index 720 (11) of the temporary correction table 700, and obtain an output value 710 (11) having a value of "14". Next, the correction module 220 sets the output value 410(4) to the output value 710(11) (as shown in FIG. 9).

From another perspective, the correction table 400 and the temporary correction table 700 can be represented by the following programs (3) and (4).

M _n ={ m _{n, 0} , m _{n ,1} ,..., g _{n ,255} }.........(3)

G _n ={ g _{n ,0} , g _{n ,1} ,..., g _{n ,255} }.........(4)

Wherein, M _n represents the n-th correction table update time 400, n is a positive integer. m _n,0 represents the 0th output value 410(0) in the correction table 400 at the nth update, and so on. G _n denotes the temporary correction table 700 is generated by the n-th update and G _{n, 0} represents the n-th update of the temporary correction value table 700 outputs 0 710 (0), and so on. The steps of synthesizing the correction table 400 and the temporary correction table 700 can be expressed by the following programs (5) to (7).

M _{n +1} ={ m _{n +1,0} , m _{n +1,1} ,..., m _{n +1,255} }...(5)

m _{n +1, j} = g _{n , t} .........(6)

t = m _{n , j} .........(7)

In other words, the jth output value m _{n,j in} the nth updated correction table 400 is treated as the index value t. The correction module 220 finds the output value g _{n,t of the} temporary correction table 700 based on the index value _t . The output value g _n,t is taken as the jth output value m _n+1,j in the correction table 400 at the time of the n+1th update.

The steps of updating the calibration table 400 described above are repeated until the recursive module 230 determines that the recursive condition has been met. In an embodiment, the recursive module 230 determines whether a number of recursive times meets a recursive threshold, and if so, determines that the recursive condition has been met. Each time the correction table 400 is updated, the number of recursions is updated. For example, the initial value of the number of recursive times is 0, and the number of recursions is incremented by one each time the correction table 400 is updated, and the recursion threshold is 3. When the number of recursive times is greater than or equal to 3, the recursive module 230 determines that the number of recursive times has met the recursion threshold. In other words, if the correction table 400 is updated three times, the recursive condition is satisfied. However, in other embodiments, the initial value of the number of recursive times is three, and the number of recursive times is reduced by one each time the correction table 400 is updated, the recursion threshold is 0, and when the number of recursive times is less than or equal to 0, The return module 230 determines that the number of recursions meets the recursion threshold. However, the recursive threshold may also be set to 5, 10 or other values, and the invention is not limited thereto.

In one embodiment, after generating or updating the calibration table 400, the recursive module 230 calculates an error value between the measured value and the standard value. If the error value is less than an error threshold, the recursive module 230 determines that the recursive condition has been met. For example, the recursive module 230 calculates an error value according to the following equation (8).

Here, R _n is an error value when the correction table 400 is updated for the nth time. j and k are integers. k represents the number of color patches in the guide table 300. i _n,j represents the measured value corresponding to the jth color patch when the correction table 400 is updated for the nth time. S _j represents the standard value corresponding to the jth color patch. It is worth noting that when calculating R ₁ , i _1,j is not adjusted by the correction table (ie, i _1,j is the first measurement). When calculating R ₂ , i _2,j will be adjusted by the correction table M ₁ (ie, i _2,j is the third measurement value, not the second measurement value). Each time the correction table M _{n is} generated, the recursive module 230 calculates and determines whether the error value R _n is less than an error threshold. If the error is smaller than this threshold error value R _n, the recursive determination module 230 will be handed back condition is met, and outputs a correction table M _n.

What is calculated by the above equation (8) is the sum of squared error. However, in other embodiments, the recursive module 230 may also calculate the error value using a mean squared error, or a sum of absolute error, which is not limited herein.

In this embodiment, when the error value is less than the error threshold or the number of recursive times meets the recursion threshold, the recursive module 230 determines that the recursion condition has been satisfied. However, in other embodiments, the recursive module 230 may also determine that the recursion condition has been satisfied when the error value is less than the error threshold and the number of recursive times meets the recursion threshold.

FIG. 10 is a flow chart illustrating a color correction method in accordance with an embodiment.

Referring to FIG. 10, in step S1002, the capture module 210 obtains the first image of the guide. This guide table includes a plurality of color patches, and each color patch corresponds to a standard value.

In step S1004, the capture module 210 obtains a plurality of first measurement values according to the first image, wherein each of the first measurement values corresponds to a standard value.

In step S1006, the correction module 220 generates a correction table according to the first measurement value and the standard value.

In step S1008, the recursive module 230 determines whether the recursive condition has been met. This process ends if the recursive condition is met.

If the recursive condition is not satisfied, in step S1010, the capture module 210 obtains the second image of the guide table, and obtains a plurality of second measurement values according to the second image. In step S1012, the correction module 220 adjusts the second measurement value according to the calibration table to generate a plurality of third measurement values. In step S1014, the correction module 220 updates the correction table according to the third measurement value and the standard value.

However, the steps in FIG. 10 have been described in detail above, and the detailed description thereof will not be repeated here.

[Second embodiment]

The second embodiment is similar to the first embodiment, and only the differences will be described herein. In the first embodiment, when the capture module 210 first obtains the measured value, the calibration module 220 does not use the correction table to adjust the first measurement values. However, in the second embodiment, after each capture module 210 obtains the measured value, the calibration module 220 adjusts the measured values according to the calibration table.

Figure 11 is a flow chart illustrating a color correction method according to a second embodiment.

Referring to FIG. 11, in step S1102, the correction module 220 and the recursive module 230 execute an initialization process. The correction module 220 sets an initial correction table, each of which will be equal to the corresponding index. That is, in the second embodiment, m _1,i in the above equation (3) will be equal to i. The recursive module 230 sets the recursive threshold and the error threshold.

In step S1104, the capture module 210 obtains an image of the guide table 300 and obtains a plurality of measured values of the image.

In step S1106, the correction module 220 adjusts the measured values according to the correction table. It should be noted that when the step S1106 is executed for the first time, since the output value in the correction table is equal to the corresponding index, the adjusted measured value is the same as the measured value before the adjustment.

In step S1108, the recursive module 230 calculates an error value based on the standard value and the adjusted measured value. That is, in the second embodiment, each of the measured values i _n,j in the above equation (8) is adjusted by the correction module 220 according to the correction table.

In step S1110, the recursive module 230 determines whether the recursive condition is satisfied. In the second embodiment, when the error value is less than the error threshold or the number of recursive times meets the recursion threshold, the recursive module 230 determines that the recursion condition has been satisfied. If the recursive condition has been met, the flow of Figure 11 will be ended.

If the recursive condition has not been met, in step S1112, the correction module 220 generates a temporary correction table based on the standard value and the adjusted measurement value. It should be noted that, when the step S1112 is performed for the first time, since the adjusted measured value is the same as the measured value before the adjustment, the generated temporary correction table is equivalent to the first generated in the first embodiment. Calibration table (shown in Figure 5).

In step S1114, the correction module 220 merges the temporary correction table and the correction table, and returns to step S1104. It is worth noting that when step S1114 is executed for the first time, the combined correction table will be the same as the temporary correction table. Specifically, the above equations (6) and (7) can be rewritten as the following equations (9) and (10).

t = m _{1, j} = j .........(9)

m _{2, j} = g _{1, t} = g _{1, j} .........(10)

In summary, the color correction method and the image processing apparatus proposed by the present invention can generate a correction table in a recursive manner. Therefore, the error between the measured value and the standard value will gradually decrease and converge.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the invention, and any one of ordinary skill in the art can make some modifications and refinements without departing from the spirit and scope of the invention. The scope of the invention is defined by the scope of the appended claims.

100‧‧‧Image processing device

110‧‧‧Image capture unit

120‧‧‧ processor

130‧‧‧ memory

132‧‧‧Color Correction System

210‧‧‧Capture module

220‧‧‧ calibration module

230‧‧‧Return module

300‧‧‧Guide

310(1)~310(A)‧‧‧Color blocks

400‧‧‧ calibration table

410(0)~410(255)‧‧‧ Output value

420 (0) ~ 420 (255) ‧ ‧ index

700‧‧‧Interim correction table

710(0)~710(255)‧‧‧ output value

720 (0) ~ 720 (255) ‧ ‧ index

S1002, S1004, S1006, S1008, S1010, S1012, S1014, S1102, S1104, S1106, S1108, S1110, S1112, S1114‧‧ ‧ steps of color correction method

1 is a block diagram illustrating an image processing apparatus in accordance with an embodiment.

2 is a schematic diagram illustrating a color correction system in accordance with an embodiment.

3A and 3B are schematic diagrams illustrating an example of a guide table according to an embodiment.

4 and 5 are schematic diagrams illustrating the creation of a correction table in accordance with an embodiment.

6 through 9 are schematic diagrams illustrating an update correction table in accordance with an embodiment.

FIG. 10 is a flow chart illustrating a color correction method in accordance with an embodiment.

Figure 11 is a flow chart illustrating a color correction method according to a second embodiment.

S1002, S1004, S1006, S1008, S1010, S1012, S1014‧‧‧ steps of color correction method

Claims (14)

A color correction method for an image processing device, the color correction method comprising: obtaining a first image of a guide table, wherein the guide table comprises a plurality of color blocks, and each of the color blocks corresponds to a standard value; Obtaining, according to the first image, a plurality of first measurement values, wherein each of the first measurement values corresponds to one of the standard values; and generating a correction according to the first measurement values and the standard values Determining whether a recursive condition is satisfied; and if the recursive condition is not satisfied, obtaining a second image of the navigation table, obtaining a plurality of second measurement values according to the second image, and adjusting the plurality of measurements according to the calibration table The second measured value is used to generate a plurality of third measured values, and the calibration table is updated according to the third measured values and the standard values until the recursive condition is satisfied. The color correction method of claim 1, wherein the correction table includes a plurality of first output values, each of the first output values corresponding to a first index, wherein the first measurement value is based on the first measurement value The step of generating the calibration table includes: setting the first index of the portion to the first measurement values; and setting the first output values corresponding to the first measurement values to the criteria a value; and generating the first output values that do not correspond to the first measurements. The color correction method of claim 1, wherein the determining whether the recursive condition is satisfied comprises: determining whether a number of recursions meets a recursion threshold; If the number of recursive times meets the recursive threshold, it is determined that the recursive condition is satisfied, and the step of updating the correction table according to the third measured values and the standard values further comprises: updating the number of recursive times. The color correction method of claim 1, wherein the step of determining whether the recursive condition is satisfied comprises: calculating an error value according to the first measurement value and the standard value, or according to the third The measured value and the standard value are used to calculate the error value; determining whether the error value is less than an error threshold; and if the error value is less than the error threshold, determining that the recursion condition is satisfied. The color correction method of claim 2, wherein the first measurement values comprise a minimum measurement value and a maximum measurement value, wherein the first measurement values are generated corresponding to the first measurement values. The step of outputting the value includes: setting the first output values corresponding to the first indexes that are smaller than the minimum measurement value to a first preset value; setting the first indexes that are greater than the maximum measurement value The corresponding first output values are a second preset value; and the remaining first output values are generated by an interpolation algorithm. The color correction method of claim 2, wherein the first index comprises a second index, and the second index corresponds to a second output value of the first output values, wherein Third measurement The step of updating the correction table with the value and the standard value includes: establishing a temporary correction table, wherein the temporary correction table includes a plurality of third output values, and each of the third output values corresponds to a third index; The third index of the portion is the third measurement value; setting the third output values corresponding to the third measurement values to the standard values; generating the third measurement values not corresponding to the third measurement values The third output value; the second output value being one of the third indexes to obtain a fourth output value of the third output values; and setting the second output value to the fourth output value. The color correction method of claim 4, wherein the step of calculating an error value according to the first measurement values and the standard values comprises: calculating the first measurement values and the standard values. The sum of the residuals is obtained by the sum of the residuals, and the step of calculating the error value according to the third measured values and the standard values includes: calculating between the third measured values and the standard values The sum of the residuals is squared to obtain the error value. An image processing device includes: an image capturing unit, configured to obtain a first image of a guide table, wherein the guide table includes a plurality of color blocks, and each of the color blocks corresponds to a standard value; and a process And coupled to the image capturing unit, according to the first Obtaining a plurality of first measurement values, wherein each of the first measurement values corresponds to one of the standard values, wherein the processor is configured to generate the standard measurement values according to the first measurement values and the standard values. a calibration table, and determining whether a recursion condition is satisfied, wherein if the recursive condition is not satisfied, the processor is configured to obtain a second image of the navigation table, and obtain a plurality of second measurement values according to the second image And adjusting the second measurement values according to the calibration table to generate a plurality of third measurement values, and updating the correction table according to the third measurement values and the standard values until the recursion condition is satisfied. The image processing device of claim 8, wherein the correction table includes a plurality of first output values, and each of the first output values corresponds to a first index, wherein the processor is further configured to set The first index of the part is the first measurement value, the first output values corresponding to the first measurement values are set to the standard values, and the first measurement is not corresponding to the first measurement The first output values of the values. The image processing device of claim 8, wherein the processor is further configured to determine whether a number of recursive times meets a recursive threshold, and if the number of recursive times meets the recursive threshold, the processor further The processor is further configured to: when the update table is updated according to the third measurement values and the standard values, the number of recursive times is updated. An image processing apparatus according to claim 8, wherein The processor is further configured to calculate an error value according to the first measured values and the standard values, or calculate the error value according to the third measured values and the standard values, wherein the processor further The processor is further configured to determine whether the error value is less than an error threshold, and if the error value is less than the error threshold, the processor is further configured to determine that the recursion condition is satisfied. The image processing device of claim 9, wherein the first measurement value comprises a minimum measurement value and a maximum measurement value, and the processor is further configured to set the smaller than the minimum measurement value. The first output values corresponding to the first indexes are a first preset value, and the first output values corresponding to the first indexes greater than the maximum measured value are set to a second predicted value. And the remaining first output values are generated by an interpolation algorithm. The image processing device of claim 9, wherein the first index comprises a second index, and the second index corresponds to a second output value of the first output values, wherein the processing The processor is further configured to establish a temporary correction table, wherein the temporary correction table includes a plurality of third output values, and each of the third output values corresponds to a third index, wherein the processor is further configured to set a portion The third index is the third measurement value, and the third output values corresponding to the third measurement values are set to the standard values, and the third measurement values are not corresponding to the third measurement values. a third output value, the second output value being one of the third indexes to obtain a fourth output value of the third output values, and setting the second output value to the fourth output value . The image processing device of claim 11, wherein the processor is further configured to calculate a sum of squared residuals between the first measured values and the standard values to obtain the error value, or Calculating a sum of squared residuals between the third measured values and the standard values to obtain the error value.