A method and system for predicting raw response values for a digital camera based on illumination interpolation

Abstract

The application is a kind of digital camera raw response value prediction method and system based on illumination interpolation, belonging to the technical field of computer digital image processing. First, the real raw response value of the test camera under different illuminations is measured under the reference light source, the predicted raw response value under the reference light source is predicted using the three-element imaging model and normalized. Then, the weight matrix between the predicted raw response value under each illumination and the real raw response value of the camera shooting is calculated, the weight matrix is linearly interpolated according to the illumination and constructed into a lookup table. Finally, the corresponding weight of the test light source in the lookup table is used to weight the predicted raw response value under the test light source, and the camera shooting raw response value under the test light source is obtained. The application effectively solves the scale inconsistency problem between the camera raw response value prediction value and the real camera shooting raw response value, effectively solves the problem of digital camera raw response value prediction and its application in open light source environment.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of computer digital image processing, and particularly relates to a raw response value prediction method and system for a digital camera based on illumination interpolation. BACKGROUND

[0002] The raw response value of a digital camera represents the most original image data, which refers to the digital value obtained by analog-to-digital conversion of the electrical signal generated by the camera sensor in response to the scene light signal without any image processing algorithm (such as white balance correction, color correction, gamma correction, noise reduction processing, etc.).

[0003] Many existing technologies need to use a digital camera, such as multispectral reconstruction, digital camera color characterization, color correction, and light source estimation. The methods used in these technologies mostly rely on shooting a color card by a digital camera, and then constructing a corresponding correction model, but in actual use, it is not possible to shoot a color card in any scene using a digital camera, and it is not possible to realize real-time construction of a correction model based on a color card in a scene. If the raw response value of a digital camera shooting a color card in any scene in the real world can be predicted in real time, and a related training model is constructed based on the raw response value, real-time training of a model based on a color card in any scene can be effectively realized. However, the current raw response value prediction model cannot be effectively applied in actual application, and the main reason is that in actual imaging, the imaging model based on the three elements of light source, object spectral reflectance, and camera sensitivity function does not consider factors such as scene illumination and camera parameters, resulting in a floating value phenomenon of the synthesized raw response value, that is, the data scale of the raw response value predicted by the imaging model is inconsistent with the raw response value actually shot by the camera. Normalization processing is generally used to realize scale consistency, but this processing method can only have good performance in a high dynamic range of scene color, and in a low-illumination environment, the normalization operation will destroy the data scale of the real scene, resulting in a decrease in the effect of the constructed correction model. In view of the above-mentioned problems, the academic and industrial circles at home and abroad have not proposed a good solution. SUMMARY

[0004] The present application belongs to the technical field of computer digital image processing, and particularly relates to a raw response value prediction method and system for a digital camera based on illumination interpolation.

[0005] To solve the problems in the above research, the application provides a method for solving the problems, first, the real raw response value of the test camera under different illuminations is measured under the reference light source, the three-element imaging model is used to predict the predicted raw response value under the reference light source and the normalization processing is carried out. Then, the weight matrix between the predicted raw response value under each illumination and the real raw response value of the camera shooting is calculated, the weight matrix is linearly interpolated according to the illumination and is constructed into a lookup table. Finally, the weight corresponding to the illumination in the lookup table is used to weight the predicted raw response value under the test light source, and the camera shooting raw response value under the test light source is obtained.

[0006] The technical scheme of the application is a digital camera raw response value prediction method based on illumination interpolation, specifically including the following steps:

[0007] Step 1, the relative spectral power distribution curve SPD of the reference light source is measured by using an illuminometer and a monochromator ref And the spectral response function S of the digital camera;

[0008] Step 2, the spectral reflectance R of the standard color card is measured by using a spectrophotometer;

[0009] Step 3, the raw response value D of the color card under multiple illuminations of the reference light source is obtained by using a digital camera shooting i , and the light source illumination value k i under the corresponding shooting scene is obtained by using an illuminometer;

[0010] Step 4, the predicted raw response value of the standard color card under the reference light source is calculated, and the raw response value is normalized D pre ;

[0011] Step 5, the weight w i between the camera shooting raw response value and the predicted raw response value under each illumination is calculated;

[0012] Step 6, the w i is linearly interpolated according to the illumination gradient to obtain the illumination weight lookup table LUT illum ;

[0013] Step 7, the relative spectral power distribution curve SPD test and the scene illumination k test of the test light source are obtained by using an illuminometer;

[0014] Step 8, the initial predicted raw response value of the reference color card under the test light source is calculated, and the raw response value is normalized D test ;

[0015] Step 9, the illumination weight w corresponding to the test light source illumination is found out by using the illumination weight lookup tableillum ;

[0016] Step 10, calculate the difference weight w between the relative spectral power distribution of the test light source and the reference light source spd ;

[0017] Step 11, calculate the final predicted raw response value D of the standard color card under the test light source by weighting the initial predicted raw response value with the difference weight and the illumination weight final。

[0018] Further, in step 4, the predicted raw response value of the standard color card under the reference light source is calculated, and the response value is normalized, the method is as follows:

[0019]

[0020] In the formula, S represents the spectral response function matrix of the digital camera, SPD ref represents the relative spectral power distribution vector of the reference light source, diag() represents the conversion of the vector to a diagonal matrix, R represents the spectral reflectance matrix of the standard color card, D w represents the predicted raw response value of the standard white plate with spectral reflectance of 1 under the reference light source, D pre represents the normalized predicted raw response value of the standard color card under the reference light source.

[0021] Further, in step 5, the weight between the raw format response value and the predicted raw response value of the camera under each illumination is calculated, and the specific method is as follows:

[0022]

[0023] In the formula, D i represents the raw format response value of the camera under the i th illumination, w i represents the weight between the raw format response value and the predicted raw response value of the camera under the i th illumination.

[0024] Further, in step 6, the illumination weight lookup table LUT illum is obtained by linear interpolation of w i according to the illumination gradient change, and the linear interpolation method is as follows:

[0025]

[0026]

[0027]

[0028] In the formula, k new is the illumination value that needs to be interpolated, ki is the illuminance value of the reference light source under the ith illuminance, satisfying k i ≤ k new ≤ k i+1 , t is an interpolation coefficient, w i is the weight between the raw format response value and the predicted raw response value of the camera under the ith illuminance, w new is the raw response value of the camera under the ith illuminance, and the predicted raw response value is obtained by interpolation calculation, LUT new is the weight between the raw format response value and the predicted raw response value of the camera under the illuminance k illum , and the constructed illuminance weight lookup table, wherein each row represents an illuminance k i and the weight w i between the raw format response value and the predicted raw response value of the camera under the corresponding illuminance.

[0029] Further, the initial predicted raw response value D test of the reference color card under the test light source in step 8 is calculated, and the specific method is as follows:

[0030]

[0031] In the formula, SPD test represents the relative spectral power distribution matrix of the test light source, D tw represents the predicted raw response value of the standard white plate with spectral reflectance of 1 under the test light source, D test represents the normalized initial predicted raw response value of the standard color card under the test light source.

[0032] Further, the lookup table LUT illum is used to find out the corresponding illuminance weight w test of the test light source illuminance k illum ;

[0033] Further, in step 10, the difference weight between the relative spectral power distribution of the test light source and the reference light source is calculated, and the specific calculation method is as follows:

[0034]

[0035] In the formula, mean() is an average function, w spd is the difference weight between the relative spectral power distribution of the test light source and the reference light source.

[0036] Further, in step 11, the initial predicted raw response value D test is weighted and calculated by using the difference weight w spd and the illuminance weight w illum , to obtain the final predicted raw response value of the standard color card under the test light source, and the specific calculation method is as follows:

[0037]

[0038] In the formula, D test is the initial predicted raw response value under the test light source, w illum is the final predicted raw response value of the standard color card under the test light source. illum is the illuminance k test corresponding to the illuminance weight, w spd is the calculated difference weight, D final is the final predicted raw response value of the standard color card under the test light source.

[0039] The application also provides a digital camera raw response value prediction system based on illuminance interpolation, comprising a processor and a memory, the memory being used to store program instructions, and the processor being used to call the stored instructions in the memory to execute a digital camera raw response value prediction method based on illuminance interpolation as described in the above technical solution.

[0040] The application aims to solve the problem of scale inconsistency between the predicted value of the raw response value of the digital camera and the actual shooting raw response value of the camera. Based on a three-element imaging model, the raw response value of the standard color card is first obtained by shooting the standard color card under multiple illuminances using the digital camera under a reference light source, and then a "illuminance-weight" lookup table under the reference light source is constructed using a linear interpolation algorithm. Under the test light source environment, only the test environment illuminance is used to perform lookup table, and the initial predicted raw response value is weighted by the weight corresponding to the illuminance, which can effectively solve the problem of scale inconsistency between the predicted value of the raw response value of the camera and the actual shooting raw response value of the camera, and effectively solve the problem of raw response value prediction of the digital camera in an open light source environment and its practical application. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 is the flowchart of the embodiment of the application.

[0042] Figure 2 is the relative spectral power distribution curve of the reference light source and the test light source in the embodiment of the application.

[0043] Figure 3 is the distribution diagram of the predicted raw response value and the actual shooting response value RGB of the camera under the 1144 lux illuminance in the embodiment of the application.

[0044] Figure 4 is the distribution diagram of the predicted raw response value and the actual shooting response value RGB of the camera under the 1240 lux illuminance in the embodiment of the application.

[0045] Figure 5This is a distribution diagram of the three RGB channels of the predicted raw response value and the actual camera shooting response value under an illumination of 1411 lux in an embodiment of the present invention, where lux is the illumination unit lux. DETAILED DESCRIPTION

[0046] When the technical solution of the present invention is specifically implemented, those skilled in the art can use computer software technology to run it.

[0047] like Figure 1 As shown, the present invention provides a method for predicting the raw response value of a digital camera based on illumination interpolation. Specifically, the method comprises the following steps:

[0048] Step 1: Use an illuminometer and a monochromator to measure the relative spectral power distribution curve SPD of the reference light source. ref and the digital camera spectral response function S;

[0049] Step 2, using a spectrophotometer to measure the spectral reflectance R of the standard color card;

[0050] Step 3: Use a digital camera to capture the raw response value D of the color card under multiple illuminations of the reference light source. i At the same time, the illuminance value k of the light source in the corresponding shooting scene is obtained using the illuminance meter i ;

[0051] Step 4: Calculate the predicted raw response value of the standard color card under the reference light source and normalize the raw response value D pre ;

[0052] In step 4, the predicted raw response value of the standard color card under the reference light source is calculated and the raw response value is normalized. The method is shown in formula (1):

[0053] (1)

[0054] Where S represents the spectral response function matrix of the digital camera, SPD ref Represents the relative spectral power distribution vector of the reference light source, diag() represents converting the vector into a diagonal matrix, R represents the spectral reflectance matrix of the standard color card, D w Represents the predicted raw response value of a standard white plate with a spectral reflectance of 1 under the reference light source, D pre Represents the predicted raw response value of the standard color card normalized under the reference light source.

[0055] Step 5: Calculate the weight w between the raw response value captured by the camera and the predicted raw response value under each illumination i ;

[0056] The weight between the raw response value and the predicted raw response value of the camera under each illumination is calculated in step 5, and the specific method is as follows:

[0057] (2)

[0058] In the formula, D i represents the raw response value of the camera under the ith illumination, w i represents the weight between the raw response value and the predicted raw response value of the camera under the ith illumination.

[0059] Step 6, linear interpolation is performed on w i according to the illumination gradient to obtain an illumination weight lookup table LUT illum .

[0060] Step 6, linear interpolation is performed on w i according to the illumination gradient to obtain an illumination weight lookup table LUT illum , and the linear interpolation method is as follows:

[0061] (3)

[0062] (4)

[0063] (5)

[0064] In the formula, k new is the illumination value that needs to be interpolated, k i is the illumination value of the reference light source under the ith illumination, which satisfies k i ≤ k new ≤ k i+1 , t is the interpolation coefficient, w i is the weight between the raw response value and the predicted raw response value of the camera under the ith illumination, w new is the weight between the raw response value and the predicted raw response value of the camera under the illumination k new calculated by interpolation, and LUT illum is the constructed illumination weight lookup table, wherein each row represents an illumination k i and the weight w i between the raw response value and the predicted raw response value of the camera under the corresponding illumination.

[0065] Step 7, the relative spectral power distribution curve SPD test of the test light source and the scene illumination k test are obtained by using an illuminometer.

[0066] Step 8, calculate the initial predicted raw response value of the reference color chart under the test light source, and normalize the raw response value D test ;

[0067] Step 8, calculate the initial predicted raw response value of the reference color chart under the test light source D test , the specific method is as follows:

[0068] (6)

[0069] In the formula, SPD test represents the relative spectral power distribution matrix of the test light source, D tw represents the predicted raw response value of the standard white plate with spectral reflectance of 1 under the test light source, D test represents the normalized initial predicted raw response value of the standard color chart under the test light source.

[0070] Step 9, find the illumination weight w illum corresponding to the test light source illumination using the illumination weight lookup table.

[0071] Step 10, calculate the difference weight w spd between the relative spectral power distributions of the test light source and the reference light source.

[0072] The difference weight between the relative spectral power distributions of the test light source and the reference light source is calculated in step 10, and the specific calculation method is as follows:

[0073] (7)

[0074] In the formula, mean() is the average function, w spd is the difference weight between the relative spectral power distributions of the test light source and the reference light source.

[0075] Step 11, use the difference weight and the illumination weight to weight the initial predicted raw response value to obtain the final predicted raw response value D final of the standard color chart under the test light source.

[0076] In step 11, the initial predicted raw response value D spd is weighted and calculated using the difference weight w illum and the illumination weight w test to obtain the final predicted raw response value of the standard color chart under the test light source, and the specific calculation method is as follows:

[0077] (8)

[0078] In the formula, D test is the initial predicted raw response value under the test light source calculated by formula (6), willum For step 9, a look-up table (LUT) is used illum The test illuminant illuminance k is looked up test The corresponding illuminant weight, w spd The difference weight, D, calculated for equation (7) final The final predicted raw response value of the standard color chart under the test illuminant.

[0079] The following is described with a specific example, and the embodiment of the present application is described as follows:

[0080] The embodiment uses the Nikon D7200 digital camera sensitivity function, LEDCube adjustable LED light source light box, international standard ColorCheck140 color chart, and the reference illuminant illuminance and spectral power distribution are measured using SPIC-300;

[0081] In the embodiment, the reference illuminant illuminance and spectral power distribution are measured using SPIC-300 (the reference illuminant relative spectral power distribution curve is shown by the blue curve in the accompanying drawings), and the standard color chart spectral reflectance is measured by the i1-pro spectrophotometer. The predicted raw response value of the standard color chart under the reference illuminant is calculated, and the raw response value is normalized, and the method is shown as equation (1): Figure 2

[0082] (1)

[0083] In the equation, S represents the spectral response function matrix of the digital camera Nikon D7200, SPD ref represents the relative spectral power distribution vector of the reference illuminant, 'diag()' represents converting the vector into a diagonal matrix, R represents the spectral reflectance matrix of the standard color chart, D w represents the predicted raw response value of the standard white plate with a spectral reflectance of 1 under the reference illuminant, D pre represents the normalized predicted raw response value of the standard color chart under the reference illuminant.

[0084] The predicted raw response value D pre of the standard color chart under the reference illuminant is obtained, and then the raw response value of the standard color chart under different illuminances of the reference illuminant is photographed using the digital camera Nikon D7200, and the weight between the raw response value photographed by the camera and the predicted raw response value under each illuminance is calculated, and the specific method is as follows:

[0085] (2)

[0086] In the equation, D i represents the raw response value photographed by the camera under the i-th illuminance, w i ​The weight between the raw response value and the predicted raw response value taken by the i-th illuminance camera.

[0087] After obtaining the weight under the known illuminance, the weight w i is linearly interpolated according to the illuminance gradient change to obtain an illuminance weight lookup table LUT illum , and the linear interpolation method is as follows:

[0088] (3)

[0089] (4)

[0090] (5)

[0091] In the formula, k new is the illuminance value that needs to be interpolated, k i is the illuminance value of the reference light source under the i-th illuminance, k i ≤ k new ≤ k i+1 , t is the interpolation coefficient, w i is the weight between the raw response value and the predicted raw response value taken by the i-th camera under the illuminance, w new is the weight between the raw response value and the predicted raw response value taken by the camera when the illuminance is k new , and LUT illum is the constructed illuminance weight lookup table, wherein each row represents an illuminance k i and the weight w i between the raw response value and the predicted raw response value taken by the camera under the corresponding illuminance.

[0092] After obtaining the relative spectral power distribution curve SPD test of the test light source and the scene illuminance k test , the initial predicted raw response value D test of the reference color card under the test light source is calculated (the relative spectral power distribution curve of the test light source is shown as a red curve in the accompanying drawings Figure 2 ), and the specific method is as follows:

[0093] (6)

[0094] In the formula, SPD test represents the relative spectral power distribution matrix of the test light source, D tw represents the predicted raw response value of the standard white plate with a spectral reflectance of 1 under the test light source, and D test represents the normalized initial predicted raw response value of the standard color card under the test light source.

[0095] Then, the illumination weight w test corresponding to the illumination k illum of the test light source is looked up from the illumination weight lookup table.

[0096] (7)

[0097] where mean() is the mean function, w spd is the difference weight between the relative spectral power distribution of the test light source and the reference light source.

[0098] Finally, the initial predicted raw response value D test is weighted calculated by the difference weight w spd and the illumination weight w illum to obtain the final predicted raw response value of the standard color card under the test light source, and the specific calculation method is as follows:

[0099] (8)

[0100] where D test is the initial predicted raw response value of the test light source calculated by formula (6), w illum is the illumination weight corresponding to the illumination k illum of the test light source looked up from the lookup table LUT test by step 9, w spd is the difference weight calculated by formula (7), and D final is the final predicted raw response value of the standard color card under the test light source.

[0101] In the example, the root-mean-square error (RMSE) is used as the evaluation index of the raw response value prediction, and the specific calculation method of the root-mean-square error is as follows:

[0102] (9)

[0103] where N represents the number of samples, D final represents the predicted raw response value, and D truth represents the raw response value obtained by camera shooting. The smaller the value of RMSE, the closer the predicted raw response value and the raw response value obtained by camera shooting, and the better the prediction effect.

[0104] The prediction raw response value error under multiple illuminations of the test light source is shown in Table 1, and the distribution of the predicted raw response value and the real camera shooting response value RGB under some illuminations is shown in the accompanying drawings. Figures 3-5

[0105] ​Table 1 The error of the raw response value predicted by the method of the present invention under multiple illuminations of the test light source

[0106]

[0107] It can be seen intuitively from the evaluation index values ​​in Table 1 that the method of the present invention has a high accuracy in predicting the raw response value under the test light source and has good performance under multiple illuminations, thus proving the superiority of the method proposed in the present invention.

[0108] On the other hand, an embodiment of the present invention also provides a digital camera raw response value prediction system based on illumination interpolation, including: a processor and a memory, the memory being used to store program instructions, and the processor being used to call the stored instructions in the memory to execute a digital camera raw response value prediction method based on illumination interpolation as described in the above technical solution.

[0109] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions merely illustrate the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for predicting raw response values ​​of a digital camera based on illumination interpolation, characterized in that: The specific steps include: Step 1, measuring the relative spectral power distribution curve of the reference light source and the spectral response function of the digital camera; Step 2, measuring the spectral reflectance of the standard color card; Step 3: Obtain the raw response values ​​of the color card under multiple illuminations of the reference light source, and simultaneously obtain the illumination value of the light source under the corresponding shooting scene; Step 4: Calculate the predicted raw response value of the standard color card under the reference light source and normalize the raw response value; Step 5: Calculate the weight between the raw response value captured by the camera and the predicted raw response value under each illumination; Step 6: Perform linear interpolation on the weights according to the illumination gradient to obtain an illumination weight lookup table; Step 7: Obtain the relative spectral power distribution curve of the test light source and the scene illumination; Step 8: Calculate the initial predicted raw response value of the standard color card under the test light source and normalize the raw response value; Step 9: Use the illumination weight lookup table to find the illumination weight corresponding to the illumination of the test light source; Step 10, calculating the difference weight between the relative spectral power distributions of the test light source and the reference light source; Step 11: Perform weighted calculation on the initial predicted raw response value using the difference weight and the illuminance weight to obtain the final predicted raw response value of the standard color card under the test light source.

2. The method for predicting raw response values ​​of a digital camera based on illumination interpolation according to claim 1, wherein: In step 1, the relative spectral power distribution curve SPD of the reference light source is measured using an illuminometer and a monochromator. ref And the digital camera spectral response function S.

3. The method for predicting raw response values ​​of a digital camera based on illumination interpolation according to claim 1, wherein: In step 2, the spectral reflectance R of the standard color card is measured using a spectrophotometer; In step 3, a digital camera is used to capture the raw response value D of the color card under multiple illuminations of the reference light source. i At the same time, the illuminance value k of the light source in the corresponding shooting scene is obtained using the illuminance meter i , where i represents the i-th illumination.

4. The method for predicting raw response values ​​of a digital camera based on illumination interpolation according to claim 1, wherein: In step 4, the predicted raw response value of the standard color card under the reference light source is calculated and the response value is normalized as follows: ； Where S represents the spectral response function matrix of the digital camera, SPD ref Represents the relative spectral power distribution vector of the reference light source, diag() represents converting the vector into a diagonal matrix, R represents the spectral reflectance matrix of the standard color card, D w Represents the predicted raw response value of a standard white plate with a spectral reflectance of 1 under the reference light source, D pre Represents the predicted raw response value of the standard color card normalized under the reference light source.

5. The method for predicting raw response values ​​of a digital camera based on illumination interpolation according to claim 1, wherein: In step 5, the weights between the raw format response values ​​captured by the camera and the predicted raw response values ​​at each illumination level are calculated as follows: ； Where D i Represents the raw format response value captured by the i-th illumination camera, w i Represents the weight between the raw format response value captured by the i-th illumination camera and the predicted raw response value, D pre Represents the predicted raw response value of the standard color card normalized under the reference light source.

6. The method for predicting raw response values ​​of a digital camera based on illumination interpolation according to claim 1, wherein: In step 6, the weight w is adjusted according to the illumination gradient. i Perform linear interpolation to obtain the illumination weight lookup table LUT illum , the linear interpolation method is as follows: ； ； ； Where k new is the illumination value that needs to be interpolated by weight, k i is the illuminance value of the reference light source under the i-th illuminance, satisfying k i <k new <k i+1 , t is the interpolation coefficient, w i is the weight between the raw format response value captured by the camera under the i-th illumination and the predicted raw response value, w new The illumination obtained by interpolation is k new The weight between the raw response value captured by the camera and the predicted raw response value, LUT illum The illumination weight lookup table is constructed, where each row represents an illumination k i And the weight w between the raw format response value captured by the camera under the corresponding illumination and the predicted raw response value i .

7. The method for predicting raw response values ​​of a digital camera based on illumination interpolation according to claim 1, wherein: In step 8, calculate the initial predicted raw response value D of the standard color card under the test light source test , the specific method is as follows: ； Where S represents the spectral response function matrix of the digital camera, diag() represents the conversion of the vector into a diagonal matrix, SPD test Represents the relative spectral power distribution matrix of the test light source, D tw Represents the predicted raw response value of a standard white plate with a spectral reflectance of 1 under the test light source, D test Represents the normalized initial predicted raw response value of the standard color card under the test light source, and R represents the spectral reflectance matrix of the standard color card.

8. The method for predicting raw response values ​​of a digital camera based on illumination interpolation according to claim 1, wherein: In step 10, the difference weight between the relative spectral power distribution of the test light source and the reference light source is calculated as follows: ； Where, mean() is the average function, w spd The SPD is the difference between the relative spectral power distribution of the test light source and the reference light source. ref Represents the relative spectral power distribution vector of the reference light source, SPD test Represents the relative spectral power distribution matrix of the test light source.

9. The method for predicting raw response values ​​of a digital camera based on illumination interpolation according to claim 1, wherein: In step 11, the difference weight w is used spd and illumination weight w illum For the initial prediction of raw response value D test Perform weighted calculation to obtain the final predicted raw response value of the standard color card under the test light source. The specific calculation method is as follows: ； Where D test is the initial predicted raw response value under the test light source, w illum Using a Lookup Table (LUT) for Step 9 illum Find out the test light source illuminance k test The corresponding illumination weight, w spd is the difference weight, D final It is the final predicted raw response value of the standard color card under the test light source.

10. A digital camera raw response value prediction system based on illumination interpolation, characterized in that: include: A processor and a memory, the memory is used to store program instructions, and the processor is used to call the stored instructions in the memory to execute a digital camera raw response value prediction method based on illumination interpolation as described in any one of claims 1 to 9.

Images