Error evaluation and prediction method and system for color constancy algorithm for outdoor scene

Abstract

The application discloses an outdoor scene-oriented color constancy algorithm error evaluation prediction method and system, and belongs to the technical field of computer and information service. The application comprises the following steps: acquiring an outdoor scene color constancy dataset under natural daylight illumination; acquiring real light source color information of a color cast image in the color constancy dataset; acquiring estimated light source color information calculated based on a color constancy algorithm to be evaluated; performing color space conversion on the light source color information; for the color constancy algorithm to be evaluated, corresponding estimated values are obtained according to real light source color attributes, estimated light source color attributes and a color constancy algorithm error evaluation quantitative model, so that the light source color estimation accuracy of different color constancy algorithms is represented. The application realizes high-precision and high-stability error evaluation of the color constancy algorithm for the outdoor scene under natural daylight illumination.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of computer and information services, and particularly relates to an error evaluation and prediction method and system for color constancy algorithm for outdoor scenes. BACKGROUND

[0002] Color constancy refers to the perceptual characteristic that the color perception of an object surface remains unchanged when the color light illuminating the object surface changes. However, a computer does not have the visual characteristics of the human eye, and therefore cannot ignore the influence of the light source on the color of the object in the image. In order to enable the computer vision system to also have such characteristics similar to the human eye, the process of restoring the color constancy of the human eye by using a computer has become an important topic in the field of computers, and has been widely applied in the fields of image processing and computer vision.

[0003] Color constancy algorithm is a method for solving the problem of computer color constancy, and its general principle is to estimate the light source color from the visual scene in order to perform image correction. There are many existing color constancy algorithms, and therefore it is particularly important to evaluate the error of the color constancy algorithm in a specific scene. The basic principle of error evaluation of the color constancy algorithm is to calculate the approximation between the original light source and the estimated light source after processing the general data set image by the algorithm. The method commonly used to quantify the approximation is to mathematically measure the difference between the original light source and the estimated light source in the normalized RGB color space. Common mathematical measurement methods include recovery angle error, reproduction angle error, and Euclidean distance. In addition, since human vision has undergone complex adaptive evolution in the long-term evolution process of natural sunlight, outdoor scene categories usually need to be specially considered and specially processed in the fields of image processing and computer vision.

[0004] However, the related research of domestic and foreign scholars at the present stage has not fully considered the particularity of outdoor scenes under natural sunlight illumination, and the error evaluation and prediction methods of existing color constancy algorithms have not overcome the interference of different light sources on the stability of error evaluation, nor have they fully considered the influence of different color spaces on the accuracy of error evaluation.

[0005] In view of the above problems, a technical scheme is urgently needed to realize an error evaluation and prediction method and system of color constancy algorithm in accordance with the subjective color visual perception of the human eye for outdoor scenes under natural sunlight illumination by using appropriate color spaces, while taking into account the stability of the error evaluation results under different light sources. SUMMARY

[0006] The purpose of the present application is to solve the problems described in the background art, and to provide an outdoor scene-oriented color constancy algorithm error evaluation prediction method and system. For outdoor scenes under natural daylight illumination, the human eye has undergone complex adaptive evolution to natural daylight during long-term evolution. Since the CIELUV color space is more accurate in representing the color of light sources including natural daylight, the CIELUV color space should be used. However, since the white point of the CIELUV color space contains zero values, and previous color constancy algorithm error evaluation methods all need to normalize the white point coordinates as the denominator, the previous color constancy algorithm error evaluation methods cannot be calculated in the CIELUV color space, and thus cannot achieve good prediction results for outdoor scenes under natural daylight illumination. The technical solution of the present application can effectively solve the above problems, thereby realizing high-precision and high-stability error evaluation of color constancy algorithms for outdoor scenes under natural daylight illumination.

[0007] The technical solution of the present application provides an outdoor scene-oriented color constancy algorithm error evaluation prediction method, comprising the following steps:

[0008] Step 1: Obtain the color constancy data set of an outdoor scene under natural daylight illumination.

[0009] Step 2: Obtain the true light source color information RGB u of the color cast image in the data set of step 1.

[0010] Step 3: Obtain the estimated light source color information RGB e calculated based on the color constancy algorithm to be evaluated.

[0011] Step 4: Convert the above light source color information to color space S1 to obtain the true light source color attribute C u and the estimated light source color attribute C e calculated based on the color constancy algorithm to be evaluated.

[0012] Step 5: Input the converted true light source color attribute C u and the converted estimated light source color attribute C e calculated based on the color constancy algorithm to be evaluated into the color constancy algorithm error evaluation quantization model F constructed by the present application to obtain the error estimation value of the color constancy algorithm to be evaluated, and thus realize the representation of the accuracy of light source color estimation for different color constancy algorithms.

[0013] The specific form of the color constancy algorithm error evaluation quantization model F is as follows:

[0014]

[0015]

[0016] wherein F is the color constancy algorithm error estimation value, C w = (L w , u w , v w ) = (100, 0, 0) is the CIELUV color space reference white point, wherein the L value represents the lightness of the color, and the u value and the v value represent the coordinates of the color in the CIE 1976 uniform color space color chart, C u = (L u , u u , v u ) is the color attribute of the real light source color information converted to the color space S1, C e = (L e , u e , v e ) is the color attribute of the estimated light source color information converted to the color space S1, which is calculated based on the color constancy algorithm to be evaluated.

[0017] Moreover, in step 2, the real light source color information RGB u is adopted in the normalized RGB color space.

[0018] Moreover, in step 3, the color constancy algorithm to be evaluated includes the Grey-Edge, GP, Grey-World, max-RGB, PCA, and Shades of Grey algorithms.

[0019] Moreover, in step 3, the estimated light source color information RGB e is adopted in the normalized RGB color space.

[0020] Moreover, in step 4, the color space S1 adopts the CIELUV color space.

[0021] The application further provides an outdoor scene-oriented color constancy algorithm error evaluation and prediction system, comprising the following modules:

[0022] A data set acquisition module is configured to acquire an outdoor scene color constancy data set under natural daylight illumination.

[0023] A real light source color acquisition module is configured to acquire real light source color information RGB u of a color cast image in the outdoor scene color constancy data set under natural daylight illumination.

[0024] An estimated light source color acquisition module is configured to acquire estimated light source color information RGBe ;

[0025] The color space conversion module is used to convert the above light source color information into the color space S1 to obtain the real light source color attribute C u and the estimated light source color attribute C calculated based on the color constancy algorithm to be evaluated e ;

[0026] The color constancy algorithm error evaluation module is used to evaluate the prediction error of the color constancy algorithm to be evaluated using the color constancy algorithm error evaluation quantification model, and to measure the light source color estimation accuracy of the color constancy algorithm to be evaluated at the subjective color perception level based on the output results. The implementation method is to convert the converted real light source color attribute C u , the estimated light source color attribute C calculated based on the color constancy algorithm to be evaluated after conversion e The error is input into the color constancy algorithm error evaluation quantization model F constructed by the present invention to obtain the error estimation value of the color constancy algorithm to be evaluated, thereby realizing the characterization of the light source color estimation accuracy of different color constancy algorithms.

[0027] The specific form of the color constancy algorithm error evaluation quantization model F is as follows:

[0028]

[0029]

[0030] Among them, F is the error estimation value of the color constancy algorithm, C w =(L w ,u w ,v w )=(100,0,0) is the reference white point of CIELUV color space, where L value represents the lightness of the color, u value and v value represent the coordinates of the color in the CIE 1976 uniform color space chromaticity diagram, C u =(L u ,u u ,v u ) is the color attribute of the real light source color information converted to the color space S1, C e =(L e ,u e ,v e ) is the color attribute converted into the color space S1 based on the estimated light source color information calculated based on the color constancy algorithm to be evaluated.

[0031] Moreover, in the real light source color acquisition module, the real light source color information RGB u Use normalized RGB color space.

[0032] Moreover, in the light source color estimation module, the color constancy algorithms to be evaluated include Grey-Edge, GP, Grey-World, max-RGB, PCA, and Shades of Grey algorithms.

[0033] Moreover, in the light source color acquisition module, the estimated light source color information RGB is calculated based on the color constancy algorithm to be evaluated. e Use normalized RGB color space.

[0034] Moreover, in the color space conversion module, the color space S1 adopts the CIELUV color space.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] The present invention proposes a color constancy algorithm error assessment and prediction technology solution for outdoor scenes. Based on the image characteristics of outdoor scenes illuminated by natural daylight and using a color constancy algorithm error assessment quantification model, it achieves a comprehensive and accurate characterization of the light source color estimation effects of different color constancy algorithms, thereby providing an effective and targeted color constancy algorithm error assessment method for the field. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a flow chart of an embodiment of the present invention;

[0038] Figure 2 This is a diagram of the visual experiment judgment operation interface in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following provides a detailed description of the embodiments of the present invention with reference to the accompanying drawings.

[0040] like Figure 1 As shown, an embodiment of the present invention provides a color constancy algorithm error evaluation and prediction technology solution for outdoor scenes. It relies on the image characteristics of outdoor scenes illuminated by natural daylight and uses a color constancy algorithm error evaluation quantization model as a means to achieve a comprehensive and accurate characterization of the light source color estimation effects of different color constancy algorithms, thereby providing an effective and targeted color constancy algorithm error evaluation method for the field.

[0041] The embodiment adopts the natural daylight illumination outdoor scene part in the Gehler dataset in the color constancy universal dataset, which contains 56 different images, each image contains a ColorChecker color card as a reference, and all pictures are saved in the Canon RAW format; and adopts six classic color constancy algorithms as the color constancy algorithms to be evaluated, including Grey-Edge, GP, Grey-World, max-RGB, PCA and Shades of Grey algorithms; the psychophysical experiment results are used as the model test basis to illustrate the accuracy of the error evaluation and prediction method of the color constancy algorithm proposed in the present application. It should be noted that the present application is not limited to the above color constancy dataset or color constancy algorithm, and the present method is also applicable to other datasets and algorithms.

[0042] The technical scheme of the present application can be automatically implemented by computer software technology when implemented by those skilled in the art. The method process provided by the embodiment includes the following steps:

[0043] 1) Obtain a color constancy dataset;

[0044] In the embodiment, the natural daylight illumination outdoor scene part in the Gehler dataset in the color constancy universal dataset is adopted, which contains 56 different images, each image contains a ColorChecker color card as a reference, and all pictures are saved in the Canon RAW format.

[0045] 2) Determine whether the type of the obtained color constancy dataset is within the scope applicable to the present application, i.e., whether the type of the obtained color constancy dataset is a natural daylight illumination outdoor scene, if not, the present application is not applicable, and if yes, the next step is performed;

[0046] In the embodiment, the images selected in step 1) are all natural daylight illumination outdoor scenes.

[0047] 3) Obtain the true light source color information RGB u of the color cast image in the color constancy dataset, and adopt a normalized RGB color space;

[0048] In the embodiment, the true light source color information RGB u of the color cast image in the color constancy dataset in step 1) can be obtained by reading the RGB value corresponding to the white block in the ColorChecker color card in the image, and the value obtained after normalization is RGB u .

[0049] 4) Obtain the estimated light source color information RGB e, and the normalized RGB color space is adopted;

[0050] In the embodiments, the color constancy algorithms to be evaluated include six classical color constancy algorithms as the color constancy algorithms to be evaluated, including Grey-Edge, GP, Grey-World, max-RGB, PCA and Shades of Grey algorithms. The six algorithms are respectively used for light source color estimation on the color cast images of the selected color constancy data set in step 1), and the values obtained after normalization are the RGB e .

[0051] 5) The light source color information is converted to a color space S1 to obtain a real light source color attribute C u and an estimated light source color attribute C e calculated based on the color constancy algorithm to be evaluated. The color space S1 adopts a CIELUV color space.

[0052] In the embodiments, the light source color information obtained in steps 3) and 4) is converted to the CIELUV color space to obtain a real light source color attribute C u and an estimated light source color attribute C e calculated based on the color constancy algorithm to be evaluated. The color space conversion method belongs to a public method, and will not be described herein.

[0053] 6) The converted real light source color attribute C u and the converted estimated light source color attribute C e calculated based on the color constancy algorithm to be evaluated in step 5) are input into the color constancy algorithm error evaluation quantification model F constructed in the present application to obtain the error estimation value of the color constancy algorithm to be evaluated, and then the accuracy of the light source color estimation of different color constancy algorithms is represented;

[0054] The specific form of the color constancy algorithm error evaluation quantification model F is as follows:

[0055]

[0056] Wherein, F is the color constancy algorithm error estimation value, C w =(L w ,u w ,v w )=(100,0,0) is a CIELUV color space reference white point, wherein the L value represents the brightness of the color, and the u value and the v value represent the coordinates of the color in the CIE 1976 uniform color space color card, C u =(L u ,u u ,vu ) is the color attribute of the real light source color information converted to the color space S1, C e = (L e , u e , v e ) is the color attribute of the estimated light source color information converted to the color space S1 based on the color constancy algorithm to be evaluated.

[0057] To further demonstrate the technical advantages of the method of the present application in the error estimation of the color constancy algorithm, a subjective comparison experiment is adopted, and the weighted correlation coefficient Pearson's r between the subjective evaluation value of the observer for the color deviation degree of the color constancy algorithm after correction obtained by the subjective experiment and the error evaluation quantitative model F of the color constancy algorithm in step 6) is calculated by the method of calculating the weighted correlation coefficient Pearson's r of all color deviation images, and is compared with the error evaluation methods of various color constancy algorithms mainly used in this field before. The specific implementation is as follows: the color deviation images of the natural daylight outdoor scene in the above 56 Gehler data sets are used as materials, and 6 classic color constancy algorithms are used as color constancy algorithms to be evaluated. The subjective evaluation experiment of the color deviation degree of the color deviation image after correction of the algorithm is carried out, and the specific experimental method is as follows:

[0058] 1) The display used in the experiment is Sanc G7c 27-inch display, the resolution is 2560*1440, the screen ratio is 16:9, the contrast is 1000:1, the response time is 1ms, the sRGB color space coverage is 129%, and the refresh rate is set to 60Hz. In order to better display the image, we use iColor Disolay3 software and i1 pro2 spectrophotometer to carry out color management and calibration on the display. In this experiment, sRGB standard preset is selected, the target white point temperature is 6500K, the gamma value is 2.2, the measured white point temperature is 6664K, the gamma value is 2.21, the white point temperature deviation is 2.52%, and the gamma value deviation is 0.45%, which is within the error range.

[0059] 2) A total of 12 same observers participated in the experiment, including 6 male observers and 6 female observers, all of whom were students and had passed the Ishihara color blindness test and had normal color vision. The age of the observers ranged from 18 to 25 years old. After the test, a questionnaire containing information such as gender, age, and skin color will be filled out outside the laboratory. When the observer enters the laboratory, he will change into a gray experimental uniform to prevent clothing reflection and clothing color interference from affecting judgment, and will be asked to turn off the mobile phone and other light-emitting devices.

[0060] 3) The visual experiment judgment operation interface is as follows: Figure 2As shown, mainly consists of image area, black background, scoring area and image switching button and other parts. The interface image is generated according to the algorithm, and the diagonal matrix is generated after the color cast image conversion, and then the pre-processing is obtained. Respectively, the correct reference light under the image and the reference light under the image obtained by the algorithm are distinguished by the inverted triangle symbol above the correct image. The scoring area has 7 levels of scoring, and the observer selects the corresponding scoring before clicking the "Next Image" button to jump to the next pair of images. The experiment progress can be viewed in real time in the lower right corner.

[0061] 4) The experiment was carried out in a dark environment, and the display was the only light source. Before the experiment started, the observer sat down under the guidance of the experimenter, adjusted the seat position and height, so that he could look straight at the center of the display at a distance of 60 cm from the display. After the observer entered the color adaptation state, the experimenter explained the experiment in detail. The observer observed the two images on the display as required, compared their similarity, and selected the score from the following seven options. The specific meaning of the seven scores is: 1 point (very dissimilar), 2 points (relatively dissimilar), 3 points (a little dissimilar), 4 points (uncertain similarity or dissimilarity), 5 points (a little similar), 6 points (relatively similar), 7 points (very similar). The observation time is not limited. After the observer determines the score, click the "Next Image" button in the lower right corner to switch to the next group of images. During the experiment, the experiment progress will be displayed in the lower right corner. After the experiment is completed, a text prompt will be displayed, and the observer's information will be recorded. For each observer, the order of the images, the left-right order of the real image and the algorithm processed image are random. A single experiment contains 6*56=336 evaluations, and takes about 20 minutes.

[0062] Through the above subjective experiment, the observer's subjective evaluation value of the color cast degree of the 56 natural daylight illuminated outdoor scene color cast images corrected by the 6 color constancy algorithms to be evaluated can be obtained, and the weighted correlation coefficient Pearson's r between the subjective evaluation value and the error evaluation quantitative model F of the color constancy algorithm constructed by the application can be calculated. Compared with the various color constancy algorithm error evaluation methods mainly used in this field before, as shown in Table 1. The results show that the correlation coefficient between the subjective evaluation value and the estimated value of the model constructed by the application is 0.922, which is significantly higher than the various color constancy algorithm error evaluation methods mainly used in this field before, proving that the color constancy algorithm error evaluation quantitative model F constructed by the application has extremely high accuracy, and further proving that the method described in the application has strong technical advantages in the accuracy evaluation of color constancy algorithm.

[0063] Table 1 Weighted correlation coefficient Pearson's r test results

[0064]

[0065] The application also provides an outdoor scene-oriented color constancy algorithm error evaluation prediction system, comprising the following modules:

[0066] A dataset acquisition module is configured to acquire a color constancy dataset;

[0067] A dataset type judgment module is configured to judge whether the type of the acquired color constancy dataset is within the scope of application of the application; if not, the application is not applicable; if yes, the next step is performed;

[0068] A real light source color acquisition module is configured to acquire real light source color information RGB u of a color cast image in the color constancy dataset;

[0069] An estimated light source color acquisition module is configured to acquire estimated light source color information RGB e calculated based on a color constancy algorithm to be evaluated;

[0070] A color space conversion module is configured to convert the light source color information to a color space S1 to obtain real light source color attributes C u and estimated light source color attributes C e calculated based on the color constancy algorithm to be evaluated;

[0071] A color constancy algorithm error evaluation module is configured to use a color constancy algorithm error evaluation quantification model to evaluate the prediction error size of the color constancy algorithm to be evaluated, and measure the light source color estimation accuracy of the color constancy algorithm to be evaluated in the subjective color perception level according to the output result. The implementation is that the converted real light source color attributes C u and the converted estimated light source color attributes C e calculated based on the color constancy algorithm to be evaluated are input into the color constancy algorithm error evaluation quantification model F constructed by the application to obtain the error evaluation value of the color constancy algorithm to be evaluated, and then the light source color estimation accuracy of different color constancy algorithms is represented.

[0072] The specific form of the color constancy algorithm error evaluation quantification model F is as follows:

[0073]

[0074]

[0075] wherein, F is the color constancy algorithm error evaluation value, C w =(L w ,u w ,vw )=(100,0,0) is the reference white point of CIELUV color space, where L value represents the lightness of the color, u value and v value represent the coordinates of the color in the CIE 1976 uniform color space chromaticity diagram, C u =(L u ,u u ,v u ) is the color attribute of the real light source color information converted to the color space S1, C e =(L e ,u e ,v e ) is the color attribute converted into the color space S1 based on the estimated light source color information calculated based on the color constancy algorithm to be evaluated.

[0076] The specific implementation of each module corresponds to each step and will not be described in detail in the present invention.

[0077] The specific embodiments described herein are merely illustrative of the spirit of the present invention. Persons skilled in the art may make various modifications, additions, or substitutions to the described specific embodiments without departing from the spirit of the present invention or exceeding the scope of the appended claims.

Claims

1. A color constancy algorithm error evaluation and prediction method for outdoor scenes, characterized by: The following steps are involved: Step 1: Obtain a color constancy dataset of outdoor scenes illuminated by natural daylight; Step 2: Get the true light source color information RGB of the color-biased image in the dataset of step 1. u ; Step 3: Obtain the estimated light source color information RGB calculated based on the color constancy algorithm to be evaluated e ; Step 4: Convert the above light source color information to color space S1 to obtain the real light source color attribute C u and the estimated light source color attribute C calculated based on the color constancy algorithm to be evaluated e ; Step 5: Set the real light source color attribute C u , estimate the light source color attribute C e Input it into the constructed color constancy algorithm error evaluation quantization model F to obtain the error estimation value of the color constancy algorithm to be evaluated; The specific form of the color constancy algorithm error evaluation quantization model F is as follows: Among them, F is the error estimation value of the color constancy algorithm, C w =(L w ,u w ,v w )=(100,0,0) is the reference white point of CIELUV color space, where L value represents the lightness of the color, u value and v value represent the coordinates of the color in the CIE 1976 uniform color space chromaticity diagram, C u =(L u ,u u ,v u ) is the color attribute of the real light source color information converted to the color space S1, C e =(L e ,u e ,v e ) is the color attribute converted into the color space S1 based on the estimated light source color information calculated based on the color constancy algorithm to be evaluated.

2. The color constancy algorithm error evaluation and prediction method for outdoor scenes according to claim 1, characterized in that: In step 2, the real light source color information RGB u Use normalized RGB color space.

3. The color constancy algorithm error evaluation and prediction method for outdoor scenes according to claim 1, characterized in that: In step 3, the color constancy algorithms to be evaluated include Grey-Edge, GP, Grey-World, max-RGB, PCA, and Shades of Grey algorithms.

4. The color constancy algorithm error evaluation and prediction method for outdoor scenes according to claim 1, characterized in that: In step 3, the estimated light source color information RGB is calculated based on the color constancy algorithm to be evaluated e Use normalized RGB color space.

5. The color constancy algorithm error evaluation and prediction method for outdoor scenes according to claim 1, characterized in that: In step 4, the color space S1 adopts the CIELUV color space.

6. A color constancy algorithm error evaluation and prediction system for outdoor scenes, characterized by: Includes the following modules: The dataset acquisition module is used to obtain the color constancy dataset of outdoor scenes illuminated by natural daylight; The real light source color acquisition module is used to obtain the real light source color information RGB of the color-biased image in the dataset u ; The estimated light source color acquisition module is used to obtain the estimated light source color information RGB calculated based on the color constancy algorithm to be evaluated e ; The color space conversion module is used to convert the above light source color information into the color space S1 to obtain the real light source color attribute C u and the estimated light source color attribute C calculated based on the color constancy algorithm to be evaluated e ; The color constancy algorithm error evaluation module is used to convert the real light source color attribute C u , estimate the light source color attribute C e Input it into the constructed color constancy algorithm error evaluation quantization model F to obtain the error estimation value of the color constancy algorithm to be evaluated; The specific form of the color constancy algorithm error evaluation quantization model F is as follows: Among them, F is the error estimation value of the color constancy algorithm, C w =(L w ,u w ,v w )=(100,0,0) is the reference white point of CIELUV color space, where L value represents the lightness of the color, u value and v value represent the coordinates of the color in the CIE 1976 uniform color space chromaticity diagram, C u =(L u ,u u ,v u ) is the color attribute of the real light source color information converted to the color space S1, C e =(L e ,u e ,v e ) is the color attribute converted into the color space S1 based on the estimated light source color information calculated based on the color constancy algorithm to be evaluated.

7. The color constancy algorithm error evaluation and prediction system for outdoor scenes according to claim 6, characterized in that: In the real light source color acquisition module, the real light source color information RGB u Use normalized RGB color space.

8. The color constancy algorithm error evaluation and prediction system for outdoor scenes according to claim 6, characterized in that: In the light source color estimation module, the color constancy algorithms to be evaluated include Grey-Edge, GP, Grey-World, max-RGB, PCA, and Shades of Grey algorithms.

9. The color constancy algorithm error evaluation and prediction system for outdoor scenes according to claim 6, characterized in that: In the light source color acquisition module, the estimated light source color information RGB is calculated based on the color constancy algorithm to be evaluated e Use normalized RGB color space.

10. The color constancy algorithm error evaluation and prediction system for outdoor scenes according to claim 6, characterized in that: In the color space conversion module, the color space S1 adopts the CIELUV color space.

Images