Indoor scene-oriented color constancy algorithm error evaluation and prediction method and system

Abstract

The invention discloses an indoor scene-oriented color constancy algorithm error evaluation and prediction method and system, and belongs to the technical field of computers and information services. The method comprises the following steps: acquiring an indoor scene color constancy data set containing portrait content; acquiring real light source color information of the color cast image in the data set; obtaining estimated light source color information calculated based on a to-be-evaluated color constancy algorithm; performing color space conversion on the light source color information; and for a to-be-evaluated color constancy algorithm, according to the real light source color attribute, the estimated light source color attribute and the color constancy algorithm error evaluation quantification model, obtaining a corresponding estimated value, and realizing characterization of light source color estimation accuracy of different color constancy algorithms. According to the invention, high-precision and high-stability error evaluation of a color constancy algorithm for an indoor scene containing portrait content is realized.

Description

Technical Field

[0001] The present invention belongs to the field of computer and information service technology, and in particular relates to a color constancy algorithm error evaluation and prediction method and system for indoor scenes. Background Art

[0002] Color constancy refers to the perceptual property that when the color of the light illuminating the surface of an object changes, the human visual system still perceives the color of the object surface unchanged. However, computers do not have the visual characteristics of the human eye, so the influence of the light source on the color of the object in the image cannot be ignored. In order to make the computer vision system have this characteristic similar to the human eye, the process of using computers to restore the color constancy of the human eye has become an important topic in the computer field and has been widely used in image processing, computer vision and other fields.

[0003] Color constancy algorithm is a general term for methods to solve computer color constancy problems. Its general principle is to estimate the color of the light source from the visual scene in order to facilitate image correction. There are many existing color constancy algorithms, so it is particularly important to evaluate the error of color constancy algorithms in specific scenarios. The basic principle of error evaluation of color constancy algorithms is to calculate the approximation between the original light source and the light source estimated by the algorithm after processing the general data set image with 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. Commonly used mathematical measurement methods include restoration angle error, reproduction angle error, and Euclidean distance. In addition, since human vision is very sensitive to the color of human faces, indoor scene categories containing human portraits usually require special consideration and special processing in the field of image processing and computer vision.

[0004] However, at present, relevant research by domestic and foreign scholars has not fully considered the particularity of indoor scenes containing portrait content, and the error assessment prediction methods of existing color constancy algorithms have not overcome the interference of different light sources on the stability of error assessment, nor have they fully considered the impact of different color spaces on the accuracy of error assessment.

[0005] In response to the above problems, it is urgent to propose a technical solution that uses appropriate color space for indoor scenes containing portrait content to implement an error evaluation and prediction method and system for the color constancy algorithm that conforms to the subjective color visual perception of the human eye, while taking into account the stability of the error evaluation results under different light sources. Summary of the invention

[0006] The purpose of the present invention is to solve the problems described in the background technology, and to propose a color constancy algorithm error evaluation prediction method and system for indoor scenes. For indoor scenes containing portrait content, the visual attention of the human eye is mainly focused on the skin color of the face and the surface color of indoor objects. Since the CIELAB color space is more accurate in representing the surface color, the CIELAB color space should be used. However, since the white point of the CIELAB color space is (100, 0, 0) and contains zero values, and the previous color constancy algorithm error evaluation methods all need to use the white point coordinates as the denominator for normalization, the previous color constancy algorithm error evaluation methods cannot be calculated in the CIELAB color space, and thus cannot achieve a good prediction effect for indoor scenes containing portrait content. The technical solution proposed by the present invention can effectively solve the above-mentioned problems, thereby realizing high-precision and high-stability error evaluation of the color constancy algorithm for indoor scenes containing portrait content.

[0007] The technical solution of the present invention is to provide a color constancy algorithm error evaluation and prediction method for indoor scenes, comprising the following steps:

[0008] Step 1, obtaining a color constancy dataset of indoor scenes containing portrait content;

[0009] Step 2: Get the real light source color information RGB of the color-biased image in the data set of step 1 u ;

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

[0011] 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 ;

[0012] Step 5: Convert the real light source color attribute C converted in step 4 u , the estimated light source color attribute C calculated based on the color constancy algorithm to be evaluated after conversion e Input into the color constancy algorithm error evaluation quantization model F constructed by the present invention, obtain the error estimation value of the color constancy algorithm to be evaluated, and then realize the characterization of the light source color estimation accuracy of different color constancy algorithms;

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

[0014]

[0015]

[0016] Where F is the error estimate of the color constancy algorithm, C w =(L w ,a w ,b w )=(100,0,0) is the reference white point of CIELAB color space, where L value represents the lightness of the color, a value represents the red-green component of the color, b value represents the yellow-blue component of the color, C u =(L u ,a u ,b u ) is the color attribute of the real light source color information converted to the color space S1, C e =(L e ,a e ,b e ) is the color attribute of the color space S1 converted from the estimated light source color information calculated based on the color constancy algorithm to be evaluated.

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

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

[0019] Furthermore, 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.

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

[0021] The present invention also provides a color constancy algorithm error evaluation and prediction system for indoor scenes, comprising the following modules:

[0022] A data set acquisition module, used to acquire a color constancy data set of indoor scenes containing portrait content;

[0023] The real light source color acquisition module is used to obtain the real light source color information RGB of the color cast images in the indoor scene color constancy dataset containing portrait content u ;

[0024] 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 ;

[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 use the color constancy algorithm error evaluation quantification model to evaluate the prediction error size of the color constancy algorithm to be evaluated, and to measure the light source color estimation accuracy of the color constancy algorithm to be evaluated at the subjective color perception level according to 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] Where F is the error estimate of the color constancy algorithm, C w =(L w ,a w ,b w )=(100,0,0) is the reference white point of CIELAB color space, where L value represents the lightness of the color, a value represents the red-green component of the color, b value represents the yellow-blue component of the color, C u =(L u ,a u ,b u ) is the color attribute of the real light source color information converted to the color space S1, C e =(L e ,a e ,b e ) is the color attribute of the color space S1 converted from 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 acquisition 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 CIELAB 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 indoor scenes, which is based on the characteristics of indoor scene images containing portrait content and uses a color constancy algorithm error assessment quantification 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 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 is an example diagram of the color constancy data set used in the embodiment of the present invention. DETAILED DESCRIPTION

[0039] In conjunction with the accompanying drawings, a detailed description of the embodiments of the present invention is provided as follows.

[0040] like Figure 1 As shown, an embodiment of the present invention provides a color constancy algorithm error evaluation and prediction technology solution for indoor scenes, which is based on the characteristics of indoor scene images containing portrait content 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 uses the indoor scene part containing portraits in the Gehler dataset in the general color constancy dataset, which contains 58 different images in total, each image contains a ColorChecker color card as a reference, and all images are saved in Canon RAW format; and 6 classic color constancy algorithms are used as the color constancy algorithms to be evaluated, including Grey-Edge, GP, Grey-World, max-RGB, PCA, and Shades of Grey algorithms; based on the results of psychophysical experiments as the model verification basis, the accuracy of the color constancy algorithm error evaluation prediction method proposed in this paper is explained. It should be noted that the present invention is not limited to the above-mentioned color constancy dataset or color constancy algorithm, and this method is also applicable to other datasets and algorithms.

[0042] When the technical solution of the present invention is specifically implemented, those skilled in the art can use computer software technology to achieve automatic operation. The method flow provided in the embodiment includes the following steps:

[0043] 1) Obtain a color constancy dataset of indoor scenes containing portrait content;

[0044] In the embodiment, the indoor scene part containing portraits in the Gehler dataset in the color constancy general dataset is used, which contains 58 different images in total, each image contains a ColorChecker color card as a reference, and all images are saved in Canon RAW format. Figure 2 shown.

[0045] 2) Obtain the true light source color information RGB of the color cast image in the indoor scene color constancy dataset containing portrait content u , using normalized RGB color space;

[0046] In the embodiment, the real light source color information RGB of the color cast image in the color constancy dataset in step 1) u It can be obtained by reading the RGB value corresponding to the white color block in the ColorChecker color card in the image. The value obtained after normalization is the RGB u .

[0047] 3) Obtain the estimated light source color information RGB calculated based on the color constancy algorithm to be evaluated e , using normalized RGB color space;

[0048] In the embodiment, the color constancy algorithm to be evaluated includes 6 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 6 algorithms are used to estimate the light source color for the color cast image of the color constancy data set selected in step 1), and the normalized value obtained is RGB e .

[0049] 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 , color space S1 adopts CIELAB color space.

[0050] In the embodiment, the light source color information obtained in the above steps 2) and 3) is converted into the CIELAB color space 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 above color space conversion method is a public method and will not be described in detail in the present invention.

[0051] 5) The real light source color attribute C converted in step 4) u , the estimated light source color attribute C calculated based on the color constancy algorithm to be evaluated after conversion e Input into the color constancy algorithm error evaluation quantization model F constructed by the present invention, obtain the error estimation value of the color constancy algorithm to be evaluated, and then realize the characterization of the light source color estimation accuracy of different color constancy algorithms;

[0052] The specific form of the color constancy algorithm error evaluation quantization model F is as follows:

[0053]

[0054]

[0055] Where F is the error estimate of the color constancy algorithm, C w =(L w ,a w ,b w )=(100,0,0) is the reference white point of CIELAB color space, where L value represents the lightness of the color, a value represents the red-green component of the color, b value represents the yellow-blue component of the color, C u =(L u ,a u ,b u) is the color attribute of the real light source color information converted to the color space S1, C e =(L e ,a e ,b e ) is the color attribute of the color space S1 converted from the estimated light source color information calculated based on the color constancy algorithm to be evaluated.

[0056] In order to further confirm the technical advantages of the method of the present invention in the error estimation of the color constancy algorithm, a subjective comparison experiment is adopted. By calculating the weighted correlation coefficient Pearson's r of all color-cast images, the weighted correlation coefficient Pearson's r between the subjective evaluation values ​​of the color cast degree after correction of the six color constancy algorithms obtained by the subjective experiment and the color constancy algorithm error evaluation quantification model F in step 5) is calculated, and compared with the various color constancy algorithm error evaluation methods previously used in this field. The specific implementation is: using the 58 color-cast images of indoor scenes containing portraits in the above-mentioned Gehler data set as materials, using the six classic color constancy algorithms as the color constancy algorithms to be evaluated, and conducting a subjective evaluation experiment on the color cast degree of the color-cast images after correction by the algorithms. The specific experimental method is as follows:

[0057] 1) The monitor used in the experiment is a Sanc G7c 27-inch monitor with a resolution of 2560*1440, a screen ratio of 16:9, a contrast ratio of 1000:1, a response time of 1ms, 129% sRGB color space coverage, and a refresh rate of 60Hz. In order to better display the image, we use iColor Disolay3 software and i1 pro2 spectrophotometer to perform color management and calibration on the monitor. In this experiment, the 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.

[0058] 2) A total of 12 identical observers participated in the experiment, including 6 male observers and 6 female observers. All observers were students in school and had passed the Ishihara color blindness test and had normal color vision ability. The age range of the observers was between 18 and 25 years old. After passing the test, a questionnaire containing information such as gender, age, and skin color will be filled out outside the laboratory. When entering the laboratory, the observer will change into a gray lab coat to prevent the reflection of the clothes and the color of the clothes themselves from interfering with judgment, and will be required to turn off light-emitting devices such as mobile phones.

[0059] 3) The visual experiment judgment operation interface is mainly composed of image area, black background, scoring area and image switching button. The images in the interface are converted from color cast images after generating diagonal matrices based on the estimated light source obtained by the algorithm and the real light source in the data set, and then preprocessed. They represent the image under the correct reference light source and the image under the reference light source obtained by the algorithm, respectively. There is an inverted triangle symbol above the correct image to distinguish it. There are 7 levels of scoring in the scoring area. After the observer selects the corresponding score, click the "Next Image" button to jump to the next pair of images. The experimental progress can be viewed in real time in the lower right corner.

[0060] 4) The experiment was conducted in a dark environment, with the display as the only light source. Before the experiment began, the observer sat down under the guidance of the experimenter, and adjusted the seat position and height so that he could look up 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 the similarity between them, and clicked to select from the following seven score options. The specific meanings of the seven scores are: 1 point (very dissimilar), 2 points (relatively dissimilar), 3 points (somewhat dissimilar), 4 points (uncertain similarity), 5 points (somewhat similar), 6 points (relatively similar), 7 points (very similar). The observation time is not limited. After the observer confirms the score, click the "Next Image" button in the lower right corner to switch to the next set of images. During the experiment, the experimental progress will be displayed in the lower right corner. After the experiment is over, a text reminder will pop up and the observer information will be recorded. For each observer, the order of image appearance and the left-right order of real images and algorithm-processed images are random. A single experiment contains 6*58=348 evaluations and takes about 20 minutes.

[0061] Through the above subjective experiment, the observers' subjective evaluation values ​​of the degree of color cast of 58 indoor scene color cast images containing portraits after correction by the six color constancy algorithms to be evaluated can be obtained, and the weighted correlation coefficient Pearson's r between the subjective evaluation value and the color constancy algorithm error evaluation quantification model F constructed by the present invention is further calculated, and compared with the various color constancy algorithm error evaluation methods previously used in this field, as shown in Table 1. The results show that the correlation coefficients between the subjective evaluation value and the model estimation value constructed by the present invention are 0.924, which are significantly higher than the various color constancy algorithm error evaluation methods previously used in this field, proving that the color constancy algorithm error evaluation quantification model F constructed by the present invention has extremely high accuracy, and further proving that the method of the present invention has a strong technical advantage in the evaluation of the accuracy of the color constancy algorithm.

[0062] Table 1 Pearson's r test results of weighted correlation coefficient

[0063]

[0064] The present invention also provides a color constancy algorithm error evaluation and prediction system for indoor scenes, comprising the following modules:

[0065] A data set acquisition module, used to acquire a color constancy data set of indoor scenes containing portrait content;

[0066] The real light source color acquisition module is used to obtain the real light source color information RGB of the color cast images in the indoor scene color constancy dataset containing portrait content u ;

[0067] 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 ;

[0068] 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 ;

[0069] The color constancy algorithm error evaluation module is used to use the color constancy algorithm error evaluation quantification model to evaluate the prediction error size of the color constancy algorithm to be evaluated, and to measure the light source color estimation accuracy of the color constancy algorithm to be evaluated at the subjective color perception level according to 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.

[0070] The specific form of the color constancy algorithm error evaluation quantization model F is as follows:

[0071]

[0072]

[0073] Where F is the error estimate of the color constancy algorithm, C w =(L w ,a w ,b w)=(100,0,0) is the reference white point of CIELAB color space, where L value represents the lightness of the color, a value represents the red-green component of the color, b value represents the yellow-blue component of the color, C u =(L u ,a u ,b u ) is the color attribute of the real light source color information converted to the color space S1, C e =(L e ,a e ,b e ) is the color attribute of the color space S1 converted from the estimated light source color information calculated based on the color constancy algorithm to be evaluated.

[0074] The specific implementation of each module corresponds to each step, which will not be elaborated in the present invention.

[0075] The specific embodiments described herein are merely examples of the spirit of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in similar ways, but they will not deviate from the spirit of the present invention or exceed the scope defined by the appended claims.

Claims

1. A color constancy algorithm error evaluation and prediction method for indoor scenes, characterized in that: The following steps are involved: Step 1, obtaining a color constancy dataset of indoor scenes containing portrait content; Step 2: Get the real light source color information RGB of the color-biased image in the data set 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 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: Where F is the error estimate of the color constancy algorithm, C w =(L w ,a w ,b w )=(100,0,0) is the reference white point of CIELAB color space, where L value represents the lightness of the color, a value represents the red-green component of the color, b value represents the yellow-blue component of the color, C u =(L u ,a u ,b u ) is the color attribute of the real light source color information converted to the color space S1, C e =(L e ,a e ,b e ) is the color attribute of the color space S1 converted from 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 indoor 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 indoor 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 indoor 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 indoor scenes according to claim 1, characterized in that: In step 4, the color space S1 adopts the CIELAB color space.

6. A color constancy algorithm error evaluation and prediction system for indoor scenes, characterized in that: Includes the following modules: A data set acquisition module, used to acquire a color constancy data set of indoor scenes containing portrait content; The real light source color acquisition module is used to obtain the real light source color information RGB of the color-biased images 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 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: Where F is the error estimate of the color constancy algorithm, C w =(L w ,a w ,b w )=(100,0,0) is the reference white point of CIELAB color space, where L value represents the lightness of the color, a value represents the red-green component of the color, b value represents the yellow-blue component of the color, C u =(L u ,a u ,b u ) is the color attribute of the real light source color information converted to the color space S1, C e =(L e ,a e ,b e ) is the color attribute of the color space S1 converted from 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 indoor 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 indoor scenes according to claim 6, characterized in that: In the light source color estimation acquisition 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 indoor scenes according to claim 6, characterized in that: In the estimated 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 indoor scenes according to claim 6, characterized in that: In the color space conversion module, the color space S1 adopts the CIELAB color space.

Images