Method for evaluating color speckle in laser display

Abstract

The application relates to an evaluation method, belongs to the technical field of laser display, and particularly relates to a method for evaluating color speckle in a laser display device. The application discloses a method for evaluating color speckle in a laser display, and the method comprises the following steps: a) obtaining a color speckle image from a laser display device; b) processing the color speckle image to extract color coordinates u' and v'; c) generating a sample covariance matrix V based on the color coordinates u' and v'; d) calculating eigenvalues alpha and beta of the sample covariance matrix V; and e) evaluating the distribution characteristics of the color speckle by using the square root of the ratio of the eigenvalues alpha and beta and the square root of the product of the eigenvalues alpha and beta. The application discloses an eigenvalue analysis method for evaluating the distribution characteristics of color speckle. By calculating the eigenvalues of the sample covariance matrix and the square root of the ratio and the square root of the product of the eigenvalues, the color speckle in the laser display device can be more accurately and intuitively evaluated, and valuable reference bases are provided for performance evaluation, fault detection and optimization improvement of the laser display device.

Description

Technical Field

[0001] This invention relates to an evaluation method, belonging to the field of laser display technology, and particularly to a method for evaluating color speckle in laser display devices. Background Technology

[0002] Laser display technology has been widely used in the display field, offering advantages such as high brightness, high contrast, and fast response speed. However, laser display devices often exhibit color speckle during use, which negatively impacts image quality and the viewing experience.

[0003] Color speckle is caused by the characteristics of the laser light source and appears as colored spots or patterns in the image. For color display devices, the speckle distribution of each color channel (such as red, green and blue) may be different.

[0004] Currently, the method defined in the international standard IEC 62906-5-4 is commonly used to evaluate color speckle in laser display devices. This method uses the variance and covariance of the color coordinates u′ and v′ in the CIE1976 color space (as shown in columns 5, 6, and 7 of Table 1) and the photometric speckle contrast to evaluate the speckle phenomenon.

[0005] Covariance is a statistic used to measure the relationship between two random variables. It describes whether the trends of change of these two variables are synchronized, that is, whether they increase or decrease simultaneously. Covariance can represent the following aspects:

[0006] ① Directional Relationship: The sign of covariance indicates whether the trends of change of the two variables are consistent. Positive covariance indicates that the two variables are positively correlated, that is, when one variable increases, the other variable also tends to increase; negative covariance indicates that the two variables are negatively correlated, that is, when one variable increases, the other variable tends to decrease.

[0007] ② Strength of relationship: The absolute value of covariance represents the strength of the association between two variables. The larger the absolute value, the stronger the relationship between the two variables; the smaller the absolute value, the weaker the relationship between the two variables.

[0008] ③ Correlation: The magnitude of covariance can be used to evaluate the correlation between two variables. When the covariance is positive, it indicates that the two variables are positively correlated, meaning that their trends of change are similar; when the covariance is negative, it indicates that the two variables are negatively correlated, meaning that their trends of change are opposite.

[0009] However, this method has some drawbacks when used for evaluation of colored speckle: First, using covariance to evaluate the distribution characteristics of speckle is not intuitive for users, making it difficult to intuitively match the speckle distribution with the covariance. Second, this method cannot comprehensively and accurately describe the dispersion of colored speckle. For example... Figure 4 The figure shows when C S-B =90, C S-G =1,C S-R When μ = 1 u′v′ =9.2, Figure 5 The figure shows when C S-B =1,C S-G =90, C S-R When μ = 1 u′v′ =-406, Figure 6 The figure shows when C S-B =1,C S-G =1,C S-R =90 μ u′v′ =82.4(C S-B Indicates blue speckle contrast, C S-G Indicates the green speckle contrast, C S-R (This represents the contrast of the red speckle). In these three cases, μ u′v′ The values ​​vary greatly, even including positive and negative values, which makes it very difficult for users to intuitively connect the relationship between the CIE 1976 color space diagram and covariance.

[0010] In the previously filed patent CN 110987382 A, an improved color speckle measurement device was used, which can achieve high-precision measurement results and fast response time. This measurement device combines advanced sensor technology and efficient data processing algorithms to provide accurate and reliable data for color speckle measurement tasks. In this patent, the measurement device from CN 110987382 A will also be used as a data acquisition device. Summary of the Invention

[0011] This invention provides a method for evaluating color speckle in laser displays, aiming to address the shortcomings of existing technologies in color speckle evaluation. This method, based on eigenvalue analysis of the sample covariance matrix, can intuitively and accurately evaluate the distribution characteristics of color speckle in laser display devices.

[0012] The technical solution adopted in this invention is a method for evaluating color speckle in laser displays, the key point of which is that it includes the following steps:

[0013] a) Acquire color speckle images from a laser display device;

[0014] b) Process the color speckle image and extract the color coordinates u′ and v′;

[0015] c) Generate the sample covariance matrix V based on the color coordinates u′ and v′;

[0016] d) Calculate the eigenvalues ​​α and β of the sample covariance matrix V;

[0017] e) The distribution characteristics of colored speckle are evaluated using the square root of the ratio of eigenvalues ​​α and β and the square root of their product.

[0018] In this invention, a colored speckle image is first acquired from a laser display device, and the image is processed to extract color coordinates u′ and v′. Next, a sample covariance matrix V is generated using the color coordinates u′ and v′, and the eigenvalues ​​of the sample covariance matrix V are calculated. The relationships between the eigenvalues ​​are analyzed, and the distribution characteristics of the colored speckle are evaluated using the magnitude and proportion of the eigenvalues.

[0019] The further calculation formula is as follows: In step c) above, the formula for generating the sample covariance matrix V based on the color coordinates u′ and v′ is as follows:

[0020]

[0021]

[0022] μ u′v′ =<(u′-<u′> )(v′-<v′> (3)

[0023]

[0024] The sample covariance matrix V describes the correlation and distribution among different color channels.

[0025] Then, by calculating the eigenvalues ​​of the sample covariance matrix V, the eigenvalues ​​α and β are obtained (as shown in Equations 5-6).

[0026]

[0027]

[0028] The eigenvalues ​​α and β reflect the distribution shape of the colored speckle. By analyzing the magnitude and proportion of these two eigenvalues, the distribution characteristics of the colored speckle can be intuitively understood.

[0029] Furthermore, based on the calculation results of the eigenvalues, the evaluation method of the present invention can evaluate the uniformity and dispersion of colored speckle. This is achieved by calculating the square root D of the ratio of eigenvalues ​​α and β. s To quantify the uniformity of the distribution of colored speckle, when β < α, When β≥α Because of the color speckle characteristics described in IEC 62906-5-4, Ds The range is approximately 0 to 0.72. When D s When the value is close to 0.7, it indicates that the speckle pattern of the three primary colors is more uniform; while when D... s When the value is close to 0, it indicates that the speckle pattern of the three primary colors is more uniform.

[0030] The aforementioned when D s Approaching 0.7 means 0.35 < D s <0.72, when D s Approaching 0 means 0 < D s ≤0.35.

[0031] Furthermore, this invention also introduces the square root of the product of eigenvalues. Used to describe the degree of dispersion of colored speckle.

[0032] Based on the above evaluation indicators, the method of the present invention can intuitively depict the distribution shape of colored speckles in the CIE 1976 color space.

[0033] This invention proposes an eigenvalue analysis method based on the sample covariance matrix to evaluate the distribution characteristics of color speckle. By calculating the eigenvalues ​​of the sample covariance matrix, as well as the square roots of their ratios and products, a more accurate and intuitive evaluation of color speckle in laser display devices is achieved, providing valuable reference for performance evaluation, fault detection, and optimization of laser display devices. Attached Figure Description

[0034] Figure 1 C S-B =100, C S-G =100, C S-R CIE 1976 color space diagram with a value of 100

[0035] Figure 2 C S-B =50, C S-G =50, C S-R CIE 1976 color space diagram with a value of 50

[0036] Figure 3 C S-B =10, C S-G =10, C S-R =10 CIE 1976 color space diagram

[0037] Figure 4 C S-B =90, C S-G =1,C S-R CIE 1976 color space diagram with 1 =

[0038] Figure 5 CS-B =1,C S-G =90, C S-R CIE 1976 color space diagram with 1 =

[0039] Figure 6 C S-B =1,C S-G =1,C S-R =90 CIE1976 color space diagram

[0040] Figure 7 C S-B =1,C S-G =90, C S-R =90 CIE1976 color space diagram

[0041] Figure 8 C S-B =90, C S-G =1,C S-R =90 CIE1976 color space diagram

[0042] Figure 9 C S-B =90, C S-G =90, C S-R CIE 1976 color space diagram with 1 =

[0043] Figure 10 A block diagram illustrating the steps of the color speckle evaluation method. Detailed Implementation

[0044] The following provides specific steps for implementing the color speckle evaluation method in laser display according to the present invention, as follows: Figure 10 As shown:

[0045] Step 1: Acquire color speckle images. Acquire color speckle images using the color speckle measuring instrument described in patent CN 110987382 A.

[0046] Step 2: Extract the color coordinates u′ and v′ of each pixel according to the CIE 1976 color space.

[0047] Step 3: Using the extracted color coordinates u′ and v′, calculate the sample covariance matrix using the following formula. This matrix describes the correlation and dispersion between different color channels.

[0048]

[0049]

[0050] μ u′v′ =<(u′-<u′> )(v′-<v′> (3)

[0051]

[0052] Step 4: Perform eigenvalue decomposition on the sample covariance matrix to obtain eigenvalues ​​α and β (as shown in Equations 5-6), and analyze the relationship between them.

[0053]

[0054]

[0055] The eigenvalues ​​α and β reflect the distribution shape of the colored speckle. By analyzing the magnitude and proportion of these two eigenvalues, the distribution characteristics of the colored speckle can be intuitively understood.

[0056] Step 5: Calculate the ratio of the square roots of the eigenvalues ​​α and β. (β<α; if β≥α, This can quantify the uniformity of the distribution of colored speckle.

[0057] This value reflects the uniformity of the colored speckle distribution; 0.35 < D s <0.72, when D s The closer the value is to 0.7, the more uniform the speckle distribution; while 0 < D s ≤0.35, when D s When the value is close to 0, it indicates that the speckle distribution is more uniform.

[0058] Step 6: Obtain the dispersion index S by calculating the square root of the product of the eigenvalues, which is the square root of the determinant of the eigenvalues. S represents the degree of dispersion of the colored speckle.

[0059] Step 7: Evaluate the distribution characteristics of the colored speckle. Based on D... s The calculation results of S are used to evaluate the distribution shape of the colored speckle.

[0060] The above steps enable the implementation of the color speckle evaluation method for laser displays according to the present invention. In practical applications, parameters can be set and adjusted according to specific needs to obtain optimal evaluation results. This method provides an effective tool and guidance for quality control, performance optimization, and fault detection of laser display devices.

[0061] Under the same conditions, the experimental results for different color channels and different speckle ratios, and the calculation results of the present invention and existing methods are shown in Table 1. Columns 5-7 show the calculation results of the variance and covariance of the color coordinates u′ and v′ in the CIE1976 color space used in the existing international standard IEC 62906-5-4. Columns 8 and 9 show the square root D of the ratio of eigenvalues ​​α and β used in the present invention. S The result of calculating the square root S of the product of the eigenvalues ​​and the eigenvalues.

[0062] Each row of data in Table 1 and Figure 1 One-to-one correspondence. C S-B Indicates blue speckle contrast, C S-G Indicates the green speckle contrast, C S-R This indicates the contrast of the red speckle pattern.

[0063] Table 1. Experimental results for different color channels and different speckle ratios.

[0064]

[0065] like Figure 1-3 As shown, corresponding to rows 1-3 of Table 1, the three primary color speckle percentages are the same, with three cases: 100%, 50%, and 10% respectively. C S-B C S-G and C S-R The percentages are the same and gradually decrease. According to the newly established evaluation index, their D... S Both are relatively large, indicating a relatively even distribution of speckle patterns. Meanwhile, S increases with C. S-B C S-G and C S-R The decrease in speckle contrast and the corresponding decrease in the value of the covariance indicate a gradual reduction in the dispersion of the speckle distribution and a gradual decrease in the coverage area of ​​the speckle. Compared with the image of the actual speckle distribution, these two indicators can more directly reflect the distribution characteristics of the speckle. Meanwhile, the absolute values ​​of the variance and covariance of u′ and v′ increase with C. S-B C S-G and C S-R The percentage decreases as v′ decreases, and the variance of v′ is less than the variance of u′. This means that as C decreases... S-B C S-G and C S-R With a reduction in percentage, the speckle distribution becomes more concentrated, and the distribution of v′ should be denser. In this case, both the present invention and the prior art can describe the speckle distribution well.

[0066] like Figure 4-6 As shown, corresponding to rows 4-6 of Table 1, when the speckle contrast of one color is 90, while the speckle contrast of the other two colors is 1, D sA value close to 0 indicates relatively low dispersion, meaning the speckle pattern is elongated and densely distributed in a certain direction. Now observe the variance and covariance of u′ and v′: Figure 4 In the figure, the variance of u′ is smaller, while the variance of v′ is larger, indicating that the distribution of speckle is close to a straight line perpendicular to u′ and is widely distributed in the direction of v′. Figure 5 In the sample, both u′ and v′ have large variances, indicating that the speckle pattern is widely distributed along the u′ and v′ directions. Their covariance is also a large negative value, therefore, u′ and v′ are negatively correlated overall. However, it is difficult to describe their specific distribution shape, and it is impossible to determine whether the speckle distribution shape is wide or narrow. Figure 6 The shape of the speckle distribution in Figure 4 Similarly, the speckle pattern is more densely distributed in the v′ direction and more dispersed in the u′ direction. It can be inferred that the speckle pattern is roughly distributed along a line perpendicular to v′. However, from the image, Figure 4 , Figure 5 and Figure 6 The overall shapes are very similar, but the original evaluation metrics (variance and covariance of u′ and v′) differ greatly and are not intuitive. Especially in Figure 5 In some cases, the shape cannot be accurately described. In these situations, the new evaluation index is better than the original evaluation method.

[0067] like Figure 7-9 As shown, corresponding to rows 7-9 of Table 1, when the speckle contrast of one color is 1, while the speckle contrast of the other two colors is 90, the square root ratio and dispersion of their corresponding eigenvalues ​​are both relatively large, indicating that the speckle distribution is relatively wide and the coverage area is large. In this case, for u′ and v′, Figure 7 , Figure 8 and Figure 9 The large variances indicate that the speckle pattern is widely distributed in both the u′ and v′ directions, but Figure 8 The covariance is much smaller. According to the definition of covariance, we know that the correlation between u′ and v′ is weak in this case. The original evaluation metrics obtained in these three cases are also not intuitive and show significant differences in some situations; therefore, using new evaluation metrics is more reasonable.

[0068] Based on the above evaluation indicators, the method of the present invention can intuitively depict the distribution shape of colored speckles in the CIE 1976 color space.

Claims

1. A method for evaluating color speckle in a laser display, characterized by, The method comprises the following steps: a) obtaining a color speckle image from a laser display device; b) processing the color speckle image to extract color coordinates u' and v'; c) generating a sample covariance matrix V based on the color coordinates u' and v'; d) calculating eigenvalues α and β of the sample covariance matrix V; e) evaluating distribution characteristics of the color speckle by using square roots of a ratio and a product of the eigenvalues α and β.

2. The method of claim 1, wherein the color speckle is evaluated in a laser display. The step c) of generating the sample covariance matrix V based on the color coordinates u' and v' means that the correlation and distribution between different color channels are described by using the following formula:

3. The method of claim 2, wherein the color speckle is evaluated in a laser display. The eigenvalues α and β reflect the distribution shape of the color speckle, and the distribution characteristics of the color speckle can be directly understood by analyzing a certain relationship between the two eigenvalues, and the calculation formula of the eigenvalues α and β is as follows:

4. The method of claim 3, wherein the color speckle is evaluated in a laser display. The analysis of the certain relationship of the two characteristic values can intuitively understand the distribution characteristics of the color speckle, and is to obtain the square root D of the ratio of the characteristic values α and β s When β < α, When β ≥ α, The square root D of the ratio of the characteristic values α and β is used to quantify the uniformity of the distribution of the color speckle s When D approaches 0.7, it indicates that the speckles of the three primary colors RGB are more uniform; and when D s approaches 0, it indicates that the speckles of the three primary colors RGB are more single.

5. The method of claim 4, wherein the color speckle is evaluated in a laser display. The D s near 0.7 means 0.35 < D s < 0.72, when D s near 0 means 0 < D s ≤ 0.

35.

6. The method of claim 3, wherein the color speckle is evaluated in a laser display. The analysis of the certain relationship of the two characteristic values can intuitively understand the distribution characteristics of the color speckle, which is to obtain the square root of the product of the characteristic values α and β which is used to describe the dispersion degree of the color speckle.

Images