A speckle quality evaluation method based on speckle gray distribution characteristic comprehensive parameter

Abstract

The application discloses a speckle quality evaluation method based on speckle gray scale distribution characteristic comprehensive parameters, and belongs to the related technical field of image measurement and optical measurement. The speckle gray scale distribution characteristic comprehensive parameters composed of speckle image average gray scale gradient, speckle image information entropy and speckle image autocorrelation parameters are used to comprehensively evaluate the quality of a speckle image. The speckle quality is comprehensively evaluated, more indexes are considered, and parameters are more comprehensive, so that the misjudgment phenomenon caused by single evaluation parameters such as average gray scale gradient and information entropy can be effectively improved. The speckle image autocorrelation parameter is used as a new parameter, and the whole speckle image is traversed, so that the contrast and randomness of the speckle image can be effectively considered, and high-quality speckles are screened out. The high-quality speckles screened out can improve the precision of processing displacement and strain of a digital image correlation system.

Description

TECHNICAL FIELD

[0001] The present application relates to a speckle quality evaluation method based on speckle gray distribution characteristic comprehensive parameters, and belongs to the technical field of image measurement and optical measurement. BACKGROUND

[0002] As an important information source in the digital image correlation measurement process, the quality of speckle pattern has different influences on the final measurement results. For the same deformation state and the same calculation parameters, different speckle patterns show different calculation accuracies. For example, when the speckle pattern shows strong regularity, it will produce false matching in the subset matching process, and thus the measurement result will be greatly different from the actual result. Therefore, it is necessary to evaluate the quality of speckle pattern to determine the randomness and regularity of speckle characteristics. The evaluation and selection of speckle pattern are very important for improving the measurement accuracy of the digital image correlation method.

[0003] In the existing data, the evaluation of the quality of speckle pattern in the digital image correlation measurement mainly includes local evaluation parameters and global evaluation parameters. The parameters such as gray gradient square sum, average gray gradient and entropy are quantitative parameters. There are also ways to evaluate and analyze the speckle pattern from the aspects of speckle particle size and image coding type. However, most of these ways use a single parameter, which is easy to misjudge and not accurate enough in evaluating the quality of speckle pattern. Therefore, the quality evaluation of speckle pattern based on the speckle gray distribution characteristic comprehensive parameters considers more indicators and more comprehensive parameters, which can more effectively consider the contrast and randomness of speckle pattern, so as to screen out higher quality speckle pattern for the equipment or system based on the digital image correlation principle in the engineering measurement process. SUMMARY

[0004] The purpose of the present application is to provide a speckle quality evaluation method based on speckle gray distribution characteristic comprehensive parameters. The speckle gray distribution characteristic comprehensive parameters composed of the average gray gradient of speckle image, the information entropy of speckle image and the autocorrelation parameter of speckle image are used to comprehensively evaluate the quality of speckle image.

[0005] The purpose of the present application is achieved by the following technical scheme:

[0006] The speckle quality evaluation method based on speckle gray distribution characteristic comprehensive parameters of the present application uses the speckle gray distribution characteristic comprehensive parameters composed of the average gray gradient of speckle image, the information entropy of speckle image and the autocorrelation parameter of speckle image to comprehensively evaluate the quality of speckle image, including the following steps:

[0007] Step 1: Obtain the digital image of speckle pattern:

[0008] The speckle has various forms, including physical speckle and digital speckle, the physical speckle includes natural pattern texture of a material surface, artificial paint, painted dots or other methods to make a physical speckle field, and the digital speckle field is generated by computer simulation;

[0009] For the physical speckle, a digital speckle image is obtained by image acquisition through a camera, so as to perform subsequent quality evaluation of the speckle image;

[0010] Step two: determining an average gray gradient parameter of the speckle image:

[0011] The average gray gradient is a global parameter, and the average gray gradient parameter δ f As shown in formula (1):

[0012]

[0013] Wherein, W and H are width and height of the speckle image respectively, is a module of a gray gradient vector of each pixel point in the speckle image, f x (x i ,y j ), f y (x i ,y j ) are gray derivatives of pixel point (x i ,y j ) in the speckle image in x and y directions respectively;

[0014] According to the relationship between the error and standard deviation of the digital image correlation measurement result of the multiple speckle images and the average gray gradient parameter of the speckle image, in most cases, the larger the average gray gradient parameter of the speckle image, the smaller the error of the digital image correlation measurement, and the better the quality of the speckle image;

[0015] Step three: determining an information entropy of the speckle image:

[0016] The information entropy of the speckle image is used to express the gray distribution in the image, the information of the speckle image content is expressed by different gray values of different pixel points in different positions, and the information entropy H(Y) of the speckle image is shown in formula (2):

[0017]

[0018] Wherein, β is the depth of the pixel, and p(a j ) is a normalized probability of each gray level;

[0019] According to the relationship between the error and standard deviation of the digital image correlation measurement result of the multiple speckle images and the information entropy of the speckle image, in most cases, the larger the information entropy of the speckle image, the smaller the error of the digital image correlation measurement, and the better the quality of the speckle image;

[0020] Step 4: Determine the speckle autocorrelation parameters of the speckle image:

[0021] The autocorrelation parameter of a speckle image reflects the degree of autocorrelation of the speckle image and is used to measure the randomness of the entire speckle image and to determine the magnitude of the correlation between different subsets in the same speckle image.

[0022] Using the center point of the speckle image as the center of the sub-region, different sub-region sizes are set, and the entire speckle image is traversed to obtain the correlation between each small sub-region and the central sub-region of the speckle image.

[0023] The size of the relevant matching sub-region is determined based on the density and size of the speckles, and the length and width of the sub-region must encompass at least three speckles.

[0024] Preferably, a zero-mean normalized cross-correlation function C, which is insensitive to changes in light intensity, is used when determining the autocorrelation degree of the speckle image. ZNCC As shown in equation (3):

[0025]

[0026] Where, f(x) i ,y j ) represents the in-frame point (x) of the reference image subset during the relevant matching process. i ,y j The grayscale value of g(x') i ,y' j ) represents the point (x') within the subset of the target image during the correlation matching process. i ,y' j The grayscale value of ) The grayscale mean of a subset of the reference image; Let M be the gray mean of a subset of the target image, and M be the radius of the subset.

[0027] After obtaining the correlation results of a subset of the entire speckle image, the mean c of the positive values ​​of the full-field correlation coefficient is determined. mean Simultaneously, the correlation results of the central sub-region and its overlapping sub-regions are removed to obtain a new full-field distribution map of correlation coefficients and the values ​​of correlation coefficients. The mean c of the positive values ​​of the full-field correlation coefficients without the central sub-region is determined. mean1 , and c mean Compare the results to determine if there are significant differences.

[0028] In most cases, the smaller the mean of the positive values ​​of the overall correlation coefficient, the lower the correlation between the speckle sub-intervals in the entire speckle image, the better the randomness of the speckle, and the better the speckle quality.

[0029] Step 5: Determine the comprehensive evaluation parameters for the speckle image;

[0030] The speckle gray scale distribution characteristic parameter N is determined as the speckle quality evaluation comprehensive parameter according to the average gray scale gradient parameter of the speckle image, the speckle image information entropy and the speckle image autocorrelation parameter, as shown in formula (4):

[0031]

[0032] Step six: using the speckle gray scale distribution characteristic comprehensive parameter to evaluate the quality of the speckle image;

[0033] The speckle gray scale distribution characteristic comprehensive parameter is used to evaluate the quality of the speckle image, the larger the speckle gray scale distribution characteristic comprehensive parameter is, the better the quality of the speckle image is, the smaller the error of digital image correlation matching and measurement is, and the better the effect is, the smaller the speckle gray scale distribution characteristic comprehensive parameter is, the worse the quality of the speckle image is, the larger the error of digital image correlation matching and measurement is, and the worse the effect is.

[0034] Beneficial effects:

[0035] 1. The speckle quality evaluation method based on the speckle gray scale distribution characteristic comprehensive parameter can comprehensively evaluate the quality of the speckle image by using the speckle gray scale distribution characteristic comprehensive parameter composed of the speckle image average gray scale gradient, the speckle image information entropy and the speckle image autocorrelation parameter, more indexes and more comprehensive parameters are considered in the evaluation parameter, and the misjudgment phenomenon caused by using single evaluation parameters such as the average gray scale gradient and the information entropy can be effectively improved.

[0036] 2. The speckle quality evaluation method based on the speckle gray scale distribution characteristic comprehensive parameter can effectively consider the contrast and randomness of the speckle image by using the new parameter of the speckle image autocorrelation parameter and adopting the full-field traversal method to perform correlation matching between the full-field image sub-area and the speckle image center sub-area, and high-quality speckles can be screened out.

[0037] 3. The speckle quality evaluation method based on the speckle gray scale distribution characteristic comprehensive parameter is more accurate and feasible compared with other speckle quality evaluation methods, so that high-quality speckle images can be screened out in the measurement or use process of the equipment or system based on the digital image correlation principle in engineering measurement, the precision in the displacement and strain measurement process of the digital image correlation system is improved, a high-quality speckle quality evaluation method is provided for the preparation of the standard speckle used in the calibration process of the digital image correlation system, and high-quality target speckle images can be screened out and prepared. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 It is a speckle quality evaluation method based on the speckle gray scale distribution characteristic comprehensive parameter;

[0039] Figure 2 is speckle pattern A and its gray level histogram;

[0040] Figure 3 is speckle pattern B and its gray level histogram;

[0041] Figure 4 is speckle pattern C and its gray level histogram;

[0042] Figure 5 is the full-field correlation coefficient distribution and full-field correlation coefficient value diagram of speckle pattern A;

[0043] Figure 6 is the full-field correlation coefficient distribution and full-field correlation coefficient value diagram of speckle pattern B;

[0044] Figure 7 is the full-field correlation coefficient distribution and full-field correlation coefficient value diagram of speckle pattern C;

[0045] Figure 8 is the MATLAB software simulation main interface diagram; DETAILED DESCRIPTION

[0046] In order to better illustrate the purpose and advantages of the present application, the content of the application will be further described below in combination with the drawings and examples.

[0047] Example 1:

[0048] The speckle quality evaluation method based on the speckle gray level distribution characteristic comprehensive parameter of the present application is used to comprehensively evaluate the speckle image, and the speckle gray level distribution characteristic comprehensive parameter of the speckle image based on the speckle image average gray level gradient, the speckle image information entropy and the speckle image autocorrelation parameter, as shown in the following formula: Figure 1 The method comprises the following steps:

[0049] Step one: obtaining the digital image of the speckle pattern:

[0050] The speckle has various forms, including physical speckle and digital speckle. The physical speckle includes the natural pattern texture on the surface of the material, the artificial paint, the painted dots or other methods to make the physical speckle field, and the digital speckle field is generated by computer simulation;

[0051] For the physical speckle, the digital speckle image is obtained by image acquisition through the camera, so as to perform the subsequent quality evaluation of the speckle image;

[0052] In the example, the digital speckle field is generated by computer simulation, three speckle patterns with different characteristics and large differences are selected, the image size is 512pixelx512pixel, and the three speckle patterns are named as speckle pattern A, speckle pattern B and speckle pattern C respectively, and the corresponding gray level histograms of the speckle patterns are obtained, as shown in the following formula:Figure 2 、 Figure 3 、 Figure 4 as shown in FIG. 1;

[0053] Step two: determine the average gray gradient parameter of the speckle image:

[0054] The average gray gradient is a global parameter, and the average gray gradient parameter δ f as shown in formula (1):

[0055]

[0056] wherein W and H are the width and height of the speckle image respectively, is the norm of the gray gradient vector of each pixel point in the speckle image, f x (x i ,y j ), f y (x i ,y j ) are the gray derivatives of the pixel point (x i ,y j ) in the speckle image in x and y directions respectively;

[0057] According to the relationship between the error and standard deviation of the digital image correlation measurement result of multiple speckle images and the average gray gradient parameter of the speckle image, in most cases, the larger the average gray gradient parameter of the speckle image, the smaller the error of the digital image correlation measurement, and the better the quality of the speckle image;

[0058] In the embodiment, the average gray gradient parameters of the speckle image A, the speckle image B and the speckle image C are shown in Table 1:

[0059] Table 1 Average gray gradient parameters of the speckle image A, the speckle image B and the speckle image C

[0060] Speckle quality evaluation item Speckle pattern A Speckle pattern B Speckle pattern C Speckle image average grey level gradient δ f ]] 14.80 18.27 15.85

[0061] The speckle quality evaluation result based on the average gray gradient parameter of the speckle image is that the speckle image B is better than the speckle image C, and the speckle image C is better than the speckle image A;

[0062] Step three: determine the information entropy of the speckle image:

[0063] The information entropy of the speckle image is used to express the gray distribution in the image, and the information of the speckle image content is expressed by different gray values of different pixel points in different positions. The information entropy H(Y) of the speckle image is shown in formula (2):

[0064]

[0065] wherein β is the depth of the pixel, p(a j) is the normalized probability of occurrence of each gray level;

[0066] According to the relationship between the error and standard deviation of the digital image correlation measurement result of the plurality of speckle images and the information entropy of the speckle image, in most cases, the greater the information entropy of the speckle image, the smaller the error of the digital image correlation measurement, and the better the quality of the speckle image;

[0067] In the embodiment, the information entropy of the speckle images of the speckle pattern A, the speckle pattern B and the speckle pattern C is shown in Table 2:

[0068] Table 2 Information entropy of the speckle pattern A, the speckle pattern B and the speckle pattern C

[0069] Speckle quality evaluation item Speckle pattern A Speckle pattern B Speckle pattern C Speckle image information entropy H(Y) 7.3318 7.2881 7.4143

[0070] The speckle quality evaluation result based on the information entropy of the speckle image is that the speckle pattern C is better than the speckle pattern A, and the speckle pattern A is better than the speckle pattern B;

[0071] Step four: determining the speckle autocorrelation parameter of the speckle image:

[0072] The autocorrelation parameter of the speckle image reflects the autocorrelation degree of the speckle image, and is used to measure the randomness of the whole speckle image and judge the correlation between different subsets in the same speckle image;

[0073] Taking the center point of the speckle image as the center of the subregion, different subregion sizes are set, and the correlation degree between each small subregion and the center subregion of the speckle image is obtained by traversing the whole speckle image;

[0074] The size of the correlation matching subregion is determined according to the density of the speckle points and the size of the speckle points, and the length and width of the subregion at least include three speckle points;

[0075] In the embodiment, the zero-mean normalized cross-correlation function C ZNCC , which is not sensitive to light intensity changes, is used to determine the autocorrelation degree of the speckle image.

[0076]

[0077] , wherein f(x i ,y j ) is the gray value of the point (x i ,y j ) in the reference image subset in the correlation matching process, g(x' i ,y' j ) is the gray value of the point (x' i ,y' j ) in the target image subset in the correlation matching process, f is the gray mean value of the reference image subset; g is the gray mean value of the target image subset, and M is the radius of the subset.

[0078] After obtaining the correlation results of the subset of the whole speckle pattern image, the mean value c of the positive values of the full-field correlation coefficient is determined mean At the same time, the correlation results of the central sub-region and the sub-regions overlapping with the central sub-region are removed, and the new full-field correlation coefficient distribution map and the correlation coefficient value are obtained, and the mean value c of the positive values of the full-field correlation coefficient without the central sub-region is determined mean1 The correlation coefficient value of the central sub-region is compared with c mean to determine whether there is a large difference.

[0079] In most cases, the smaller the mean value of the positive values of the full-field correlation coefficient, the lower the correlation degree between the speckle sub-regions of the whole speckle pattern image, the better the randomness of the speckle, and the better the quality of the speckle.

[0080] In the embodiment, the full-field correlation coefficient distribution of the three speckle pattern images and the speckle pattern image full-field correlation coefficient value are shown in Figure 5 , Figure 6 , Figure 7 In each image, the image numbered ① is a full-field correlation coefficient distribution diagram of each speckle pattern image, the image numbered ② is a full-field correlation coefficient value diagram of each speckle pattern image, the image numbered ③ is a full-field correlation coefficient distribution diagram of each speckle pattern image without the sub-regions overlapping with the central sub-region, and the image numbered ④ is a full-field correlation coefficient value diagram of each speckle pattern image without the sub-regions overlapping with the central sub-region.

[0081] In the embodiment, the mean value c of the positive values of the speckle pattern image full-field correlation coefficient of the speckle pattern A, the speckle pattern B and the speckle pattern C is shown in Table 3. mean As shown in Table 3:

[0082] Table 3: Speckle pattern image autocorrelation parameters of speckle pattern A, speckle pattern B and speckle pattern C

[0083] Speckle quality evaluation item Speckle pattern A Speckle pattern B Speckle pattern C Speckle image autocorrelation parameter c mean ]]> 0.2151 0.1581 0.2355

[0084] The speckle quality evaluation result based on the autocorrelation parameters of the speckle pattern image is that the speckle pattern B is better than the speckle pattern A, and the speckle pattern A is better than the speckle pattern C.

[0085] Step five: determining the comprehensive evaluation parameter of the speckle pattern image.

[0086] Based on the speckle pattern image average gray gradient parameter, the speckle pattern image information entropy and the speckle pattern image autocorrelation parameter, the speckle gray distribution characteristic parameter N is determined as the speckle quality evaluation comprehensive parameter, as shown in formula (4):

[0087]

[0088] In the embodiment, the speckle quality evaluation comprehensive parameter N of the speckle pattern A, the speckle pattern B and the speckle pattern C is shown in Table 4.

[0089] Table 4. Comprehensive parameters N for speckle quality evaluation of speckle patterns A, B, and C.

[0090] Speckle quality evaluation item Speckle pattern A Speckle pattern B Speckle pattern C Speckle gray scale distribution characteristic comprehensive parameter N / 100 5.0450 8.4221 4.9901

[0091] Step 6: Evaluate the image quality of speckle images using comprehensive parameters of speckle grayscale distribution characteristics;

[0092] The examples use MATLAB for software simulation, such as... Figure 8 As shown, the top left corner shows the speckle image to be evaluated;

[0093] The quality of speckle images is evaluated using a comprehensive parameter of speckle gray-scale distribution characteristics. The larger the comprehensive parameter of speckle gray-scale distribution characteristics, the better the speckle image quality, the smaller the error in digital image correlation matching and measurement, and the better the effect. Conversely, the smaller the comprehensive parameter of speckle gray-scale distribution characteristics, the worse the speckle image quality, the larger the error in digital image correlation matching and measurement, and the worse the effect.

[0094] In the embodiments, when the method of the present invention is used to evaluate the quality of speckle images, speckle image B is better than speckle image A, and speckle image A is better than speckle image C.

[0095] The results of speckle image quality assessment using different single parameters differ from those using individual parameters versus a combination of parameters. The following tests are used to verify the speckle image quality assessment results:

[0096] Each speckle image was shifted by 1.26 pixels and 2.66 pixels respectively, and the mean, error, and standard deviation of the displacement measured using speckle image A, speckle image B, and speckle image C were obtained, as shown in Table 5:

[0097] Table 5. Mean, error, and standard deviation of displacement measured by speckle plot.

[0098] Speckle quality evaluation item Speckle pattern A Speckle pattern B Speckle pattern C 1.26 pixel calculation displacement mean (pixel) 1.2599 1.2599 1.2597 1.26 pixel calculation mean error (pixel) 0.0001 0.0001 0.0003 1.26 pixel calculation standard deviation (pixel) 0.0013 0.0009 0.0013 2.66 pixel calculation displacement mean (pixel) 2.6603 2.6601 2.6603 2.66 pixel calculation mean error (pixel) 0.0003 0.0001 0.0003 2.66 pixel calculation standard deviation (pixel) 0.0013 0.0009 0.0014

[0099] As shown in Table 5, based on the mean, error, and standard deviation of the displacement measured in the speckle image, speckle image B is superior to speckle image A, and speckle image A is superior to speckle image C. This is consistent with the results of the present invention using the comprehensive parameter of speckle gray-level distribution characteristics for speckle image quality evaluation. In contrast, speckle image quality evaluation methods using the average gray-level gradient parameter or the single parameter of information entropy of speckle image both exhibit misjudgment and the evaluation results are not accurate enough. In summary, the present invention can accurately predict and evaluate the quality of speckle images.

[0100] The above detailed description of the specific description, the purpose, technical scheme and beneficial effects of the application are further described in detail, it should be understood that the above description is only a specific embodiment of the application, and is not used to limit the protection scope of the application, any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the application should be included in the protection scope of the application.

Claims

1. A method for evaluating speckle quality based on comprehensive parameters of speckle grayscale distribution characteristics, characterized in that: A comprehensive evaluation of speckle image quality is achieved using a speckle image gray-level distribution characteristic parameter composed of the average gray-level gradient, information entropy, and autocorrelation parameters of the speckle image. This evaluation includes the following steps: Step 1: Obtain the digital image of the speckle pattern: Speckle patterns exist in various forms, including physical speckle and digital speckle. Physical speckle includes natural patterns and textures on the material surface, artificial spraying, dotting, or other methods to create physical speckle fields, while digital speckle fields are generated through computer simulation. For physical speckle patterns, digital speckle images are acquired using a camera to facilitate subsequent quality evaluation of the speckle images. Step 2: Determine the average gray-level gradient parameters of the speckle image: The average gray-level gradient is a global parameter, δ. f As shown in equation (1): Where W and H are the width and height of the speckle image, respectively. f is the magnitude of the gray-level gradient vector of each pixel in the speckle image. x (x i ,y j ),f y (x i ,y j ) represents the pixel (x) in the speckle image. i ,y j The gray-scale derivatives in the x and y directions, respectively; Based on the relationship between the error and standard deviation of digital image correlation measurement results of multiple speckle images and the average gray-level gradient parameter of the speckle image, in most cases, the larger the average gray-level gradient parameter of the speckle image, the smaller the error of digital image correlation measurement and the better the quality of the speckle image. Step 3: Determine the information entropy of the speckle image: The information entropy of a speckle image is used to express the gray-level distribution in the image. The information of the speckle image content is expressed by the different gray-level values ​​of pixels at different locations. The information entropy H(Y) of the speckle image is shown in Equation (2): Where β is the depth of the pixel, p(a j The standardized probability of each gray level occurring; Based on the relationship between the error and standard deviation of digital image correlation measurement results of multiple speckle images and the information entropy of speckle images, in most cases, the larger the information entropy of the speckle image, the smaller the error of digital image correlation measurement, and the better the quality of the speckle image. Step 4: Determine the speckle autocorrelation parameters of the speckle image: The autocorrelation parameter of a speckle image reflects the degree of autocorrelation of the speckle image and is used to measure the randomness of the entire speckle image and to determine the magnitude of the correlation between different subsets in the same speckle image. Using the center point of the speckle image as the center of the sub-region, different sub-region sizes are set, and the entire speckle image is traversed to obtain the correlation between each small sub-region and the central sub-region of the speckle image. The size of the relevant matching sub-region is determined based on the density and size of the speckles, and the length and width of the sub-region must encompass at least three speckles. After obtaining the correlation results of a subset of the entire speckle image, the mean c of the positive values ​​of the full-field correlation coefficient is determined. mean Simultaneously, the correlation results of the central sub-region and its overlapping sub-regions are removed to obtain a new full-field distribution map of correlation coefficients and the values ​​of correlation coefficients. The mean c of the positive values ​​of the full-field correlation coefficients without the central sub-region is determined. mean1 , and c mean Compare the results to determine if there are significant differences. In most cases, the smaller the mean of the positive values ​​of the overall correlation coefficient, the lower the correlation between the speckle sub-intervals in the entire speckle image, the better the randomness of the speckle, and the better the speckle quality. Step 5: Determine the comprehensive evaluation parameters for the speckle image; Based on the above parameters of average gray-level gradient, information entropy, and autocorrelation of speckle image, the speckle gray-level distribution characteristic parameter N is determined as the comprehensive parameter for speckle quality evaluation, as shown in equation (3): Step 6: Evaluate the image quality of speckle images using comprehensive parameters of speckle grayscale distribution characteristics; The quality of speckle images is evaluated using a comprehensive parameter of speckle gray-scale distribution characteristics. The larger the comprehensive parameter, the better the speckle image quality, the smaller the error in digital image correlation matching and measurement, and the better the effect. Conversely, the smaller the comprehensive parameter, the worse the speckle image quality, the larger the error in digital image correlation matching and measurement, and the worse the effect.

2. The speckle quality evaluation method based on comprehensive parameters of speckle grayscale distribution characteristics as described in claim 1, characterized in that: When determining the autocorrelation degree of speckle images, a normalized cross-correlation parameter C with zero mean that is insensitive to changes in light intensity is used. ZNCC As shown in equation (4): Where, f(x) i ,y j ) represents the in-frame point (x) of the reference image subset during the relevant matching process. i ,y j The grayscale value of g(x') i ,y' j ) represents the point (x') within the subset of the target image during the correlation matching process. i ,y' j The grayscale value of ) The grayscale mean of a subset of the reference image; Let M be the gray mean of a subset of the target image, and M be the radius of the subset.

3. The speckle quality evaluation method based on comprehensive parameters of speckle grayscale distribution characteristics as described in claim 1, characterized in that: The autocorrelation parameter of the speckle image is used to reflect the degree of autocorrelation of the speckle image, so as to measure the randomness of the entire speckle image and judge the magnitude of the correlation between different subsets in the same speckle image. The autocorrelation parameter of a speckle image is the average of the positive values ​​of the correlation coefficients between the sub-regions of the entire image and the central sub-region of the speckle image, obtained by establishing sub-regions of different sizes with the center point of the speckle image as the center point and traversing the entire speckle image.

Images