A method and system for evaluating the visual quality of washed stone surface

Abstract

The present invention relates to a method and system for evaluating the visual quality of a washed stone surface layer. The method includes: performing morphological processing on the image of the washed stone surface layer to extract the geometric morphological characteristics of the exposed aggregates in the image; the geometric morphological characteristics include the geometric morphology, geometric center, and relative distance between each aggregate exposed; based on the geometric morphological characteristics of the exposed aggregates in the image, defining the longest diagonal as the maximum diameter L of the exposed aggregates max , the distance D between the geometric centers of any two exposed aggregates e , calculating the relative spacing ξ of the exposed aggregates = L max / D e ; calculating the probability density of the normal distribution of ξ. If ξ conforms to the normal distribution and the mean value of ξ satisfies the 95% confidence interval, it is determined that the washed stone surface layer meets the requirements and has the highest visual quality. The present invention can realize real-time, efficient, and quantifiable visual detection and evaluation of the exposure degree of the aggregates on the washed stone surface layer, can significantly improve the construction efficiency of the washed stone, improve the construction quality, and achieve uniform exposure of the aggregates on the washed stone surface layer and the optimal visual quality.

Description

Technical Field

[0001] The invention belongs to the technical field of decoration engineering, and in particular relates to a method and system for evaluating the visual quality of a washed stone surface layer. Background Art

[0002] Washed stone is a construction process in which pebbles or gravel are mixed with cement in a certain proportion. After being smoothed and initially dried, the cement on the surface of the aggregate is washed away with water to expose the surface of the aggregate (the pebbles used in washed stone are also called aggregates or particles, referred to as aggregates in this article) for decorative purposes. It has the characteristics of wear resistance and corrosion resistance, and is often used in building surface decoration, park ecological pavement and other projects. For example, a new wetland park is planned to be built in a certain place, and the park is designed to have an ecological washed stone pavement of nearly 10 km. The construction of this ecological pavement will significantly improve the environmental quality of the park. However, during the construction of the ecological pavement, due to differences in the water washing construction process, the distribution of exposed aggregates on the surface of some pavements is uneven (such as Figure 2 As shown in the figure, if the distance between aggregates is not uniform and the aggregate distribution is uneven, it will not meet the requirements of high-quality construction process control and the appearance quality. It will also reduce the visual quality of the entire ecological pavement. Figure 1 As shown in the figure. After the washed stone surface layer is initially washed, the exposure degree of aggregate particles gradually increases with the intensity of washing. As for what degree can meet the design standards of the washed stone surface layer, it is one of the problems faced by engineering technicians. Therefore, a real-time, efficient and quantifiable method for visually detecting and evaluating the exposure degree of washed stone surface aggregate is urgently needed. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for evaluating the visual quality of a washed stone surface layer to solve the above-mentioned technical problems.

[0004] The present invention provides a method for evaluating the perceptual quality of a washed stone surface layer, comprising:

[0005] Step 1, morphologically processing the washed stone surface layer image to extract the geometric features of the exposed aggregates in the image; the geometric features include the exposed geometric shape of the aggregates, the geometric center, and the relative distance between the aggregates;

[0006] Step 2: Based on the geometric features of the exposed aggregate in the image, define the longest diagonal angle as the maximum diameter L of the exposed aggregate. max , the geometric center distance D between any two exposed aggregates e , calculate the relative spacing of exposed aggregate ξ=L max / D e ;

[0007] Step 3, calculate the probability density of the normal distribution of ξ. If ξ conforms to the normal distribution and the mean of ξ satisfies the 95% confidence interval, it is determined that the washed stone surface layer meets the requirements and has the highest visual quality.

[0008] Furthermore, the washed stone surface layer image in step 1 is acquired using a portable imaging device.

[0009] Furthermore, the morphological processing in step 1 includes image preprocessing, aggregate particle size verification, image segmentation, and particle morphological processing.

[0010] Furthermore, the image preprocessing includes:

[0011] The washed stone surface layer image is corrected, and after the image correction is completed, a new coordinate system is re-established for the corrected image.

[0012] Furthermore, the image segmentation includes:

[0013] The target area image is divided into two parts, one part contains only the scale part; the other part contains only the aggregate information part. The scale part is cut off to avoid affecting the calculation accuracy. Subsequent calculations are all performed using the part containing only the aggregate information for calculation and analysis, and output in proportion.

[0014] Furthermore, in step 1, the image optimal threshold segmentation method is used to perform threshold segmentation processing on the aggregate morphology, including:

[0015] 1) Convert color images to grayscale images

[0016] In the new coordinate system, by establishing an m×n-order matrix, each element in the washed stone surface layer is changed into a two-dimensional function f(x,y) of a digital image and expressed in matrix form;

[0017] The weighted method is used to process the grayscale of the washed stone surface layer image and convert the color image into a grayscale image as follows:

[0018] f g (x,y)=w i f i (x,y),i=1…3

[0019] In the formula, w i is the weight of chromaticity; i represents color, 1 to 3 represent red, blue and green respectively; f i (x,y) is the grayscale image function;

[0020] 2) According to the gray value characteristics of the image, the aggregate morphology in the image is segmented by threshold, the point-like, isolated smaller targets are deleted, the holes are filled, and the perimeter area information of each connected area is obtained; the morphological threshold is used to select the optimal solution, and the gray image function is set as f i(x, y), the grayscale image gradient function is g(x, y), the set of all pixels is defined as g(i), the grayscale threshold of the target area image is k, and its corresponding pixel set is C(k j ), where K j is the grayscale increase value, then K j The pixel set is:

[0021] C(k j )=[g(x,y)≤k]

[0022] Taking j as the reference, the geometry of pixels less than a certain threshold is expressed as follows:

[0023] g(k)=g(i)∪C(k j )

[0024] By continuously adjusting the set of k, g(x,y) = k, and calculating the maximum diameter and shape coefficient of the aggregate, and comparing them with the image verification results, the dimensionless parameters are as follows:

[0025]

[0026] Where L is the maximum diameter, S is the shape factor, and the subscript m is the parameter value of the standard ellipse, mk Parameter values ​​for different threshold values, and then determine the optimal threshold k=20.

[0027] Further, the particle morphology processing includes:

[0028] The image is further processed to achieve single particle cutting. First, the connected regions with an area smaller than a specified value are removed, the missing regions are filled, and then the connected regions in the image are merged.

[0029] The present invention also provides a system for evaluating the perceptual quality of a washed stone surface layer, comprising a perceptual quality evaluation module, wherein the perceptual quality evaluation module is used to execute the method for evaluating the perceptual quality of a washed stone surface layer.

[0030] The present invention also provides a non-transitory computer-readable storage medium, which stores computer instructions. When the computer instructions are executed by a processor, the method for evaluating the visual quality of a washed stone surface layer is implemented.

[0031] The present invention also provides an electronic device, comprising:

[0032] A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for evaluating the visual quality of a washed stone surface layer by executing the computer instructions.

[0033] Through the above scheme, through the visible quality evaluation method and system of the washed stone surface layer, it is possible to realize real-time, efficient and quantifiable visible detection and evaluation of the exposure of the washed stone surface layer aggregate, which can significantly improve the construction efficiency of the washed stone, improve the construction quality, and achieve uniform exposure of the washed stone surface layer aggregate and optimal appearance.

[0034] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and implement it according to the contents of the specification, the following is a detailed description of the preferred embodiments of the present invention in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a flow chart of the method for evaluating the perceptible quality of the washed stone surface layer of the present invention;

[0036] Figure 2 It is a schematic diagram of uneven distribution of concrete aggregate;

[0037] Figure 3 Schematic diagram of geometric morphological characteristics of exposed aggregates of the present invention, wherein (a) is the geometric morphological characteristics, and (b) is the normal distribution of relative distance;

[0038] Figure 4 is a schematic diagram of image preprocessing in one embodiment of the present invention, wherein (a) is the original image, and (b) is the coordinate system;

[0039] Figure 5 1 is a schematic diagram of image recognition and ratio adjustment in one embodiment of the present invention, wherein (a) only includes a scale image, and (b) only includes a scale image;

[0040] Figure 6 is a schematic diagram of image recognition and ratio adjustment in one embodiment of the present invention;

[0041] Figure 7 It is the verification of particle morphology parameters exposed at different thresholds in one embodiment of the present invention;

[0042] Figure 8 It is the characteristic of particle morphology parameters exposed at different thresholds in one embodiment of the present invention;

[0043] Fig. 9 It is an RGB image extracted by different threshold point particles in one embodiment of the present invention;

[0044] Fig.10 The images of the surface layer of washed stone at various process stages in one embodiment of the present invention, wherein (a) is the initial wash, (b) is the secondary wash, and (c) is to be formed;

[0045] Fig.11It is the evolution characteristic of the shape coefficient of the exposed aggregate of the washed stone surface layer in one embodiment of the present invention;

[0046] Fig.12 is the normal distribution characteristic of aggregate exposure in one embodiment of the present invention;

[0047] Fig.13 It is a schematic diagram of an electronic device of the present invention. DETAILED DESCRIPTION

[0048] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0049] Ginseng Figure 1 As shown, this embodiment provides a method for evaluating the perceptual quality of a washed stone surface layer, comprising:

[0050] Step S1, morphologically processing the washed stone surface layer image to extract geometric features of the exposed aggregates in the image; the geometric features include the exposed geometric shape of the aggregates, the geometric center, and the relative distance between the aggregates;

[0051] Step S2: Based on the geometric features of the exposed aggregate in the image, define the longest diagonal angle as the maximum diameter L of the exposed aggregate. max , the geometric center distance D between any two exposed aggregates e , calculate the relative spacing of exposed aggregate ξ=L max / D e ;

[0052] Step S3, calculate the probability density of the normal distribution of ξ. If ξ conforms to the normal distribution and the mean of ξ satisfies the 95% confidence interval, it is determined that the washed stone surface layer meets the requirements and has the highest visual quality.

[0053] The method for evaluating the visible quality of washed stone surface layer can realize real-time, efficient and quantifiable visible detection and evaluation of washed stone surface layer aggregate exposure, which can significantly improve the washed stone construction efficiency, improve the construction quality, and achieve uniform exposure of washed stone surface layer aggregate and optimal appearance.

[0054] The present invention is described in further detail below.

[0055] Washed stone surface layers have advantages such as durability and are often used in scenes such as building surface decoration and ecological trails. Affected by the construction process of washed stone, the degree of aggregate exposure in the surface layer is difficult to grasp, and there is a lack of quantitative evaluation indicators. Its morphological characteristics include particle size, shape coefficient, and surface coefficient. According to the characteristics of particle shape, particles can be divided into two categories: star-shaped particles and non-star-shaped particles. This embodiment is based on the traditional image segmentation method, and is optimized to propose an optimal threshold algorithm based on the detection of aggregate exposure in the washed stone surface layer. By extracting the shape contour of the aggregate exposed in the washed stone surface layer image, using mathematical means to calculate, and using the basic principles of statistics to perform statistical analysis, evaluation is performed based on the statistical results to guide engineering construction and improve construction efficiency.

[0056] 1. Evaluation model of aggregate exposure of washed stone surface layer

[0057] The geometric shape of exposed aggregate refers to the geometric shape of aggregate exposure, geometric center and relative distance between aggregates. Its contour information is a series of discrete point sets in the Cartesian coordinate system. Based on these discrete two-dimensional point set data P (x, y), the method of polar Fourier series expansion can be used to reconstruct the contour of connected particles. Taking the exposure of surface aggregate after the construction of washed stone surface layer as an example, see Figure 3 (a). The white area is the aggregate exposed from the surface layer, and the gray area is the concrete. It can be seen that the shape of the exposed aggregate particles is relatively random. From the plane morphology, it basically presents a type of ellipse, or a two-dimensional non-star-shaped particle. Therefore, this embodiment uses a type of ellipse morphology for analysis. The maximum diameter is calculated by calculating the distance between any two points on the edge of the pebble outline, and then the maximum pixel distance of the pebble outline is selected by comparison, and then the maximum pixel distance is converted into the actual size as the maximum diameter of the pebble. According to the geometric morphological characteristics of the exposed aggregate in the figure, the longest diagonal is defined as the diameter of the exposed aggregate, and L is used. max In other words, the distance between any two exposed aggregate geometric centers is expressed as D e Indicated by. Under the condition of known particle outline pixels, by calculating the maximum center distance between adjacent particles, that is, Figure 3 The distance between O1 and O2 in (a) is obtained. The dimensionless parameter ξ = L max / D e The relative spacing of exposed aggregates is described. This parameter is used to determine the uniformity of aggregate distribution and to guide the construction of washed stone surface layers.

[0058] Based on the statistical principle of normal distribution, a quantitative index for judging the beauty of particle exposure is proposed. That is, the aggregate exposure spacing conforms to the normal distribution and meets the 95% criterion. The normal distribution formula is shown in formula (1), and the distribution is shown in Figure 3 (b).

[0059]

[0060] Where ξ is the sample population mean and is the population standard deviation.

[0061] If the relative distances of the surface aggregate exposure are all within the 95% range, the visible quality of the washed stone surface layer is the highest under this condition, and this is used to quantify the control parameters of the washed stone surface aggregate exposure and to quantitatively evaluate the visible quality of the surface layer.

[0062] 2. Principle and Equipment

[0063] The acquisition of washed stone surface layer images is the basis for the acquisition of the two-dimensional geometric contour of the exposed aggregate, the extraction of geometric morphological features, and the visual evaluation of the exposed aggregate on the surface layer. The CCD (Charge-coupled Device) is used to collect scattered signals, and the image processing technology is applied to extract the actual scattered light energy distribution. The proposed algorithm can be used to calculate the particle size distribution of the national standard particulate matter. The handheld portable device used in this embodiment. The device has the functions of target area image acquisition, image processing, morphological feature analysis, etc., and also has the characteristics of portability and easy operation. The shell of the device is made of high-strength ABS material and is printed in one piece using 3D printing technology. The internal arrangement includes a camera, a Raspberry Pi motherboard system, a 5-inch display screen, and a lithium battery. The image resolution is greater than 1080p, and the delay rate of taking photos is less than 1s; the working time of the device is not less than 4h. According to the construction process of the washed stone surface layer, the imaging device is directly used based on the device to obtain the washed stone surface layer image, and image recognition and processing are performed.

[0064] The staff at the construction site hold the device to take photos of the target area of ​​the washed stone surface layer. The photos are stored in the motherboard flash card. The motherboard will then call the image, pre-process the image, analyze and calculate the image, etc. After the morphological feature calculation is completed, the statistical feature analysis is further performed, and then the normal distribution characteristics of the aggregate exposure in the target area are output. The construction personnel determine and adjust the construction process according to the feedback of the image analysis results until the optimal visual perception is achieved and the design visual perception requirements are met.

[0065] 3. Particle morphology processing

[0066] 1. Image preprocessing and verification

[0067] First, the equipment is used to take photos of the concrete aggregate to be put into use. When taking photos, a ruler is placed in the target area for subsequent image preprocessing and aggregate particle size verification.

[0068] 1) Image preprocessing

[0069] Affected by environmental and human factors, image distortion may occur due to tilt during image capture. To reduce the impact, the image is preprocessed. The most commonly used preprocessing operations include image cropping, correction, smoothing, and enhancement. This embodiment uses a similar method to correct the image, mainly rotating and cropping the captured image. After the image is corrected, a new coordinate system is re-established for the corrected image to more accurately define the scope of the study area. That is, a certain scale line is used as the relative coordinate origin, the x direction is defined as the image width direction; the y direction is the image height direction. The establishment of new coordinates provides a basis for subsequent particle morphology feature extraction, exposure degree calculation, etc. See Figure 4 shown.

[0070] 2) Aggregate particle size verification

[0071] According to the analysis of the original aggregate grading to be used in the washed stone surface layer, the average particle size D 50 =6.7mm. The ruler comparison method is used to verify the aggregate size to be put into use, and the concrete aggregate is analyzed according to the elliptical shape. According to its contour shape, any two points on the shape are extracted, and the maximum distance between the two points is the exposed aggregate diameter. After calculation and analysis, the median aggregate size D 50 =6.44mm. Compared with the original grading diameter, it is about 3.8% smaller, which is related to the mutual covering between particles.

[0072] The shape coefficient of the particles is calculated by the ratio of the perimeter p of the contour line of a two-dimensional object to the same area A. After calculation, the average value of the shape coefficient of the aggregate in the figure is about 2.05.

[0073] The verification results of particle diameter and shape coefficient show that the image proportions are correct and the calculation method has a small error, so the next step of analysis and calculation can be carried out.

[0074] 3) Image Segmentation

[0075] Since there are still rulers in the processed image, the image needs to be further cut and segmented. That is, the target area image is divided into two parts, one part contains the ruler; the other part only contains aggregate information. The ruler part is cut off to avoid affecting the calculation accuracy. Subsequent calculations are all based on the part containing only aggregate information and are output in proportion. Figure 5 shown.

[0076] 2. Image Optimal Threshold Segmentation Method

[0077] 1) Grayscale image of target area

[0078] After preprocessing, the target area is still a color image, which contains aggregate and other information. In order to simplify the complex process of image processing and improve computational efficiency, the color image needs to be converted into a grayscale image. By covering different grayscale levels, the detailed information in the image is then reflected. In the new coordinate system, by establishing an m×n matrix, each element in the washed stone surface layer is changed into a two-dimensional function f(x, y) of the digital image and expressed in matrix form. There are many grayscale processing methods, such as the maximum value method, the average value method, etc. This embodiment uses the weighted method for the washed stone surface layer image, that is, Figure 5 (a) Perform grayscale processing to convert the color image into a grayscale image. See formula (2):

[0079] f g (x,y)=w i f i (x,y),i=1…3 (2)

[0080] In the formula, w i is the weight of chromaticity. i represents color, such as 1 to 3 represent red, blue and green respectively. i (x, y) is the grayscale image function. The image after grayscale processing is shown in Figure 6 shown.

[0081] 2) Optimal Threshold Segmentation Method

[0082] According to the gray value characteristics of the image, the aggregate morphology in the image is further subjected to noise reduction, namely threshold segmentation processing, to delete some small, isolated targets in point form, and to fill the voids. After processing, the perimeter area information of each connected area can be obtained, and then the shape parameters of the exposed aggregate can be calculated. Therefore, it is particularly necessary to select a reasonable and optimal threshold. Generally speaking, the morphological threshold is close to the lower limit, and the morphological feature extraction of the exposed aggregate in the image is more detailed. However, overlapping and adhesion features appear for exposed aggregates that are closer, resulting in extreme values ​​of parameter values ​​and reduced calculation accuracy. When the morphological threshold is close to the upper limit, although the overlapping and adhesion features of the exposed aggregate are improved, some small particles are corroded by the image, which also reduces the calculation accuracy and causes errors in particle size and shape coefficient. By adjusting the adaptive threshold, the adhesion phenomenon between particles can be reduced, and the threshold segmentation processing of adhesion cells can be realized. In view of this, the present invention proposes a method for selecting the optimal solution for morphological threshold discrimination. It can be seen from formula (2) that the gray image function is f i (x, y), grayscale image gradient function g(x, y). Define the set of all pixels as g(i), the grayscale threshold of the target area image as k; the corresponding pixel set is C(k j ), K j is the grayscale increase value. Then K j The pixel set is shown in the following formula:

[0083] C(k j )=[g(x,y)≤k] (3)

[0084] Taking j as the reference, the geometry of pixels less than a certain threshold is expressed as follows.

[0085] g(k)=g(i)∪C(k j ) (4)

[0086] By continuously adjusting the set of k, g(x,y) = k, and calculating the maximum diameter and shape coefficient of the aggregate, and comparing them with the above image verification results, the dimensionless parameters are shown in formula (5):

[0087]

[0088] Where L is the maximum diameter and S is the shape factor. m is the parameter value of the standard ellipse, mk The parameter values ​​for different threshold values ​​are then determined to obtain the optimal threshold k.

[0089] The threshold range is defined as k = 5 to 85. The results of different threshold processing are compared below. Figure 7 .

[0090] After optimization, the optimal segmentation threshold value for this calculation is no more than 30. In this embodiment, k=20 is selected. This threshold value also ensures that particles in the image are over-segmented or under-segmented during segmentation, and the relative error of the calculation is minimized.

[0091] 3. Particle morphology processing

[0092] After thresholding, some aggregate images tend to stick together, that is, the white areas representing the target particles in the image are interconnected, and the image needs to be further processed to achieve single particle cutting, such as corrosion and dilation operations in morphological processing. Corrosion and dilation can improve the regional structural morphology in the image, but will also cause a certain reduction in effective information and enhancement of irrelevant information. Therefore, corrosion and dilation are often used as a combination to reduce the loss of image information. First, connected areas with an area less than the specified value are removed, and the missing areas are filled; then the connected areas in the image are merged. After calculation and processing, the relationship between the area of ​​the merged area and the shape coefficient under different threshold conditions can be seen. Figure 8 .Depend on Figure 8 It can be seen that after exceeding a certain threshold, the merging range no longer significantly affects the change of the shape coefficient. This result also proves that the calculation and analysis using k = 20 is reasonable.

[0093] After selecting the threshold, extract the attributes of different regions in the image. Then extract the geometric center points of the merged regions. Fig. 9 As shown, the distance between adjacent areas can be calculated.

[0094] 4. Evolution characteristics of aggregate exposure

[0095] According to the construction process of the washed stone surface layer, there are three stages: initial washing, secondary washing and waiting for forming, which are represented by symbols t1, t2 and t3 respectively.

[0096] First, clean up the dust and debris on the surface of the construction area. Screen the crushed stone at the same time, mix the screened crushed stone with M25 cement mortar, and transport the mixed washed stone mixture to the construction paving area for standby. The washed stone mixture is evenly spread in the construction area. After the washed stone surface is paved, the washed stone surface layer is leveled with a leveling roller, and the mud in the washed stone pavement surface layer is squeezed out with the vibration function, and then manually compacted and smoothed. This can be repeated several times to make its density meet the design requirements. Curing for half an hour. Then use a brush dipped in water to brush off the cement slurry, and rinse the surface cement slurry to expose the aggregate. The on-site staff will place a ruler in the target area for photography, use the equipment to take pictures of each stage respectively, and store the pictures on the mainboard flash card. The mainboard program runs to complete the image call, image processing, i.e. analysis and calculation. After the calculation is completed, the normal distribution characteristics of the aggregate exposure in the target area are output. The staff will carry out the next step of washing work based on the feedback of the exposure results.

[0097] The other two stages of construction work are fine washing and shaping of the washed stone surface layer. The processing process is the same as above and will not be repeated one by one. The images of the washed stone surface layers at each stage are shown in Fig.10 shown.

[0098] Depend on Fig.10 It can be seen that the aggregate exposure is small in the initial stage. As the degree of water washing increases, the aggregate exposure gradually increases. According to the above processing flow, the images of each stage are processed. Such as image preprocessing, image scale adjustment, image processing and morphological feature extraction using the optimal threshold k = 20, etc., to obtain the evolution characteristics of the geometric characteristics of the exposed aggregate at each stage. The evolution characteristics of the morphological coefficient of the exposed aggregate at each stage are shown in Fig.11 .

[0099] Depend on Fig.11 It can be seen that in the initial water washing stage, the exposed aggregates are less and the aggregate shape is smaller, so the shape coefficient is smaller. As the degree of water washing increases, the number and degree of exposed aggregates gradually increase, and the shape coefficient also increases accordingly. In the waiting forming stage, the shape coefficient increases further, with an average value of about 1.75, gradually approaching the original aggregate shape coefficient, which is about 16% smaller, which is related to the concrete filling between aggregates. In fact, the evolution of the maximum diameter of the exposed aggregate also conforms to similar characteristics, which will not be repeated.

[0100] 5. Evaluation of the visual quality of washed stone surface

[0101] According to the above statistics, the maximum diameter L m , Exposed aggregate geometric center distance D e Parameters, calculate the probability density of the ξ normal distribution. Based on the normal distribution, when the mean ξ satisfies the 95% confidence interval, the washed stone surface layer meets the requirements and has the highest visual quality, which is used as an evaluation index for the visual quality of aggregate exposure. Fig.12 shown.

[0102] Depend on Fig.12 It can be seen that its evolution characteristics are similar to the shape coefficient. With the increase of water washing degree, the aggregate exposure gradually increases, and the mean value of ξ also increases accordingly.

[0103] Based on the normal distribution, with a 95% confidence interval as the criterion, the mean is about ξ = 0.8. The aggregate spacing is about 0.8 to 1.1 times the particle diameter. The washed stone surface layer is washed and formed, and the appearance can be optimized.

[0104] This embodiment also provides a system for evaluating the perceptual quality of a washed stone surface layer, including a perceptual quality evaluation module, wherein the perceptual quality evaluation module is used to execute the method for evaluating the perceptual quality of a washed stone surface layer.

[0105] This embodiment also provides a non-transitory computer-readable storage medium, which stores computer instructions. When the computer instructions are executed by a processor, the method for evaluating the visual quality of a washed stone surface layer is implemented.

[0106] Ginseng Fig.13 As shown, this embodiment also provides an electronic device, including:

[0107] A memory 201 and a processor 202, wherein the memory 201 and the processor 202 are communicatively connected to each other, the memory 201 stores computer instructions, and the processor 202 executes the method for evaluating the visual quality of a washed stone surface layer by executing the computer instructions.

[0108] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the technical principles of the present invention, and these improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for evaluating the visual quality of a washed stone surface layer, characterized in that: include: Step 1, morphological processing is performed on the washed stone surface layer image to extract the geometric features of the exposed aggregate in the image; The geometric features include the exposed geometric shape of the aggregate, the geometric center, and the relative distance between the aggregates; Step 2: Based on the geometric features of the exposed aggregate in the image, define the longest diagonal angle as the maximum diameter L of the exposed aggregate. max , the geometric center distance D between any two exposed aggregates e , calculate the relative spacing of exposed aggregate ξ=L max / D e ; Step 3, calculate the probability density of the normal distribution of ξ. If ξ conforms to the normal distribution and the mean of ξ satisfies the 95% confidence interval, it is determined that the washed stone surface layer meets the requirements and has the highest visual quality. The morphological processing in step 1 includes image preprocessing, aggregate particle size verification, image segmentation, and particle morphological processing; The image preprocessing comprises: Correct the washed stone surface layer image, and after the image correction is completed, re-establish a new coordinate system for the corrected image; In step 1, the image optimal threshold segmentation method is used to perform threshold segmentation processing on the aggregate morphology, including: 1) Convert color images to grayscale images In the new coordinate system, by establishing an m×n-order matrix, each element in the washed stone surface layer is changed into a two-dimensional function f(x,y) of a digital image and expressed in matrix form; The weighted method is used to process the grayscale of the washed stone surface layer image and convert the color image into a grayscale image as follows: f g (x,y)=w i f i (x,y),i=1…3 In the formula, w i is the weight of chromaticity; i represents color, 1 to 3 represent red, blue and green respectively; f i (x,y) is the grayscale image function; 2) According to the gray value characteristics of the image, the aggregate morphology in the image is segmented by threshold value, the point-like, isolated smaller targets are deleted, the holes are filled, and the perimeter area information of each connected area is obtained; the morphological threshold is used to distinguish and select the optimal solution. Let the gray image function be f i (x, y), the grayscale image gradient function is g(x, y), the set of all pixels is defined as g(i), the grayscale threshold of the target area image is k, and its corresponding pixel set is C(k j ), where k j is the grayscale increase value, then k j The pixel set is: C(k j )=[g(x,y)≤k] Taking j as the reference, the geometry of pixels less than a certain threshold is expressed as follows: g(k)=g(i)∪C(k j ) By continuously adjusting the set of k, g(x,y) = k, and calculating the maximum diameter and shape coefficient of the aggregate, and comparing them with the image verification results, the dimensionless parameters are as follows: Where L is the maximum diameter, S is the shape factor, and the subscript m is the parameter value of the standard ellipse, mk Parameter values ​​for different threshold values, and then determine the optimal threshold k=20.

2. The method for evaluating the perceptual quality of washed stone surface layer according to claim 1, characterized in that: The washed stone surface layer image in step 1 is acquired using a portable imaging device.

3. The method for evaluating the perceptual quality of washed stone surface layer according to claim 1, characterized in that: The image segmentation comprises: The target area image is divided into two parts, one part contains only the scale part; the other part contains only the aggregate information part. The scale part is cut off to avoid affecting the calculation accuracy. Subsequent calculations are all performed using the part containing only the aggregate information for calculation and analysis, and output in proportion.

4. The method for evaluating the perceptual quality of washed stone surface layer according to claim 1, characterized in that: The image segmentation comprises: The particle morphology processing includes: The image is further processed to achieve single particle cutting. First, the connected regions with an area smaller than a specified value are removed, the missing regions are filled, and then the connected regions in the image are merged.

5. A system for evaluating the perceived quality of washed stone surface, characterized in that: It comprises a perceptible quality evaluation module, and the perceptible quality evaluation module is used to execute the perceptible quality evaluation method of a washed stone surface layer as described in any one of claims 1-4.

6. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, which, when executed by a processor, implement a method for evaluating the perceptual quality of a washed stone surface layer as described in any one of claims 1 to 4.

7. An electronic device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes a method for evaluating the visual quality of a washed stone surface layer as described in any one of claims 1 to 4 by executing the computer instructions.

Images