Limestone decomposition effect detection method based on machine vision
Through machine vision technology, the surface image of cement clinker is analyzed and the limestone decomposition effect is evaluated, which solves the problem of difficulty in detecting limestone decomposition effect in the existing technology, improves detection accuracy and efficiency, and helps improve the quality of cement clinker.
Patent Information
- Application Number
- CN202411304835.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-19
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-09-19
AI Technical Summary
The prior art is difficult to effectively detect the decomposition effect of limestone, which affects the quality of cement clinker, and the manual inspection cost is high and inaccurate.
Using a machine vision-based method, the complexity of the crack and pore areas is determined by the grayscale value and gradient value of the surface image of the cement clinker sample under a microscope, the complexity of the crack and pore areas is calculated, the target parameter value is calculated, and the decomposition effect of limestone is evaluated.
It improves the detection accuracy and efficiency of limestone decomposition effect, reduces the detection cost, provides a reference for the production environment of cement clinker and the adjustment of raw material ratio, and improves the quality of cement clinker.
Smart Images

Figure CN118823022B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of image data processing, and in particular to a method for detecting limestone decomposition effects based on machine vision. Background Art
[0002] Cement clinker is a key raw material in the cement production process. The characteristics of cement clinker have a decisive influence on the quality and performance of cement. The main mineral components of cement clinker include tricalcium silicate, dicalcium silicate, tricalcium aluminate and tetracalcium aluminoferrite. The raw materials for the production of cement clinker mainly include limestone and clay. Limestone, whose main component is calcium carbonate, decomposes into calcium oxide and carbon dioxide at high temperature; silicate minerals in clay decompose to release silicon oxide and aluminum oxide; calcium oxide reacts with silicon oxide and aluminum oxide to form minerals such as calcium silicate, and finally forms cement clinker.
[0003] For example, in a Chinese patent application document with publication number CN112521033A, a cement clinker is provided, which is made into blocks through processing equipment, and carbon dioxide is filled into the cement clinker during the preparation process to form pores. The cement clinker is composed of the following raw materials in parts by weight: 100-110 parts of limestone, 40-50 parts of magnesite, 30-40 parts of water-based resin, 20-26 parts of carbide slag, 4-6 parts of yellow sandstone, 2.5-4.5 parts of clay, 1.5-3 parts of coal gangue, 1-2 parts of glass slag, 0.2-0.5 parts of vanadium slag and 0.1-0.2 parts of stabilizer; the specifications of the carbide slag, vanadium slag and glass slag are all 30-60 mesh; the storage environment of the cement clinker is dry and ventilated.
[0004] When the ventilation conditions and the ratio of raw materials in the cement clinker production process are fixed, since the calcium oxide in the cement clinker production process mainly comes from the decomposition of limestone, the quality of cement clinker is directly affected by the decomposition effect of limestone. Therefore, it is necessary to test the decomposition effect of limestone in order to improve the quality of the produced cement clinker. Summary of the invention
[0005] In order to detect the decomposition effect of limestone so as to improve the quality of the produced cement clinker, the present application provides a limestone decomposition effect detection method based on machine vision, comprising: obtaining a surface image of a sample of target cement clinker under a microscope, wherein the target cement clinker is produced using limestone of a target proportion; determining crack pixels from the surface image according to the grayscale value and gradient value of the pixel points in the surface image, and forming a crack line from adjacent crack pixels, so as to determine a first parameter value according to the crack line in the surface image, and the first parameter value is used to characterize the complexity of the cracks in the surface image; determining multiple closed areas in the surface image, and taking the closed areas whose similarity to the class ellipse is greater than a preset similarity threshold as pore areas, so as to determine a second parameter value according to the area and number of the pore areas, and the second parameter value is used to characterize the complexity of the pore areas in the surface image; determining a target parameter value according to the sum of the first parameter value and the second parameter value, and determining an evaluation value of the decomposition effect of the limestone according to the target parameter value and a preset corresponding relationship, and the preset corresponding relationship is used to characterize the corresponding relationship between the evaluation value and the target parameter value under the target proportion.
[0006] In this way, a surface image of a target cement clinker sample is obtained under a microscope; the complexity of cracks on the surface of the cement clinker sample and the complexity of the pore area are determined based on the surface image, and the target parameter value can be determined to determine the evaluation value of the decomposition effect of limestone, thereby realizing the detection of the decomposition effect of limestone in the production process of cement clinker. Since the calcium oxide in the production process of cement clinker mainly comes from the decomposition of limestone, the quality of cement clinker is directly affected by the decomposition effect of limestone. Therefore, by detecting the decomposition effect of limestone, it is possible to provide valuable reference for the production environment of cement clinker or the adjustment of raw material ratio, which is helpful to improve the quality of the produced cement clinker.
[0007] At the same time, since the decomposition effect of limestone is determined based on the surface image of the target cement clinker sample under a microscope, compared with manual observation of the surface image of cement clinker under a microscope, it can not only reduce the manual work intensity, but also improve the accuracy of the detection of the decomposition effect of limestone and reduce the cost of detecting the decomposition effect of limestone.
[0008] Optionally, determining crack pixels from the surface image based on the grayscale values and gradient values of pixels in the surface image includes: determining an edge line in the surface image, and for a target pixel on the edge line, determining a first difference value in grayscale value between the target pixel and adjacent pixels in the target direction, and determining a second difference value in gradient value between the target pixel and adjacent pixels in the target direction; determining a third parameter value of the target pixel based on the first difference value and the second difference value, the third parameter value being used to characterize the probability that the target pixel is a crack pixel; and taking a pixel on the edge line of the surface image whose third parameter value is greater than a preset parameter threshold as a crack pixel in the surface image.
[0009] Optionally, the target direction is a gradient direction of a target pixel point, and determining a third parameter value of the target pixel point according to the first difference degree value and the second difference degree value includes: , P is the third parameter value of the target pixel, is the normalization function, H is the average gray value of the target pixel and its adjacent pixels in the target direction, h is the gray value of the target pixel, is the first difference value, T is the average gradient value between the target pixel and the adjacent pixels in the target direction, t is the gradient value of the target pixel, is the second difference value, n is the number of pixels on the edge line where the target pixel is located, is the gradient direction angle of the i-th pixel on the edge line where the target pixel is located, r is the gradient direction angle of the target pixel, To take the absolute value operation.
[0010] In this way, the target pixel point is compared with the adjacent pixel points in the gradient direction of the target pixel point, such as the degree of difference between the grayscale values, or the degree of difference between the grayscale gradients; combined with the difference between the grayscale values of other pixels on the edge line where the target pixel point is located, it is helpful to determine a more accurate third parameter value of the target pixel point, so that the cracks in the surface image can be more accurately determined through the third parameter value.
[0011] Optionally, determining the first parameter value according to the crack line in the surface image includes: ,in, is the first parameter value, norm is the normalization function, m is the number of crack pixels in the surface image, M is the number of pixels in the surface image, is the number of crack lines in the surface image.
[0012] In this way, by combining the number of crack lines in the surface image and the area ratio occupied by the cracks, more accurate crack features in the surface image can be determined, so that the crack complexity of the surface image can be better characterized by the obtained first parameter value.
[0013] Optionally, the similarity between the closed area and the ellipse-like region is determined in the following manner: ), where K is the similarity between the closed area and the ellipse-like area, exp is an exponential function with a natural constant as the base, and S is the number of pixels located at the edge of the closed area. is the horizontal coordinate of the f-th pixel point on the closed area in the target coordinate system, is the ordinate of the f-th pixel point on the closed area in the target coordinate system, is the distance between the two pixel points where the edge of the closed area intersects the horizontal axis of the target coordinate system. It is the distance between the two pixel points where the edge of the closed area intersects with the vertical axis of the target coordinate system; the target coordinate system takes the line segment corresponding to the maximum distance between the two pixel points located at the edge of the closed area as the horizontal axis, and the target coordinate system takes the midpoint of the line segment as the coordinate origin.
[0014] In this way, by establishing a coordinate system for the closed area and comparing the pixel points on the edge of the closed area with the ellipse formula, the similarity between the closed area and the ellipse-like area can be better determined, thereby better distinguishing between the pore area and the closed area outside the pores. Since a more accurate pore area can be determined, it is helpful to determine a more accurate decomposition effect of limestone.
[0015] Optionally, determining the second parameter value according to the area and number of the pore regions includes: ,in, is the second parameter value, norm is the normalization function, a is the total area of all pore regions in the surface image, A is the area of the surface image, is the number of pore areas in the surface image.
[0016] In this way, by combining the total number of all pore regions and the total area of all pore regions, the pore characteristics of the surface of cement clinker can be better characterized.
[0017] Optionally, determining the target parameter value according to the sum of the first parameter value and the second parameter value includes: , where W is the target parameter value, exp is an exponential function with a natural constant as the base, Q1 is the first parameter value, and Q2 is the second parameter value.
[0018] In this way, since the first parameter value is used to characterize the complexity of cracks in the surface image and the second parameter value is used to characterize the complexity of the pore area in the surface image, combining the first parameter value and the second parameter value can combine the pore characteristics and crack characteristics of the cement clinker obtained according to the target ratio of limestone, so as to determine the evaluation value of the decomposition effect of the limestone.
[0019] Optionally, the preset corresponding relationship is obtained in the following manner: obtaining an evaluation value determined based on a historical inspection result of a target sample, wherein the target sample is a sample of cement clinker produced from limestone according to a target ratio; obtaining a target parameter value determined based on a historical surface image of the target sample under a microscope; and obtaining the preset corresponding relationship based on the target parameter values corresponding to target samples with different evaluation values.
[0020] In this way, a preset correspondence can be obtained based on the evaluation value determined by the historical test results of the target sample and the target evaluation value corresponding to the target sample. When the limestone decomposition effect is subsequently detected in the production process of cement clinker, the preset correspondence and the surface image of the cement clinker can be used to obtain the evaluation value of the limestone decomposition effect without the need to conduct chemical composition analysis, petrographic analysis, and physical property testing on the cement clinker sample again; or, there is no need to test the undecomposed lime in the cement clinker, which can improve the detection efficiency of the limestone decomposition effect and reduce the cost of detecting the limestone decomposition effect.
[0021] The technical solution provided by the embodiments of the present application may include the following beneficial effects: obtaining a surface image of a target cement clinker sample under a microscope; determining the complexity of cracks on the surface of the cement clinker sample and the complexity of the pore area based on the surface image, and being able to determine the target parameter value to determine the evaluation value of the decomposition effect of limestone, thereby realizing the detection of the decomposition effect of limestone in the production process of cement clinker. Since the calcium oxide in the production process of cement clinker mainly comes from the decomposition of limestone, the quality of cement clinker is directly affected by the decomposition effect of limestone. Therefore, by detecting the decomposition effect of limestone, it is possible to provide valuable reference for the production environment of cement clinker or the adjustment of the raw material ratio, which is helpful to improve the quality of the produced cement clinker.
[0022] At the same time, since the decomposition effect of limestone is determined based on the surface image of the target cement clinker sample under a microscope, compared with manual observation of the surface image of cement clinker under a microscope, it can not only reduce the manual work intensity, but also improve the accuracy of the detection of the decomposition effect of limestone and reduce the cost of detecting the decomposition effect of limestone.
[0023] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0025] Figure 1 The present invention is a flow chart showing a method for detecting limestone decomposition effect based on machine vision according to an exemplary embodiment. DETAILED DESCRIPTION
[0026] Here, exemplary embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.
[0027] First, the application scenario of the embodiment of the present application is briefly introduced. In the application scenario of the present application, in the process of producing cement clinker using limestone and clay, since the calcium oxide in the cement clinker production process mainly comes from the decomposition of limestone, the quality of cement clinker is directly affected by the decomposition effect of limestone. Therefore, it is necessary to detect the decomposition effect of limestone in order to improve the quality of the produced cement clinker.
[0028] In response to the above technical problems, an embodiment of the present application provides a method for detecting the decomposition effect of limestone based on machine vision, which obtains a surface image of a target cement clinker sample under a microscope; the complexity of cracks on the surface of the cement clinker sample and the complexity of the pore area are determined according to the surface image, and the target parameter value can be determined to determine the evaluation value of the decomposition effect of the limestone, thereby realizing the detection of the decomposition effect of limestone in the production process of cement clinker. Since the calcium oxide in the production process of cement clinker mainly comes from the decomposition of limestone, the quality of cement clinker is directly affected by the decomposition effect of limestone. Therefore, by detecting the decomposition effect of limestone, it is possible to provide valuable reference for the production environment of cement clinker or the adjustment of raw material ratio, which is helpful to improve the quality of the produced cement clinker.
[0029] At the same time, since the decomposition effect of limestone is determined based on the surface image of the target cement clinker sample under a microscope, compared with manual observation of the surface image of cement clinker under a microscope, it can not only reduce the intensity of manual work, but also improve the accuracy of detection of the decomposition effect of limestone.
[0030] Figure 1 is a flow chart of a method for detecting limestone decomposition effect based on machine vision according to an exemplary embodiment. Figure 1As shown, the method includes the following steps.
[0031] In step S101 , a surface image of a target cement clinker sample under a microscope is acquired.
[0032] The target cement clinker is produced using limestone of a target proportion. For example, cement clinker is produced using production raw materials including limestone of a target proportion. Limestone and clay react at high temperature to obtain cement clinker composed of tricalcium silicate, dicalcium silicate, tricalcium aluminate and tetracalcium aluminoferrite.
[0033] The surface image of the target cement clinker sample under a microscope can be obtained by using equipment such as a scanning electron microscope and a transmission electron microscope. Since the calcium oxide required for the production of cement clinker mainly comes from the decomposition of limestone, the characteristics of the cement clinker obtained under different decomposition effects of limestone are different. Therefore, obtaining the surface image of the target cement clinker sample under a microscope is helpful to determine the decomposition effect of limestone in the production process of cement clinker through the surface image.
[0034] In step S102, crack pixels are determined from the surface image according to the grayscale values and gradient values of the pixels in the surface image, and adjacent crack pixels are combined into a crack line to determine a first parameter value according to the crack line in the surface image.
[0035] The first parameter value is used to characterize the complexity of cracks in the surface image. There may be some cracks on the surface of the produced cement clinker. The content of calcium oxide obtained under different decomposition effects of limestone is different, which makes the cracks on the surface of the produced cement clinker different.
[0036] Alternatively, in the production process of cement clinker, the decomposition effect of limestone may be affected by factors such as the temperature of the production environment and the calcination time; different decomposition effects of limestone indicate that the temperature and calcination time of the cement clinker production process are different, and at different production temperatures and calcination times, the cracks on the surface of the cement clinker are different.
[0037] Here, the crack lines existing in the surface image are determined, and the first parameter value is determined according to the crack lines in the surface image. The first parameter value is used to characterize the complexity of the cracks in the surface image, and the first parameter value can provide a reference for determining the decomposition effect of limestone.
[0038] In step S103, multiple closed areas in the surface image are determined, and closed areas whose similarity with the ellipse-like image is greater than a preset similarity threshold are taken as pore areas, so as to determine the second parameter value according to the area and number of the pore areas.
[0039] The second parameter value is used to characterize the complexity of the pore area in the surface image. Under different decomposition effects of limestone, the same content of limestone produces different amounts of calcium oxide and carbon dioxide after decomposition, resulting in different pore characteristics on the surface of the obtained cement clinker.
[0040] Alternatively, the different decomposition effects of limestone are affected by the production temperature and the calcination time of the limestone. Under the different decomposition effects of limestone, the production temperature of cement clinker and the calcination time of limestone are different, resulting in different pore characteristics on the surface of cement clinker.
[0041] Here, multiple closed areas in the surface image are determined, and the closed areas whose similarity with the class ellipse is greater than a preset similarity threshold are taken as pore areas, so as to determine the second parameter value according to the area and number of the pore areas. Since the second parameter value is determined according to the area and number of the pore areas in the surface image, the second parameter value can better characterize the characteristics of the pores of the cement clinker sample. Therefore, the second parameter value can provide a reference for determining the decomposition effect of limestone.
[0042] The preset similarity threshold may be between 0.5 and 0.75.
[0043] In step S104, a target parameter value is determined according to the sum of the first parameter value and the second parameter value, and an evaluation of the decomposition effect of limestone is determined according to the target parameter value and a preset corresponding relationship.
[0044] The preset corresponding relationship is used to characterize the corresponding relationship between the evaluation value and the target parameter value under the target ratio. For example, the preset corresponding relationship can be obtained in the following way: obtaining the evaluation value determined according to the historical inspection result of the target sample, wherein the target sample is a sample of cement clinker produced by limestone according to the target ratio; obtaining the target parameter value determined according to the historical surface image of the target sample under a microscope; obtaining the preset corresponding relationship according to the target parameter value corresponding to the target sample with different evaluation values produced by limestone according to the target ratio.
[0045] Multiple cement clinker samples such as L1, L2, L3, L4 and L5 can be acquired in advance. The multiple cement clinkers are obtained by producing cement clinker with production raw materials composed of limestone in the same proportion. Referring to the process of obtaining the target parameter values of the cement clinker samples in the embodiment of the present application, the target parameter values corresponding to the multiple cement clinker samples such as L1, L2, L3, L4 and L5 can be obtained respectively according to the historical surface images of the target samples under a microscope.
[0046] Multiple cement clinker samples such as L1, L2, L3, L4 and L5 can be inspected. For example, chemical composition analysis, petrographic analysis and physical property testing can be performed on the cement clinker samples. Since the chemical composition, surface characteristics and physical properties of cement clinker obtained under different decomposition effects of limestone are different, the limestone decomposition effect corresponding to the cement clinker sample can be determined.
[0047] Alternatively, under different decomposition effects of limestone, the amount of undecomposed limestone remaining in the cement clinker is different. Therefore, the undecomposed lime in the cement clinker can be tested to obtain the decomposition effect of limestone in the production process of the cement clinker sample. For example, an evaluation value of the decomposition effect of the cement clinker sample can be obtained, and different evaluation values can correspond to different decomposition effects.
[0048] In this way, a preset correspondence can be obtained based on the evaluation value determined by the historical test results of the target sample and the target evaluation value corresponding to the target sample. When the limestone decomposition effect is subsequently detected in the production process of cement clinker, the preset correspondence and the surface image of the cement clinker can be used to obtain the evaluation value of the limestone decomposition effect without the need to conduct chemical composition analysis, petrographic analysis, and physical property testing on the cement clinker sample again; or, there is no need to test the undecomposed lime in the cement clinker, which can improve the detection efficiency of the limestone decomposition effect and reduce the cost of detecting the limestone decomposition effect.
[0049] Through the machine vision-based limestone decomposition effect detection method provided in the embodiment of the present application, a surface image of a target cement clinker sample under a microscope is obtained; the complexity of cracks on the surface of the cement clinker sample and the complexity of the pore area are determined based on the surface image, and the target parameter value can be determined to determine the evaluation value of the limestone decomposition effect, thereby realizing the detection of the limestone decomposition effect in the cement clinker production process. Since the calcium oxide in the cement clinker production process mainly comes from the decomposition of limestone, the quality of the cement clinker is directly affected by the decomposition effect of the limestone. Therefore, by detecting the decomposition effect of the limestone, it is possible to provide a valuable reference for the production environment of the cement clinker or the adjustment of the raw material ratio, which is helpful to improve the quality of the produced cement clinker.
[0050] At the same time, since the decomposition effect of limestone is determined based on the surface image of the target cement clinker sample under a microscope, compared with manual observation of the surface image of cement clinker under a microscope, it can not only reduce the intensity of manual work, but also improve the accuracy of detection of the decomposition effect of limestone.
[0051] In one embodiment, the method of determining crack pixels from a surface image based on grayscale values and gradient values of pixels in the surface image includes: determining an edge line in the surface image, and for a target pixel on the edge line, determining a first difference value in grayscale values between the target pixel and adjacent pixels in the target direction, and determining a second difference value in gradient values between the target pixel and adjacent pixels in the target direction; determining a third parameter value of the target pixel based on the first difference value and the second difference value, the third parameter value being used to characterize the probability that the target pixel is a crack pixel; and taking a pixel on the edge line of the surface image whose third parameter value is greater than a preset parameter threshold as a crack pixel in the surface image.
[0052] The gradient value of the pixel where the crack is located in the surface image of cement clinker will be larger than the gradient value of the pixel where no crack occurs. The gradient value of the pixel can reflect the change in the grayscale value of the pixel. Moreover, the texture complexity of the pixel where the crack is located will be higher than the texture complexity of the pixel where no crack occurs. Therefore, the crack pixel can be determined from the surface image through the grayscale value and gradient value of the pixel in the surface image, so as to analyze the crack characteristics of the cement clinker.
[0053] The edge lines or edge pixels in the image can be determined by processing the image with gradient operators such as the Sobel operator, Prewitt operator, Roberts operator, and Laplacian operator; or, the image can be subjected to Hough transform to determine the edge pixels or edge lines in the image.
[0054] The cracks existing on the surface of cement clinker will be more dispersed than normal edges, or, compared with the pixels at the normal edges of the surface of cement clinker, the pixels at the cracks will have greater differences in texture features from the pixels at the normal edges. Therefore, in the embodiment of the present application, the target pixel point is compared with the adjacent pixel points in the target direction, which can more accurately determine the probability that the pixel point in the surface image of the cement clinker belongs to a crack pixel point.
[0055] The greater the difference between the target pixel point and other adjacent pixel points in the target direction, the greater the probability that the target pixel point is a crack pixel point; on the contrary, the smaller the difference between the target pixel point and other adjacent pixel points in the target direction, the smaller the probability that the target pixel point is a crack pixel point; taking the pixel point whose third parameter value is greater than the preset parameter threshold on the edge line of the surface image as the crack pixel point in the surface image, the edge pixel points in the surface image can be screened to avoid the influence of other edge pixel points in the surface image that do not have cracks on the crack characteristics of the cement clinker.
[0056] The preset parameter threshold may be determined according to the value range of the target parameter value. For example, when the target parameter value is normalized to between 0 and 1, the preset parameter threshold may be between 0.4 and 0.6.
[0057] The first difference degree value in grayscale value between the target pixel and the adjacent pixels in the target direction may be, for example, the variance or standard deviation of the grayscale values of a plurality of pixels including the target pixel and the adjacent pixels of the target pixel in the target direction.
[0058] The second difference degree value in gradient values between the target pixel and the adjacent pixels in the target direction may be, for example, the variance or standard deviation of the gradient values of a plurality of pixels including the target pixel and the adjacent pixels of the target pixel in the target direction.
[0059] In one embodiment, the target direction is the gradient direction of the target pixel point, and determining the third parameter value of the target pixel point according to the first difference degree value and the second difference degree value includes: , P is the third parameter value of the target pixel, is the normalization function, H is the average gray value of the target pixel and its adjacent pixels in the target direction, h is the gray value of the target pixel, is the first difference value, T is the average gradient value between the target pixel and the adjacent pixels in the target direction, t is the gradient value of the target pixel, is the second difference value, n is the number of pixels on the edge line where the target pixel is located, is the gradient direction angle of the i-th pixel on the edge line where the target pixel is located, r is the gradient direction angle of the target pixel, To take the absolute value operation.
[0060] The normalization function may adopt a minimum-maximum normalization function and an inverse tangent function, etc., and the value of the normalization result is made between 0 and 1 while keeping the change direction of the variable of the normalization function consistent with that of the normalization result, so as to facilitate comparison between parameters.
[0061] Comparing the target pixel with the adjacent pixels in the gradient direction of the target pixel, such as the degree of difference between the grayscale values, or the degree of difference between the grayscale gradients; combining the difference between the grayscale values of other pixels on the edge line where the target pixel is located, helps to determine a more accurate third parameter value of the target pixel, so that the cracks in the surface image can be more accurately determined through the third parameter value.
[0062] In one embodiment, determining the first parameter value according to the crack line in the surface image includes: ,in, is the first parameter value, norm is the normalization function, m is the number of crack pixels in the surface image, M is the number of pixels in the surface image, is the number of crack lines in the surface image.
[0063] Adjacent crack pixel points in a surface image can form a crack line. The more crack lines there are in a surface image, the more complex the texture of the cracks in the surface image may be. The greater the proportion of the area of the crack lines in the surface image to the area of the entire surface image, the deeper the cracks in the surface image. Combining the number of crack lines in the surface image and the proportion of the area occupied by the cracks can determine more accurate crack features in the surface image, thereby better characterizing the crack complexity of the surface image through the obtained first parameter value.
[0064] In one embodiment, the similarity between the closed region and the ellipse-like region is determined by: ), where K is the similarity between the closed area and the ellipse-like area, exp is an exponential function with a natural constant as the base, and S is the number of pixels located at the edge of the closed area. is the horizontal coordinate of the f-th pixel point on the closed area in the target coordinate system, is the ordinate of the f-th pixel point on the closed area in the target coordinate system, is the distance between the two pixel points where the edge of the closed area intersects the horizontal axis of the target coordinate system. It is the distance between the two pixel points where the edge of the closed area intersects with the vertical axis of the target coordinate system; the target coordinate system takes the line segment corresponding to the maximum distance between the two pixel points located at the edge of the closed area as the horizontal axis, and the target coordinate system takes the midpoint of the line segment as the coordinate origin.
[0065] In the production process of cement clinker, carbon dioxide obtained from the decomposition of limestone will cause pores to form on the surface of the cement clinker. Limestone with the same ratio will have different contents of carbon dioxide under different decomposition effects, resulting in different characteristics of the pores produced on the surface of the cement clinker. For example, the number of pores on the surface of the cement clinker may be different, or the area occupied by the pores on the surface of the cement clinker may be different.
[0066] The pores formed on the surface of cement clinker due to the carbon dioxide produced by the decomposition of limestone will appear as circles or ellipses. The surface image of cement clinker may also have some closed areas other than the pores, for example, irregular polygonal closed areas of the cement clinker itself. In order to avoid the influence of closed areas other than the pores on the pore characteristics of cement clinker, the similarity between the closed area and the ellipse-like area can be determined.
[0067] For example, when the shape of the outer contour of the closed area is close to an ellipse, when a plane rectangular coordinate system is established with the line segment corresponding to the maximum distance between two pixel points of the closed area as the horizontal axis and the midpoint of the line segment as the coordinate origin, the plane rectangular coordinate system will have the major axis of the elliptical outer contour of the closed area as the horizontal axis, and the plane rectangular coordinate system will have the minor axis of the elliptical outer contour of the closed area as the vertical coordinate; in the calculation formula for the similarity K between the closed area and the ellipse-like in the embodiment of the present application, the difference between the coordinates of the pixel points located in the outer contour of the closed area and the coordinate points in the ellipse equation is small, so that the value of the obtained similarity K becomes larger.
[0068] For another example, when the shape of the outer contour of the closed area is an irregular polygon, when a plane rectangular coordinate system is established with the line segment corresponding to the maximum distance between two pixel points in the closed area as the horizontal axis and the midpoint of the line segment as the origin of the coordinates, since the shape of the outer contour of the closed area is quite different from that of an ellipse, the pixel points located on the outer contour of the closed area are not included in the calculation formula of the similarity K between the closed area and the ellipse-like area. The larger the value of this item, the smaller the value of the similarity K.
[0069] In this way, by establishing a coordinate system for the closed area and comparing the pixel points on the edge of the closed area with the ellipse formula, the similarity between the closed area and the ellipse-like area can be better determined, thereby better distinguishing between the pore area and the closed area outside the pores. Since a more accurate pore area can be determined, it is helpful to determine a more accurate decomposition effect of limestone.
[0070] In one embodiment, determining the second parameter value according to the area and number of the pore regions includes: ,in, is the second parameter value, norm is the normalization function, a is the total area of all pore regions in the surface image, A is the area of the surface image, is the number of pore areas in the surface image.
[0071] The pores formed by the carbon dioxide produced by the decomposition of limestone on the surface of cement clinker will appear circular or elliptical. The same proportion of limestone will produce different amounts of carbon dioxide under different decomposition effects. By combining the total number of all pore areas and the total area of all pore areas, the pore characteristics of the cement clinker surface can be better characterized.
[0072] In one embodiment, determining the target parameter value according to the sum of the first parameter value and the second parameter value includes: , where W is the target parameter value, exp is an exponential function with a natural constant as the base, Q1 is the first parameter value, and Q2 is the second parameter value.
[0073] In this way, since the first parameter value is used to characterize the complexity of cracks in the surface image and the second parameter value is used to characterize the complexity of the pore area in the surface image, combining the first parameter value and the second parameter value can combine the pore characteristics and crack characteristics of the cement clinker obtained according to the target ratio of limestone, so as to determine the evaluation value of the decomposition effect of the limestone.
[0074] It should be understood that the features of some embodiments of the various applications described herein may be combined with each other, unless specifically stated otherwise.
[0075] Although terms such as "first", "second", and "third" may be used herein to describe various components, parts, regions, layers, or sections, these components, parts, regions, layers, or sections are not limited to these terms. Instead, these terms are only used to distinguish one component, part, region, layer, or section from another component, part, region, layer, or section. Therefore, without departing from the teachings of the examples described herein, the first component, part, region, layer, or section mentioned in the examples may also be referred to as the second component, part, region, layer, or section.
[0076] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description herein, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0077] In addition, the word "exemplary" is used herein to mean serving as an example, instance, or diagram. Any aspect or design described herein as "exemplary" is not necessarily to be understood as being advantageous over other aspects or designs. Instead, the use of the word exemplary is intended to present concepts in a concrete manner. As used herein, the term "or" is intended to mean an inclusive "or" rather than an exclusive "or".
[0078] Likewise, although the present application has been shown and described with respect to one or more implementations, equivalent variations and modifications will occur to those skilled in the art upon reading and understanding this specification and the accompanying drawings. With particular regard to the various functions performed by the components (e.g., elements, resources, etc.) described above, unless otherwise noted, the terms used to describe such components are intended to correspond to any component (functionally equivalent) that performs the specific functions of the described components, even if not structurally equivalent to the disclosed structures.
[0079] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary technical means in the art that are not disclosed in the present application, and the specification and embodiments are only considered as exemplary.
[0080] It should be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.
Claims
1. A method for detecting limestone decomposition effect based on machine vision, characterized in that: include: Acquire a surface image of a sample of target cement clinker under a microscope, wherein the target cement clinker is produced using limestone of a target proportion; Determine crack pixels from the surface image according to gray values and gradient values of pixels in the surface image, and combine adjacent crack pixels into a crack line, so as to determine a first parameter value according to the crack line in the surface image, wherein the first parameter value is used to characterize the crack complexity of the surface image; Determine multiple closed areas in the surface image, and use the closed areas whose similarity with the ellipse-like image is greater than a preset similarity threshold as pore areas, so as to determine a second parameter value according to the area and number of the pore areas, and the second parameter value is used to characterize the complexity of the pore areas in the surface image; Determine a target parameter value according to the sum of the first parameter value and the second parameter value, and determine an evaluation value of the decomposition effect of the limestone according to the target parameter value and a preset corresponding relationship, wherein the preset corresponding relationship is used to characterize the corresponding relationship between the evaluation value and the target parameter value under the target ratio; Wherein, the method of determining crack pixels from the surface image according to the grayscale values and gradient values of the pixels in the surface image comprises: determining an edge line in the surface image, and for a target pixel on the edge line, determining a first difference degree value in grayscale values between the target pixel and adjacent pixels in the target direction, and determining a second difference degree value in gradient values between the target pixel and adjacent pixels in the target direction; determining a third parameter value of the target pixel according to the first difference degree value and the second difference degree value, the third parameter value being used to characterize the probability that the target pixel is a crack pixel; and taking a pixel on the edge line of the surface image whose third parameter value is greater than a preset parameter threshold as a crack pixel in the surface image; The target direction is the gradient direction of the target pixel point. According to the first difference degree value and the second difference degree value, a third parameter value of the target pixel point is determined, including: , P is the third parameter value of the target pixel, is the normalization function, H is the average gray value of the target pixel and its adjacent pixels in the target direction, h is the gray value of the target pixel, is the first difference value, T is the average gradient value between the target pixel and the adjacent pixels in the target direction, t is the gradient value of the target pixel, is the second difference value, n is the number of pixels on the edge line where the target pixel is located, is the gradient direction angle of the i-th pixel on the edge line where the target pixel is located, r is the gradient direction angle of the target pixel, To take the absolute value operation; The similarity between the closed region and the ellipse-like region is determined in the following manner: ), where K is the similarity between the closed area and the ellipse-like area, exp is an exponential function with a natural constant as the base, and S is the number of pixels located at the edge of the closed area. is the horizontal coordinate of the f-th pixel point on the closed area in the target coordinate system, is the ordinate of the f-th pixel point on the closed area in the target coordinate system, is the distance between the two pixel points where the edge of the closed area intersects the horizontal axis of the target coordinate system. It is the distance between the two pixel points where the edge of the closed area intersects with the vertical axis of the target coordinate system; the target coordinate system takes the line segment corresponding to the maximum distance between the two pixel points located at the edge of the closed area as the horizontal axis, and the target coordinate system takes the midpoint of the line segment as the coordinate origin.
2. The method for detecting limestone decomposition effect based on machine vision according to claim 1, characterized in that: The step of determining the first parameter value according to the crack line in the surface image comprises: ,in, is the first parameter value, norm is the normalization function, m is the number of crack pixels in the surface image, M is the number of pixels in the surface image, is the number of crack lines in the surface image.
3. The method for detecting limestone decomposition effect based on machine vision according to claim 1, characterized in that: Determining the second parameter value according to the area and number of the pore regions includes: ,in, is the second parameter value, norm is the normalization function, a is the total area of all pore regions in the surface image, A is the area of the surface image, is the number of pore areas in the surface image.
4. The method for detecting limestone decomposition effect based on machine vision according to claim 1, characterized in that: The step of determining the target parameter value according to the sum of the first parameter value and the second parameter value comprises: , where W is the target parameter value, exp is an exponential function with a natural constant as the base, Q1 is the first parameter value, and Q2 is the second parameter value.
5. The method for detecting limestone decomposition effect based on machine vision according to any one of claims 1 to 4, characterized in that: The preset corresponding relationship is obtained in the following manner: Obtaining an evaluation value determined according to a historical test result of a target sample, wherein the target sample is a sample of cement clinker produced from limestone according to a target proportion; acquiring a target parameter value determined according to a historical surface image of the target sample under a microscope; The preset corresponding relationship is obtained according to the target parameter values corresponding to the target samples with different evaluation values.
