Method for detecting grading of dam material of rock-fill dam

Abstract

The application discloses a method for detecting grading of dam construction material of a rock-fill dam, and relates to the field of image technology.The method comprises the following steps: performing gray scale enhancement processing on an infrared gray scale image to obtain an enhanced gray scale image; performing segmentation on target images and background images in the enhanced gray scale image by using an automatic threshold method to obtain a binary image; performing peak value processing on the binary image to obtain a first binary three-dimensional image; performing watershed segmentation processing on particles in a conglutination state in the first binary three-dimensional image to obtain a particle segmentation line image; performing processing on each particle in the particle segmentation line image by using a minimum circumscribed rectangle to obtain a particle size interval of each particle in the particle segmentation line image; and fitting the particle size interval and a temperature rise characteristic curve equation of the dam construction material to obtain a particle grading curve.The above technical scheme can identify the grading curve of particles in the dam construction material regardless of the range of particle sizes in the dam construction material, and can improve the accuracy of the grading curve.

Description

Technical Field

[0001] This invention relates to the field of image technology, specifically to a method for detecting the gradation of materials used in rockfill dam construction. Background Technology

[0002] The layered filling and compaction of soil and rock materials in a rockfill dam is a crucial process in the quality control of rockfill dam construction. The quality control of dam material compaction is related to the overall structural stability and seepage stability of the dam body. Effectively controlling the dam filling quality is key to ensuring the safe and stable operation of the dam. Therefore, compaction quality testing is a key aspect of rockfill dam filling quality control. Various compaction parameters of the fill material are measured directly or indirectly using appropriate technical means to control the compaction quality. Among these testing methods, the gradation analysis of the materials is a core component.

[0003] In related technical methods, image processing techniques are typically used to extract the particle size distribution of sediment particles or to examine the gradation of asphalt mixtures for road surfaces. However, when these methods are used to detect the gradation of materials, the particle sizes detected are all large, exceeding 1 mm. When the particle size is small, the gradation curves of small particles cannot be identified, thus reducing the accuracy of the identification. Summary of the Invention

[0004] The purpose of this invention is to provide a method for testing the gradation of materials used in rockfill dam construction, in order to solve the technical problems existing in related technologies.

[0005] To achieve the above objectives, the present invention provides a method for testing the gradation of rockfill dam construction materials, comprising:

[0006] Obtain infrared grayscale images and temperature rise characteristic curve equations of the rockfill dam construction materials;

[0007] An enhanced grayscale image is obtained by performing grayscale enhancement processing on an infrared grayscale image.

[0008] The target image and background image in the enhanced grayscale image are segmented by an automatic thresholding method to obtain a binary image;

[0009] Peaking is performed on the binary image to obtain the first binary 3D image;

[0010] Watershed segmentation processing is performed on the particles in the first binary 3D image that are in an adhesive state to obtain a particle segmentation line image;

[0011] The minimum bounding rectangle method is used to process each particle in the particle segmentation image to obtain the particle size range of each particle in the particle segmentation image.

[0012] By fitting the particle size range of each particle and the temperature rise characteristic curve equation of the rockfill dam construction material, the particle size distribution curve is obtained.

[0013] Optionally, the step of performing grayscale enhancement processing on the infrared grayscale image to obtain an enhanced grayscale image includes:

[0014] The enhanced grayscale image is obtained by mapping the grayscale value of each pixel in the infrared grayscale image to a grayscale range using the gamma transform function.

[0015] Optionally, the step of segmenting the target image and the background image in the enhanced grayscale image using an automatic thresholding method to obtain a binary image includes:

[0016] The target grayscale value is calculated using an automatic thresholding method based on the grayscale value of each pixel in the enhanced grayscale image.

[0017] For each pixel in the enhanced grayscale image, pixels with grayscale values ​​greater than the target grayscale value are assigned a value of 1, and pixels with grayscale values ​​less than or equal to the target grayscale value are assigned a value of 0, thus obtaining a binary image.

[0018] Optionally, the step of calculating the target grayscale value using an automatic thresholding method based on the grayscale value of each pixel in the enhanced grayscale image includes:

[0019] Multiple preset grayscale values ​​are determined, wherein the preset grayscale values ​​are used to segment background pixels and target pixels;

[0020] For each preset grayscale value, perform the following processing:

[0021] The pixels in the enhanced grayscale image are divided into background pixels and target pixels based on the preset grayscale values;

[0022] Based on the grayscale value of each pixel in the background pixels, calculate the first average grayscale value of the background pixels and the first probability of the background pixels appearing; based on the grayscale value of each pixel in the target pixels, calculate the second average grayscale value of the target pixels and the second probability of the target pixels appearing.

[0023] Calculate the third average gray value of all pixels in the enhanced grayscale image;

[0024] Based on the first probability, the second probability, the first average gray value, the second average gray value, and the third average gray value, calculate the variance between the background pixels and the target pixels;

[0025] After processing all preset grayscale values, the preset grayscale value corresponding to the largest variance is determined as the target grayscale value.

[0026] Optionally, the variance is expressed by the following formula:

[0027] ;

[0028] in, For variance, The first probability, The third average gray value, The first average gray value, The second probability, The second average gray value, This is the preset grayscale value.

[0029] Optionally, the step of peaking the binary image to obtain the first binary three-dimensional image includes:

[0030] Determine the first set of pixels corresponding to the target image and the second set of pixels corresponding to the background image in the binary image;

[0031] For the first pixel in the first set, determine the second pixel in the second set that is closest to the target pixel, calculate the distance between the first pixel and the second pixel, and assign the distance as a pixel value to the first pixel to obtain the second binary 3D image, wherein the first pixel is any pixel in the first set;

[0032] The region to which the target image belongs in the second binary 3D image is processed by the local minimum with a threshold, and the pixels corresponding to the region where the local minimum is located are determined as the third set.

[0033] The pixel values ​​of the pixels in the second binary 3D image that correspond to the third set are assigned the local minimum value to obtain the first binary 3D image.

[0034] Optionally, the step of processing each particle in the particle segmentation image using the minimum bounding rectangle method to obtain the particle size range of each particle in the particle segmentation image includes:

[0035] For each particle in the particle segmentation image, perform the following processing:

[0036] The boundary pixels and their corresponding coordinates of the particles are extracted using a region boundary function.

[0037] Select the first boundary pixel as the starting point, select the second boundary pixel adjacent to the first boundary pixel in a counterclockwise direction, and calculate the target vector from the first boundary pixel to the second boundary pixel based on the coordinates of the first boundary pixel and the second boundary pixel. Use the direction of the target vector as the direction of one side of the rectangle, solve the sub-boundary rectangle of the particle corresponding to the target vector, and calculate the area of ​​the sub-boundary rectangle.

[0038] Traverse all boundary pixels to obtain multiple sub-boundary rectangles and their corresponding areas. Determine the bounding rectangle of the particle with the smallest area and determine the vertex coordinate data of the bounding rectangle. Calculate the particle size range based on the vertex coordinate data.

[0039] After processing all particles, the particle size range of each particle in the particle segmentation image is obtained.

[0040] Optionally, the equation for the heating characteristic curve is expressed by the following calculation formula:

[0041] ;

[0042] in, This represents the temperature rise of the sand and gravel particles. The temperature difference between the fluid and the solid. The conversion factor between particle surface area and volume of the building material for rockfill dams. The particle volume of the dam construction material for rockfill dams. The particle mass density of the material used in constructing a rockfill dam. Specific heat capacity of particles in the dam construction material for rockfill dams.

[0043] The above technical solution involves enhancing the infrared grayscale image of the rockfill dam construction material to obtain an enhanced grayscale image. An automatic thresholding method is used to segment the target image from the background image in the enhanced grayscale image, resulting in a binary image. Peaking of the binary image yields a first binary three-dimensional image. Watershed segmentation is then performed on the particles in the first binary three-dimensional image to obtain particle segmentation line images. Subsequently, the minimum bounding rectangle method is used to process each particle in the particle segmentation line images to obtain the particle size range of each particle. Finally, the particle size range of each particle is fitted to the temperature rise characteristic curve equation of the rockfill dam construction material to obtain the particle gradation curve. Through the segmentation processing of the infrared image, regardless of the particle size range in the dam construction material, the particle gradation curve can be identified, and the accuracy of the gradation curve can be improved.

[0044] Other features and advantages of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0045] Figure 1 This is a schematic diagram illustrating a method for detecting the gradation of rockfill dam construction materials according to an exemplary embodiment of the present invention.

[0046] Figure 2 This is a schematic diagram of the transformation curve when the gamma coefficient is 1, according to an exemplary embodiment of the present invention.

[0047] Figure 3 This is a diagram illustrating the effect of the transformation curve when the gamma coefficient is 1, according to an exemplary embodiment of the present invention.

[0048] Figure 4 This is an illustration of the effect of segmenting a binary image using an automatic thresholding method according to an exemplary embodiment of the present invention.

[0049] Figure 5 This is a rendering of a first binary three-dimensional image according to an exemplary embodiment of the present invention.

[0050] Figure 6 This is a rendering of a particle segmentation line image according to an exemplary embodiment of the present invention.

[0051] Figure 7 This is a schematic diagram illustrating the minimum bounding rectangle algorithm according to an exemplary embodiment of the present invention.

[0052] Figure 8 This is a schematic diagram illustrating the effect of the minimum bounding rectangle marker in a particle segmentation image according to an exemplary embodiment of the present invention. Detailed Implementation

[0053] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, so as to provide a better understanding of the concept of the present invention, the technical problem solved, the technical features constituting the technical solution, and the technical effects brought about.

[0054] like Figure 1 As shown, Figure 1 This is a schematic diagram illustrating a method for testing the gradation of rockfill dam construction materials according to an exemplary embodiment of the present invention. (Refer to...) Figure 1 The method includes;

[0055] S101: Obtain infrared grayscale images and temperature rise characteristic curve equations of the rockfill dam construction materials;

[0056] S102: Perform grayscale enhancement processing on the infrared grayscale image to obtain an enhanced grayscale image;

[0057] S103: The target image and the background image in the enhanced grayscale image are segmented by an automatic thresholding method to obtain a binary image;

[0058] S104: Perform peaking processing on the binary image to obtain the first binary 3D image;

[0059] S105: Perform watershed segmentation processing on the particles in the first binary 3D image that are in an adhesive state to obtain a particle segmentation line image;

[0060] S106: Process each particle in the particle segmentation image using the minimum bounding rectangle method to obtain the particle size range of each particle in the particle segmentation image;

[0061] S107: Fit the particle size range of each particle and the temperature rise characteristic curve equation of the rockfill dam construction material to obtain the particle size distribution curve.

[0062] The above technical solution involves enhancing the infrared grayscale image of the rockfill dam construction material to obtain an enhanced grayscale image. An automatic thresholding method is used to segment the target image from the background image in the enhanced grayscale image, resulting in a binary image. Peaking of the binary image yields a first binary three-dimensional image. Watershed segmentation is then performed on the particles in the first binary three-dimensional image to obtain particle segmentation line images. Subsequently, the minimum bounding rectangle method is used to process each particle in the particle segmentation line images to obtain the particle size range of each particle. Finally, the particle size range of each particle is fitted to the temperature rise characteristic curve equation of the rockfill dam construction material to obtain the particle gradation curve. Through the segmentation processing of the infrared image, regardless of the particle size range in the dam construction material, the particle gradation curve can be identified, and the accuracy of the gradation curve can be improved.

[0063] To enable those skilled in the art to better understand the gradation testing method for rockfill dam construction materials provided by this invention, the above steps are illustrated in detail below.

[0064] For example, in the process of image gradation recognition, the purpose of processing particle sampling images using digital image technology is to extract effective particle size information. This processing includes steps such as image enhancement, threshold segmentation, and size data extraction. Infrared grayscale images can convert invisible infrared temperature field signals into two-dimensional images of black and white grayscale pixels. The temperature rise characteristic curve equation can be used to establish a quantitative mapping relationship between temperature rise and particle size and mass, realizing accurate conversion from the apparent features of infrared images to the actual particle size and mass parameters of materials.

[0065] For example, after obtaining an infrared grayscale image, the image may have problems such as high image noise and low contrast. Filtering, grayscale transformation and other methods are used to process the infrared grayscale image to reduce interference with image quality.

[0066] In one possible manner, the grayscale enhancement processing of the infrared grayscale image to obtain an enhanced grayscale image includes:

[0067] The enhanced grayscale image is obtained by mapping the grayscale value of each pixel in the infrared grayscale image to a grayscale range using the gamma transform function.

[0068] It should be understood that the infrared grayscale image can first be transformed using the gamma transform function. Grayscale transformation is a processing method that calculates and reassigns values ​​to each pixel in the image. Grayscale transformation can expand or compress the grayscale range of the original image. Appropriate grayscale transformation calculation methods can enhance the contrast of target areas in the image and improve defects such as flat grayscale and low contrast.

[0069] In this embodiment of the invention, grayscale transformation is performed using a gamma transform function. Specifically, it can be expressed by the following calculation formula:

[0070] ;

[0071] in, To enhance the grayscale values ​​of a grayscale image, The grayscale values ​​of the infrared grayscale image. For pixel coordinates, For compensation coefficient, The gamma coefficient can be used to adjust the shape of the calculation curve. When When the time is right, the gamma transform becomes a linear transform.

[0072] In this embodiment of the invention, the grayscale value of each pixel in the infrared grayscale image is mapped to a target grayscale range using a gamma transform function, thereby achieving grayscale stretching or compression, and ultimately obtaining an enhanced grayscale image. In this embodiment, the grayscale values ​​of the infrared grayscale image are mostly concentrated in the range of 15 to 60, so the purpose of the gamma transform is mainly to stretch the image's grayscale range. Specifically, in this embodiment, .when At times, such as Figure 2 and Figure 3 As shown, after the grayscale values ​​of the infrared grayscale image are linearly mapped to a larger target grayscale range, the overall contrast of the image is improved, and the particles are easier to identify. It can be the grayscale value distribution range of an infrared grayscale image. It can enhance the grayscale value distribution range of a grayscale image.

[0073] For example, after enhancing the original grayscale image, in order to extract dimensional features such as rock particle area, minimum bounding rectangle, and side length, image segmentation techniques are needed to segment individual target objects. In the image of this invention, the target image and background image in the enhanced grayscale image can first be segmented to obtain a binary image. The segmentation of the target image and background image can be a method and process for extracting and separating the target of interest based on the differences in pixel features of different target regions in the enhanced grayscale image. In this embodiment of the invention, segmentation is performed using an automatic thresholding method (OSTU).

[0074] In one possible manner, the step of segmenting the target image and the background image in the enhanced grayscale image using an automatic thresholding method to obtain a binary image includes:

[0075] The target grayscale value is calculated using an automatic thresholding method based on the grayscale value of each pixel in the enhanced grayscale image.

[0076] For each pixel in the enhanced grayscale image, pixels with grayscale values ​​greater than the target grayscale value are assigned a value of 1, and pixels with grayscale values ​​less than or equal to the target grayscale value are assigned a value of 0, thus obtaining a binary image.

[0077] It should be understood that in a digital image, the target image and the background image often have different grayscale values. Therefore, the image can be viewed as a combination of two regions with different grayscale characteristics. Thresholding segmentation involves finding a suitable threshold value, i.e., the target grayscale value, and using this as a criterion to separate regions with different grayscale characteristics. After determining the target grayscale value, the algorithm uses this target grayscale value as a standard to judge the grayscale value of each pixel in the image. Pixels with grayscale values ​​greater than the target grayscale value will be reassigned a value of 1, while pixels with grayscale values ​​less than the target grayscale value will be reassigned a value of 0. After processing, the enhanced grayscale image will be segmented into two regions with pixel values ​​of 0 and pixel values ​​of 1, producing corresponding binary images, as shown in the diagram. Figure 4 As shown.

[0078] Specifically, it can be expressed by the following calculation formula:

[0079] ;

[0080] in, The target grayscale value.

[0081] In one possible manner, calculating the target grayscale value using an automatic thresholding method based on the grayscale value of each pixel in the enhanced grayscale image includes:

[0082] Multiple preset grayscale values ​​are determined, wherein the preset grayscale values ​​are used to segment background pixels and target pixels;

[0083] For each preset grayscale value, perform the following processing:

[0084] The pixels in the enhanced grayscale image are divided into background pixels and target pixels based on the preset grayscale values;

[0085] Based on the grayscale value of each pixel in the background pixels, calculate the first average grayscale value of the background pixels and the first probability of the background pixels appearing; based on the grayscale value of each pixel in the target pixels, calculate the second average grayscale value of the target pixels and the second probability of the target pixels appearing.

[0086] Calculate the third average gray value of all pixels in the enhanced grayscale image;

[0087] Based on the first probability, the second probability, the first average gray value, the second average gray value, and the third average gray value, calculate the variance between the background pixels and the target pixels;

[0088] After processing all preset grayscale values, the preset grayscale value corresponding to the largest variance is determined as the target grayscale value.

[0089] It should be understood that the automatic thresholding method is also known as the Otsu method. The OSTU method is based on the least squares principle, and its goal is to maximize the gray-level variance between the two segmented regions of the image. Preset gray-level values ​​can be used to distinguish between background pixels and target pixels. The specific processing procedure is as follows:

[0090] When the size of the enhanced grayscale image is Its pixel coordinates The corresponding grayscale value is , The enhanced grayscale image includes a total of A number of gray levels.

[0091] For any gray level It can be done Indicates the first The number of pixels contained in the i-th gray level, then the i-th The probability of each gray level appearing This can be expressed by the following formula:

[0092] ;

[0093] in, ,and ,

[0094] When the preset grayscale value is The two parts after segmentation are the background pixels. and target pixel ; The grayscale range is , The grayscale range is Background pixels The first probability of occurrence This can be expressed by the following calculation formula:

[0095] .

[0096] target pixel The second probability of occurrence It can be expressed by the following calculation formula;

[0097] .

[0098] Background pixels First average gray value This can be expressed by the following formula:

[0099] .

[0100] target pixel Second average gray value This can be expressed by the following formula:

[0101] .

[0102] Enhance the third average gray value of all pixels in a grayscale image. This can be expressed by the following calculation formula:

[0103] .

[0104] The variance is then expressed by the following formula:

[0105] ;

[0106] in, For variance, The first probability, The third average gray value, The first average gray value, The second probability, The second average gray value, This is the preset grayscale value.

[0107] Then, the above steps are repeated for each preset grayscale value to obtain multiple variances. The preset grayscale value corresponding to the largest variance is determined as the target grayscale value.

[0108] For example, in the obtained binary image, particles and the surrounding background are effectively segmented. However, due to the positional adjacency between particles during sampling, many particles in the segmented binary image also exhibit adhesion. Therefore, it is necessary to effectively segment the adhered particles in the binary image to ensure the accuracy of the gradation calculation.

[0109] In one possible manner, the peaking process of the binary image to obtain a first binary three-dimensional image includes:

[0110] Determine the first set of pixels corresponding to the target image and the second set of pixels corresponding to the background image in the binary image;

[0111] For the first pixel in the first set, determine the second pixel in the second set that is closest to the target pixel, calculate the distance between the first pixel and the second pixel, and assign the distance as a pixel value to the first pixel to obtain the second binary 3D image, wherein the first pixel is any pixel in the first set;

[0112] The region to which the target image belongs in the second binary 3D image is processed by the local minimum with a threshold, and the pixels corresponding to the region where the local minimum is located are determined as the third set.

[0113] The pixel values ​​of the pixels in the second binary 3D image that correspond to the third set are assigned the local minimum value to obtain the first binary 3D image.

[0114] It should be understood that before performing watershed segmentation on an image, the binary image needs to be processed into three dimensions. This involves representing the relative positional information of each pixel in the granular region onto the corresponding pixel value, thus peaking the flattened binary image. In conventional image processing, this can be done using the following method, expressed as:

[0115] ;

[0116] in, To prepare the preprocessed matrix, preprocessing can include peaking, with its size and... Same size. Operation The calculation rules for Distance Transform are as follows: assuming for The set of all elements with a pixel value of 1, that is, the first set of all corresponding pixels in the target image. For The median coordinate is The point to be found in the second set. Let the coordinates of the nearest 0-pixel point be set as ,but The median coordinate is The pixel value is calculated using the following formula:

[0117] And iterate through all pixels in the first set. Assume... for The set of all pixels with a value of 0, which is the second set of pixels corresponding to the background image. For The median coordinate is point, The median coordinate is The rules for assigning pixel values ​​are as follows: .

[0118] The second binary 3D image is obtained after the above processing. Then, considering... The computational principle of the operation is that the closer to the geometric center of the particle, the smaller the value obtained after calculating and taking the opposite number. Local minima are generally located in the geometric center region of the particle. Therefore, the local minima algorithm with threshold is first used to find the region where the local minimum value of the particle's geometric center is located. This region is represented by a third set. This means that after processing, all pixel values ​​within the area that meets the conditions are assigned a value of 1, and the remaining pixel values ​​are assigned a value of 0.

[0119] Will pass Image after processing Mid position and The pixels in the corresponding region are reassigned to the minimum value of the region, that is... middle The meaning of "smoothing out" a region is that it can make... middle Pixel values ​​within the region are assigned a uniform and smaller local minimum, thereby eliminating multiple local minima that might generate redundant segmentation lines within the geometric center region of the particles, thus obtaining a first binary 3D image. The specifically processed image is shown below. Figure 5 As shown.

[0120] For example, the watershed segmentation method can be a method that treats the target particle region as a catchment basin and achieves particle segmentation by finding the watershed lines between adherent particles. In the watershed segmentation method, the segmented pixels include three types of feature points, namely... Class feature pixels, Class feature pixels and Class feature pixels. Among them, the pixel at the center of the particle region becomes the local pixel minimum. Class feature pixels. Make non-center pixels in the granular region become... A class of feature pixels, whose pixel gradient points only to that region. Class feature pixels. Let the pixels at the adhesion points of adjacent particles become... Class feature pixels, which have pointers to neighboring regions Pixel gradient of class feature pixels.

[0121] In the watershed algorithm, the goal is to find the watershed line between different "basins," where each "basin" represents a target object. The watershed line separates adhered target objects. The basic idea of ​​watershed line calculation is as follows: assuming each region's minimum point is a water outlet, as water continuously flows out from this point, the water level rises and eventually floods the entire terrain. When water from different basins is about to converge due to the rising water level, a "wall" is built at the point where the waterlines are about to meet to prevent the convergence. As the water level rises further, the size of the "wall" expands, eventually forming a continuous boundary wall, which is the watershed line.

[0122] In the specific calculation process, it is assumed that the set of local minimum point coordinates in the first binary 3D image is as follows: For each local minimum point, there exists a catchment basin containing the minimum point. The set of coordinates of points within the catchment basin is used... Indicated. Additionally, for the maximum and minimum values ​​within the global scope, respectively, use... and Indication. During the watershed segmentation process, using... Indicates water level height. For the coordinates that were submerged by water The set of This can be expressed by the following formula:

[0123] .

[0124] As the water level rises, the algorithm continuously counts the number of submerged points in the first binary 3D image. When the water level increases with a positive increment, the 3D effect of the image is mapped onto... The image on the plane is a binary image. When the water level is... When, use This represents the set of submerged points within a catchment basin. It can be expressed by the following formula: .

[0125] During algorithm execution, as the water level rises... The continuous rise, and The number of elements contained in these two sets will also increase accordingly. Therefore, based on the water level height... When used as an independent variable yes a subset of The water level is At that time, the set of points that were submerged. Therefore, the water level height is... At that time, the connected components in the set of points submerged by the water basin are also Connected components in.

[0126] When the algorithm calculates the watershed height, the initial water level is: At this time there is ,in, Let be the set of points submerged at the initial water level. The water level is At that time, the set of points that were submerged. As the water level continues to rise, the water level height is... Time calculation According to calculate When, use represent The set of connected components in the middle. And for each of these connected components... There are three possible scenarios: It is an empty set. There is one and only one connected component. There are two or more connected components.

[0127] In use structure During the process, its construction method is mainly influenced by and The relationship between them has an impact. Specifically, when there is only one connected component, take... and The union of the sets constructs Conversely, if When there are multiple connected components, the water level will be further increased, causing the water levels in the basins on both sides of the ridgeline to merge and reach the same height. At this point, a "wall" will be built at the water confluence. This wall will rise as the water level increases, consistently preventing the water from merging. Eventually, this continuously rising wall will form a watershed line, resulting in a particle segmentation line image, such as... Figure 6 As shown.

[0128] For example, after obtaining the particle segmentation line image, the particle size range of each particle in the particle segmentation line image can be calculated using the minimum bounding rectangle method, and then a gradation curve can be drawn based on the particle size range.

[0129] In one possible manner, the process of processing each particle in the particle segmentation image using the minimum bounding rectangle method to obtain the particle size range of each particle in the particle segmentation image includes:

[0130] For each particle in the particle segmentation image, perform the following processing:

[0131] The boundary pixels and their corresponding coordinates of the particles are extracted using a region boundary function.

[0132] Select the first boundary pixel as the starting point, select the second boundary pixel adjacent to the first boundary pixel in a counterclockwise direction, and calculate the target vector from the first boundary pixel to the second boundary pixel based on the coordinates of the first boundary pixel and the second boundary pixel. Use the direction of the target vector as the direction of one side of the rectangle, solve the sub-boundary rectangle of the particle corresponding to the target vector, and calculate the area of ​​the sub-boundary rectangle.

[0133] Traverse all boundary pixels to obtain multiple sub-boundary rectangles and their corresponding areas. Determine the bounding rectangle of the particle with the smallest area and determine the vertex coordinate data of the bounding rectangle. Calculate the particle size range based on the vertex coordinate data.

[0134] After processing all particles, the particle size range of each particle in the particle segmentation image is obtained.

[0135] It should be understood that the region boundary function can be the Bwboundaries function in the program software. In specific processing, such as... Figure 7 As shown, assume the set of boundary pixel coordinates is ,in, This represents the number of boundary pixels. (Choose arbitrarily) The first boundary pixel in the image is assumed to be... Select in a counter-clockwise direction Adjacent second boundary pixels A vector passing through two points can be represented as , with vector The direction is taken as the direction of one side of the rectangle, and the vector corresponding to the particle is calculated. child bounding rectangle And calculate the area of ​​the sub-circumscribed rectangle. Solve for the sub-circumscribed rectangle by following the steps described above. and the corresponding rectangular area .

[0136] exist The minimum value is determined, and the sub-boundary rectangle corresponding to the minimum value is determined as the bounding rectangle of the particle. After calculating the minimum bounding rectangle of the particle according to the above algorithm, the coordinate data of the four vertices of the minimum bounding rectangle can be obtained. For the first... For each particle, the vertex coordinates can be obtained from a set. Given the form, set It includes five coordinate points, where the first and fifth coordinate points are the same. The specific form can be expressed in the following way: ;

[0137] After solving, draw the minimum bounding rectangle of each particle according to the vertex data, and mark the result as shown in the figure. Figure 8 As shown.

[0138] For the found minimum bounding rectangle, the side lengths of two adjacent sides of the rectangle are... and It can be calculated using the following formula:

[0139] ;

[0140] In the calculation and Then, the minimum value is the shortest side of the smallest bounding rectangle of the particle. When calculating the gradation curve, with This serves as the basis for determining whether a particle can pass through a sieve with a certain aperture, thereby identifying the particle size range of each particle. After processing all particles, the particle size range of each particle in the particle segmentation image is obtained.

[0141] For example, after calculating the particle size range of each particle, the particle size range of each particle and the temperature rise characteristic curve equation of the rockfill dam construction material can be fitted to obtain the particle size distribution curve.

[0142] Specifically, the heating characteristic curve equation can be obtained by conducting heating characteristic tests on particles of the same material. These heating characteristic tests are conducted using methods found in relevant technologies. Therefore, the heating characteristic curve equation can be expressed by the following calculation formula:

[0143] ;

[0144] in, The temperature rise of particles in the dam construction material for rockfill dams. The temperature difference between the fluid and the solid. The conversion factor between particle surface area and volume of the building material for rockfill dams. The particle volume of the dam construction material for rockfill dams. Mass density of particles in the dam construction material for rockfill dams. Specific heat capacity of particles in the dam construction material for rockfill dams.

[0145] The above calculation formula shows that, under the same heating environment and heating time, the larger the volume of the rock particles, the smaller the temperature rise.

[0146] The above technical solution allows for the detection of gradation curves for particles of any size range in dam construction materials, thereby improving the accuracy of the detection. Furthermore, when using these detection results to test the compaction quality of dam materials, the accuracy of the compaction quality test can be improved.

[0147] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for testing the gradation of materials used in rockfill dam construction, characterized in that, include: Obtain infrared grayscale images and temperature rise characteristic curve equations of the rockfill dam construction materials; An enhanced grayscale image is obtained by performing grayscale enhancement processing on an infrared grayscale image. The target image and background image in the enhanced grayscale image are segmented by an automatic thresholding method to obtain a binary image; Peaking is performed on the binary image to obtain the first binary 3D image; Watershed segmentation processing is performed on the particles in the first binary 3D image that are in an adhesive state to obtain a particle segmentation line image; The minimum bounding rectangle method is used to process each particle in the particle segmentation image to obtain the particle size range of each particle in the particle segmentation image. By fitting the particle size range of each particle and the temperature rise characteristic curve equation of the rockfill dam construction material, the particle size distribution curve is obtained.

2. The method for detecting the gradation of rockfill dam construction materials according to claim 1, characterized in that, The process of performing grayscale enhancement processing on the infrared grayscale image to obtain an enhanced grayscale image includes: The enhanced grayscale image is obtained by mapping the grayscale value of each pixel in the infrared grayscale image to a grayscale range using the gamma transform function.

3. The method for testing the gradation of rockfill dam construction materials according to claim 1, characterized in that, The step of segmenting the target image and background image in the enhanced grayscale image using an automatic thresholding method to obtain a binary image includes: The target grayscale value is calculated using an automatic thresholding method based on the grayscale value of each pixel in the enhanced grayscale image. For each pixel in the enhanced grayscale image, pixels with grayscale values ​​greater than the target grayscale value are assigned a value of 1, and pixels with grayscale values ​​less than or equal to the target grayscale value are assigned a value of 0, thus obtaining a binary image.

4. The method for detecting the gradation of rockfill dam construction materials according to claim 3, characterized in that, The step of calculating the target grayscale value using an automatic thresholding method based on the grayscale value of each pixel in the enhanced grayscale image includes: Multiple preset grayscale values ​​are determined, wherein the preset grayscale values ​​are used to segment background pixels and target pixels; For each preset grayscale value, perform the following processing: The pixels in the enhanced grayscale image are divided into background pixels and target pixels based on the preset grayscale values; Based on the grayscale value of each pixel in the background pixels, calculate the first average grayscale value of the background pixels and the first probability of the background pixels appearing; based on the grayscale value of each pixel in the target pixels, calculate the second average grayscale value of the target pixels and the second probability of the target pixels appearing. Calculate the third average gray value of all pixels in the enhanced grayscale image; Based on the first probability, the second probability, the first average gray value, the second average gray value, and the third average gray value, calculate the variance between the background pixels and the target pixels; After processing all preset grayscale values, the preset grayscale value corresponding to the largest variance is determined as the target grayscale value.

5. The method for detecting the gradation of rockfill dam construction materials according to claim 4, characterized in that, The variance is expressed by the following formula: ； in, For variance, The first probability, The third average gray value, The first average gray value, The second probability, The second average gray value, This is the preset grayscale value.

6. The method for testing the gradation of rockfill dam construction materials according to claim 1, characterized in that, The step of peaking the binary image to obtain the first binary three-dimensional image includes: Determine the first set of pixels corresponding to the target image and the second set of pixels corresponding to the background image in the binary image; For the first pixel in the first set, determine the second pixel in the second set that is closest to the target pixel, calculate the distance between the first pixel and the second pixel, and assign the distance as a pixel value to the first pixel to obtain the second binary 3D image, wherein the first pixel is any pixel in the first set; The region to which the target image belongs in the second binary 3D image is processed by the local minimum with a threshold, and the pixels corresponding to the region where the local minimum is located are determined as the third set. The pixel values ​​of the pixels in the second binary 3D image that correspond to the third set are assigned the local minimum value to obtain the first binary 3D image.

7. The method for detecting the gradation of rockfill dam construction materials according to claim 6, characterized in that, The step of processing each particle in the particle segmentation image using the minimum bounding rectangle method to obtain the particle size range of each particle in the particle segmentation image includes: For each particle in the particle segmentation image, perform the following processing: The boundary pixels and their corresponding coordinates of the particles are extracted using a region boundary function. Select the first boundary pixel as the starting point, select the second boundary pixel adjacent to the first boundary pixel in a counterclockwise direction, and calculate the target vector from the first boundary pixel to the second boundary pixel based on the coordinates of the first boundary pixel and the second boundary pixel. Use the direction of the target vector as the direction of one side of the rectangle, solve the sub-boundary rectangle of the particle corresponding to the target vector, and calculate the area of ​​the sub-boundary rectangle. Traverse all boundary pixels to obtain multiple sub-boundary rectangles and their corresponding areas. Determine the bounding rectangle of the particle with the smallest area and determine the vertex coordinate data of the bounding rectangle. Calculate the particle size range based on the vertex coordinate data. After processing all particles, the particle size range of each particle in the particle segmentation image is obtained.

8. The method for testing the gradation of rockfill dam construction materials according to any one of claims 1-7, characterized in that, The equation for the heating characteristic curve is expressed by the following calculation formula: ； in, The temperature rise of particles in the dam construction material for rockfill dams. The temperature difference between the fluid and the solid. The conversion factor between particle surface area and volume of the building material for rockfill dams. The particle volume of the dam construction material for rockfill dams. Mass density of particles in the dam construction material for rockfill dams. Specific heat capacity of particles in the dam construction material for rockfill dams.

Images