Method and device for determining the expansion rate of steel slag asphalt mixture

Abstract

This invention discloses a method and apparatus for determining the expansion rate of steel slag asphalt mixture based on image processing, comprising: acquiring visual morphology data of steel slag asphalt mixture specimens during the expansion process according to a preset time sequence, wherein the visual morphology data includes a sequence of images that at least reflect the surface contour and steel slag particle distribution of the steel slag asphalt mixture specimens; determining characteristic parameters characterizing the expansion state of the steel slag asphalt mixture specimens using a computer vision algorithm based on the visual morphology data, wherein the computer vision algorithm includes at least image denoising, geometric correction, and target region segmentation; and calculating the expansion rate of the steel slag asphalt mixture specimens based on the characteristic parameters and a pre-established mapping relationship between pixel space and physical space.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically, to a method and apparatus for determining the expansion rate of steel slag asphalt mixture. Background Technology

[0002] Traditional methods for testing the expansion rate of steel slag asphalt mixtures mainly rely on physical contact measurement methods, such as using dial gauges or vernier calipers to directly measure the volume change of the specimen. These methods require the measuring element to contact the specimen surface, which is cumbersome, time-consuming, and the measurement results are easily affected by human reading errors and unevenness of the specimen surface. In addition, existing indirect testing methods based on mass and density also require repeated weighing and immersion treatment, which cannot achieve continuous and dynamic monitoring of the expansion process and cannot meet the high-precision and automated evaluation requirements of the expansion behavior of steel slag modified mixtures.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This invention provides a method, apparatus, and storage medium for determining the expansion rate of steel slag asphalt mixtures, thereby at least solving the technical problems of existing steel slag asphalt mixture expansion rate testing methods, which rely on physical contact measurement and suffer from complex operation, low automation, large human error, and inability to achieve non-contact continuous dynamic monitoring.

[0005] According to one aspect of the present invention, in order to achieve the above-mentioned objective, a method for determining the expansion rate of steel slag asphalt mixture is provided, comprising the following steps: Visual morphology data of steel slag asphalt mixture specimens during the expansion process are collected according to a preset time sequence. The visual morphology data includes a sequence of images that at least reflect the surface contour of the steel slag asphalt mixture specimens and the distribution of steel slag particles. Based on visual morphology data, characteristic parameters representing the expansion state of steel slag asphalt mixture specimens are determined by computer vision algorithms. The computer vision algorithms include at least image denoising, geometric correction, and target region segmentation. Based on the feature parameters and the pre-established mapping relationship between pixel space and physical space, the expansion rate of steel slag asphalt mixture specimens is calculated.

[0006] Furthermore, visual morphological data of steel slag asphalt mixture specimens during the expansion process were collected according to a preset time sequence, including: Generate an image acquisition command, which carries preset resolution and frame rate parameters; Obtain the original image; Convert the original image to a grayscale image; Add timestamp labels to grayscale images according to a preset time sequence; Generate a sequence of images based on the grayscale images with added timestamp tags.

[0007] Furthermore, based on visual morphology data, computer vision algorithms were used to determine characteristic parameters representing the expansion state of steel slag asphalt mixture specimens, including: Gaussian filtering is used to denoise each frame of grayscale image in the image sequence to obtain the first intermediate image; The first intermediate image is denoised using median filtering to obtain the second intermediate image; A perspective transformation algorithm is used to perform geometric correction on the second intermediate image to obtain the third intermediate image; An adaptive histogram equalization algorithm is used to enhance the contrast of the third intermediate image to obtain the fourth intermediate image; An adaptive threshold segmentation algorithm is used to binarize the fourth intermediate image to obtain a binary image. Based on the binary image, the main outline and main area of ​​the steel slag asphalt mixture specimen are extracted.

[0008] Furthermore, the method also includes: Register the sequence images with the template images of the standard calibration blocks; Based on the registration results, the pixel span of the standard calibration block in the image sequence is identified; Obtain the known physical dimensions of the standard calibration block; Based on the known ratio of physical size to pixel span, establish a mapping relationship between pixel space and physical space.

[0009] Furthermore, based on the feature parameters and the pre-established mapping relationship between pixel space and physical space, the expansion rate of the steel slag asphalt mixture specimen is calculated, including: Based on the mapping relationship, the main body area of ​​the steel slag asphalt mixture specimen before expansion is converted into the first physical area; Based on the mapping relationship, the main body area of ​​the expanded steel slag asphalt mixture specimen is converted into the second physical area; Calculate the difference between the second physical area and the first physical area to obtain the amount of physical area expansion; Calculate the percentage of the physical area expansion to the first physical area, and determine the percentage as the expansion rate.

[0010] Furthermore, the method also includes: Obtain multiple expansion rates of steel slag asphalt mixture specimens at multiple different times; Multiple expansion rates were smoothed using the moving average method; Expansion rates that deviate from the moving average by more than a predetermined threshold are identified as abnormal data; Abnormal data are corrected by using linear interpolation or the mean of adjacent time points.

[0011] Furthermore, the method also includes: Obtain information on the modification type of steel slag used in steel slag asphalt mixture specimens; In response to modification type information, at least one adjustable parameter in the computer vision algorithm is adaptively adjusted; The adjustable parameters include: kernel size of the filtering algorithm, truncation threshold of the contrast enhancement algorithm, dual threshold of the edge detection algorithm, and area threshold of the contour screening.

[0012] Furthermore, the method also includes: The measured values ​​of the expansion rate of steel slag asphalt mixture specimens at multiple sampling times during the expansion process were obtained; Input the measured expansion rate into the pre-built logistic growth curve model; The parameters of the logistic growth curve model were determined by fitting the model using the least squares method. Using the logistic growth curve model with determined parameters, the predicted long-term expansion rate of steel slag asphalt mixture specimens is output.

[0013] According to one embodiment of the present invention, a device for determining the expansion rate of steel slag asphalt mixture is also provided, comprising: a data acquisition module, used to acquire visual morphology data of steel slag asphalt mixture specimens during the expansion process according to a preset time sequence, the visual morphology data including a sequence of images reflecting at least the surface contour of the steel slag asphalt mixture specimens and the distribution of steel slag particles; a determination module, used to determine characteristic parameters characterizing the expansion state of the steel slag asphalt mixture specimens based on the visual morphology data using a computer vision algorithm, the computer vision algorithm including at least image denoising, geometric correction, and target region segmentation; and a calculation module, used to calculate the expansion rate of the steel slag asphalt mixture specimens based on the characteristic parameters and a pre-established mapping relationship between pixel space and physical space.

[0014] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is executed, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of the present invention.

[0015] In this embodiment of the invention, a sequence of images containing the surface contour of the specimen and the distribution of steel slag particles is acquired according to a preset time sequence as visual morphology data. Then, a computer vision algorithm including at least image denoising, geometric correction and target region segmentation is used to automatically extract feature parameters representing the expansion state. The expansion rate is calculated by combining the pre-established mapping relationship between pixel space and physical space. This achieves non-contact, fully automatic and high-precision quantification of the expansion process of steel slag asphalt mixture, avoids human error, significantly improves test efficiency and data reliability, and solves the technical problems of existing steel slag asphalt mixture expansion rate test methods that rely on physical contact measurement, such as complex operation, low degree of automation, large human error and inability to achieve non-contact continuous dynamic monitoring. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a flowchart of a method for determining the expansion rate of steel slag asphalt mixture according to one embodiment of the present invention; Figure 2 This is a structural block diagram of a device for determining the expansion rate of steel slag asphalt mixture according to one embodiment of the present invention. Detailed Implementation

[0017] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0018] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] According to an embodiment of the present invention, a method for determining the expansion rate of steel slag asphalt mixture is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0020] This method embodiment can be executed in an electronic device or similar computing device that includes a memory and a processor. Taking operation on a vehicle terminal as an example, the vehicle terminal may include one or more processors (processors may include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), digital signal processing (DSP) chips, microcontroller units (MCUs), field-programmable gate arrays (FPGAs), neural network processors (NPUs), tensor processors (TPUs), artificial intelligence (AI) type processors, etc.) and a memory for storing data. Optionally, the vehicle terminal may also include transmission devices, input / output devices, and display devices for communication functions. Those skilled in the art will understand that the above structural description is merely illustrative and does not limit the structure of the vehicle terminal. For example, the vehicle terminal may include more or fewer components than described above, or have a different configuration than described above.

[0021] The memory can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the method for determining the expansion rate of steel slag asphalt mixture in this embodiment of the invention. The processor executes various functional applications and data processing by running the computer program stored in the memory, thereby realizing the aforementioned method for determining the expansion rate of steel slag asphalt mixture. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to a mobile terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0022] The transmission device is used to receive or send data via a network. Specific examples of the network mentioned above may include a wireless network provided by the mobile terminal's communication provider. In one example, the transmission device includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0023] Display devices can be, for example, touchscreen liquid crystal displays (LCDs) and touch displays (also referred to as "touchscreens" or "touch displays"). The LCD allows users to interact with the user interface of the mobile terminal. In some embodiments, the mobile terminal has a graphical user interface (GUI), which allows users to interact with the GUI through finger contact and / or gestures on a touch-sensitive surface. Optional human-computer interaction functions include: creating web pages, drawing, word processing, creating electronic documents, playing games, video conferencing, instant messaging, sending and receiving emails, call interfaces, playing digital video, playing digital music, and / or web browsing, etc. Executable instructions for performing the above human-computer interaction functions are configured / stored in one or more processor-executable computer program products or readable storage media.

[0024] Figure 1 This is a flowchart of a method for determining the expansion rate of steel slag asphalt mixture according to one embodiment of the present invention, as shown below. Figure 1 As shown, the method includes the following steps: Step S110: Collect visual morphology data of steel slag asphalt mixture specimens during the expansion process according to a preset time sequence. The visual morphology data includes a sequence of images that at least reflect the surface contour of the steel slag asphalt mixture specimens and the distribution of steel slag particles. The specific content is as follows: In step S110, visual morphology data acquisition is performed. Based on a predetermined test plan, the timing parameters for image acquisition are determined. These timing parameters include at least the sampling interval and the total acquisition duration. For example, for the expansion process of steel slag asphalt mixture specimens under immersion or humid heat conditions, the sampling interval can be set to 10 minutes, 30 minutes, or 1 hour, and the total acquisition duration can cover the complete cycle from the start of expansion to its stabilization, typically 72 hours or longer, which can be adjusted according to the type of steel slag modification and the expected expansion rate.

[0025] After the acquisition is initiated, image acquisition commands are generated periodically according to the sampling interval. These commands carry preset resolution parameters, such as 1920×1080 pixels, to ensure that the image can clearly distinguish the surface contours of the specimen and the boundaries of the steel slag particles. In response to the image acquisition commands, the image acquisition device acquires the original image of the steel slag asphalt mixture specimen at the corresponding moment. The original image is a color image, with each pixel composed of values ​​from three color channels: red (R), green (G), and blue (B). The value range of each component is typically from 0 to 255.

[0026] After acquiring the original color image, it is immediately converted to a grayscale image. The purpose of grayscale conversion is to reduce the data dimensionality of subsequent processing while preserving the surface contour of the specimen and the brightness difference between the steel slag particles and the background. The conversion process uses a weighted average algorithm, specifically calculated as follows: Grayscale value = 0.299 × R + 0.587 × G + 0.114 × B, where R, G, and B are the values ​​of the red, green, and blue channels of the same pixel in the original color image, respectively. This weighting coefficient is determined based on prior knowledge of human eye sensitivity to different colors, ensuring that the converted grayscale image has good visual contrast.

[0027] After conversion, a timestamp is added to each grayscale image frame according to the sampling time sequence. The timestamp label must contain at least the offset of the image relative to the start time of the test, such as "0 hours", "0.5 hours", "1.0 hours", etc., or use an absolute time format. The timestamp label is used to identify the position of different frames in the dilation time series, providing a temporal reference for subsequent calculation of the dilation rate over time.

[0028] Based on this, the grayscale images with added timestamps are organized into a sequence of images in ascending order of time. This sequence of images constitutes the visual morphology data of this embodiment. This visual morphology data needs to reflect information in at least two dimensions: the first dimension is the surface contour information of the steel slag asphalt mixture specimen, i.e., the change in the shape and area enclosed by its outer boundary during the expansion process. This information is used to quantify the overall degree of expansion. The second dimension is the distribution information of steel slag particles exposed on or near the surface of the specimen, including the number of steel slag particles, the projected area of ​​a single particle, the equivalent diameter of the particle, and the degree of aggregation between particles. This information is used to analyze the influence of steel slag modification (such as different modifiers and different dosages) on the evolution of the internal structure of the mixture.

[0029] To ensure the validity of the aforementioned visual morphology data, the specimen must be placed under uniform and constant illumination during image acquisition to avoid non-target fluctuations in image grayscale values ​​caused by changes in light angle or intensity. Simultaneously, a standard calibration block with known actual physical dimensions, such as a square metal block with a side length of 10 mm or a circular calibration plate with a diameter of 20 mm, should be placed within the specimen's field of view. The standard calibration block and the steel slag asphalt mixture specimen should be located in the same imaging plane, ensuring that each frame of the image sequence contains a clear image of the calibration block. The image of the calibration block will provide a crucial reference for subsequently establishing the mapping relationship between pixel space and physical space.

[0030] Through the above steps, the sequence of images acquired according to the preset time sequence not only fully records the continuous evolution of the specimen morphology from the initial state to the stable state during the expansion process, but also embeds a fixed reference object for scale calibration, thus providing a reliable and complete data foundation for subsequent image denoising, geometric correction, feature extraction and accurate calculation of expansion rate.

[0031] Step S120: Based on the visual morphology data, determine the characteristic parameters representing the expansion state of the steel slag asphalt mixture specimen using a computer vision algorithm. The computer vision algorithm includes at least image denoising, geometric correction, and target region segmentation, as detailed below: In step S120, feature parameters characterizing the expansion state of the specimen are further determined using a computer vision algorithm based on the visual morphology data. The computer vision algorithm includes at least three core steps: image denoising, geometric correction, and target region segmentation, which are described in detail below.

[0032] Due to unavoidable interference from factors such as ambient lighting fluctuations, camera sensor noise, and dust or moisture on the specimen surface during image acquisition, the original image sequence often contains Gaussian noise and salt-and-pepper noise. Gaussian noise manifests as random fluctuations in the gray values ​​of each pixel in the image, following a normal distribution around the true value, while salt-and-pepper noise appears as randomly occurring black or white dots. If these noises are not removed, they will directly affect the accuracy of subsequent edge detection and contour extraction.

[0033] This implementation employs a dual denoising strategy combining Gaussian filtering and median filtering. First, Gaussian filtering is performed on each grayscale image frame in the image sequence. Gaussian filtering is achieved through convolution of a two-dimensional Gaussian kernel with the image. The core idea is: for each pixel, a neighborhood window is taken centered on it; the grayscale values ​​of pixels within the window are weighted and averaged according to weights determined by a Gaussian function; this weighted average replaces the original pixel's grayscale value. The size and standard deviation σ of the Gaussian kernel are key parameters affecting the denoising effect. For example, the convolution kernel size is set to 5×5 pixels, and the standard deviation σ is equal to 1.5. This parameter combination effectively removes Gaussian noise while preserving the edge details of the specimen and steel slag particles, avoiding information loss due to over-smoothing. The image after Gaussian filtering is denoted as the first intermediate image.

[0034] Then, median filtering is performed on the first intermediate image. Median filtering uses a 3×3 pixel square sliding window, sorting the gray values ​​of all pixels within the window's coverage area by size, and taking the median of the sorted values ​​as the new gray value of the center pixel of the window. Median filtering has excellent suppression capabilities for salt-and-pepper noise while maintaining edge sharpness. After median filtering, isolated black and white noise points in the image are effectively eliminated, while the clarity of the specimen's surface contour is not significantly compromised. The image after median filtering is recorded as the second intermediate image. Through the above dual filtering, a denoised image with a significantly improved signal-to-noise ratio is obtained, providing clean input data for subsequent geometric correction and edge detection.

[0035] In actual image acquisition, the camera's optical axis is difficult to guarantee is perfectly perpendicular to the specimen's surface, and the lens may exhibit optical distortion, causing perspective distortion of the specimen's actual geometry in the acquired image. For example, a circular specimen may appear as an ellipse in the image, and a square calibration block may appear as a trapezoid. This geometric distortion disrupts the linear mapping between pixel size and actual physical size and must be corrected before feature extraction.

[0036] This implementation employs a perspective transformation algorithm based on reference points for geometric correction. Specifically, four standard calibration blocks are pre-set on the test mold or specimen placement platform. These calibration blocks have known actual physical dimensions, such as squares with sides of 10 mm, and form a rectangular layout. In each frame of the image sequence, the center coordinates or corner coordinates of these four calibration blocks are automatically identified using a Hough circle detection or corner detection algorithm, and these coordinates are used as the source control points for the perspective transformation. These four source control points typically form an irregular quadrilateral in the image.

[0037] Then, based on the layout of the four calibration blocks in actual space, the coordinates of the target control points after perspective transformation are set. The target control points should form a rectangle consistent with the actual calibration block layout, and the ratio of the pixel size corresponding to the side length of the rectangle to the actual size should remain uniform throughout the image. The `getPerspectiveTransform` function from the OpenCV library is used to calculate a 3×3 perspective transformation matrix based on the source control points and the target control points. This matrix describes the mapping relationship from the distorted image space to the corrected image space.

[0038] Next, a perspective transformation is applied to the second intermediate image, that is, the `warpPerspective` function is called to reproject each pixel in the image to a new coordinate position according to the perspective transformation matrix. In the transformed image, the calibration block is restored to a rectangle, and the circular or rectangular outline of the specimen is also restored to a geometric shape consistent with the actual shape. Through the above geometric correction, the measurement errors caused by camera angle and lens distortion are eliminated, so that the mapping relationship between pixel space and physical space remains linear and uniform across the entire image.

[0039] Even after denoising and geometric correction, the grayscale values ​​of the steel slag asphalt mixture specimen area and the background area may still overlap or become blurred, especially since the grayscale difference between steel slag particles and the asphalt matrix is ​​relatively small. To accurately extract the specimen's outline and area, the specimen area needs to be separated from the background area; this process is called target region segmentation. The specimen area is also known as the foreground area.

[0040] This implementation method combines contrast enhancement with adaptive thresholding to achieve target region segmentation. First, contrast enhancement is performed on the geometrically corrected image. An adaptive histogram equalization algorithm is used to divide the image into multiple local sub-regions, for example, 8×8 squares, with each sub-region being one-eighth the size of the original image. The gray-level histogram of each sub-region is calculated and histogram equalization is performed to ensure the gray-level distribution within each sub-region is as uniform as possible. Based on this, bilinear interpolation is used to eliminate boundary effects between sub-regions. Adaptive histogram equalization effectively improves the local contrast between the specimen edges and the steel slag particles and the background, while avoiding excessive detail enhancement or noise amplification that might occur with global histogram equalization. The contrast-enhanced image is designated as the third intermediate image.

[0041] Then, binarization is performed on the third intermediate image. The Otsu adaptive thresholding segmentation algorithm is employed. This algorithm, based on the principle of maximizing the inter-class variance of the image's gray-level histogram, automatically calculates an optimal global threshold. Pixels with gray values ​​greater than or equal to this threshold are set as foreground pixels (e.g., assigned a value of 255, i.e., white), and pixels with gray values ​​less than this threshold are set as background pixels (e.g., assigned a value of 0, i.e., black). The advantage of the Otsu algorithm is that it eliminates the need for manual threshold setting and adaptively calculates the optimal segmentation threshold based on the actual gray-level distribution of each frame, thus adapting to different lighting conditions and different specimen surface conditions.

[0042] After binarization, a binary image is obtained. In the binary image, the white areas correspond to the main body of the steel slag asphalt mixture specimen and the steel slag particles inside, while the black areas correspond to the background. However, due to the possibility of small holes or shadows on the specimen surface, small black dots (holes) often appear in the white areas or small white dots (isolated noise) in the background areas in the binary image. To optimize the segmentation results, a morphological opening operation is performed on the binary image. The opening operation is a process of erosion followed by dilation, using a 3×3 pixel square structuring element as the convolution kernel. The erosion operation can remove isolated white noise and small protrusions, while the dilation operation can fill in the small holes in the foreground area. After morphological optimization, the binary image has a continuous and complete foreground area and a clean and neat background, accurately reflecting the true contour of the specimen.

[0043] After obtaining the optimized binary image, characteristic parameters representing the expansion state of the steel slag asphalt mixture specimens are extracted based on the binary image. The characteristic parameters include at least two categories: the first category is the basic morphological parameters used to calculate the expansion rate, including the main outline and main area of ​​the specimen; the second category is the microstructural parameters used to analyze the effect of steel slag modification, including the number of steel slag particles, average particle size, and area ratio.

[0044] When extracting the main contour, a combination of the Canny edge detection algorithm and a contour finding algorithm is used. First, Canny edge detection is applied to the binary image, with a low threshold of 50 and a high threshold of 150, to extract the edge point set of the specimen region. Then, the contour finding function is called, using a mode that retrieves the outermost contour, and a contour approximation algorithm is used to compress contour points to reduce redundant data. Among all the found contours, they are filtered based on their area: a minimum area threshold of 500 pixels is set, eliminating noisy contours with excessively small areas, and retaining the contour with the largest area as the main contour of the specimen. Based on the main contour, the pixel area occupied by the specimen in the image is obtained using a pixel area calculation function.

[0045] When extracting features from steel slag particles, the individual contours of steel slag particles are further detected within the main contour area of ​​the specimen. In binary images, steel slag particles appear as relatively bright regions with approximately circular or elliptical shapes. By setting a smaller area threshold, such as 20 pixels, and a roundness constraint, such as a contour area to convex hull area ratio greater than 0.7, the contours of individual steel slag particles are identified. The number of all identified steel slag particle contours is counted and recorded as the number of steel slag particles. For each steel slag particle contour, the diameter of its minimum circumscribed circle is calculated, and the average of all particle diameters is taken as the average particle size. Simultaneously, the sum of the pixel areas of all steel slag particle contours is calculated and divided by the pixel area of ​​the main contour of the specimen to obtain the area ratio of the steel slag particles. These feature parameters are stored in association with the corresponding frame timestamp label, specimen modification type, and other information, forming the input data for subsequent expansion rate calculation.

[0046] Through the above steps of image denoising, geometric correction, target region segmentation, and feature parameter extraction, this embodiment transforms the original visual shape data into a quantified and structured set of feature parameters, laying a reliable data foundation for the accurate calculation of the dilation rate.

[0047] Step S140: Based on the feature parameters and the pre-established mapping relationship between pixel space and physical space, calculate the expansion rate of the steel slag asphalt mixture specimen. The specific details are as follows: In step S140, the expansion rate of the steel slag asphalt mixture specimen is further calculated based on the feature parameters and the pre-established mapping relationship between pixel space and physical space. This calculation process consists of four sub-steps: establishing the mapping relationship, converting the physical area, calculating the expansion amount, and determining the expansion rate, which are described in detail below.

[0048] Since all length and area values ​​in image processing are in pixels, while engineering practice requires physical dimensions in millimeters or square millimeters, a mapping relationship between pixel space and physical space must be established beforehand. The core of this mapping relationship is a conversion factor, which represents the actual physical length corresponding to the side length of each pixel.

[0049] This implementation utilizes a standard calibration block to establish a mapping relationship. Within the image acquisition field of view, a standard calibration block with known actual physical dimensions, such as a square metal block with a side length of ten millimeters, is placed on the same plane as the specimen. In the geometrically corrected image sequence, the pixel region occupied by the calibration block is automatically identified using template matching or edge detection algorithms. The pixel span of this pixel region in the horizontal direction is calculated; for example, the pixel distance between the left and right edges of the calibration block is one hundred pixels. Since the actual side length of the calibration block is ten millimeters, the actual physical length corresponding to each pixel is ten millimeters divided by one hundred pixels, i.e., 0.1 millimeters per pixel. Correspondingly, the actual physical area corresponding to each square pixel is 0.01 square millimeters.

[0050] To verify the accuracy of the mapping relationship, two other reference objects with known actual sizes can be randomly selected from the image, such as other markers of the same size as the calibration block. The conversion error between their image pixel size and actual physical size can then be calculated. If the error is less than one percent, the current conversion coefficient is considered valid; if the error exceeds one percent, the calibration block identification and conversion coefficient calculation are repeated until the accuracy requirements are met. Once this mapping relationship is established, it can be used for area and length conversion of all frames in the image sequence.

[0051] In the feature parameter extraction step, the main pixel area of ​​the specimen at the initial moment before expansion, and the main pixel area at any current moment during the expansion process, have been obtained. Based on the aforementioned mapping relationship, the pixel area is converted into the actual physical area.

[0052] The specific conversion method is as follows: multiply the pixel area at the initial moment by the actual physical area corresponding to each square pixel to obtain the actual physical area at the initial moment. Similarly, multiply the pixel area at the current moment by the same conversion factor to obtain the actual physical area at the current moment. This actual physical area directly reflects the projected area of ​​the specimen in real space, eliminating the scale effects caused by image resolution and camera distance. For example, if each square pixel corresponds to 0.01 square millimeters, and the pixel area at the initial moment is 50,000 square pixels, then the actual physical area at the initial moment is 500 square millimeters.

[0053] After obtaining the actual physical area at the initial and current times, the area expansion of the specimen is calculated. The area expansion is defined as the current actual physical area minus the initial actual physical area. This difference describes the absolute increase in the projected area of ​​the specimen during the expansion process.

[0054] If the specimen maintains uniform radial expansion during the expansion process, there is an approximately proportional relationship between the area expansion and the volume expansion. Since the specimen is a standard cylindrical specimen, its height is constrained by the mold or limited by itself, and the expansion mainly manifests as radial expansion. Therefore, the area expansion can reasonably characterize the degree of volume expansion. In this embodiment, the area expansion is used as an intermediate variable for calculating the volume expansion rate, and its error can be controlled within an acceptable range based on previous experimental calibration, for example, not exceeding five percent.

[0055] The expansion rate is defined as the ratio of the area expansion to the initial actual physical area, expressed as a percentage. Specifically, the area expansion is divided by the initial actual physical area, and the quotient is multiplied by 100% to obtain an approximate value of the specimen's volume expansion rate at the current moment. The calculation result is rounded to two decimal places.

[0056] According to the preset time sequence of image acquisition, the above-mentioned steps of physical area conversion, area expansion calculation, and expansion rate calculation are repeatedly performed on each frame of the image to obtain the expansion rate data sequence of the specimen at different times. For example, at the initial time, the expansion rate is 0%. At the first sampling time, the expansion rate is 0.32%. At the second sampling time, the expansion rate is 0.58%. And so on, until the expansion tends to stabilize. The expansion rate at each time is associated with the corresponding timestamp and stored to form a dynamic dataset of expansion rate changing over time.

[0057] To facilitate subsequent analysis and demonstration, this implementation method also automatically generates a graph of the expansion rate changing over time. With time represented on the horizontal axis and the expansion rate on the vertical axis, the calculation results at each sampling time are plotted as points or lines on a coordinate system. Simultaneously, the peak and stable values ​​of the expansion rate are labeled, where the peak value represents the maximum expansion rate reached during the entire expansion process, and the stable value represents the average value when the change in expansion rate is less than a predetermined threshold in the later stages of expansion, for example, a change of no more than 0.01% per hour. These results can be exported as data tables or graphical files for engineering evaluation.

[0058] Through the above steps of establishing mapping relationships, converting physical areas, calculating expansion amount, and determining expansion rate, this embodiment transforms pixel-level feature parameters in the image domain into expansion rate indicators with clear physical meaning, realizing non-contact, fully automated quantitative evaluation of the expansion process of steel slag asphalt mixture.

[0059] Based on steps S110 to S140 above, in this embodiment of the invention, a sequence of images containing the surface contour of the specimen and the distribution of steel slag particles are collected according to a preset time sequence as visual morphology data. Then, a computer vision algorithm including at least image denoising, geometric correction and target region segmentation is used to automatically extract feature parameters representing the expansion state. The expansion rate is calculated by combining the pre-established mapping relationship between pixel space and physical space. This achieves non-contact, fully automatic and high-precision quantification of the expansion process of steel slag asphalt mixture, avoids human error and significantly improves test efficiency and data reliability.

[0060] The method for determining the expansion rate of steel slag asphalt mixture according to an embodiment of the present invention collects visual morphology data of steel slag asphalt mixture specimens during the expansion process according to a preset time sequence. The method includes: generating an image acquisition command, which carries preset resolution and frame rate parameters; acquiring the original image; converting the original image into a grayscale image; adding timestamp tags to the grayscale image according to a preset time sequence; and generating a sequence of images based on the grayscale image with added timestamp tags. This method automates, standardizes, and makes the data acquisition of the expansion process traceable, significantly improving the stability and efficiency of subsequent processing. Furthermore, based on the visual morphology data, the characteristic parameters representing the expansion state of the steel slag asphalt mixture specimen are determined using computer vision algorithms. These include: denoising each frame of the grayscale image sequence using Gaussian filtering to obtain a first intermediate image; denoising the first intermediate image using median filtering to obtain a second intermediate image; geometrically correcting the second intermediate image using a perspective transformation algorithm to obtain a third intermediate image; enhancing the contrast of the third intermediate image using an adaptive histogram equalization algorithm to obtain a fourth intermediate image; binarizing the fourth intermediate image using an adaptive threshold segmentation algorithm to obtain a binary image; and extracting the main outline and area of ​​the steel slag asphalt mixture specimen based on the binary image. This robustly and accurately extracts the main outline and area features of the steel slag asphalt mixture specimen from the visual morphology data.

[0061] Furthermore, the method also includes: registering the sequence images with the template image of the standard calibration block; identifying the pixel span of the standard calibration block in the sequence images based on the registration result; obtaining the known physical size of the standard calibration block; and establishing a mapping relationship between the pixel space and the physical space based on the ratio of the known physical size to the pixel span, thereby realizing high-precision scale calibration from the pixel space to the physical space and laying a reliable measurement benchmark for the accurate quantification of the dilation rate.

[0062] Furthermore, based on feature parameters and the pre-established mapping relationship between pixel space and physical space, the expansion rate of the steel slag asphalt mixture specimen is calculated, including: converting the main area of ​​the steel slag asphalt mixture specimen before expansion into a first physical area according to the mapping relationship; converting the main area of ​​the steel slag asphalt mixture specimen after expansion into a second physical area according to the mapping relationship; calculating the difference between the second physical area and the first physical area to obtain the physical area expansion amount; calculating the percentage of the physical area expansion amount to the first physical area, and determining the percentage as the expansion rate. This achieves end-to-end accurate automatic quantization from image features to actual expansion indicators, eliminating errors in manual measurement and conversion.

[0063] Furthermore, the method also includes: obtaining multiple expansion rates of steel slag asphalt mixture specimens at multiple different times; smoothing multiple expansion rates using the moving average method; identifying expansion rates that deviate from the moving average by more than a predetermined threshold as abnormal data; and correcting abnormal data by using linear interpolation or the mean of adjacent times, effectively eliminating random errors and fluctuations in the expansion rate time series data, and significantly improving the stability and reliability of the test results.

[0064] Furthermore, the method also includes: obtaining information on the modification type of steel slag used in the steel slag asphalt mixture specimens; and adaptively adjusting at least one adjustable parameter in the computer vision algorithm in response to the modification type information. The adjustable parameters include: kernel size of the filtering algorithm, truncation threshold of the contrast enhancement algorithm, dual threshold of the edge detection algorithm, and area threshold of the contour screening algorithm. This achieves dynamic adaptation to image features of different modified steel slag asphalt mixtures, significantly improving the accuracy of feature extraction and the generalization ability of the method.

[0065] Furthermore, the method also includes: obtaining the measured expansion rate values ​​of steel slag asphalt mixture specimens at multiple sampling times during the expansion process; inputting the measured expansion rate values ​​into a pre-constructed logistic growth curve model; determining the parameters of the logistic growth curve model through least squares fitting; and using the logistic growth curve model with determined parameters to output the predicted long-term expansion rate values ​​of the steel slag asphalt mixture specimens. This achieves long-term expansion trend prediction based on finite short-term data, providing a reliable forward-looking basis for the engineering life assessment of steel slag asphalt mixtures.

[0066] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0067] This invention also provides a device for determining the expansion rate of steel slag asphalt mixture, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0068] Figure 2 According to one embodiment of the present invention, a device for determining the expansion rate of steel slag asphalt mixture includes: The acquisition module 201 is used to acquire visual morphology data of steel slag asphalt mixture specimens during the expansion process according to a preset time sequence. The visual morphology data includes a sequence of images that at least reflect the surface contour of the steel slag asphalt mixture specimens and the distribution of steel slag particles. The determination module 202 is used to determine the characteristic parameters representing the expansion state of the steel slag asphalt mixture specimen based on visual morphology data and through computer vision algorithms. The computer vision algorithms include at least image denoising, geometric correction and target region segmentation. The calculation module 203 is used to calculate the expansion rate of steel slag asphalt mixture specimens based on feature parameters and the pre-established mapping relationship between pixel space and physical space.

[0069] It should be noted that the above modules can be implemented by software or hardware. For the latter, they can be implemented in the following ways, but are not limited to: all the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0070] According to one embodiment of the present invention, an electronic device is also provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the above-described method for determining the expansion rate of steel slag asphalt mixture during runtime.

[0071] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program: Step S1: Collect visual morphology data of steel slag asphalt mixture specimens during the expansion process according to a preset time sequence. The visual morphology data includes a sequence of images that at least reflect the surface contour of the steel slag asphalt mixture specimens and the distribution of steel slag particles. Step S2: Based on the visual morphology data, determine the characteristic parameters representing the expansion state of the steel slag asphalt mixture specimen using a computer vision algorithm. The computer vision algorithm includes at least image denoising, geometric correction, and target region segmentation. Step S3: Based on the feature parameters and the pre-established mapping relationship between pixel space and physical space, calculate the expansion rate of the steel slag asphalt mixture specimen.

[0072] According to one embodiment of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the storage medium is located to execute the above-described method for determining the expansion rate of steel slag asphalt mixture.

[0073] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps: Step S1: Collect visual morphology data of steel slag asphalt mixture specimens during the expansion process according to a preset time sequence. The visual morphology data includes a sequence of images that at least reflect the surface contour of the steel slag asphalt mixture specimens and the distribution of steel slag particles. Step S2: Based on the visual morphology data, determine the characteristic parameters representing the expansion state of the steel slag asphalt mixture specimen using a computer vision algorithm. The computer vision algorithm includes at least image denoising, geometric correction, and target region segmentation. Step S3: Based on the feature parameters and the pre-established mapping relationship between pixel space and physical space, calculate the expansion rate of the steel slag asphalt mixture specimen.

[0074] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0075] According to one embodiment of the present invention, a computer program product is also provided, including a computer program that, when executed by a processor, implements the above-described method for determining the expansion rate of steel slag asphalt mixture.

[0076] Optionally, in this embodiment, the above-mentioned computer program product can be configured as a computer program that performs the following steps: Step S1: Collect visual morphology data of steel slag asphalt mixture specimens during the expansion process according to a preset time sequence. The visual morphology data includes a sequence of images that at least reflect the surface contour of the steel slag asphalt mixture specimens and the distribution of steel slag particles. Step S2: Based on the visual morphology data, determine the characteristic parameters representing the expansion state of the steel slag asphalt mixture specimen using a computer vision algorithm. The computer vision algorithm includes at least image denoising, geometric correction, and target region segmentation. Step S3: Based on the feature parameters and the pre-established mapping relationship between pixel space and physical space, calculate the expansion rate of the steel slag asphalt mixture specimen.

[0077] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0078] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0079] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0080] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0081] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0082] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0083] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for determining the expansion rate of steel slag asphalt mixture, characterized in that, include: Visual morphology data of steel slag asphalt mixture specimens during the expansion process are collected according to a preset time sequence. The visual morphology data includes a sequence of images that at least reflect the surface contour of the steel slag asphalt mixture specimens and the distribution of steel slag particles. Based on the visual morphology data, the characteristic parameters representing the expansion state of the steel slag asphalt mixture specimen are determined by computer vision algorithms. The computer vision algorithms include at least image denoising, geometric correction, and target region segmentation. Based on the aforementioned feature parameters and the pre-established mapping relationship between pixel space and physical space, the expansion rate of the steel slag asphalt mixture specimen is calculated.

2. The method according to claim 1, characterized in that, Visual morphology data of steel slag asphalt mixture specimens during the expansion process were collected according to a preset time sequence, including: Generate an image acquisition command, wherein the image acquisition command carries preset resolution parameters and frame rate parameters; Obtain the original image; Convert the original image to a grayscale image; Add timestamp tags to the grayscale image according to the preset time sequence; The sequence of images is generated based on the grayscale image with added timestamp tags.

3. The method according to claim 2, characterized in that, Based on the visual morphology data, characteristic parameters representing the expansion state of the steel slag asphalt mixture specimens are determined using computer vision algorithms, including: Gaussian filtering is used to denoise each frame of grayscale image in the sequence of images to obtain the first intermediate image; The first intermediate image is denoised using median filtering to obtain the second intermediate image; A perspective transformation algorithm is used to perform geometric correction on the second intermediate image to obtain a third intermediate image; An adaptive histogram equalization algorithm is used to enhance the contrast of the third intermediate image to obtain a fourth intermediate image. The fourth intermediate image is binarized using an adaptive threshold segmentation algorithm to obtain a binary image. Based on the binary image, the main outline and main area of ​​the steel slag asphalt mixture specimen are extracted.

4. The method according to claim 1 or 2, characterized in that, The method further includes: The sequence of images is registered with the template image of the standard calibration block; Based on the registration results, the pixel span of the standard calibration block in the image sequence is identified; Obtain the known physical dimensions of the standard calibration block; A mapping relationship between the pixel space and the physical space is established based on the ratio of the known physical size to the pixel span.

5. The method according to claim 1, characterized in that, Based on the aforementioned feature parameters and the pre-established mapping relationship between pixel space and physical space, the expansion rate of the steel slag asphalt mixture specimen is calculated, including: Based on the mapping relationship, the main body area of ​​the steel slag asphalt mixture specimen before expansion is converted into the first physical area; Based on the mapping relationship, the main body area of ​​the expanded steel slag asphalt mixture specimen is converted into a second physical area; Calculate the difference between the second physical area and the first physical area to obtain the physical area expansion. Calculate the percentage of the expansion of the physical area to the first physical area, and determine the percentage as the expansion rate.

6. The method according to claim 5, characterized in that, The method further includes: Obtain multiple expansion rates of the steel slag asphalt mixture specimen at multiple different times; The multiple expansion rates are smoothed using the moving average method; Expansion rates that deviate from the moving average by more than a predetermined threshold are identified as abnormal data; The abnormal data are corrected by using linear interpolation or the mean of adjacent time points.

7. The method according to claim 1, characterized in that, The method further includes: Obtain information on the modification type of the steel slag used in the steel slag asphalt mixture specimens; In response to the modification type information, at least one adjustable parameter in the computer vision algorithm is adaptively adjusted; The adjustable parameters include: kernel size of the filtering algorithm, truncation threshold of the contrast enhancement algorithm, dual threshold of the edge detection algorithm, and area threshold of the contour screening.

8. The method according to claim 1 or 7, characterized in that, The method further includes: The measured expansion rate of the steel slag asphalt mixture specimen was obtained at multiple sampling times during the expansion process; The measured value of the expansion rate is input into the pre-constructed logistic growth curve model; The parameters of the logistic growth curve model were determined by fitting using the least squares method. Using the logistic growth curve model with determined parameters, the predicted long-term expansion rate of the steel slag asphalt mixture specimen is output.

9. A device for determining the expansion rate of steel slag asphalt mixture, characterized in that, include: The acquisition module is used to acquire visual morphology data of steel slag asphalt mixture specimens during the expansion process according to a preset time sequence. The visual morphology data includes a sequence of images that at least reflect the surface contour of the steel slag asphalt mixture specimens and the distribution of steel slag particles. The determination module is used to determine, based on the visual morphology data, the characteristic parameters representing the expansion state of the steel slag asphalt mixture specimen using a computer vision algorithm, wherein the computer vision algorithm includes at least image denoising, geometric correction, and target region segmentation. The calculation module is used to calculate the expansion rate of the steel slag asphalt mixture specimen based on the feature parameters and the pre-established mapping relationship between pixel space and physical space.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the storage medium is located to perform the method for determining the expansion rate of steel slag asphalt mixture according to any one of claims 1 to 8.

Images