Hot recycled asphalt mixture heating monitoring system for road and bridge construction

Abstract

The application relates to the technical field of heating control, in particular to a hot regeneration asphalt mixture heating monitoring system for road and bridge construction. The system comprises an acquisition module, a first processing module and a second processing module. The acquisition module is used for acquiring a thermal infrared grayscale image in an asphalt heating tank, determining local abnormal pixel points, and further determining an initial distribution uniformity of the thermal infrared grayscale image. The first processing module is used for obtaining a local abnormal connected domain and determining a temperature distribution complexity of the local abnormal connected domain. The second processing module is used for determining a real distribution uniformity according to the temperature distribution complexity and the initial distribution uniformity. The monitoring module is used for determining an adaptive enhancement weight value, obtaining a target image, and obtaining a monitoring result of each monitoring moment in a heating process according to the target image. The application can realize adaptive image enhancement, and effectively solve the problems of low contrast, fuzzy visual effect and large interference information of the thermal infrared grayscale image.

Description

Technical Field

[0001] This application relates to the field of industrial monitoring technology, specifically to a monitoring system for the heating and mixing of hot recycled asphalt used in road and bridge construction. Background Technology

[0002] Hot recycled asphalt mixing involves milling off old asphalt pavement, transporting it to recycling equipment, adding appropriate amounts of new asphalt and additives to form a new asphalt mixture, and then using pavers and other equipment to lay it on bridges and roads to create a new asphalt pavement. This process significantly reduces the amount of new asphalt and other raw materials used, thereby saving resources, reducing costs, and mitigating environmental impact. During the hot recycled asphalt mixing process, it is necessary to stir and mix the old and new asphalt with other auxiliary materials to ensure uniform temperature distribution and material homogeneity. Uneven material temperatures or localized excessively high or low temperatures during heating will affect the performance and quality of the asphalt mixture.

[0003] In the industrialization process of hot recycled asphalt heating, monitoring equipment in a monitoring system is needed to monitor the heating process for safe production and raw material addition. Currently, thermal imagers are used to acquire images of the recycling equipment during the heating process. These images are then enhanced, and heating monitoring is performed based on the enhanced images. However, this method suffers from significant variations in image enhancement results due to differences in heating conditions at different stages. This leads to low-quality images with low contrast and blurred visual effects, negatively impacting the monitoring system's effectiveness in monitoring the heating process. Summary of the Invention

[0004] To address the technical problems of low contrast, blurred visual effects, and high levels of interference in enhanced images, this application provides a hot recycled asphalt mixing and heating monitoring system for road and bridge construction. The specific technical solution adopted is as follows:

[0005] This application proposes a hot recycled asphalt mixing heating monitoring system for road and bridge construction, the system comprising:

[0006] The acquisition module is used to monitor the acquisition of thermal infrared grayscale images at each monitoring moment during the hot recycling asphalt mixing and heating process, determine local abnormal pixels based on the grayscale values ​​of the pixels in the thermal infrared grayscale images, and determine the initial distribution uniformity of the thermal infrared grayscale images based on the distribution uniformity of the local abnormal pixels and the grayscale differences between the local abnormal pixels.

[0007] The first processing module is used to perform region growing on each of the local abnormal pixels to obtain a local abnormal connected region, obtain the neighboring pixels of each pixel in the local abnormal connected region, and determine the temperature distribution complexity of the local abnormal connected region based on the gray value variance of all pixels in the local abnormal connected region and the gradient direction between the pixel and its neighboring pixels.

[0008] The second processing module is used to determine the true uniformity of the thermal infrared grayscale image acquired at each monitoring time based on the complexity of the temperature distribution of all local abnormal connected regions in the thermal infrared grayscale image acquired at each monitoring time and the initial uniformity of the thermal infrared grayscale image.

[0009] The monitoring module is used to enhance the thermal infrared grayscale image acquired at each monitoring time according to the actual distribution uniformity to obtain the target image; and to determine the heating monitoring result at each monitoring time during the hot recycled asphalt mixing heating process based on the target image obtained at each monitoring time.

[0010] Preferably, determining local abnormal pixels includes:

[0011] The quartiles of the set of gray values ​​of all pixels in the thermal infrared grayscale image obtained at each monitoring time are statistically analyzed, and the preset first grayscale threshold and preset second grayscale threshold are determined based on the quartiles.

[0012] Pixels with gray values ​​greater than a preset first gray value threshold in the thermal infrared grayscale image obtained at each monitoring time are designated as high-temperature pixels, and pixels with gray values ​​less than a preset second gray value threshold in the thermal infrared grayscale image obtained at each monitoring time are designated as low-temperature pixels.

[0013] Both the high-temperature pixel and the low-temperature pixel are considered as local abnormal pixels.

[0014] Preferably, determining the initial distribution uniformity of the thermal infrared grayscale image includes:

[0015] The absolute value of the difference between the gray value of each local anomalous pixel in the thermal infrared grayscale image and the mean of the gray values ​​of all local anomalous pixels is taken as the mean of the sum of the results of all local anomalous pixels.

[0016] The first influence factor is determined based on the uniformity of the position distribution of all local abnormal pixels in the thermal infrared grayscale image;

[0017] Calculate the product of the first influencing factor and the abnormal temperature parameter, and use the normalized result of the negative of the product as the initial distribution uniformity of the thermal infrared grayscale image.

[0018] Preferably, the method for obtaining the first impact factor is as follows:

[0019] The thermal infrared grayscale image is uniformly divided into a preset number of grayscale sub-images;

[0020] The sum of the variances of the number of high-temperature pixels and the variances of the number of low-temperature pixels in all grayscale sub-images is used as the first influencing factor.

[0021] Preferably, obtaining the neighboring pixels of each pixel within the local abnormal connected component includes:

[0022] For any local anomalous pixel, each local anomalous pixel is used as an initial seed point, and the region growing algorithm is used to obtain the extraction result of the local anomalous connected region in the thermal infrared grayscale image.

[0023] The direction perpendicular to the gradient direction of any pixel within the local abnormal connected region is determined as the target direction;

[0024] In the target direction, the two other pixels that are closest to the pixel are identified as the neighboring pixels of the pixel.

[0025] Preferably, determining the complexity of the temperature distribution of the local anomalous connected domain includes:

[0026] The intersection points of the line containing the gradient direction of any pixel within the local abnormal connected region and the lines containing the gradient directions of adjacent pixels are counted respectively. The distance influence factor is determined based on the number of intersection points and the distance between the intersection points.

[0027] The normalized result of the inverse of the mean of the distance influence factors of all pixels in the local abnormal connected region is used as the temperature distribution complexity of the local abnormal connected region.

[0028] Preferably, determining the distance influence factor based on the number of intersections and the distance between intersections includes:

[0029] When the number of intersections is less than 2, the distance influence factor is determined to be 1;

[0030] When the number of intersection points is equal to 2, the distance influence factor is obtained by normalizing the Euclidean distance between the two intersection points.

[0031] Preferably, determining the true uniformity of the distribution of the thermal infrared grayscale image acquired at each monitoring time includes:

[0032] Calculate the mean of the temperature distribution complexity of all local abnormal connected regions in the obtained thermal infrared grayscale image at each monitoring time.

[0033] The sum of the normalized mean and the constant parameter is used as the denominator;

[0034] The ratio of the initial distribution uniformity of the thermal infrared grayscale image obtained at each monitoring time to the denominator is taken as the true distribution uniformity of the thermal infrared grayscale image obtained at each monitoring time.

[0035] Preferably, the method for obtaining the target image is as follows:

[0036] The ratio of the inverse proportional mapping result of the true distribution uniformity of the thermal infrared grayscale image obtained at each monitoring time to the sum of the inverse proportional mapping results of the preset number of monitoring times before the current time is used as the weight parameter of the thermal infrared grayscale image obtained at each monitoring time.

[0037] The product of the first preset parameter and the weight parameter, plus the sum of the second preset parameter, is used as the adaptive enhancement weight of the thermal infrared grayscale image obtained at each monitoring time.

[0038] The thermal infrared grayscale image is layered based on a bilateral filtering method to obtain a background layer image and a detail layer image.

[0039] The adaptive enhancement weights are used as the weights of the detail layer image, and the difference between 1 and the adaptive enhancement weights is used as the weights of the background layer image. Weighted fusion is then performed to obtain the target image.

[0040] Preferably, the determination of the heating monitoring results at each monitoring moment during the hot recycled asphalt mixing and heating process includes:

[0041] Based on the target image at each monitoring moment, the temperature distribution results at each monitoring moment during the hot recycled asphalt mixing and heating process are accurately visualized. By comparing and analyzing with the standard temperature distribution results, the heating monitoring results at each monitoring moment are obtained. The monitoring results include two categories: qualified and unqualified.

[0042] This application has the following beneficial effects:

[0043] This application acquires thermal infrared grayscale images of asphalt heating tanks and determines the initial distribution uniformity based on the number and distribution of local abnormal pixels and their corresponding grayscale values. This enables effective analysis of the grayscale distribution of local abnormal pixels. Furthermore, it performs region growing on these local abnormal pixels. Based on the variance of grayscale values ​​of all pixels within the local abnormal connected region and the gradient direction between the pixel and its neighboring pixels, it determines the temperature distribution complexity of the local abnormal connected region. Using the gradient direction information of neighboring pixels, it obtains the temperature distribution complexity of the corresponding local abnormal connected region. Combining the temperature distribution complexity of the local abnormal connected region with the initial distribution uniformity, it obtains the true distribution uniformity. The initial distribution uniformity is then adjusted based on the temperature distribution complexity to obtain the true distribution uniformity. This allows for adaptive enhancement weights to be obtained based on the true distribution uniformity, enabling adaptive enhancement of the thermal infrared grayscale image and thus achieving monitoring of the heating of asphalt mixtures. This application combines the grayscale values, grayscale gradients, and grayscale distributions of local abnormal pixels with those of neighboring pixels to analyze the entire thermal infrared grayscale image. This allows subsequent image enhancement to effectively consider the grayscale distribution of the thermal infrared grayscale image itself, thereby achieving adaptive image enhancement. This effectively solves the problems of low contrast, blurred visual effects, and large interference information in thermal infrared grayscale images, and improves the accuracy of monitoring results during the heating process. Attached Figure Description

[0044] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a structural diagram of a hot recycled asphalt mixing heating monitoring system for road and bridge construction, provided in one embodiment of this application. Detailed Implementation

[0046] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the hot recycled asphalt mixing heating monitoring system for road and bridge construction proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0048] The following description, in conjunction with the accompanying drawings, details the specific scheme of the hot recycled asphalt mixing heating monitoring system for road and bridge construction provided in this application.

[0049] Please see Figure 1 The diagram illustrates a structural diagram of a hot recycled asphalt mixing heating monitoring system for road and bridge construction provided in an embodiment of this application. The monitoring system includes: an acquisition module 101, a first processing module 102, a second processing module 103, and a monitoring module 104.

[0050] The acquisition module 101 is used to monitor the acquisition of thermal infrared grayscale images at each monitoring moment during the hot recycling asphalt mixing and heating process, determine local abnormal pixels based on the grayscale values ​​of the pixels in the thermal infrared grayscale images, and determine the initial distribution uniformity of the thermal infrared grayscale images based on the distribution uniformity of the local abnormal pixels and the grayscale differences between the local abnormal pixels.

[0051] In the asphalt heating tank, the inconsistent size of the asphalt blocks during heating leads to varying degrees of uniform heating among the different sized blocks. As heating progresses, the heating becomes more uniform. This process of decreasing heating variation corresponds to the different stages of reheating a mixture of new and old asphalt. Because old asphalt is solid and more difficult to melt than new asphalt after prolonged use, the different degrees of melting between the old and new asphalt mixtures result in a prolonged solid-liquid coexistence during reheating.

[0052] For the boundary region of solid-liquid contact, it will exhibit obvious grayscale value changes and relatively complex grayscale textures in the infrared grayscale image acquired by the monitoring equipment. At different reheating stages, it is easy to understand that as the degree of reheating increases, boundary regions with varying degrees of grayscale change and texture complexity will be generated. These differences in grayscale change and texture complexity directly affect the optimal enhancement level for the current infrared grayscale image. The monitoring equipment is a thermal imager in the monitoring system, used to achieve safety monitoring of the heating process.

[0053] Furthermore, as the degree of reheating increases, the proportion of solid asphalt in the mixture of new and old asphalt gradually decreases, that is, the area of ​​solid asphalt gradually decreases, and thus the area of ​​the aforementioned boundary region also gradually decreases. The boundary region represents the area of ​​the region that exhibits changes in grayscale values ​​and relatively complex grayscale textures, so its change will also affect the value of the current infrared grayscale image enhancement.

[0054] In summary, it can be seen that during the reheating process of the new and old asphalt mixture, due to the constant change in the solid-liquid ratio and the gradual reduction of the solid area, the optimal image enhancement level for the infrared grayscale images collected by the monitoring system at different times will also change accordingly. The embodiments of this application provide a method for monitoring the industrial process of reheating the new and old asphalt mixture before road and bridge pavement construction. This method involves determining the size of the aforementioned edge regions, the magnitude of grayscale changes, and the complexity of grayscale textures in the infrared grayscale images collected by the monitoring equipment, thereby determining the optimal enhancement level for the infrared grayscale images collected by the monitoring equipment to achieve accurate monitoring of the mixing and heating process.

[0055] In this embodiment of the application, an infrared thermal imager can be used to periodically acquire infrared thermal images, and each frame of infrared thermal image can be preprocessed to obtain a thermal infrared grayscale image.

[0056] Image preprocessing can specifically include image denoising and image grayscale processing. It can be understood that in thermal infrared grayscale images, the higher the temperature, the higher the grayscale value of the pixel corresponding to the pixel, and the lower the temperature, the lower the grayscale value of the pixel corresponding to the pixel.

[0057] Specifically, in this embodiment, an infrared thermal image of the asphalt heating tank is acquired every minute, and each infrared thermal image is subjected to image denoising and grayscale processing to obtain several thermal infrared grayscale images.

[0058] For any thermal infrared grayscale image, in this embodiment, local abnormal pixels are determined based on the grayscale values ​​of the pixels in the thermal infrared grayscale image, including: pixels with grayscale values ​​greater than a preset first grayscale threshold are designated as high-temperature pixels, and pixels with grayscale values ​​less than a preset second grayscale threshold are designated as low-temperature pixels, wherein the preset first grayscale threshold is greater than the preset second grayscale threshold; both high-temperature pixels and low-temperature pixels are designated as local abnormal pixels.

[0059] Among them, local abnormal pixels are pixels with excessively high or low temperatures. It can be understood that during the heating process of the asphalt heating tank, some areas may be overheated, while other areas may be underheated due to the excessive size of the asphalt blocks. These situations can be called abnormal situations, and the corresponding pixels can be called local abnormal pixels.

[0060] Therefore, in this embodiment of the application, pixels with gray values ​​greater than a preset first gray value threshold are designated as high-temperature pixels, pixels with gray values ​​less than a preset second gray value threshold are designated as low-temperature pixels, and high-temperature pixels and low-temperature pixels are collectively designated as local abnormal pixels.

[0061] In this embodiment, both the preset first grayscale threshold and the preset second grayscale threshold are threshold values ​​for grayscale values, with the preset first grayscale threshold being greater than the preset second grayscale threshold. In this embodiment, the quartiles of the set of grayscale values ​​of all pixels in the thermal infrared grayscale image are statistically analyzed, and anomaly analysis is performed based on these quartiles to determine the preset first grayscale threshold and the preset second grayscale threshold. The calculation of the quartiles is a well-known technique in the field of data processing, and the specific process will not be elaborated further.

[0062] The quartiles include Q1 and Q3, and IQR = Q3 - Q1. Q1, Q3, and IQR are all well-known concepts in the quartile method. The preset second gray threshold is Q1 - 1.5 × IQR, and the preset first gray threshold is Q3 + 1.5 × IQR. Here, 1.5 is an empirical coefficient set based on experimental conditions and can be adjusted according to the detection situation in different scenarios. There are no restrictions on this.

[0063] Furthermore, in this embodiment, the initial distribution uniformity of the thermal infrared grayscale image is determined based on the total number of local abnormal pixels, their distribution in the thermal infrared grayscale image, and the grayscale values ​​of the local abnormal pixels, including:

[0064] The absolute value of the difference between the gray value of any local abnormal pixel and the mean gray value of all local abnormal pixels is taken as the gray value mean difference of the local abnormal pixel; the mean of the gray value differences of all local abnormal pixels is taken as the abnormal temperature parameter; the thermal infrared grayscale image is divided into a preset number of grayscale sub-images, and the sum of the variances of the number of high-temperature pixels and the number of low-temperature pixels in all grayscale sub-images is calculated as the first influence factor; the inversely proportional normalized value of the product of the abnormal temperature parameter and the first influence factor is taken as the initial distribution uniformity.

[0065] Among them, the abnormal temperature parameter and the first influencing factor are negatively correlated with the initial distribution uniformity.

[0066] Among them, a negative correlation means that the dependent variable decreases as the independent variable increases, and the dependent variable increases as the independent variable decreases. It can be a subtraction relationship, a division relationship, etc., which is determined by the actual application.

[0067] In this embodiment, the thermal infrared grayscale image is uniformly divided into several grayscale sub-images. The number of grayscale sub-images can be adjusted according to the size of the thermal infrared grayscale image. A larger thermal infrared grayscale image may contain richer image information, allowing for a relatively larger number of grayscale sub-images; conversely, a smaller thermal infrared grayscale image only requires dividing it into a few grayscale sub-images to characterize the image features. Here, an empirical value of 100 is set, uniformly dividing the thermal infrared grayscale image into 100 grayscale sub-images.

[0068] Preferably, in one embodiment of this application, the process for obtaining the initial distribution uniformity is as follows:

[0069] First, an abnormal temperature parameter is calculated based on the grayscale values ​​of locally anomalous pixels to characterize the overall temperature difference at different locations in the thermal infrared grayscale image. A larger value for the abnormal temperature parameter indicates a greater difference in temperature values ​​between locally anomalous pixels at different locations, and a more uneven temperature distribution; conversely, a smaller value indicates a relatively smaller difference in temperature values ​​between locally anomalous pixels at different locations, and a more uniform overall temperature distribution.

[0070]

[0071] In the formula, This represents the abnormal temperature parameter of the thermal infrared grayscale image, where n represents the number of locally abnormal pixels. This represents the grayscale value of the i-th locally abnormal pixel. This represents the average grayscale value of all locally abnormal pixels. This indicates taking the absolute value.

[0072] Secondly, based on the difference in the number of high-temperature pixels and the number of low-temperature pixels in the grayscale sub-images divided by the thermal infrared grayscale image, and combined with the aforementioned abnormal temperature parameter, the initial distribution uniformity A of the thermal infrared grayscale image is determined:

[0073]

[0074] In the formula, Represents the normalization function. , These are the variances of the number of high-temperature pixels and the number of low-temperature pixels in all grayscale sub-images, respectively.

[0075] It should be noted that normalization aims to make the calculation result range from 0 to 1. This is a common technique in data processing, including but not limited to Max-Min normalization and Z-score normalization. Preferably, Max-Min normalization is used in this embodiment.

[0076] Among them, the more discrete the location distribution of each local high temperature and local low temperature region within the asphalt heating tank, that is, the variance of the number of high temperature pixels in all grayscale sub-images... The variance of the number of low-temperature pixels A larger value indicates a lower degree of fusion between different materials within the asphalt heating tank and a more uneven temperature distribution; this is the primary influencing factor. The larger the value of A, the more uniform the temperature difference and temperature distribution inside the asphalt heating tank, and the higher the degree of fusion between different materials.

[0077] It is understood that, since the embodiments of this application are intended to monitor the fusion process of various materials in the asphalt heating tank, the number of local abnormal pixels n cannot be 0 in the embodiments of this application. When the number of local abnormal pixels is 0, it is considered that the fusion process has not started or has been completed, and no monitoring is performed at this time.

[0078] The first processing module 102 is used to perform region growing on each of the local abnormal pixels to obtain a local abnormal connected region, obtain the neighboring pixels of each pixel in the local abnormal connected region, and determine the temperature distribution complexity of the local abnormal connected region based on the gray value variance of all pixels in the local abnormal connected region and the gradient direction of the pixel and its neighboring pixels.

[0079] Region growing is a well-known image processing technique in this field. It uses local anomalous pixels as seed points and merges adjacent pixels with similar gray levels to each seed point into the corresponding region to obtain a local anomalous connected region. It can be understood that multiple local anomalous pixels can be merged into the same local anomalous connected region.

[0080] Furthermore, for each locally anomalous connected component, the neighboring pixels of each pixel within the locally anomalous connected component are determined, including: determining the direction perpendicular to the gradient direction of any pixel within the locally anomalous connected component as the target direction; and determining the two other pixels closest to the pixel within the target direction as the neighboring pixels of the pixel.

[0081] In other words, by drawing a perpendicular line along the gradient direction of each pixel in the local abnormal connected domain, the corresponding target direction is obtained. It can be understood that since there are two directions along the perpendicular line, either one can be chosen as the target direction in this embodiment without restriction.

[0082] In this embodiment, a straight line can be drawn along the target direction through the pixel point. The two other pixels on the straight line that are closest to the pixel point are taken as the neighboring pixels of the pixel point. In this way, the neighboring pixels of each pixel point in the local abnormal connected region are obtained.

[0083] Furthermore, in this embodiment of the application, the temperature distribution complexity of the local abnormal connected region is determined based on the gray value variance of all pixels in the local abnormal connected region and the gradient direction of the pixel and its neighboring pixels. This includes: counting the intersection points of the line where the gradient direction of any pixel in the local abnormal connected region is located and the line where the gradient direction of the neighboring pixels is located; determining the distance influence factor based on the number of intersection points and the distance between the intersection points; and calculating the normalized value of the negative of the mean of the distance influence factor of all pixels in the local abnormal connected region as the temperature distribution complexity of the local abnormal connected region.

[0084] In this embodiment, any pixel within a locally abnormal connected region is taken as the pixel to be tested. The neighboring pixels of the pixel to be tested are determined, and the gradient directions of the pixel to be tested and its neighboring pixels are obtained. Along the gradient direction, straight lines are drawn through the pixel to be tested and its neighboring pixels respectively. The intersection point of the straight lines corresponding to the two neighboring pixels and the straight line corresponding to the pixel to be tested is determined. That is to say, the intersection point is the intersection point of the straight lines corresponding to the neighboring pixels and the straight lines corresponding to the pixel to be tested. Since there are 2 neighboring pixels, the number of corresponding intersection points can be 0, 1 or 2, which will be discussed in this embodiment.

[0085] Furthermore, in this embodiment of the application, the distance influence factor is determined based on the number of intersections and the distance between the intersections, including: when the number of intersections is less than 2, the distance influence factor is determined to be 1; when the number of intersections is equal to 2, the Euclidean distance between the two intersections is normalized to obtain the distance influence factor.

[0086] In this embodiment of the application, when the number of intersection points is less than 2, it means that the number of intersection points of the gradient direction of the pixel and the straight line of its neighboring pixel does not meet the intersection condition. The distance influence factor is set to 1. When the number of intersection points is 2, the Euclidean distance between the two intersection points is normalized to obtain the distance influence factor.

[0087] In this embodiment of the invention, the Euclidean distance between two intersection points is normalized. Specifically, a threshold truncation method can be used, that is, the minimum value of the Euclidean distance between the two intersection points is set to 1, and the maximum value is a preset constant, such as 300. That is, when the Euclidean distance is greater than 300, the distance influence factor is directly set to 1. When the Euclidean distance is [1, 300], linear normalization is performed. For example, when the Euclidean distance is 150, the distance influence factor is 0.5.

[0088] It is understandable that the smaller the Euclidean distance between two intersection points, the smaller the corresponding distance influence factor, which can characterize the complex temperature changes in the region where the pixel is located, and the greater the influence of the unstable temperature around the local abnormal connected domain.

[0089] In this embodiment, the complexity of temperature distribution characterizes the degree of disorder in the temperature distribution within the region corresponding to the local abnormal connected domain. The normalized value of the inverse of the mean of the distance influence factor of all pixels in the local abnormal connected domain is calculated as the complexity of the temperature distribution of the local abnormal connected domain. The smaller the distance influence factor, the greater the influence of the unstable temperature around the local abnormal connected domain on the pixel, and the more complex the corresponding temperature distribution, and the greater the complexity of the temperature distribution.

[0090] The second processing module 103 is used to determine the true uniformity of the thermal infrared grayscale image acquired at each monitoring time based on the complexity of the temperature distribution of all local abnormal connected regions in the thermal infrared grayscale image acquired at each monitoring time and the initial uniformity of the thermal infrared grayscale image.

[0091] It is understandable that a thermal infrared grayscale image may contain multiple locally anomalous connected components, each with a corresponding degree of temperature distribution complexity. Therefore, the mean of the temperature distribution complexity of all locally anomalous connected components is calculated, and the true distribution uniformity P of the thermal infrared grayscale image is determined based on this mean and the initial distribution uniformity of the thermal infrared grayscale image.

[0092]

[0093] In the formula, P represents the true uniformity of the thermal infrared grayscale image distribution, and A represents the initial uniformity of the thermal infrared grayscale image distribution. Represents the normalization function. The mean value representing the complexity of the temperature distribution of all locally anomalous connected regions in a thermal infrared grayscale image. This represents a constant coefficient, used to avoid a denominator of 0, in order to minimize the impact on the calculation results. The value range is set to (0, 0.01]. Preferably, in this embodiment... Take the empirical value of 0.001.

[0094] In thermal infrared grayscale images, the greater the complexity of the temperature distribution of all local abnormal connected regions, the greater the influence of unstable temperatures around the local abnormal connected regions on the pixels in each local temperature abnormal region. The temperature unevenness at different locations will be affected by the abnormal temperatures of multiple local abnormal pixels, and the true distribution uniformity should be smaller. Conversely, the smaller the mean of the temperature distribution complexity, the smaller the influence of local abnormal pixels on the surrounding areas, the higher the overall temperature uniformity, and the true distribution uniformity should be larger.

[0095] The monitoring module 104 is used to enhance the thermal infrared grayscale image acquired at each monitoring time according to the actual distribution uniformity to obtain the target image; and to determine the heating monitoring result at each monitoring time during the hot recycled asphalt mixing heating process based on the target image obtained at each monitoring time.

[0096] Furthermore, in this embodiment of the application, the adaptive enhancement weight is determined based on the true distribution uniformity. Specifically, the true distribution uniformity is inversely mapped, and the ratio of the inverse mapping result of the true distribution uniformity of each thermal infrared grayscale image to the sum of all the inverse mapping results is used as the weight parameter of each thermal infrared grayscale image.

[0097] When the uniformity of the true distribution is greater, it indicates that the gray-scale distribution in the thermal infrared gray-scale image is more uniform. In this case, a smaller weight needs to be set to avoid over-sharpening of the image due to excessive enhancement of details in the thermal infrared gray-scale image. Conversely, when the uniformity of the true distribution is smaller, a larger weight can be set to highlight the detailed features of the thermal infrared gray-scale image. Thus, the adaptive enhancement weight is determined based on the uniformity of the true distribution.

[0098] Specifically, during the heating process of hot recycled asphalt mixing, a sliding window is used for real-time calculation. When the temperature distribution is chaotic and the uniformity of the actual distribution is low, there is more image detail information, so a larger adaptive enhancement weight is assigned to obtain clearer detail information. When the temperature distribution is uniform and the uniformity of the actual distribution is high, there is less image detail information, so a smaller fusion weight is assigned to prevent the image from being over-sharpened.

[0099] Furthermore, for any thermal infrared grayscale image, taking the thermal infrared grayscale image acquired at the t-th monitoring time as an example, the weight parameters of the thermal infrared grayscale image acquired at the t-th monitoring time are expressed as follows: :

[0100]

[0101] In the formula, This represents an exponential function with the natural constant as its base. It represents the true uniformity of the distribution of the thermal infrared grayscale image acquired at the t-th monitoring time, and N is the preset sliding window size, representing the number of historical monitoring times participating in the weight calculation, such as 10 monitoring times.

[0102] It should be noted that if the complete sliding window size is not met before the t-th monitoring time, that is, if the corresponding 10 monitoring times are not included, the mean of the uniformity of all real distributions before the t-th monitoring time can be used as the missing element to fill in the missing elements of the sliding window, and the weight parameters can be calculated based on the filling results.

[0103] Furthermore, in this embodiment of the invention, the adaptive enhancement weight of the thermal infrared grayscale image acquired at each monitoring moment is determined according to the weight parameter. Since the enhancement method used subsequently is bilateral filtering layered enhancement, the value range of the adaptive enhancement weight should be controlled to be greater than or equal to 0 and less than 1. In this embodiment, the value range of the adaptive enhancement weight is controlled to be [0.5, 0.7].

[0104] In other embodiments, the value range can be adjusted according to the resolution of the monitoring device. Specifically, the higher the resolution of the monitoring device, the richer the detail information of the high-frequency detail layer obtained by bilateral filtering. In this case, the upper limit of the value range can be increased to make the detail information in the enhanced result more significant. The lower the resolution of the monitoring device, the lower the contrast between low and high frequencies obtained by bilateral filtering. In this case, the lower limit of the value range can be increased to improve the contrast in the enhanced result.

[0105] Specifically, the weighting parameters of the thermal infrared grayscale image acquired at the t-th monitoring time are expressed as: :

[0106]

[0107] Furthermore, in this embodiment of the application, the image enhancement processing of the thermal infrared grayscale image according to the adaptive enhancement weight to obtain the target image includes: performing image layering on the thermal infrared grayscale image based on the bilateral filtering processing method to obtain a background layer image and a detail layer image; using the adaptive enhancement weight as the weight of the detail layer image, and the difference between 1 and the adaptive enhancement weight as the weight of the background layer image, and performing weighted fusion to obtain the target image.

[0108] The bilateral filtering method is a commonly used image layering method in this field, and will not be elaborated further. The bilateral filtering method decomposes the thermal infrared grayscale image into a high-frequency detail layer and a low-frequency background layer, where the high-frequency detail layer corresponds to the detail layer image and the low-frequency background layer corresponds to the background layer image. In this embodiment, the adaptive enhancement weight can be used as the weight of the detail layer image, and the difference between 1 and the adaptive enhancement weight can be used as the weight of the background layer image. Because the thermal infrared grayscale image contains many texture details, assigning a larger adaptive enhancement weight can obtain clearer detail information. Therefore, based on the adaptive enhancement weight as the weight of the detail layer image, reliable image enhancement is achieved. The weighted fusion image enhancement technique is also well-known in this field and will not be elaborated further.

[0109] In this embodiment, after obtaining the target image, the temperature distribution at each monitoring moment during the asphalt mixing process can be accurately visualized based on the target image. By comparing and analyzing with the standard temperature distribution results, the monitoring results at each monitoring moment are obtained. The monitoring results include two categories: qualified and unqualified. Thus, variables such as heating intensity, heating area, and heating time can be controlled according to the target image. Preferably, in this embodiment, based on the target image at each monitoring moment, the maximum, minimum, and average grayscale values ​​in the target image are statistically analyzed and compared with the standard temperature distribution results. If the temperature values ​​corresponding to the maximum and minimum grayscale values ​​are within the standard temperature range, and the temperature value corresponding to the average grayscale value is within the preset ideal temperature range at each monitoring moment, then the mixing and heating state of the hot recycled asphalt at that monitoring moment is considered normal, and the monitoring result is qualified; otherwise, the monitoring result at that monitoring moment is considered unqualified.

[0110] This application acquires thermal infrared grayscale images of asphalt heating tanks and determines the initial distribution uniformity based on the number and distribution of local abnormal pixels and their corresponding grayscale values. This enables effective analysis of the grayscale distribution of local abnormal pixels. Furthermore, it performs region growing on these local abnormal pixels. Based on the variance of grayscale values ​​of all pixels within the local abnormal connected region and the gradient direction between the pixel and its neighboring pixels, it determines the temperature distribution complexity of the local abnormal connected region. Using the gradient direction information of neighboring pixels, it obtains the temperature distribution complexity of the corresponding local abnormal connected region. Combining the temperature distribution complexity of the local abnormal connected region with the initial distribution uniformity, it obtains the true distribution uniformity. The initial distribution uniformity is then adjusted based on the temperature distribution complexity to obtain the true distribution uniformity. This allows for adaptive enhancement weights to be obtained based on the true distribution uniformity, enabling adaptive enhancement of the thermal infrared grayscale image and thus achieving monitoring of the heating of asphalt mixtures. This application combines the grayscale values, grayscale gradients, and grayscale distributions of local abnormal pixels with those of neighboring pixels to analyze the entire thermal infrared grayscale image. This enables subsequent image enhancement to effectively consider the grayscale distribution of the thermal infrared grayscale image itself, thereby achieving adaptive image enhancement and effectively solving the problems of low contrast, blurred visual effects, and large interference information in thermal infrared grayscale images.

[0111] It should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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 of the technical features. Such 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 this application, and should all be included within the protection scope of this application.

[0112] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A monitoring system for the mixing and heating of hot recycled asphalt used in road and bridge construction, characterized in that, The system includes: The acquisition module is used to monitor the acquisition of thermal infrared grayscale images at each monitoring moment during the hot recycling asphalt mixing and heating process, determine local abnormal pixels based on the grayscale values ​​of the pixels in the thermal infrared grayscale images, and determine the initial distribution uniformity of the thermal infrared grayscale images based on the distribution uniformity of the local abnormal pixels and the grayscale differences between the local abnormal pixels. The first processing module is used to perform region growing on each of the local abnormal pixels to obtain a local abnormal connected region, obtain the neighboring pixels of each pixel in the local abnormal connected region, and determine the temperature distribution complexity of the local abnormal connected region based on the gray value variance of all pixels in the local abnormal connected region and the gradient direction between the pixel and its neighboring pixels. The second processing module is used to determine the true uniformity of the thermal infrared grayscale image acquired at each monitoring time based on the complexity of the temperature distribution of all local abnormal connected regions in the thermal infrared grayscale image acquired at each monitoring time and the initial uniformity of the thermal infrared grayscale image. The monitoring module is used to enhance the thermal infrared grayscale image acquired at each monitoring time according to the actual distribution uniformity to obtain the target image; and to determine the heating monitoring result at each monitoring time during the hot recycled asphalt mixing heating process based on the target image obtained at each monitoring time. The determination of local abnormal pixels includes: The quartiles of the set of gray values ​​of all pixels in the thermal infrared grayscale image obtained at each monitoring time are statistically analyzed, and the preset first grayscale threshold and preset second grayscale threshold are determined based on the quartiles. Pixels with gray values ​​greater than a preset first gray value threshold in the thermal infrared grayscale image obtained at each monitoring time are designated as high-temperature pixels, and pixels with gray values ​​less than a preset second gray value threshold in the thermal infrared grayscale image obtained at each monitoring time are designated as low-temperature pixels. Both the high-temperature pixel and the low-temperature pixel are considered as local abnormal pixels; Determining the initial distribution uniformity of the thermal infrared grayscale image includes: The absolute value of the difference between the gray value of each local anomalous pixel in the thermal infrared grayscale image and the mean of the gray values ​​of all local anomalous pixels is taken as the mean of the sum of the results of all local anomalous pixels. The first influence factor is determined based on the uniformity of the position distribution of all local abnormal pixels in the thermal infrared grayscale image; Calculate the product of the first influencing factor and the abnormal temperature parameter, and use the normalized result of the negative of the product as the initial distribution uniformity of the thermal infrared grayscale image. Obtaining the neighboring pixels of each pixel within the local abnormal connected component includes: For any local anomalous pixel, each local anomalous pixel is used as an initial seed point, and the region growing algorithm is used to obtain the extraction result of the local anomalous connected region in the thermal infrared grayscale image. The direction perpendicular to the gradient direction of any pixel within the local abnormal connected region is determined as the target direction; In the target direction, determine the two other pixels that are closest to the pixel as the neighboring pixels of the pixel; Determining the complexity of the temperature distribution of the local anomalous connected domain includes: The intersection points of the line containing the gradient direction of any pixel within the local abnormal connected region and the lines containing the gradient directions of adjacent pixels are counted respectively. The distance influence factor is determined based on the number of intersection points and the distance between the intersection points. The normalized result of the inverse of the mean of the distance influence factor of all pixels in the local abnormal connected region is used as the temperature distribution complexity of the local abnormal connected region. Determining the true uniformity of the distribution of the thermal infrared grayscale image acquired at each monitoring time includes: Calculate the mean of the temperature distribution complexity of all local abnormal connected regions in the obtained thermal infrared grayscale image at each monitoring time. The sum of the normalized mean and the constant parameter is used as the denominator; The ratio of the initial distribution uniformity of the thermal infrared grayscale image obtained at each monitoring time to the denominator is taken as the true distribution uniformity of the thermal infrared grayscale image obtained at each monitoring time. The method for obtaining the target image is as follows: The ratio of the inverse proportional mapping result of the true distribution uniformity of the thermal infrared grayscale image obtained at each monitoring time to the sum of the inverse proportional mapping results of the preset number of monitoring times before the current time is used as the weight parameter of the thermal infrared grayscale image obtained at each monitoring time. The product of the first preset parameter and the weight parameter, plus the sum of the second preset parameter, is used as the adaptive enhancement weight of the thermal infrared grayscale image obtained at each monitoring time. The thermal infrared grayscale image is layered based on a bilateral filtering method to obtain a background layer image and a detail layer image. The adaptive enhancement weights are used as the weights of the detail layer image, and the difference between 1 and the adaptive enhancement weights is used as the weights of the background layer image. Weighted fusion is then performed to obtain the target image.

2. The hot recycled asphalt mixing and heating monitoring system for road and bridge construction as described in claim 1, characterized in that, The method for obtaining the first impact factor is as follows: The thermal infrared grayscale image is uniformly divided into a preset number of grayscale sub-images; The sum of the variances of the number of high-temperature pixels and the variances of the number of low-temperature pixels in all grayscale sub-images is used as the first influencing factor.

3. The hot recycled asphalt mixing and heating monitoring system for road and bridge construction as described in claim 1, characterized in that, The step of determining the distance influence factor based on the number of intersections and the distance between intersections includes: When the number of intersections is less than 2, the distance influence factor is determined to be 1; When the number of intersection points is equal to 2, the distance influence factor is obtained by normalizing the Euclidean distance between the two intersection points.

4. The hot recycled asphalt mixing and heating monitoring system for road and bridge construction as described in claim 1, characterized in that, The determination of the heating monitoring results at each monitoring moment during the hot recycled asphalt mixing and heating process includes: Based on the target image at each monitoring moment, the temperature distribution results at each monitoring moment during the hot recycled asphalt mixing and heating process are accurately visualized. By comparing and analyzing with the standard temperature distribution results, the heating monitoring results at each monitoring moment are obtained. The monitoring results include two categories: qualified and unqualified.

Images