A method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition

Abstract

The present invention discloses a method for detecting the diffusion range of grouting slurry in pavement cracks based on intelligent device recognition, which relates to the field of grouting engineering. The method includes: extracting a top-view contour area of ​​the crack; obtaining a target top-view contour area; obtaining the actual volume of the local area; obtaining at least one local target area and a target volume of the local target area; modeling a three-dimensional model of the pavement crack; obtaining the real-time filling volume of the local target area; calculating the real-time diffusion ratio of the slurry; identifying local target areas with insufficient grouting as suspicious areas, and visually displaying the filling status of the suspicious areas in the three-dimensional model of the pavement crack. By extracting the top-view contour area of ​​the crack, obtaining the target top-view contour area, obtaining the actual volume of the local area, modeling a three-dimensional model of the pavement crack, and performing a visual display, the distribution of the slurry in the three-dimensional model can be determined in combination with the three-dimensional modeling.

Description

Technical Field

[0001] The present invention relates to the field of grouting engineering, and in particular to a method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition. Background Art

[0002] Pavement crack grouting is a commonly used repair method, primarily used to repair cracks in cement concrete pavements. Grouting works by injecting a specially formulated grouting material into the cracks, filling and solidifying them, thereby restoring the integrity and stability of the pavement.

[0003] Because the cracks extend downward and their angle with the vertical direction varies, ultrasonic ranging can be used to determine the location of the cracks, but it is difficult to perform three-dimensional modeling of the cracks. Therefore, it is difficult to determine the specific diffusion range of the grouting slurry in the cracks. Summary of the Invention

[0004] In order to solve the above technical problems, a method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device identification is provided. This technical solution solves the problems raised in the above background technology.

[0005] In order to achieve the above objects, the technical solution adopted by the present invention is:

[0006] The method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition includes:

[0007] Obtain sample gas, the sample gas is hydrogen, and the temperature of the sample gas is 25 degrees;

[0008] Obtaining the water content of the slurry for grouting pavement cracks as a target content;

[0009] Use the sample gas to fill the cracks in the road surface, and after filling, use a film to seal the road surface;

[0010] Obtain an infrared image of the road surface from a bird's-eye view, and extract the top-view contour area of ​​the crack from the infrared image of the road surface;

[0011] In the top-view contour area of ​​the crack, hydrogen is detected using near-infrared spectroscopy. Based on the detection results, the top-view contour area of ​​the crack is corrected to obtain the target top-view contour area.

[0012] Evenly dividing the target top-view contour area into at least one local area, detecting using a mass spectrometer to obtain a first mass of hydrogen elements in the local area, and obtaining an actual volume of the local area based on the first mass of hydrogen elements;

[0013] Based on the actual volume, the local regions are merged to obtain at least one local target region and a target volume of the local target region;

[0014] Based on the target volume of the local target area, a three-dimensional model of the pavement cracks is obtained;

[0015] The film is removed, and high-pressure air is used to expel hydrogen from the pavement cracks. When the hydrogen content detected in the exhausted gas is less than one part per hundred thousand, slurry is used to inject grout into the pavement cracks. During the grouting process, a mass spectrometer is used to detect the second mass of hydrogen in the local target area at preset intervals. Based on the second mass of hydrogen, the real-time filling volume of the local target area is obtained. The preset time is set based on the measurement requirements.

[0016] Based on the real-time filling volume and the target volume, the real-time diffusion ratio of the slurry is calculated;

[0017] The required filling ratio of crack grouting is obtained. When the real-time diffusion ratio is less than the required filling ratio, the local target area with insufficient grouting is identified as a suspicious area, and the filling status of the suspicious area is visualized in the 3D model of the pavement crack.

[0018] Preferably, the step of extracting the top view contour area of ​​the crack from the infrared image of the road surface comprises the following steps:

[0019] Obtain the pixel value of the pixel point on the road surface under infrared recognition as the feature value;

[0020] In the infrared image of the road surface, pixel points with pixel values ​​and feature values ​​are identified as feature points;

[0021] The feature points are aggregated to form a road surface contour area, and the portion outside the road surface contour area in the infrared image of the road surface is used as the top-view contour area of ​​the crack.

[0022] Preferably, the correction of the top view contour area of ​​the crack based on the detection result comprises the following steps:

[0023] Evenly dividing the top view contour area of ​​the crack into at least one sampling area, and judging whether hydrogen exists in the sampling area based on the detection result. If so, the sampling area is regarded as a necessary area; if not, the sampling area is regarded as a non-essential area.

[0024] All unnecessary areas are deleted from the top-view contour area of ​​the crack to obtain the target top-view contour area.

[0025] Preferably, obtaining the actual volume of the local area based on the first mass of the hydrogen element comprises the following steps:

[0026] The first mass of hydrogen element divided by the relative molecular mass of hydrogen gas gives the amount of substance of hydrogen gas;

[0027] The amount of hydrogen substance is multiplied by the molar volume of the gas to obtain the actual volume of the local area.

[0028] Preferably, merging the local regions to obtain at least one local target region and a target volume of the local target region comprises the following steps:

[0029] When the actual volumes of adjacent local regions are consistent, the adjacent local regions are marked as associated attributes;

[0030] Continuous local regions with associated attributes are merged to obtain at least one local target region, and actual volumes of the continuous local regions with associated attributes are accumulated to obtain a target volume of the local target region.

[0031] Preferably, the modeling of the three-dimensional model of the pavement cracks based on the target volume of the local target area includes the following steps:

[0032] obtaining a portion of the crack opening on the pavement surface located in a local target area as a characteristic opening;

[0033] Obtain the length and width of the characteristic opening, and multiply the length and width of the characteristic opening to obtain the characteristic area;

[0034] The target volume of the local target area is divided by the feature area to obtain the feature height;

[0035] At least one identification point is selected on the characteristic opening, and the slope of the tangent line of the characteristic opening at the identification point is averaged to obtain a characteristic slope, and a straight line with a slope equal to the characteristic slope is used as a characteristic straight line;

[0036] The direction perpendicular to the direction pointed by the characteristic line is taken as the characteristic direction;

[0037] The length of the local target area along the characteristic direction is taken as the characteristic length;

[0038] Compare the characteristic length with the characteristic height to obtain the characteristic coefficient, and substitute the characteristic coefficient into the inverse tangent function to obtain the characteristic angle, which is the angle between the local target area and the vertical direction;

[0039] The feature height is divided by the cosine of the feature angle to obtain the length of the local target area;

[0040] Generate a feature opening of the same size as the feature opening in the three-dimensional model, arrange the feature openings according to the connection relationship and direction of the feature openings, stretch the feature openings downward in the vertical direction to form a feature cube, the height of the feature cube is equal to the length of the local target area corresponding to the feature opening, and match the feature cube to the local target area corresponding to the feature opening to which it is connected;

[0041] A target direction is generated in the three-dimensional model. The relative relationship between the target direction and the feature opening is consistent with the relative relationship between the feature direction and the feature opening. The feature cube is rotated, and the rotation direction is coplanar with the target direction. After the rotation, the top of the feature cube is connected to the feature opening, and the angle between the feature cube and the vertical direction is equal to the feature angle.

[0042] Preferably, obtaining the real-time filling volume of the local target area based on the second mass of hydrogen element comprises the following steps:

[0043] The second mass of the hydrogen element is divided by the relative molecular mass of hydrogen gas to obtain the amount of water molecules in the local target area;

[0044] The amount of water molecules in the local target area is multiplied by the relative molecular mass of the water molecules to obtain the mass of the water molecules in the local target area;

[0045] The mass of water molecules in the local target area is divided by the target content to obtain the mass of the slurry in the local target area. The mass of the slurry in the local target area is divided by the density of the slurry to obtain the real-time filling volume of the local target area.

[0046] Preferably, the calculation of the real-time diffusion ratio of the slurry based on the real-time filling volume and the actual volume comprises the following steps:

[0047] The sum of all real-time filling volumes divided by the sum of all actual volumes gives the real-time diffusion ratio of the slurry.

[0048] Preferably, the identifying of the local target area with insufficient grouting as a suspicious area comprises the following steps:

[0049] The real-time filling volume of the local target area is divided by the target volume of the local target area to obtain a judgment value, and the local target area with a judgment value smaller than the filling requirement ratio is regarded as a suspicious area.

[0050] Preferably, visually displaying the filling status of the suspicious area in the three-dimensional model of the road surface cracks includes the following steps:

[0051] In the three-dimensional model, a feature cube corresponding to the suspicious area is obtained;

[0052] Evenly divide the suspicious area to obtain at least one suspicious partition;

[0053] Evenly divide the feature cube corresponding to the suspicious area in the same way to obtain at least one feature partition;

[0054] Number the suspicious partitions from left to right and from top to bottom, and number the characteristic partitions from left to right and from top to bottom, and associate the suspicious partitions with the characteristic partitions with the same numbers.

[0055] Near-infrared spectroscopy is used to detect hydrogen elements in suspicious partitions, and characteristic partitions corresponding to suspicious partitions where hydrogen elements do not exist are marked as unfilled in the three-dimensional model.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] By extracting the top-view contour area of ​​the cracks, obtaining the target top-view contour area, obtaining the actual volume of the local area, modeling a three-dimensional model of the pavement cracks and performing visualization, and using infrared detection and element quality detection methods, the crack volume and inclination angle in each local area can be determined, and then a more accurate three-dimensional model of the pavement cracks can be performed. Therefore, when grouting is performed, the three-dimensional modeling can be combined to determine the distribution of the slurry in the three-dimensional modeling, and then visualization processing can be achieved. The grouting position can be targetedly adjusted according to the results of the visualization processing, thereby achieving better grouting effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 Schematic diagram of the flow of the method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device identification of the present invention;

[0059] Figure 2 This is a schematic diagram of the process of extracting the top view contour area of ​​a crack from an infrared image of a road surface according to the present invention;

[0060] Figure 3 This is a schematic diagram of a process for correcting a top-view contour area of ​​a crack based on a detection result according to the present invention;

[0061] Figure 4 This is a schematic diagram of a process for obtaining the actual volume of a local area based on the first mass of hydrogen element according to the present invention;

[0062] Figure 5 A schematic diagram of a process for merging local regions to obtain at least one local target region and a target volume of the local target region according to the present invention;

[0063] Figure 6 A schematic diagram of a process for modeling a three-dimensional model of pavement cracks based on a target volume of a local target area according to the present invention;

[0064] Figure 7 This is a schematic diagram of a process for obtaining a real-time filling volume of a local target area based on the second mass of hydrogen element according to the present invention;

[0065] Figure 8 A schematic diagram of a process for visually displaying the filling status of a suspicious area in a three-dimensional model of a road crack according to the present invention;

[0066] Figure 9 It is a schematic diagram of the geometric relationship of the present invention. DETAILED DESCRIPTION

[0067] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are merely examples, and those skilled in the art may conceive of other obvious variations.

[0068] Reference Figure 1 As shown, the method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition includes:

[0069] Obtain sample gas, the sample gas is hydrogen, and the temperature of the sample gas is 25 degrees;

[0070] Obtaining the water content of the slurry for grouting pavement cracks as a target content;

[0071] Use the sample gas to fill the cracks in the road surface, and after filling, use a film to seal the road surface;

[0072] Obtain an infrared image of the road surface from a bird's-eye view, and extract the top-view contour area of ​​the crack from the infrared image of the road surface;

[0073] In the top-view contour area of ​​the crack, hydrogen is detected using near-infrared spectroscopy. Based on the detection results, the top-view contour area of ​​the crack is corrected to obtain the target top-view contour area.

[0074] Evenly dividing the target top-view contour area into at least one local area, detecting using a mass spectrometer to obtain a first mass of hydrogen elements in the local area, and obtaining an actual volume of the local area based on the first mass of hydrogen elements;

[0075] Based on the actual volume, the local regions are merged to obtain at least one local target region and a target volume of the local target region;

[0076] Based on the target volume of the local target area, a three-dimensional model of the pavement cracks is obtained;

[0077] The film is removed, and high-pressure air is used to expel hydrogen from the pavement cracks. When the hydrogen content detected in the exhausted gas is less than one part per hundred thousand, slurry is used to inject grout into the pavement cracks. During the grouting process, a mass spectrometer is used to detect the second mass of hydrogen in the local target area at preset intervals. Based on the second mass of hydrogen, the real-time filling volume of the local target area is obtained. The preset time is set based on the measurement requirements.

[0078] Based on the real-time filling volume and the target volume, the real-time diffusion ratio of the slurry is calculated;

[0079] The required filling ratio of crack grouting is obtained. When the real-time diffusion ratio is less than the required filling ratio, the local target area with insufficient grouting is identified as a suspicious area, and the filling status of the suspicious area is visualized in the 3D model of the pavement crack.

[0080] The difficulty in determining the slurry diffusion range is that the cracks extend below the road surface. Therefore, it is difficult to accurately model and determine the volume and shape of each crack underground. Therefore, when grouting, it is difficult to judge the grouting situation based on the actual situation of the cracks, because whether the grouting is sufficient must depend on the cracks. Traditional ultrasonic testing can only determine the specific location of the cracks, but it is difficult to determine the situation inside the cracks. Therefore, it is difficult to model them. In this solution, a series of steps are set up to model the cracks, and the grouting range is determined based on the modeling results. At the same time, the exact location of insufficient grouting is determined.

[0081] Reference Figure 2 As shown, extracting the top view contour area of ​​the crack from the infrared image of the road surface includes the following steps:

[0082] Obtain the pixel value of the pixel point on the road surface under infrared recognition as the feature value;

[0083] In the infrared image of the road surface, pixel points with pixel values ​​and feature values ​​are identified as feature points;

[0084] The feature points are aggregated to form a road surface contour area, and the portion outside the road surface contour area in the infrared image of the road surface is used as the top-view contour area of ​​the crack.

[0085] When hydrogen at 25 degrees is injected into the crack, the gas has good fluidity and can therefore fully fill the crack. However, the temperature of the hydrogen will affect the road surface in contact with it. Therefore, the top-down contour area of ​​the crack includes non-crack parts, so further processing is required. It should be noted that this is a top-down projection of the crack and cannot fully reflect the situation of the crack.

[0086] Reference Figure 3 As shown, based on the detection results, the correction of the top view contour area of ​​the crack includes the following steps:

[0087] Evenly dividing the top view contour area of ​​the crack into at least one sampling area, and judging whether hydrogen exists in the sampling area based on the detection result. If so, the sampling area is regarded as a necessary area; if not, the sampling area is regarded as a non-essential area.

[0088] All unnecessary areas are deleted from the top-view contour area of ​​the crack to obtain the target top-view contour area.

[0089] In the top-view contour area of ​​the crack, hydrogen elements will only appear in the crack part, and will not appear in the road surface part. Therefore, the top-view contour area of ​​the crack can be corrected in this way to obtain the target top-view contour area.

[0090] Reference Figure 4 As shown, obtaining the actual volume of the local area based on the first mass of the hydrogen element includes the following steps:

[0091] The first mass of hydrogen element divided by the relative molecular mass of hydrogen gas gives the amount of substance of hydrogen gas;

[0092] The amount of hydrogen substance is multiplied by the molar volume of the gas to obtain the actual volume of the local area.

[0093] The local area is a top view of the local part of the crack. The crack cannot be described by this image alone. For example, if a pole is placed at a 45-degree angle, its actual length is 1 meter, but when viewed from above, its length is only 0.5 meters. Therefore, the crack cannot be modeled by the local area alone. It should be noted that the local area is a very small area. Therefore, the crack part in the local area can be approximated as a tilted pole. As long as the volume and tilt angle of the pole are determined, the crack part in the local area can be modeled, and thus the overall crack can be modeled.

[0094] When determining the actual volume of a local area, it is determined based on the mass of hydrogen in the local area. Since the ambient atmospheric pressure can be approximately considered as the standard atmospheric pressure, and the temperature of the sample gas is 25 degrees, the molar volume of hydrogen can be determined. Based on chemical knowledge and the first mass of the hydrogen element, the amount of hydrogen substance is determined. Thus, the volume of hydrogen in the local area can be determined, and hydrogen completes the filling of the local area. Therefore, the actual volume of the local area can be obtained, that is, the volume of the crack in the local area.

[0095] Reference Figure 5 As shown, merging the local regions to obtain at least one local target region and a target volume of the local target region includes the following steps:

[0096] When the actual volumes of adjacent local regions are consistent, the adjacent local regions are marked as associated attributes;

[0097] Continuous local regions with associated attributes are merged to obtain at least one local target region, and actual volumes of the continuous local regions with associated attributes are accumulated to obtain a target volume of the local target region.

[0098] Since the trends of adjacent local areas are consistent, that is, the inclination angles and openings of the cracks in the local areas are consistent, therefore, as long as the depths of crack extension are determined to be consistent, the adjacent local areas can be merged, which can reduce the workload of differential treatment. Whether the depths of crack extension in the local areas are consistent is equivalent to whether the volumes of the cracks in the local areas are consistent.

[0099] Reference Figure 6 As shown, based on the target volume of the local target area, the three-dimensional model of the pavement cracks is modeled and obtained, including the following steps:

[0100] obtaining a portion of the crack opening on the pavement surface located in a local target area as a characteristic opening;

[0101] Obtain the length and width of the characteristic opening, and multiply the length and width of the characteristic opening to obtain the characteristic area;

[0102] The target volume of the local target area is divided by the feature area to obtain the feature height;

[0103] At least one identification point is selected on the characteristic opening, and the slope of the tangent line of the characteristic opening at the identification point is averaged to obtain a characteristic slope, and a straight line with a slope equal to the characteristic slope is used as a characteristic straight line;

[0104] The direction perpendicular to the direction pointed by the characteristic line is taken as the characteristic direction;

[0105] The length of the local target area along the characteristic direction is taken as the characteristic length;

[0106] Compare the characteristic length with the characteristic height to obtain the characteristic coefficient, and substitute the characteristic coefficient into the inverse tangent function to obtain the characteristic angle, which is the angle between the local target area and the vertical direction;

[0107] The feature height is divided by the cosine of the feature angle to obtain the length of the local target area;

[0108] Generate a feature opening of the same size as the feature opening in the three-dimensional model, arrange the feature openings according to the connection relationship and direction of the feature openings, stretch the feature openings downward in the vertical direction to form a feature cube, the height of the feature cube is equal to the length of the local target area corresponding to the feature opening, and match the feature cube to the local target area corresponding to the feature opening to which it is connected;

[0109] A target direction is generated in the three-dimensional model. The relative relationship between the target direction and the feature opening is consistent with the relative relationship between the feature direction and the feature opening. The feature cube is rotated, and the rotation direction is coplanar with the target direction. After the rotation, the top of the feature cube is connected to the feature opening, and the angle between the feature cube and the vertical direction is equal to the feature angle.

[0110] The cracks extending underground in the local target area can be obtained by stretching the crack openings on the ground in the local target area. Since the target volume of the local target area is known, the stretching length can be determined by geometric relationships. Therefore, the shape of the cracks extending underground in the local target area can be determined. However, its inclination angle also needs to be determined, which is also calculated based on geometric relationships.

[0111] Reference Figure 9 As shown in the figure, since the local target area is very small, the corresponding underground cracks can be approximately regarded as poles, so they are replaced by line segments. The geometric relationship is expressed as follows:

[0112] AC is the crack underground in the local target area, A is the ground, and BC is the overhead projection of the crack. Since the volume of the crack in the local target area is measured, the length of AB can be obtained according to the calculation formula that the bottom area multiplied by the height equals the volume. Since the local target area is known, the length of BC can be obtained, and then the inclination angle can be determined. At the same time, AC can be obtained, and then a three-dimensional model can be constructed based on these data.

[0113] Reference Figure 7 As shown, obtaining the real-time filling volume of the local target area based on the second mass of the hydrogen element includes the following steps:

[0114] The second mass of the hydrogen element is divided by the relative molecular mass of hydrogen gas to obtain the amount of water molecules in the local target area;

[0115] The amount of water molecules in the local target area is multiplied by the relative molecular mass of the water molecules to obtain the mass of the water molecules in the local target area;

[0116] The mass of water molecules in the local target area is divided by the target content to obtain the mass of the slurry in the local target area. The mass of the slurry in the local target area is divided by the density of the slurry to obtain the real-time filling volume of the local target area.

[0117] In order to avoid the interference of the injected hydrogen, high-pressure air is used to discharge it. Since the hydrogen content in the air is less than one part per million, when the hydrogen content detected in the discharged gas is less than one part per hundred thousand, slurry is used to grout the cracks in the pavement. When the slurry is filled, since the slurry contains water, the number of water molecules is determined by the mass of the hydrogen element, and then the mass of the water is determined. According to the water content of the slurry, the mass of the slurry can be obtained, and thus the volume of the slurry can be obtained. Since the slurry needs to completely fill the local target area, the volume of the slurry is compared with the target volume of the local target area to obtain the filling situation.

[0118] Based on the real-time filling volume and the actual volume, the real-time diffusion ratio of the slurry is calculated, including the following steps:

[0119] The sum of all real-time filling volumes divided by the sum of all actual volumes gives the real-time diffusion ratio of the slurry.

[0120] Identifying local target areas of insufficient grouting as suspect areas involves the following steps:

[0121] The real-time filling volume of the local target area is divided by the target volume of the local target area to obtain a judgment value, and the local target area with a judgment value smaller than the filling requirement ratio is regarded as a suspicious area.

[0122] Reference Figure 8 As shown, visualizing the filling status of the suspicious area in the three-dimensional model of the pavement cracks includes the following steps:

[0123] In the three-dimensional model, a feature cube corresponding to the suspicious area is obtained;

[0124] Evenly divide the suspicious area to obtain at least one suspicious partition;

[0125] Evenly divide the feature cube corresponding to the suspicious area in the same way to obtain at least one feature partition;

[0126] Number the suspicious partitions from left to right and from top to bottom, and number the characteristic partitions from left to right and from top to bottom, and associate the suspicious partitions with the characteristic partitions with the same numbers.

[0127] Near-infrared spectroscopy is used to detect hydrogen elements in suspicious partitions, and characteristic partitions corresponding to suspicious partitions where hydrogen elements do not exist are marked as unfilled in the three-dimensional model.

[0128] The suspicious area is equivalent to the top-view image of the feature cube. Therefore, there is a corresponding relationship between the two. If the two are divided in the same way, that is, both are divided evenly, the number of horizontal division blocks corresponds to the same number of vertical division blocks, then the division partitions have a corresponding relationship. Therefore, the detection results can be matched, so that the filling situation can be displayed in three-dimensional visualization. The grouting position can be adjusted according to the display results, and the areas with insufficient filling can be pressurized.

[0129] Furthermore, the present solution also proposes a storage medium on which a computer-readable program is stored. When the computer-readable program is called, the above-mentioned pavement crack grouting slurry diffusion range detection method based on intelligent device recognition is executed.

[0130] It is understandable that the storage medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state disk (SSD).

[0131] To sum up, the advantages of the present invention are: by extracting the top-view contour area of ​​the crack, obtaining the target top-view contour area, obtaining the actual volume of the local area, modeling a three-dimensional model of the pavement crack and performing visual display, using infrared detection and element quality detection methods, the crack volume and inclination angle in each local area can be determined, and then the pavement cracks can be more accurately three-dimensionally modeled. Therefore, when grouting is performed, the three-dimensional modeling can be combined to determine the distribution of the slurry in the three-dimensional modeling, and then visualization processing can be realized. The grouting position can be targetedly adjusted according to the results of the visualization processing, so as to achieve better grouting effect.

[0132] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions merely illustrate the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition, characterized in that: include: Obtain sample gas, the sample gas is hydrogen, and the temperature of the sample gas is 25 degrees; Obtaining the water content of the slurry for grouting pavement cracks as a target content; Use the sample gas to fill the cracks in the road surface, and after filling, use a film to seal the road surface; Obtain an infrared image of the road surface from a bird's-eye view, and extract the top-view contour area of ​​the crack from the infrared image of the road surface; In the top-view contour area of ​​the crack, hydrogen is detected using near-infrared spectroscopy. Based on the detection results, the top-view contour area of ​​the crack is corrected to obtain the target top-view contour area. Evenly dividing the target top-view contour area into at least one local area, detecting using a mass spectrometer to obtain a first mass of hydrogen elements in the local area, and obtaining an actual volume of the local area based on the first mass of hydrogen elements; Based on the actual volume, the local regions are merged to obtain at least one local target region and a target volume of the local target region; Based on the target volume of the local target area, a three-dimensional model of the pavement cracks is obtained; The film is removed, and high-pressure air is used to expel hydrogen from the pavement cracks. When the hydrogen content detected in the exhausted gas is less than one part per hundred thousand, slurry is used to inject grout into the pavement cracks. During the grouting process, a mass spectrometer is used to detect the second mass of hydrogen in the local target area at preset intervals. Based on the second mass of hydrogen, the real-time filling volume of the local target area is obtained. The preset time is set based on the measurement requirements. Based on the real-time filling volume and the target volume, the real-time diffusion ratio of the slurry is calculated; The required filling ratio of crack grouting is obtained. When the real-time diffusion ratio is less than the required filling ratio, the local target area with insufficient grouting is identified as a suspicious area, and the filling status of the suspicious area is visualized in the 3D model of the pavement crack.

2. The method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition according to claim 1 is characterized in that: The method of extracting the top view contour area of ​​the crack through the infrared image of the road surface comprises the following steps: Obtain the pixel value of the pixel point on the road surface under infrared recognition as the feature value; In the infrared image of the road surface, pixel points with pixel values ​​and feature values ​​are identified as feature points; The feature points are aggregated to form a road surface contour area, and the portion outside the road surface contour area in the infrared image of the road surface is used as the top-view contour area of ​​the crack.

3. The method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition according to claim 2 is characterized in that: The correction of the top view contour area of ​​the crack based on the detection results includes the following steps: Evenly dividing the top view contour area of ​​the crack into at least one sampling area, and judging whether hydrogen exists in the sampling area based on the detection result. If so, the sampling area is regarded as a necessary area; if not, the sampling area is regarded as a non-essential area. All unnecessary areas are deleted from the top-view contour area of ​​the crack to obtain the target top-view contour area.

4. The method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition according to claim 3 is characterized in that: The obtaining of the actual volume of the local area based on the first mass of the hydrogen element comprises the following steps: The first mass of hydrogen element divided by the relative molecular mass of hydrogen gas gives the amount of substance of hydrogen gas; The amount of hydrogen substance is multiplied by the molar volume of the gas to obtain the actual volume of the local area.

5. The method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition according to claim 4 is characterized in that: Merging the local regions to obtain at least one local target region and a target volume of the local target region comprises the following steps: When the actual volumes of adjacent local regions are consistent, the adjacent local regions are marked as associated attributes; Continuous local regions with associated attributes are merged to obtain at least one local target region, and actual volumes of the continuous local regions with associated attributes are accumulated to obtain a target volume of the local target region.

6. The method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition according to claim 5 is characterized in that: The method of modeling a three-dimensional model of pavement cracks based on the target volume of the local target area includes the following steps: obtaining a portion of the crack opening on the pavement surface located in a local target area as a characteristic opening; Obtain the length and width of the characteristic opening, and multiply the length and width of the characteristic opening to obtain the characteristic area; The target volume of the local target area is divided by the feature area to obtain the feature height; At least one identification point is selected on the characteristic opening, and the slope of the tangent line of the characteristic opening at the identification point is averaged to obtain a characteristic slope, and a straight line with a slope equal to the characteristic slope is used as a characteristic straight line; The direction perpendicular to the direction pointed by the characteristic line is taken as the characteristic direction; The length of the local target area along the characteristic direction is taken as the characteristic length; Compare the characteristic length with the characteristic height to obtain the characteristic coefficient, and substitute the characteristic coefficient into the inverse tangent function to obtain the characteristic angle, which is the angle between the local target area and the vertical direction; The feature height is divided by the cosine of the feature angle to obtain the length of the local target area; Generate a feature opening of the same size as the feature opening in the three-dimensional model, arrange the feature openings according to the connection relationship and direction of the feature openings, stretch the feature openings downward in the vertical direction to form a feature cube, the height of the feature cube is equal to the length of the local target area corresponding to the feature opening, and match the feature cube to the local target area corresponding to the feature opening to which it is connected; A target direction is generated in the three-dimensional model. The relative relationship between the target direction and the feature opening is consistent with the relative relationship between the feature direction and the feature opening. The feature cube is rotated, and the rotation direction is coplanar with the target direction. After the rotation, the top of the feature cube is connected to the feature opening, and the angle between the feature cube and the vertical direction is equal to the feature angle.

7. The method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition according to claim 6 is characterized in that: The method of obtaining the real-time filling volume of the local target area based on the second mass of the hydrogen element comprises the following steps: The second mass of the hydrogen element is divided by the relative molecular mass of hydrogen gas to obtain the amount of water molecules in the local target area; The amount of water molecules in the local target area is multiplied by the relative molecular mass of the water molecules to obtain the mass of the water molecules in the local target area; The mass of water molecules in the local target area is divided by the target content to obtain the mass of the slurry in the local target area. The mass of the slurry in the local target area is divided by the density of the slurry to obtain the real-time filling volume of the local target area.

8. The method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition according to claim 7 is characterized in that: The method of calculating the real-time diffusion ratio of the slurry based on the real-time filling volume and the actual volume comprises the following steps: The sum of all real-time filling volumes divided by the sum of all actual volumes gives the real-time diffusion ratio of the slurry.

9. The method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition according to claim 8, characterized in that: The identification of the local target area with insufficient grouting as a suspicious area includes the following steps: The real-time filling volume of the local target area is divided by the target volume of the local target area to obtain a judgment value, and the local target area with a judgment value smaller than the filling requirement ratio is regarded as a suspicious area.

10. The method for detecting the diffusion range of grouting slurry for pavement cracks based on intelligent device recognition according to claim 9 is characterized in that: The visual display of the filling condition of the suspicious area in the three-dimensional model of the road cracks includes the following steps: In the three-dimensional model, a feature cube corresponding to the suspicious area is obtained; Evenly divide the suspicious area to obtain at least one suspicious partition; Evenly divide the feature cube corresponding to the suspicious area in the same way to obtain at least one feature partition; Number the suspicious partitions from left to right and from top to bottom, and number the characteristic partitions from left to right and from top to bottom, and associate the suspicious partitions with the characteristic partitions with the same numbers. Near-infrared spectroscopy is used to detect hydrogen elements in suspicious partitions, and characteristic partitions corresponding to suspicious partitions where hydrogen elements do not exist are marked as unfilled in the three-dimensional model.

Images