Bridge inspection system and method thereof

Abstract

The invention discloses a bridge inspection system and method, the bridge inspection system comprises a control module and an execution module, and the control module is wirelessly connected with the execution module; the control module comprises an information acquisition unit, a crack detection unit, a crack trend analysis unit, a crack width calculation unit, a crack depth calculation unit and an execution control unit. And the width and depth of the crack are automatically calculated according to the identification result, and the trend of the crack is analyzed, so that automatic detection of the bridge pavement crack is realized, the data precision is high, the detection efficiency is high, professional personnel can judge the safety of a bridge according to the data, and a targeted decision is made according to the judgment result.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge detection, and in particular, to a bridge inspection system and method thereof. Background Art

[0002] As a key infrastructure for transportation, the long-term safe use of the road surface of a bridge is an important guarantee for ensuring smooth traffic and public safety. In order to maintain the good condition of the bridge and prevent potential safety risks, regular inspection and maintenance of the bridge road surface are indispensable links.

[0003] However, when detecting bridge cracks currently, traditional detection methods mainly rely on manual observation and the use of simple measuring tools, such as crack width gauges, crack length measuring rulers, etc. Although these tools can help inspectors obtain basic data such as the length and width of cracks to a certain extent, there are problems with insufficient data accuracy.

[0004] In addition, traditional crack detection methods also have many inconveniences in the operation process. Inspectors need to carry a variety of tools for on-site measurement, which is not only cumbersome to operate but also inefficient. More importantly, after obtaining the crack data, additional equipment is required for crack repair work, which further increases the complexity and time cost of the operation. Summary of the Invention

[0005] Therefore, to solve the above deficiencies, the present invention provides a bridge inspection system and method thereof here, so as to realize the automatic detection of bridge cracks, improve the accuracy of detection data, and at the same time improve the detection efficiency and is convenient to use.

[0006] On the one hand, the present invention provides a bridge inspection system, which is characterized in that it includes a control module and an execution module, and the control module is wirelessly connected to the execution module;

[0007] The control module includes:

[0008] An information acquisition unit for acquiring information data, where the information data includes image information data and ultrasonic information data;

[0009] A crack detection unit for performing crack detection according to the image information data to obtain a crack detection result;

[0010] A crack trend analysis unit for extracting a target feature boundary according to the crack detection result, extracting a target feature skeleton pixel set of the image according to the target feature boundary, performing curve fitting on the target feature skeleton pixel set to obtain a target feature skeleton curve, and performing crack trend analysis on the target feature skeleton curve to obtain a crack trend analysis result;

[0011] A crack width calculation unit, which is used to calculate the normal line of the pixel points on the skeleton curve in the skeleton curve, calculate the crack width according to the intersection points of the normal line and the target feature boundary, and obtain the crack width calculation result;

[0012] A crack depth calculation unit, which is used to calculate the crack depth according to the ultrasonic information data and obtain the crack depth calculation result;

[0013] An execution control unit, which obtains a user instruction or a crack detection result, generates an execution control signal according to the user instruction or the crack detection result, and sends the execution control signal to an execution module, so that the execution module executes an action corresponding to the execution control signal according to the execution control signal.

[0014] Furthermore; the execution module includes:

[0015] A walking unit, through which the whole execution module moves on a plane;

[0016] A visual detection unit, which is mounted on the walking unit and is used to acquire image information data;

[0017] An ultrasonic detection unit, which is mounted on the walking unit and is used to acquire ultrasonic information data;

[0018] Under the control of the execution control unit, the walking unit walks on the bridge road surface, and at the same time, the visual detection unit continuously collects the bridge road surface image information data and uploads it to the control module. When a road surface crack is detected, the execution control unit controls the walking unit to move to the crack point position. At this time, the ultrasonic detection unit starts to work and collects ultrasonic information data. Under the information data processing, analysis and calculation of the control module, the width, depth and trend of the crack are obtained, realizing the automatic detection of the bridge surface crack and the automatic measurement of the crack parameters.

[0019] Furthermore; the ultrasonic detection unit includes:

[0020] A first power device;

[0021] An ultrasonic detection mechanism, which is driven by the first power device to linearly move, and the movement direction of the ultrasonic detection mechanism is perpendicular to the movement direction of the walking unit;

[0022] The ultrasonic detection unit further includes a first installation part, the first power device is fixedly installed on the first installation part, the whole ultrasonic detection unit is fixedly installed at the bottom of the walking unit through the first installation part, the first power device can be a motor, and a pulley transmission mechanism can also be installed on the first installation part. The pulley transmission mechanism is connected to the motor, and the motor drives the pulley transmission mechanism to move to realize the linear movement of the ultrasonic detection mechanism;

[0023] The ultrasonic detection unit further includes a second power device, which drives the ultrasonic detection mechanism to rotate horizontally.

[0024] The first power device drives the ultrasonic detection mechanism to rotate, facilitating the adjustment of the detection direction of the ultrasonic detection mechanism according to the crack trend.

[0025] The ultrasonic detection unit further includes a second mounting portion, and the ultrasonic detection unit is mounted on the second power device through the second mounting portion; the second power device can be slidably connected to the first mounting portion through a guiding structure, and the second power device remains connected to the belt of the pulley transmission mechanism.

[0026] Furthermore, the ultrasonic detection mechanism includes:

[0027] A motion module;

[0028] A first ultrasonic detector, which is fixed outside one end of the motion module;

[0029] A second ultrasonic detector, which is mounted on the motion module, and the second ultrasonic detector is driven by the motion module to linearly move, so that the second ultrasonic detector approaches or moves away from the first ultrasonic detector;

[0030] The second ultrasonic detector is driven by the motion module to linearly move, so that the second ultrasonic detector approaches or moves away from the first ultrasonic detector, facilitating the adjustment of the distance between the ultrasonic detectors according to the crack width.

[0031] When the crack detection unit detects a crack on the bridge road surface, the control unit will determine the detection starting point according to the skeleton curve, and the determination method of the detection starting point is as follows:

[0032] Calculate the distance between the skeleton pixel points to obtain the alternative starting points with the maximum distance;

[0033] Calculate the distance between the alternative starting point and the walking unit, and obtain the alternative starting point closest to the walking unit as the detection starting point;

[0034] Based on the detection starting point, obtain the detection tangent and the detection normal of the detection starting point on the skeleton curve, re - establish a coordinate system based on the detection tangent and the detection normal to form a detection coordinate system, generate the detection coordinates of each pixel point on the skeleton curve based on the detection coordinate system, calculate the maximum horizontal distance of the skeleton curve, and calculate the starting point coordinates of the first ultrasonic detector and the starting point coordinates of the second ultrasonic detector according to the two pixel points with the maximum horizontal distance;

[0035] The calculation method of the starting point coordinates of the first ultrasonic detector is:

[0036] (a, b) = (x - Δx, y0 -Δy);

[0037] where (a, b) are the starting coordinates of the first ultrasonic detector; x is the abscissa of the left limit point; Δx is the horizontal coordinate compensation; y 0 is the ordinate of the detection starting point; Δy is the vertical coordinate compensation;

[0038] The horizontal coordinate compensation is determined as follows:

[0039]

[0040] where d 0 is the crack width at the left limit point, in millimeters; l 0 is the distance between the first ultrasonic detector and the crack boundary, in millimeters;

[0041] The calculation method of the starting coordinates of the second ultrasonic detector is as follows:

[0042] (c, d) = (a + A + Δx, y 0 -Δy);

[0043] where (c, d) are the starting coordinates of the second ultrasonic detector; A is the maximum horizontal distance, in millimeters;

[0044] The starting coordinates of the first ultrasonic detector and the second ultrasonic detector are calculated in the above manner, and the movement of the walking unit is controlled by the execution control unit, so that the first ultrasonic detector moves to its own starting coordinates, and then the movement module is controlled to drive the second ultrasonic detector to move to the starting coordinates of the second ultrasonic detector.

[0045] Furthermore; the inspection system further includes a repair unit, which is carried on the walking unit, and the repair material is poured into the crack through this repair unit.

[0046] Furthermore; the repair unit includes:

[0047] A third power device;

[0048] A perfusion head, which is driven by the third power device to move horizontally linearly, and the movement direction of the perfusion head is perpendicular to the walking direction of the walking unit. By driving the perfusion head to move linearly through the third power device, the perfusion position of the perfusion head can be adjusted according to the trend of the crack;

[0049] The repair unit further includes a third mounting portion. The entire repair unit is mounted on the tail of the traveling unit through the third mounting portion. The third power device is fixedly installed at the end of the third mounting portion. The third power device can be a motor. On the third mounting portion, a transmission mechanism such as a lead screw, a pulley drive mechanism, or a rack and pinion mechanism connected to the motor is installed, which can realize the linear movement of the perfusion head.

[0050] Furthermore; the repair unit further includes:

[0051] A fourth power device;

[0052] A second sliding portion, which drives the second sliding portion to perform a vertical linear movement through the fourth power device. The perfusion head is slidably installed on the second sliding portion;

[0053] A fifth power device, which drives the perfusion head to perform a relative vertical linear movement through the fifth power device;

[0054] By adjusting the height of the perfusion head in a two-stage movement manner, the structure of the repair unit can be made more compact;

[0055] The fourth and fifth power devices include but are not limited to cylinders, electric cylinders, hydraulic cylinders, or other devices that can realize the linear movement of the second sliding portion and the perfusion head. The repair unit further includes a first sliding portion 2540, which is fixed on the third mounting portion. The second sliding portion can be slidably connected to the first sliding portion through a guiding structure. The fourth power device can be fixedly installed on the second sliding portion, and its output part is connected to the first sliding portion. The fifth power device can be fixedly installed on the second sliding portion, and its output part is connected to the perfusion head. The perfusion head can be connected to the second sliding portion through a guiding structure.

[0056] Furthermore; the inspection system further includes a slurry storage unit. The entire slurry storage unit is carried on the traveling unit and remains connected to the repair unit; the repair unit is provided with repair materials through the slurry storage unit to achieve self-sufficiency of repair materials.

[0057] A bridge inspection method, which uses the above-mentioned inspection system. The inspection method includes:

[0058] Obtain a user instruction, generate a first execution control signal according to the user instruction, and send the first execution control signal to the execution module so that the execution module executes an action corresponding to the first execution control signal according to the first execution control signal;

[0059] Obtain information data, where the information data includes image information data and ultrasonic information data;

[0060] Perform crack detection based on the image information data to obtain a crack detection result;

[0061] Generate a second execution control signal based on the crack detection result, and send the second execution control signal to the execution module, so that the execution module performs an action corresponding to the second execution control signal according to the second execution control signal;

[0062] According to the crack detection result, extract the target feature boundary, extract the target feature skeleton pixel set of the image according to the target feature boundary, perform curve fitting according to the target feature skeleton pixel set, obtain the target feature skeleton curve, and perform crack trend analysis on the target feature skeleton curve to obtain the crack trend analysis result;

[0063] Specifically, after the crack detection unit detects cracks on the road surface, the crack trend analysis unit will extract the boundary of the crack features in the road surface image processed by the image processing algorithm, so as to form a crack boundary line image, extract the target feature edge pixel set based on the boundary line image, and at the same time, perform skeletonization processing on the crack features in the road surface image processed by the image processing algorithm, so as to extract the skeleton pixel set of the crack features, perform curve fitting on the basis of the skeleton pixel set, so as to obtain the skeleton curve of the target feature, and finally perform trend analysis through the skeleton curve. The trend analysis includes length calculation, curve shape matching, etc.;

[0064] Calculate the normal line of the pixel points on the skeleton curve in the skeleton curve, and calculate the crack width according to the intersection points of the normal line and the target feature boundary to obtain the crack width calculation result;

[0065] Specifically, the target feature boundary includes a first boundary and a second boundary. The area between the first boundary and the second boundary is the crack area. Calculate the normal line of the target point on the skeleton curve, and calculate the width of the crack according to the intersection points of the normal line and the first boundary and the second boundary. The target point is a point with a certain interval on the skeleton curve. The calculation method of the crack width can refer to the crack width calculation method described in Embodiment 1;

[0066] Perform crack depth calculation according to the ultrasonic information data to obtain the crack depth calculation result;

[0067] Before detecting the crack, when the crack detection unit detects a crack on the bridge road surface, the execution control unit will determine the detection starting point according to the skeleton curve. The determination method of the detection starting point is as follows:

[0068] Calculate the distance between the skeleton pixel points to obtain the alternative starting points with the farthest distance;

[0069] Calculate the distance between the alternative starting points and the walking unit, and obtain the alternative starting point closest to the walking unit as the detection starting point;

[0070] Based on the detection starting point, obtain the detection tangent and the detection normal of the detection starting point on the skeleton curve, re - establish a coordinate system based on the detection tangent and the detection normal to form a detection coordinate system, generate the detection coordinates of each pixel point on the skeleton curve based on the detection coordinate system, calculate the maximum lateral distance of the skeleton curve, and calculate the starting point coordinates of the first ultrasonic detector and the starting point coordinates of the second ultrasonic detector according to the two pixel points (hereinafter referred to as limit points) of the maximum lateral distance.

[0071] Control the movement of the walking unit through the execution control unit so that the first ultrasonic detector moves to its own starting point coordinates, and then control the movement module to drive the second ultrasonic detector to move to the starting point coordinates of the second ultrasonic detector.

[0072] The calculation method of the crack length is realized through the following method:

[0073]

[0074] Where l is the crack length, with the unit of mm; n is the number of pixel points in the skeleton pixel set; x n and y n are the horizontal and vertical coordinates of the nth pixel point in the skeleton pixel set; x n+1 and y n+1 are the horizontal and vertical coordinates of the (n + 1)th pixel point in the skeleton pixel set.

[0075] The calculation of the crack width can be realized through the following method:

[0076]

[0077] Where d is the crack width, with the unit of millimeter; i is the number of target points; x Ri and y Ri are the horizontal and vertical coordinates of the intersection point of the normal line at the ith pixel point in the skeleton curve and the first boundary; x Li and y Li are the horizontal and vertical coordinates of the intersection point of the normal line at the ith pixel point in the skeleton curve and the second boundary.

[0078] The calculation of the crack depth can be realized through the following method:

[0079]

[0080] Where h is the crack depth, with the unit of millimeter; L is the ultrasonic propagation distance on the road surface without cracks, with the unit of millimeter; t is the time taken for the ultrasonic wave to bypass the crack, with the unit of second; T is the time used for the ultrasonic wave to propagate the distance L on the road surface without cracks, with the unit of second.

[0081] The present invention has the following advantages: By collecting images of the bridge surface, automatically identifying the crack features in the images, automatically calculating the width and depth of the cracks according to the recognition results, and analyzing the trend of the cracks, the automatic detection of bridge pavement cracks is realized, with high data accuracy and fast detection efficiency, which is convenient for professionals to judge the safety of the bridge based on the above data, and then make targeted decisions according to the judgment results. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 is a schematic diagram of the logical structure of the inspection system;

[0083] Figure 2 is Figure 1 a schematic diagram of the structure of the execution module in the inspection system shown;

[0084] Figure 3 is Figure 2 a side view of the execution module shown;

[0085] Figure 4 is Figure 3 a schematic diagram of the structure of the repair unit in the execution module shown;

[0086] Figure 5 is Figure 4 a side view of the repair unit shown;

[0087] Figure 6 is Figure 3 a schematic diagram of the structure of the ultrasonic detection unit in the execution module shown;

[0088] Figure 7 is Figure 6 a schematic diagram of the structure of the ultrasonic detection mechanism in the ultrasonic detection unit shown;

[0089] In the figure: 1000, control module; 1100, information acquisition unit; 1200, crack detection unit; 1300, execution control unit; 140, crack trend analysis unit; 1500, crack depth calculation unit; 1600, crack width calculation unit; 2000, execution module; 2100, walking unit; 2200, visual detection unit; 2300, slurry storage unit; 2400, ultrasonic detection unit; 2410, first power device; 2420, first installation part; 2430, pulley transmission mechanism; 2440, second power device; 2450, ultrasonic detection mechanism; 2451, first ultrasonic detector; 2452, second ultrasonic detector; 2453, second installation part; 2454, motion module; 2500, repair unit; 2510, third power device; 2520, third installation part; 2530, perfusion head; 2540, first sliding part; 2550, fourth power device; 2560, second sliding part; 2570, fifth power device. Detailed implementation manners

[0090] The embodiments of the present application will be described in detail below. Examples of the embodiments are shown in the drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and are intended to explain the present application, and should not be construed as a limitation to the present application.

[0091] As described in the background art, when detecting bridge cracks currently, traditional detection methods mainly rely on manual observation and the use of simple measurement tools, such as crack width gauges, crack length measuring rulers, etc. Although these tools can help detection personnel obtain basic data such as the length and width of cracks to a certain extent, there are problems with insufficient data accuracy.

[0092] Embodiment 1:

[0093] Therefore, in order to solve the above technical problems existing in the prior art, this embodiment provides a bridge inspection system herein, as Figure 1 shown, the inspection system includes a control module 1000 and an execution module 2000, and the control module is wirelessly connected to the execution module;

[0094] As Figure 1 shown, the control module includes:

[0095] An information acquisition unit 1100, configured to acquire information data, where the information data includes image information data and ultrasonic information data;

[0096] A crack detection unit 1200, configured to perform crack detection based on the image information data to obtain a crack detection result;

[0097] Specifically, before detecting cracks in the image information data, the image information data can be preprocessed through image processing algorithms (such as digital image processing methods), including binarization, grayscale conversion, denoising, segmentation, extraction, etc., so as to improve the processing and analysis efficiency of the subsequent image information data and improve the target recognition accuracy. In addition, the UNet network model can be used as the basic framework, and the MobileNetV3 network can be used to replace the encoder part of the UNet for feature extraction. Subsequently, the algorithm is used for training to obtain the optimal training parameters and the optimal model parameters, and the establishment of the crack recognition model is completed. The crack features in the image are recognized through the crack recognition model, and finally the detection of road surface cracks is realized.

[0098] The crack trend analysis unit 1400 is used to extract the target feature boundary according to the crack detection result, extract the target feature skeleton pixel set of the image according to the target feature boundary, perform curve fitting according to the target feature skeleton pixel set to obtain the target feature skeleton curve, and perform crack trend analysis on the target feature skeleton curve to obtain the crack trend analysis result;

[0099] Specifically, after the crack detection unit detects that there are cracks on the road surface, the crack trend analysis unit will extract the boundary of the crack features in the road surface image processed by the image processing algorithm, so as to form a crack boundary line image, extract the target feature edge pixel set based on the boundary line image. At the same time, it will also skeletonize the crack features in the road surface image processed by the image processing algorithm to extract the skeleton pixel set of the crack features, perform curve fitting on the basis of the skeleton pixel set to obtain the skeleton curve of the target feature, and finally perform trend analysis through the skeleton curve. The trend analysis includes length calculation, curve shape matching, etc.

[0100] In this embodiment, the length calculation method of the crack can be realized through the following method:

[0101]

[0102] where l is the crack length, in mm; n is the number of pixel points in the skeleton pixel set; x n and y n are the horizontal and vertical coordinates of the nth pixel point in the skeleton pixel set; x n+1 and y n+1 are the horizontal and vertical coordinates of the (n + 1)th pixel point in the skeleton pixel set.

[0103] In this embodiment, the curve judgment can be obtained by performing shape matching between the skeleton curve and the sample set in the database to obtain the shape matching result.

[0104] The crack width calculation unit 1600 is configured to calculate the normal line of the pixel points on the skeleton curve, calculate the crack width based on the intersection points of the normal line and the target feature boundary, and obtain the crack width calculation result.

[0105] Specifically, the target feature boundary includes a first boundary and a second boundary. The area between the first boundary and the second boundary is the crack area. Calculate the normal line of the target points on the skeleton curve, and calculate the width of the crack based on the intersection points of the normal line and the first boundary and the second boundary. The target points are the points with a certain interval on the skeleton curve (the size of this interval is determined comprehensively according to factors such as image resolution and crack type).

[0106] In this embodiment, the calculation of the crack width can be implemented in the following manner:

[0107]

[0108] where d is the crack width, in millimeters; i is the number of target points; x Ri and y Ri are the abscissa and ordinate of the intersection point of the normal line at the i-th pixel point on the skeleton curve and the first boundary; x Li and y Li are the abscissa and ordinate of the intersection point of the normal line at the i-th pixel point on the skeleton curve and the second boundary.

[0109] The crack depth calculation unit 1500 is configured to calculate the crack depth based on the ultrasonic information data and obtain the crack depth calculation result.

[0110] In this embodiment, the calculation of the crack depth can be implemented in the following manner:

[0111]

[0112] where h is the crack depth, in millimeters; L is the ultrasonic propagation distance on the crack-free road surface, in millimeters; t is the time taken for the ultrasonic wave to bypass the crack, in seconds; T is the time used for the ultrasonic wave to propagate a distance of L on the crack-free road surface, in seconds.

[0113] The execution control unit 1300 obtains a user instruction or a crack detection result, generates an execution control signal according to the user instruction or the crack detection result, and sends the execution control signal to the execution module, so that the execution module performs an action corresponding to the execution control signal according to the execution control signal.

[0114] In this embodiment, the images of the bridge surface are collected to automatically identify the crack features in the images, and the width and depth of the cracks are automatically calculated according to the recognition results, and the trend of the cracks is analyzed, so as to realize the automatic detection of the cracks on the bridge pavement, with high data accuracy and fast detection efficiency, which is convenient for professionals to judge the safety of the bridge based on the above data, and then make targeted decisions according to the judgment results.

[0115] This embodiment provides one of the structures of the execution module as follows Figure 2 , 3 shown, the execution module includes:

[0116] A walking unit 2100, through which the whole execution module moves on a plane;

[0117] A visual detection unit 2200, which is mounted on the walking unit and used to obtain image information data;

[0118] An ultrasonic detection unit 2400, which is mounted on the walking unit and used to obtain ultrasonic information data.

[0119] The walking unit can walk on the bridge pavement under the control of the execution control unit. At the same time, the visual detection unit continuously collects the image information data of the bridge pavement and uploads it to the control module. When a crack on the pavement is detected, the execution control unit controls the walking unit to move to the crack point position. At this time, the ultrasonic detection unit starts to work and collects ultrasonic information data. Under the information data processing, analysis and calculation of the control module, the width, depth and trend of the crack are obtained, realizing the automatic detection of the cracks on the bridge surface and the automatic measurement of the crack parameters.

[0120] In this embodiment, the walking unit can be a mobile mechanism with a power device, such as a tracked mobile trolley with a motor or a tire mobile trolley with a motor. By sending a control signal from the execution control unit to the power device of the walking unit itself, the power device drives the whole inspection system to move on the bridge, realizing the automatic inspection of the system. In addition, the visual detection unit is a device with image or video acquisition function, such as a high-definition camera, a CCD camera, etc. In addition, the visual detection unit may further include a 3D pan-tilt head, which drives the image acquisition or video acquisition device to move through the 3D pan-tilt head to increase the detection range of the visual detection unit.

[0121] In this embodiment, as Figure 6 shown, the ultrasonic detection unit may include:

[0122] A first power device 2410;

[0123] The ultrasonic detection mechanism 2450 is driven by a first power device to move linearly, and the movement direction of the ultrasonic detection mechanism is perpendicular to the movement direction of the walking unit.

[0124] In this embodiment, the ultrasonic detection unit may further include a first mounting portion 2420. The first power device is fixedly mounted on the first mounting portion. The entire ultrasonic detection unit is fixedly mounted on the bottom of the walking unit through the first mounting portion. The first power device may be a motor. Additionally, a pulley transmission mechanism 2430 may be installed on the first mounting portion. This pulley transmission mechanism is connected to the motor, and the motor drives the pulley transmission mechanism to move, so as to achieve the linear movement of the ultrasonic detection mechanism.

[0125] As Figure 7 shown, the ultrasonic detection mechanism includes:

[0126] A motion module 2454;

[0127] A first ultrasonic detector 2451, which is fixed outside one end of the motion module;

[0128] A second ultrasonic detector 2452, which is installed on the motion module. The second ultrasonic detector is driven by the motion module to move linearly, so that the second ultrasonic detector approaches or moves away from the first ultrasonic detector.

[0129] In this embodiment, the motion module includes but is not limited to a cylinder motion module, a hydraulic cylinder motion module, a motor-screw motion module, etc. The first ultrasonic detector is an ultrasonic transmitter, and the second ultrasonic detector is an ultrasonic receiver.

[0130] In this embodiment, the motion module drives the second ultrasonic detector to move linearly, so that the second ultrasonic detector approaches or moves away from the first ultrasonic detector, which is convenient for adjusting the distance between the ultrasonic detectors according to the crack width.

[0131] When the crack detection unit detects a crack on the bridge road surface, the execution control unit will determine the detection starting point according to the skeleton curve. The determination method of the detection starting point is as follows:

[0132] Calculate the distance between the skeleton pixel points to obtain the alternative starting points with the farthest distance;

[0133] Calculate the distance between the alternative starting points and the walking unit, and obtain the alternative starting point closest to the walking unit as the detection starting point;

[0134] Based on the detection starting point, obtain the detection tangent and the detection normal line of the detection starting point on the skeleton curve, and re - establish a coordinate system based on the detection tangent and the detection normal line to form a detection coordinate system (with the detection tangent as the abscissa and the detection normal line as the ordinate). Generate the detection coordinates of each pixel point on the skeleton curve based on the detection coordinate system, calculate the maximum lateral distance of the skeleton curve, and calculate the starting coordinates of the first ultrasonic detector and the starting coordinates of the second ultrasonic detector according to the two pixel points of the maximum lateral distance (hereinafter referred to as the limiting points).

[0135] The calculation method of the starting coordinates of the first ultrasonic detector is as follows:

[0136] (a, b)=(x - Δx, y 0 - Δy);

[0137] Among them, (a, b) are the starting coordinates of the first ultrasonic detector; x is the abscissa of the left - hand limiting point; Δx is the lateral coordinate compensation; y 0 is the ordinate of the detection starting point; Δy is the longitudinal coordinate compensation;

[0138] The lateral coordinate compensation is determined by the following method:

[0139]

[0140] Among them, d 0 is the crack width at the left - hand limiting point, in millimeters; l 0 is the distance between the first ultrasonic detector and the crack boundary, in millimeters;

[0141] The calculation method of the starting coordinates of the second ultrasonic detector is as follows:

[0142] (c, d)=(a + A+ Δx, y 0 - Δy);

[0143] Among them, (c, d) are the starting coordinates of the second ultrasonic detector; A is the maximum lateral distance, in millimeters.

[0144] Calculate the starting coordinates of the first ultrasonic detector and the second ultrasonic detector in the above - mentioned manner. Control the movement of the walking unit through the execution control unit so that the first ultrasonic detector moves to its own starting coordinates, and then control the movement module to drive the second ultrasonic detector to move to the starting coordinates of the second ultrasonic detector.

[0145] In addition, the ultrasonic detection unit may also include a second power device 2440, which drives the ultrasonic detection mechanism to rotate horizontally through this second power device.

[0146] Drive the ultrasonic detection mechanism to rotate through the first power device, which is convenient for adjusting the detection direction of the ultrasonic detection mechanism according to the trend of the crack.

[0147] In this embodiment, the second power device includes, but is not limited to, a rotary cylinder, a rotary hydraulic cylinder, and other devices capable of driving the entire ultrasonic detection unit to rotate. The ultrasonic detection unit may further include a second mounting portion 2453, and the ultrasonic detection unit is mounted on the second power device through the second mounting portion.

[0148] Exemplarily, as Figure 6 shown, the second power device may be slidably connected to the first mounting portion through a guiding structure (such as a guide rail, a guide rod, etc.), and the second power device remains connected to the belt of the pulley transmission mechanism.

[0149] As Figure 3 shown, the inspection system may further include a repair unit 2500. The repair unit is carried on the traveling unit, and the repair material is poured into the crack through this repair unit.

[0150] By pouring the repair material into the crack through the repair unit, while realizing the crack detection, the crack can also be repaired, which is simple and convenient to operate and improves the work efficiency.

[0151] As Figure 4 shown, the repair unit includes:

[0152] A third power device 2510;

[0153] A pouring head 2530, which is driven by the third power device to move linearly horizontally, and the moving direction of the pouring head is perpendicular to the traveling direction of the traveling unit.

[0154] By driving the pouring head to move linearly through the third power device, the pouring position of the pouring head can be adjusted according to the trend of the crack.

[0155] In this embodiment, the repair unit may further include a third mounting portion 2520. The whole repair unit is mounted on the tail of the traveling unit through the third mounting portion. The third power device is fixedly mounted at the end of the third mounting portion. The third power device may be a motor, and a transmission mechanism capable of realizing the linear movement of the pouring head, such as a lead screw, a pulley transmission mechanism, a gear rack mechanism, etc., connected to the motor may be mounted on the third mounting portion. In addition, the third power device may also be a cylinder, an electric cylinder, a hydraulic cylinder, etc., which can directly drive the pouring head to move linearly.

[0156] As Figure 5 shown, the repair unit may further include:

[0157] A fourth power device 2550;

[0158] The second sliding part 2560 drives the second sliding part to move linearly in the vertical direction through the fourth power device, and the perfusion head is slidably mounted on the second sliding part;

[0159] A fifth power device drives the perfusion head to move linearly relative to the movable vertical line through the fifth power device.

[0160] By adjusting the height of the perfusion head in the way of secondary movement, the structure of the repair unit can be made more compact.

[0161] In this embodiment, the fourth and fifth power devices include but are not limited to cylinders, electric cylinders, hydraulic cylinders or other devices capable of realizing the linear movement of the second sliding part and the perfusion head. The repair unit may further include a first sliding part 2540, which is fixed on the third installation part. The second sliding part can be slidably connected to the first sliding part through a guiding structure. The fourth power device can be fixedly installed on the second sliding part, and its output part is connected to the first sliding part. The fifth power device 2570 can be fixedly installed on the second sliding part, and its output part is connected to the perfusion head. The perfusion head can be connected to the second sliding part through a guiding structure.

[0162] In addition, the inspection system may further include a slurry storage unit 2300, which is integrally carried on the traveling unit and is connected to the repair unit.

[0163] The slurry storage unit provides repair materials for the repair unit to achieve self-sufficiency of repair materials.

[0164] Embodiment 2:

[0165] This embodiment provides a bridge inspection method here. This inspection method uses an inspection system described in Embodiment 1. The inspection method includes:

[0166] Obtain a user instruction, generate a first execution control signal according to the user instruction, and send the first execution control signal to the execution module so that the execution module executes an action corresponding to the first execution control signal according to the first execution control signal;

[0167] Obtain information data, where the information data includes image information data and ultrasonic information data;

[0168] Perform crack detection according to the image information data to obtain a crack detection result;

[0169] Specifically, before detecting cracks in the image information data, the image information data can be preprocessed through image processing algorithms (such as digital image processing methods), including binarization, grayscale conversion, denoising, segmentation, extraction, etc., so as to improve the processing and analysis efficiency of the subsequent image information data and improve the target recognition accuracy. In addition, the UNet network model can be used as the basic framework, and the MobileNetV3 network can be used to replace the encoder part of the UNet for feature extraction. Subsequently, the algorithm is used for training to obtain the optimal training parameters and the optimal model parameters, and the establishment of the crack recognition model is completed. The crack features in the image are recognized through the crack recognition model, and finally the detection of road surface cracks is realized.

[0170] Generate a second execution control signal according to the crack detection result, and send the second execution control signal to the execution module, so that the execution module executes the action corresponding to the second execution control signal according to the second execution control signal;

[0171] According to the crack detection result, extract the target feature boundary, extract the target feature skeleton pixel set of the image according to the target feature boundary, perform curve fitting according to the target feature skeleton pixel set to obtain the target feature skeleton curve, and perform crack trend analysis on the target feature skeleton curve to obtain the crack trend analysis result;

[0172] Specifically, after the crack detection unit detects that there are cracks on the road surface, the crack trend analysis unit will extract the boundary of the crack features in the road surface image processed by the image processing algorithm, so as to form a crack boundary line image, extract the target feature edge pixel set based on the boundary line image. At the same time, it will also skeletonize the crack features in the road surface image processed by the image processing algorithm to extract the skeleton pixel set of the crack features, perform curve fitting on the basis of the skeleton pixel set to obtain the skeleton curve of the target feature, and finally perform trend analysis through the skeleton curve. The trend analysis includes length calculation, curve shape matching, etc. The calculation method of the crack length can refer to the calculation method of the crack length recorded in Embodiment 1.

[0173] Calculate the normal line of the pixel points on the skeleton curve in the skeleton curve, and calculate the crack width according to the intersection points of the normal line and the target feature boundary to obtain the crack width calculation result;

[0174] Specifically, the target feature boundary includes a first boundary and a second boundary. The crack area is between the first boundary and the second boundary. Calculate the normal line of the target point on the skeleton curve, and calculate the width of the crack according to the intersection points of the normal line and the first boundary and the second boundary. The target point is a point with a certain interval on the skeleton curve (the size of this interval is determined comprehensively according to factors such as image resolution and crack type). The calculation method of the crack width can refer to the calculation method of the crack width recorded in Embodiment 1.

[0175] Calculate the crack depth based on the ultrasonic information data to obtain the crack depth calculation result.

[0176] The calculation method of the crack depth can refer to the crack depth calculation method described in Embodiment 1.

[0177] Before detecting the crack, when the crack detection unit detects a crack on the bridge road surface, the execution control unit will determine the detection starting point according to the skeleton curve. The determination method of the detection starting point is as follows:

[0178] Calculate the distance between the skeleton pixel points to obtain the alternative starting points with the maximum distance.

[0179] Calculate the distance between the alternative starting point and the walking unit, and obtain the alternative starting point closest to the walking unit as the detection starting point.

[0180] Based on the detection starting point, obtain the detection tangent and the detection normal of the detection starting point on the skeleton curve, re - establish the coordinate system based on the detection tangent and the detection normal to form a detection coordinate system (taking the detection tangent as the abscissa and the detection normal as the ordinate), generate the detection coordinates of each pixel point on the skeleton curve based on the detection coordinate system, calculate the maximum horizontal distance of the skeleton curve, calculate the starting point coordinates of the first ultrasonic detector and the starting point coordinates of the second ultrasonic detector according to the two pixel points with the maximum horizontal distance (hereinafter referred to as the limit points). The specific calculation method can refer to the calculation method of the starting point coordinates of the first ultrasonic detector and the starting point coordinates of the second ultrasonic detector described in Embodiment 1.

[0181] Control the movement of the walking unit through the execution control unit so that the first ultrasonic detector moves to its own starting point coordinates, and then control the movement module to drive the second ultrasonic detector to move to the starting point coordinates of the second ultrasonic detector.

[0182] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A bridge inspection system, characterized in that: It includes a control module and an execution module, and the control module is wirelessly connected to the execution module; The control module includes: An information acquisition unit, used to acquire information data, wherein the information data includes image information data and ultrasound information data; A crack detection unit, used for performing crack detection according to the image information data to obtain a crack detection result; A crack trend analysis unit is used to extract a target feature boundary according to the crack detection result, extract a target feature skeleton pixel set according to the target feature boundary extraction image, perform curve fitting according to the target feature skeleton pixel set to obtain a target feature skeleton curve, perform crack trend analysis on the target feature skeleton curve, and obtain a crack trend analysis result; A crack width calculation unit is used to calculate the normal line of a pixel point in the skeleton curve on the skeleton curve, and calculate the crack width according to the intersection of the normal line and the target feature boundary to obtain a crack width calculation result; A crack depth calculation unit is used to calculate the crack depth according to the ultrasonic information data to obtain a crack depth calculation result; The execution control unit obtains user instructions or crack detection results, generates an execution control signal according to the user instructions or crack detection results, and sends the execution control signal to the execution module so that the execution module executes the action corresponding to the execution control signal according to the execution control signal.

2. A bridge inspection system according to claim 1, characterized in that: The execution module includes: A walking unit, through which the execution module is moved as a whole on a plane; A visual detection unit, which is mounted on the walking unit and is used to obtain image information data; An ultrasonic detection unit, which is mounted on the walking unit and is used to obtain ultrasonic information data; The walking unit walks on the bridge pavement under the control of the execution control unit. At the same time, the visual detection unit continuously collects image information data of the bridge pavement and uploads it to the control module. When a pavement crack is detected, the execution control unit controls the walking unit to move to the crack point. At this time, the ultrasonic detection unit starts to work and collects ultrasonic information data. Under the information data processing, analysis and calculation of the control module, the width, depth and trend of the crack are obtained, thereby realizing automatic detection of cracks on the bridge surface and automatic measurement of crack parameters.

3. A bridge inspection system according to claim 2, characterized in that: The ultrasonic detection unit comprises: The first power unit; An ultrasonic detection mechanism, driven by a first power device to move linearly, wherein the movement direction of the ultrasonic detection mechanism is perpendicular to the movement direction of the walking unit; The ultrasonic detection unit also includes a first mounting portion, the first power device is fixedly mounted on the first mounting portion, the ultrasonic detection unit as a whole is fixedly mounted on the bottom of the walking unit through the first mounting portion, the first power device can be a motor, and a pulley transmission mechanism can also be installed on the first mounting portion, the pulley transmission mechanism is connected to the motor, and the pulley transmission mechanism is driven by the motor to move, so as to realize the linear movement of the ultrasonic detection mechanism; The ultrasonic detection unit also includes a second power device, through which the ultrasonic detection mechanism is driven to rotate horizontally; The ultrasonic detection mechanism is driven to rotate by the first power device, so as to adjust the detection direction of the ultrasonic detection mechanism according to the direction of the crack; The ultrasonic detection unit also includes a second mounting portion, and the ultrasonic detection unit is mounted on a second power device through the second mounting portion; the second power device can be slidably connected to the first mounting portion through a guide structure, and the second power device is connected to the belt of the pulley transmission mechanism.

4. A bridge inspection system according to claim 1, characterized in that: The ultrasonic detection mechanism comprises: Sports module; A first ultrasonic detector, the first ultrasonic detector is fixed on the outside of one end of the motion module; A second ultrasonic detector is mounted on the motion module, and the motion module drives the second ultrasonic detector to move linearly, so that the second ultrasonic detector is close to or away from the first ultrasonic detector; The second ultrasonic detector is driven to move linearly by the motion module so that the second ultrasonic detector is close to or away from the first ultrasonic detector, so as to adjust the distance between the ultrasonic detectors according to the crack width; When the crack detection unit detects cracks on the bridge pavement, the execution control unit determines the detection starting point according to the skeleton curve. The detection starting point is determined as follows: Calculate the distance between skeleton pixels and obtain the candidate starting point with the farthest distance; Calculate the distance between the candidate starting point and the walking unit, and obtain the candidate starting point closest to the walking unit as the detection starting point; Based on the detection starting point, a detection tangent and a detection normal of the detection starting point on the skeleton curve are obtained, a coordinate system is re-established based on the detection tangent and the detection normal to form a detection coordinate system, detection coordinates of each pixel point on the skeleton curve are generated based on the detection coordinate system, a maximum lateral distance of the skeleton curve is calculated, and the starting point coordinates of the first ultrasonic detector and the starting point coordinates of the second ultrasonic detector are calculated based on the two pixel points of the maximum lateral distance; The coordinates of the starting point of the first ultrasonic detector are calculated as follows: (a, b) = (x-Δx, y0-Δy); Wherein, (a, b) is the starting point coordinate of the first ultrasonic detector; x is the horizontal coordinate of the left limit point; Δx is the horizontal coordinate compensation; y0 is the vertical coordinate of the detection starting point; Δy is the vertical coordinate compensation; The lateral coordinate compensation is determined by: Wherein, d0 is the crack width at the left limit point, in millimeters; l0 is the distance between the first ultrasonic detector and the crack boundary, in millimeters; The starting point coordinates of the second ultrasonic detector are calculated as follows: (c, d)=(a+A+Δx, y0-Δy); Wherein, (c, d) are the coordinates of the starting point of the second ultrasonic detector; A is the maximum lateral distance, in millimeters; The starting point coordinates of the first ultrasonic detector and the second ultrasonic detector are calculated in the above manner, and the walking unit is controlled to move by the execution control unit so that the first ultrasonic detector moves to its own starting point coordinates, and then the motion module is controlled to drive the second ultrasonic detector to move to the starting point coordinates of the second ultrasonic detector.

5. A bridge inspection system according to claim 1, characterized in that: The inspection system further comprises a repair unit, which is carried on the walking unit and through which repair materials are poured into the cracks.

6. A bridge inspection system according to claim 5, characterized in that: The repair unit comprises: The third power unit; The grouting head is driven to move horizontally and linearly by the third power device. The movement direction of the grouting head is perpendicular to the walking direction of the walking unit. The grouting head is driven to move linearly by the third power device. The grouting position of the grouting head can be adjusted according to the trend of the crack. The repair unit also includes a third mounting part, and the repair unit as a whole is installed on the tail of the walking unit through the third mounting part. The third power device is fixedly installed at the end of the third mounting part. The third power device can select a motor, and a screw rod, a pulley transmission mechanism, a gear rack mechanism, etc. connected to the motor are installed on the third mounting part to realize the linear movement of the injection head.

7. A bridge inspection system according to claim 6, characterized in that: The repair unit also includes: Fourth power unit; A second sliding part, driven by a fourth power device to move vertically linearly, the injection head being slidably mounted on the second sliding part; a fifth power device, through which the injection head is driven to move in a vertical linear motion relative to the movable body; The height of the injection head can be adjusted by means of secondary movement, which can make the structure of the repair unit more compact; The fourth and fifth power devices include but are not limited to cylinders, electric cylinders, hydraulic cylinders or other devices that can realize linear movement of the second sliding part and the injection head. The repair unit also includes a first sliding part 2540, which is fixed on the third mounting part. The second sliding part can be slidably connected to the first sliding part through a guide structure. The fourth power device can be fixedly installed on the second sliding part, and its output part is connected to the first sliding part. The fifth power device can be fixedly installed on the second sliding part, and its output part is connected to the injection head. The injection head can be connected to the second sliding part through a guide structure.

8. A bridge inspection system according to claim 1, characterized in that: The inspection system further comprises a slurry storage unit, which is integrally mounted on the walking unit and connected to the repair unit. The slurry storage unit provides repair materials to the repair unit, thereby achieving self-sufficiency of the repair materials.

9. A bridge inspection method, characterized in that: The inspection method uses the inspection system according to any one of claims 1 to 8, and the inspection method includes: Obtaining a user instruction, generating a first execution control signal according to the user instruction, and sending the first execution control signal to an execution module, so that the execution module executes an action corresponding to the first execution control signal according to the first execution control signal; Acquiring information data, wherein the information data includes image information data and ultrasound information data; Perform crack detection according to the image information data to obtain crack detection results; Generate a second execution control signal according to the crack detection result, and send the second execution control signal to the execution module, so that the execution module executes an action corresponding to the second execution control signal according to the second execution control signal; According to the crack detection results, the target feature boundary is extracted, the target feature skeleton pixel set is extracted from the target feature boundary extraction image, the target feature skeleton pixel set is subjected to curve fitting to obtain the target feature skeleton curve, the crack trend analysis is performed on the target feature skeleton curve to obtain the crack trend analysis result; Specifically, after the crack detection unit detects the existence of cracks on the road surface, the crack trend analysis unit will extract the boundary of the crack feature in the road surface image processed by the image processing algorithm, thereby forming a crack boundary line image, and extract the target feature edge pixel set based on the boundary line image. At the same time, the crack features in the road surface image processed by the image processing algorithm will be skeletonized to extract the skeleton pixel set of the crack features, and curve fitting will be performed on the basis of the skeleton pixel set to obtain the skeleton curve of the target feature, and finally the trend analysis will be performed through the skeleton curve, and the trend analysis includes length calculation, curve shape matching, etc. Calculate the normal of the pixel point in the skeleton curve on the skeleton curve, calculate the crack width according to the intersection of the normal line and the target feature boundary, and obtain the crack width calculation result; Specifically, the target feature boundary includes a first boundary and a second boundary, the first boundary and the second boundary are the crack area, the normal of the target point on the skeleton curve is calculated, and the width of the crack is calculated according to the intersection of the normal and the first boundary and the second boundary. The target point is a point with a certain interval on the skeleton curve. The calculation method of the crack width can refer to the calculation method of the crack width recorded in Example 1; Calculate the crack depth according to the ultrasonic information data to obtain the crack depth calculation result; Before crack detection, when the crack detection unit detects cracks on the bridge pavement, the execution control unit determines the detection starting point according to the skeleton curve. The detection starting point is determined as follows: Calculate the distance between skeleton pixels and obtain the candidate starting point with the farthest distance; Calculate the distance between the candidate starting point and the walking unit, and obtain the candidate starting point closest to the walking unit as the detection starting point; Based on the detection starting point, a detection tangent and a detection normal of the detection starting point on the skeleton curve are obtained, a coordinate system is reestablished based on the detection tangent and the detection normal to form a detection coordinate system, detection coordinates of each pixel point on the skeleton curve are generated based on the detection coordinate system, a maximum lateral distance of the skeleton curve is calculated, and the starting point coordinates of the first ultrasonic detector and the starting point coordinates of the second ultrasonic detector are calculated based on two pixel points (hereinafter referred to as limit points) of the maximum lateral distance. The execution control unit controls the movement of the walking unit so that the first ultrasonic detector moves to its own starting point coordinates, and then controls the motion module to drive the second ultrasonic detector to move to the starting point coordinates of the second ultrasonic detector; The length of the crack is calculated in the following way: Where l is the crack length in mm; n is the number of pixels in the skeleton pixel set; x n ,y n is the horizontal and vertical coordinates of the nth pixel in the skeleton pixel set; n+1 ,y n+1 is the horizontal and vertical coordinates of the n+1th pixel point in the skeleton pixel set; The crack width can be calculated by: Where d is the crack width in millimeters; i is the number of target points; x Ri ,y Ri x is the horizontal and vertical coordinates of the intersection point between the normal line and the first boundary at the i-th pixel point in the skeleton curve; Li ,y Li are the horizontal and vertical coordinates of the intersection point between the normal line and the second boundary at the i-th pixel point in the skeleton curve; Crack depth calculation can be achieved by: Among them, h is the crack depth, in millimeters; L is the ultrasonic propagation distance on a crack-free road surface, in millimeters; t is the time taken for the ultrasonic wave to bypass the crack, in seconds; T is the time taken for the ultrasonic wave to propagate at L on a crack-free road surface, in seconds.

Images