Defect detection and early warning system and device for bridge

Abstract

The invention discloses a defect detection and early warning system and device for a bridge, and relates to the technical field of bridge crack defect detection.The detection and early warning system comprises a first detection unit, a second detection unit and a third detection unit, the first detection unit is used for obtaining a grey-scale map of a bridge pavement within a certain area range, then extracting size information of a crack and marking the position of the crack relative to the bridge pavement; the second detection unit is used for receiving the position, output by the first detection unit, of the small crack relative to the bridge pavement, obtaining echo data of the position after the ultrasonic waves are applied, and analyzing the echo data to obtain crack depth information of the small crack. According to the method, rapid and large-scale initial detection of the bridge pavement cracks is realized through an image processing technology, and classification is carried out according to the identified crack size features; and for the detected and classified fine cracks, the depth information of the fine cracks is supplemented through ultrasonic detection, so that dangerous cracks with relatively large depths are prevented from being missed.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge crack defect detection, in particular to a bridge defect detection and early warning system and a device thereof. Background Art

[0002] Current bridge defect detection systems mostly rely on image detection technology, which uses high-resolution cameras to photograph the bridge surface and then uses image processing algorithms to detect and identify defects such as cracks in the bridge.

[0003] This method has good recognition for larger cracks, but lacks sufficient detection capability for narrow cracks, especially deep cracks hidden in high stress areas (such as beam end areas). These narrow but deep cracks can easily become the cause of local damage to bridge structures, so it is difficult to fully assess the impact of cracks on bridge safety. Summary of the invention

[0004] 1) Technical issues solved

[0005] The present invention provides a defect detection and early warning system and device for bridges, which can comprehensively consider the three dimensions of crack length, width and depth, and include large cracks and dangerous cracks with small surface size but large depth into the maintenance scope. By clearly assigning priority levels to cracks of different sizes and depths, it is convenient to formulate a graded repair strategy later.

[0006] 2) Technical solution

[0007] To achieve the above-mentioned purpose, the present invention provides the following technical solution: a defect detection and early warning system for bridges, suitable for detecting cracks in bridge pavement, comprising:

[0008] A first detection unit is used to obtain a grayscale image of the bridge pavement within the detection area, identify cracks after edge processing of the grayscale image, extract size information of each crack and mark its position information relative to the bridge pavement, and then output the size information and position information of each crack; wherein a first size threshold is set in the first detection unit, and cracks whose size information is smaller than the first size threshold are classified as small cracks, and the remaining cracks are classified as large cracks;

[0009] A second detection unit is used to receive the size information and position information of each small crack output by the first detection unit, obtain the echo data of each small crack on the bridge pavement after applying ultrasonic waves, and obtain the crack depth information of the small crack at the corresponding position after analyzing the echo data;

[0010] an early warning unit, configured to receive the size information and position information of the large crack and the small crack output by the first detection unit, and receive the crack depth information of the small crack output by the second detection unit; wherein a second size threshold and a first depth threshold are set in the early warning unit,

[0011] For a large crack whose size information is greater than the second size threshold, assigning a first priority;

[0012] For a large crack whose size information is smaller than the second size threshold, a third priority is assigned;

[0013] For small cracks whose crack depth information is greater than the first depth threshold, assign a second priority;

[0014] For small cracks whose crack depth information is less than the first depth threshold, a fourth priority is assigned;

[0015] Among them, the levels to be repaired of the first priority, the second priority, the third priority and the fourth priority decrease in sequence.

[0016] Furthermore, the first detection unit performs the edge processing on the grayscale image after removing noise and enhancing contrast; wherein:

[0017] The noise removal operation includes smoothing the image and suppressing the interference of local light fluctuations;

[0018] The contrast enhancement operation includes applying histogram equalization to increase the contrast between the cracks and the background.

[0019] Furthermore, the first detection unit identifies cracks after performing canny edge detection on the grayscale image, specifically:

[0020] Calculate the gradient amplitude and direction of the grayscale image, and find the area in the grayscale image where the intensity changes dramatically;

[0021] Use non-maximum suppression to eliminate false edges in the gradient magnitude map;

[0022] A double threshold method is used to distinguish strong edges from weak edges, and the weak edges are connected to extract a complete contour, which characterizes the crack.

[0023] Further, the first detection unit generates a binary edge map after performing canny edge detection on the grayscale image, and extracts the size information of each crack after performing skeleton extraction and geometric analysis on the binary edge map, wherein the size information includes length information and width information of the crack; and

[0024] A spatial coordinate system is constructed based on the bridge pavement, and after extracting the plane pixel coordinates of the crack in the binary edge map, the plane pixel coordinates of the crack are mapped to the spatial coordinate system of the bridge pavement to generate the position information of the crack relative to the bridge pavement.

[0025] Furthermore, the first size threshold set in the first detection unit specifically includes a first length threshold and a first width threshold. For the length information and width information of the crack identified and calculated from the binary edge map, if the maximum length value in the crack length information is less than the preset first length threshold, and the maximum width value in the crack width information is less than the preset first width threshold, the corresponding crack is classified as a small crack; otherwise, it is classified as a large crack.

[0026] Further, after identifying the small cracks, the first detection unit sends the size information and position information corresponding to each small crack to the second detection unit, and the second inspection unit is used to detect and obtain the crack depth information of each small crack; specifically,

[0027] At each of the small cracks, an ultrasonic emission point is selected and an ultrasonic wave with fixed specification parameters is emitted, and the second detection unit receives echo data at the corresponding ultrasonic emission point, and calculates the crack depth value of the small crack at the ultrasonic emission point by analyzing the echo flight time difference and echo intensity characteristics in the echo data;

[0028] A plurality of ultrasonic emission points are selected at different positions of the small crack, the echo data are received and calculated point by point, and the crack depth value corresponding to each ultrasonic emission point is integrated into the crack depth information of the small crack.

[0029] Further, the second size threshold set in the early warning unit specifically includes a second length threshold and a second width threshold. If the maximum length value in the crack length information is greater than the second length threshold, and the maximum width value in the crack width information is greater than the second width threshold, the corresponding large crack is assigned the first priority, and the position information of the large crack is recorded;

[0030] If the maximum length value in the crack length information is less than the second length threshold, or the maximum width value in the crack width information is less than the second width threshold, the corresponding large crack is assigned the third priority and the position information of the large crack is recorded.

[0031] Further, after the early warning unit receives the crack depth information of each small crack output from the second detection unit and compares it with the first depth threshold, if there is at least one crack depth value in the crack depth information that is greater than the first depth threshold, the corresponding small crack is assigned the second priority and the position information of the small crack is recorded;

[0032] If all the crack depth values ​​in the crack depth information are smaller than the first depth threshold, the corresponding small cracks are assigned the fourth priority and the position information of the small cracks is recorded.

[0033] A defect detection and early warning device for a bridge, comprising:

[0034] Moving parts;

[0035] A supporting component, located on the moving component, and the supporting component can move in a direction perpendicular to the bridge pavement;

[0036] A visual detection device, an ultrasonic detection device, an alarm device and a memory are located on the supporting component, and the visual detection device, the ultrasonic detection device, the alarm device and the memory are configured to jointly perform the functions of any of the above-mentioned defect detection and early warning systems; wherein the ultrasonic detection device also includes a spraying device, and the spraying device is used to spray an oil-based coupling agent onto the bridge pavement before performing ultrasonic detection on cracks in the bridge pavement.

[0037] 3) Beneficial effects:

[0038] Compared with the prior art, the invention has the following beneficial effects:

[0039] The present invention realizes rapid and large-scale initial inspection of cracks in bridge pavement through image processing technology, and classifies them according to the identified crack size characteristics; for the small cracks after detection and classification, ultrasonic detection is used to supplement the depth information of the small cracks to avoid missing dangerous cracks with larger depths.

[0040] Taking into account the three dimensions of crack length, width and depth, large cracks and dangerous cracks with small surface size but large depth are included in the maintenance scope. By assigning clear priority levels to cracks of different sizes and depths, it is convenient to formulate a graded repair strategy later. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A principle block diagram of a bridge defect detection and early warning system provided by an embodiment of the present invention;

[0042] Figure 2 A three-dimensional schematic diagram of a bridge defect detection and early warning device provided by an embodiment of the present invention;

[0043] Figure 3 A grayscale image of a bridge pavement within a certain area obtained by a first detection unit in a bridge defect detection and early warning system provided by an embodiment of the present invention;

[0044] Figure 4 The first detection unit will Figure 3 The grayscale image shown is converted into a two-dimensional edge image;

[0045] In the figure:

[0046] 10. Moving part; 20. Supporting part; 30. Visual detection device; 40. Ultrasonic detection device; 50. Push rod; 60. Spraying device; 70. Alarm device;

[0047] 100. a first detection unit;

[0048] 200, a second detection unit;

[0049] 300. Early warning unit. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0051] In the description of the present invention, it should be understood that the terms "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0052] In addition, the terms “first”, “second”, etc., if used, are merely used to distinguish between the descriptions and should not be understood as indicating or implying relative importance.

[0053] It should be noted that, in the absence of conflict, the features in the embodiments of the present invention may be combined with each other.

[0054] In the actual process of bridge pavement defect detection, the inventors found that most current detection devices or systems rely mainly on image detection technology to identify surface defects. The core of this technology is to use a high-resolution camera to take a comprehensive or regional photo of the bridge surface, and then use an image processing algorithm to analyze the captured image to detect and identify surface defects such as cracks and damage on the bridge.

[0055] Although image detection technology has the advantages of wide coverage, convenient operation and fast processing in bridge surface crack detection, it also exposes some limitations in practical applications. In particular, in the identification process of defect detection, this technology can usually detect larger cracks more accurately, such as cracks with longer lengths or wider widths. These cracks are easy to be identified by algorithms because of their significant appearance characteristics. These large-size cracks do have a higher potential for harm in bridge structures and are usually regarded as key maintenance targets.

[0056] However, in actual operation, the inventors found that when faced with small cracks with narrow width, the recognition accuracy of large-scale image detection technology is significantly reduced. More importantly, in some high-stress areas of bridges, such as beam ends, near supports, or bridge deck joints, these narrow but deep cracks have great hidden dangers. Although they are small in width and difficult to detect, they may penetrate the local structure due to their depth, and are prone to cause local damage under the long-term action of external loads or environmental erosion, thus having an impact on the overall structural safety of the bridge that cannot be ignored. If these cracks are not identified and repaired in time, they may cause further expansion of the cracks, ultimately causing serious bridge damage or even safety accidents.

[0057] Therefore, for these narrow cracks, especially cracks in high stress areas, image detection technology based on surface recognition can no longer meet actual needs. To this end, the embodiments of the present invention provide a defect detection and early warning system and device for bridges, which realizes rapid and large-scale initial inspection of cracks on bridge pavement through image processing technology, and classifies them according to the identified crack size characteristics; for small cracks after detection and classification, ultrasonic detection is used to supplement the depth information of the small cracks to avoid missing dangerous cracks with larger depths. The system and device divide cracks of different sizes and depths into clear priority levels after comprehensive evaluation of crack size and depth, and include large cracks and small cracks with potential hazards in the maintenance scope while formulating a reasonable construction plan, which is convenient for formulating a graded repair strategy.

[0058] Specifically, refer to Figure 1 , Figure 1 This is a principle block diagram of a bridge defect detection and early warning system provided by the present invention, wherein the first detection unit 100 is used to obtain a grayscale image of the bridge pavement within a certain area. Figure 2 , Figure 2 A three-dimensional schematic diagram of a bridge defect detection and early warning device provided by an embodiment of the present invention. In some embodiments of the present invention, Figure 2 The visual inspection device 30 specifically includes a high-resolution camera, that is, the first inspection unit 100 uses a high-resolution camera to shoot a section of the bridge pavement to generate a corresponding grayscale image. The grayscale image represents the brightness distribution of the bridge surface with the grayscale value (0-255) of each pixel, and cracks are usually shown as low-grayscale (dark) lines in the image.

[0059] Specific as Figure 3 As shown, Figure 3 A grayscale image is generated by photographing a section of the bridge surface. The grayscale image shows that the crack area has a low grayscale value (such as a grayscale value of 20-50), while the background has a high grayscale value (such as a grayscale value of 150-200).

[0060] After obtaining the grayscale image, it is first necessary to remove the noise. In some embodiments of the present invention, a Gaussian filter is used to smooth the grayscale image to suppress the interference of noise and local illumination changes. The Gaussian filter uses a weighted kernel to calculate the weighted average of the pixels around each pixel to reduce sharp noise signals.

[0061] After smoothing the grayscale image, the grayscale image is histogram-equalized to redistribute the grayscale value range and enhance the contrast of the image. In other words, the gap between the low grayscale value area (crack) and the high grayscale value area (background) is widened to make the crack more prominent.

[0062] In summary, it can be understood that the crack area in the original grayscale image is blurred under noise interference. After Gaussian filtering and histogram equalization, the cracks in the grayscale image are clearly visible, and the grayscale range is extended from 20-50 (crack) and 150-200 (background) to 10-30 and 180-255.

[0063] After the grayscale image is preprocessed by removing noise and enhancing contrast, the cracks in the grayscale image are identified by edge processing.

[0064] Specifically, in some embodiments of the present invention, the canny edge detection technique is used to identify cracks in the preprocessed grayscale image. First, the Sobel operator is used to calculate the gradient amplitude (edge ​​strength) and direction of each pixel in the grayscale image, where the gradient amplitude represents the severity of the grayscale change, and the gradient amplitude at the crack edge is high. The Sobel operator extracts the horizontal and vertical gradients of the image through a convolution operation, and then calculates the gradient amplitude and direction.

[0065] After finding the area with drastic intensity changes in the grayscale image through the Sobel operator, the gradient amplitude map is subjected to non-maximum suppression processing to retain the local maximum value along the gradient direction and suppress false edges. Specifically, if the gradient amplitude of a pixel is smaller than the pixel values ​​on both sides along the gradient direction, it is set to zero.

[0066] By setting high and low thresholds, the edges are divided into strong edges and weak edges. A strong edge is one with a gradient amplitude greater than the high threshold, and a weak edge is one with a gradient amplitude between the low and high thresholds. If it is connected to a strong edge, it is retained as an edge. This method is used to disconnect weak edges of noise or isolated points.

[0067] Through Canny edge detection, a binary edge map is generated, as follows Figure 4 As shown in Figure 2, the crack outline is reduced to a single-pixel-wide skeleton line using morphological operations, preserving the shape characteristics of the crack.

[0068] Regarding the geometric calculation of the crack, specifically, along the skeleton line, find the distance between the two farthest points as the length of the crack, such as Figure 4 The crack length in is L; in the edge diagram of the crack, the skeleton line is scanned vertically, the distance between the left and right edges of each skeleton point is measured, and the maximum width of the crack is calculated. That is, the maximum width output is the width of the crack, such as Figure 4 The crack width in is W.

[0069] Regarding how to obtain the position of the crack relative to the bridge pavement, specifically, first determine the position of the crack in the two-dimensional coordinate system in the image, find all the pixel points on the crack edge or skeleton line, record and output their image coordinates (u, v), and store the coordinate information of these pixel points as a list or array to form a position set of the crack in the image coordinate system.

[0070] Among them, the image coordinate system is a two-dimensional discrete coordinate system with pixels as units. The coordinate values ​​of the pixel points are determined by the image resolution. The origin of the image coordinate system is usually set at the upper left corner of the image. Here, it should be noted that in order to ensure the continuity and consistency of the crack recognition results, the grayscale image obtained by the system and device provided in the embodiments of the present invention needs to meet the requirements of continuous splicing.

[0071] During defect detection, a three-dimensional spatial coordinate system needs to be constructed on the bridge pavement. Usually, the longitudinal direction of the bridge surface (parallel to the length of the bridge) is the X-axis, the transverse direction of the bridge surface (parallel to the width of the bridge) is the Y-axis, and the height direction perpendicular to the bridge pavement is the Z-axis. The starting position of the detection device is used as the origin to determine the axial direction and unit scale of the coordinate system.

[0072] Map the crack locations in the image coordinate system to the actual physical coordinates on the bridge surface.

[0073] Regarding how to determine the mapping relationship between the image pixel coordinates and the actual three-dimensional space coordinates, specifically, in some feasible embodiments of the present invention, a calibration plate or a laser ranging device is used to obtain the intrinsic and extrinsic parameters of the camera, that is, to obtain the focal length, position and orientation of the camera in the visual detection device 30, and the image coordinates (x, v) are converted into the camera coordinates (x c ,y c , z c ), and then convert the camera coordinates to the three-dimensional space coordinates of the bridge surface through the calibration matrix (X w , Y w , Z w ), and then the crack is mapped to the two-dimensional coordinates (X w , Y w ), all the pixels of the crack are mapped to the two-dimensional coordinate data of the bridge surface and integrated into a spatial curve to describe the actual position and shape of the crack and store the corresponding crack spatial size information.

[0074] In summary, it can be understood that after performing Canny edge detection on the grayscale image, a binary edge map is generated, skeleton extraction and geometric analysis are performed on the binary edge map, and the length and width of the crack are calculated; the pixel position of the crack is extracted from the binary edge map, and combined with the calibration data, the pixel coordinates of the crack are converted into a physical position in the bridge surface coordinate system to generate the actual position data of the crack relative to the bridge pavement.

[0075] Regarding how to compare the extracted crack size features with preset values ​​and classify them into large cracks and small cracks, specifically, according to the bridge engineering specifications or specific project requirements, length and width thresholds are set to distinguish large cracks from small cracks.

[0076] In some embodiments of the present invention, in the first detection unit 100, a crack length threshold L is set. threshold and the crack width threshold W threshold , for the length and width information of the cracks identified and calculated from the grayscale image, specifically the global length of the crack, that is, the straight-line distance L between the farthest two points on the crack skeleton, and at each point of the crack skeleton, the vertical distance between the left and right boundaries is measured, and the maximum value W of all measured values ​​is selected, if:

[0077] L≤L threshold And W≤W threshold

[0078] The crack corresponding to this size is a small crack, otherwise,

[0079] L>L threshold or W>W threshold

[0080] The crack corresponding to this size is a large crack.

[0081] Regarding how to set the priority level of large cracks, specifically, the early warning unit 300 receives the size information of the large cracks output by the first detection unit 100, including the length and width data of each large crack, and extracts the maximum length value and the maximum width value of each large crack. The early warning unit 300 sets the second length threshold and the second width threshold, and performs the following judgment on each large crack:

[0082] If the maximum length value > the second length threshold, and the maximum width value > the second width threshold, it is determined to be a high priority crack and is assigned the first priority.

[0083] If the maximum length value is ≤ the second length threshold, or the maximum width value is ≤ the second width threshold, it is determined to be a low priority crack and is assigned the third priority.

[0084] Regardless of the crack priority, the physical location information of the corresponding crack will be recorded in the early warning unit for subsequent maintenance planning.

[0085] For small cracks, the first detection unit 100 sends the crack size characteristics classified as small cracks and their positions relative to the bridge pavement to the second detection unit 200, and enters the ultrasonic depth detection process to determine whether its depth exceeds the danger standard.

[0086] Specifically, after receiving the position information of the small crack, the second detection unit 200 refers to Figure 2 The visual inspection device 30 in the device includes an ultrasonic probe, which moves to the crack area according to the physical position of the crack provided by the first inspection unit 100, sprays an oil-based coupling agent to optimize the transmission of ultrasonic signals, and then performs inspection.

[0087] The ultrasonic probe emits high-frequency sound waves to the crack area. When the sound waves penetrate the crack, part of the signal is reflected at the bottom of the crack to form an echo signal. The echo signal data contains the echo flight time difference and echo intensity characteristics. The flight time difference in the echo signal and the propagation speed of the ultrasonic wave in the coupling agent and the material can be used to accurately calculate the depth of the crack. The echo intensity characteristics can be used to assist in the analysis of the crack's penetration, width and reflection surface properties.

[0088] Specifically, the echo flight time refers to the time interval from the ultrasonic wave being emitted from the probe to the ultrasonic wave being reflected from the crack surface back to the probe, and the flight time difference usually refers to the time difference between the crack reflection signal and the bottom reflection signal. If the crack does not penetrate the material, two echoes will appear: one is the reflection signal from the crack surface, and the other is the reflection signal from the bottom below the crack. The flight time difference is used to calculate the distance from the crack surface to the bottom, thereby determining the depth of the crack.

[0089] Echo intensity refers to the amplitude and energy of the echo signal received by the probe, reflecting the material interface's ability to reflect ultrasound. Echo intensity can help determine the size, width, and complete penetration of the crack. When the crack width is large, the reflected signal intensity will decrease because part of the ultrasound is scattered or attenuated. If the crack penetrates the entire material, the bottom reflected signal will completely disappear, leaving only a strong echo on the crack surface. It can be understood that the wider the crack, the lower the echo intensity. If the crack depth is large, the reflected signal intensity at the bottom of the crack will be significantly weakened.

[0090] In some feasible embodiments of the invention, for each small crack identified by the first detection unit 100, because cracks usually have different depth variations at different positions, in order to accurately evaluate the crack depth information, the system adopts a point-by-point detection method of an ultrasonic probe, calculates the crack depth value by analyzing the key features of the echo data, and finally generates complete crack depth information.

[0091] Specifically, multiple ultrasonic emission points are selected in each small crack area, wherein the position of each emission point can be evenly distributed based on the contour, length and width characteristics of the crack to ensure that the detection covers the entire crack. For example, the longer the crack or the more complex the shape, the denser the emission points are. Ultrasonic waves with fixed specification parameters are emitted from each selected emission point to ensure the consistency and controllability of the emission waveform. Generally, ultrasonic parameters include frequency (such as 2MH or 5MHz), wavelength and emission power, which can be comprehensively adjusted according to the bridge material and detection accuracy.

[0092] The second detection unit 200 transmits ultrasonic waves point by point and receives corresponding echo data, and generates a depth value d for each detection point. i , forming crack depth information data.

[0093] The early warning unit 300 receives the small crack depth data provided by the second detection unit 200, including the location of the small crack and a depth value set (i.e., depth information data). A depth threshold is set in the early warning unit 300, and the crack depth information data is compared point by point with the preset dangerous depth threshold to determine whether there is a crack depth value that exceeds the threshold.

[0094] Specifically, the early warning unit 300 assigns a priority to the small crack based on the depth information and compares it with the first depth threshold to determine its potential harmfulness.

[0095] If at least one depth value d in the crack depth information i Satisfy i >D threshold , it is judged as a deep critical crack and given the second priority; if all the depth values ​​d i ≤Dthreshold , it is determined to be a common crack and is assigned the fourth priority. Regardless of the crack priority, the physical location information of the corresponding crack is recorded in the early warning unit 300 for maintenance planning.

[0096] Regarding crack priority, among which:

[0097] Priority 1 (large cracks + high-size hazards): prioritize repairs to prevent overall damage to the bridge;

[0098] Second priority (small cracks + high depth hazard): avoid local damage caused by small crack depth and intervene quickly;

[0099] Priority 3 (large cracks + low-size hazards): Second priority to ensure the structural integrity of the bridge;

[0100] Fourth priority (small cracks + low depth damage): Low priority cracks only require regular monitoring and no emergency repair is required.

[0101] Here, it is understandable that the main reason for not inspecting the depth of large cracks is that their size information is sufficient to determine the harmfulness and priority, and the depth inspection has little impact on the priority sorting, but the time and economic costs are high. It is necessary to inspect the depth of small cracks because their potential harmfulness depends on the supplementary analysis of the depth information. Allocating resources in this way can maximize the inspection efficiency and ensure that the maintenance planning is scientific and reasonable.

[0102] For a bridge defect detection and early warning device provided in an embodiment of the present invention, please refer to Figure 2 Specifically, it includes a moving part 10 for supporting the entire device. The detection device provides stable mobility and adjusts the detection area according to the detection requirements. In some embodiments of the present invention, the moving part 10 is configured as a part similar to a mobile car, equipped with a drive motor and rollers so that it can move along the bridge pavement. The moving part 10 is equipped with a positioning sensor, such as a laser ranging sensor or GPS, to locate the specific position of the device on the bridge in real time, ensuring that the moving path of the device covers the bridge detection area.

[0103] The support component 20 is connected to the moving component 10 through multiple connecting rods, and in some embodiments is designed as a mechanical arm structure, which can move freely in the horizontal and vertical directions. The support component 20 uses a servo motor to drive the connecting rod to achieve horizontal and vertical movement. It can be understood that the movement range of the support component 20 covers the bridge surface area outside the range of the moving component 10, and the distance between the detection equipment and the bridge surface can be adjusted to adapt to the slope or uneven surface of the bridge.

[0104] In some embodiments, the support member 20 is designed so that its end can rotate in a plane perpendicular to the bridge pavement to achieve rapid switching of the detection equipment between visual detection and ultrasonic detection.

[0105] Based on the design of the supporting component 20 described above, a visual detection device 30 and an ultrasonic detection device 40 are respectively installed on two opposite sides of the supporting component 20 .

[0106] Specifically, the visual inspection device 30 includes a high-resolution industrial camera, a light source system and an image processing unit, which are used to capture a grayscale image of the bridge pavement and identify the size and position of the cracks through image processing algorithms (such as Canny edge detection, skeleton extraction, etc.).

[0107] The ultrasonic detection device 40 includes an ultrasonic probe, a spraying device 60 and a signal processing unit. When the visual detection device 30 identifies a small crack, the support component 20 rotates so that the ultrasonic detection device 40 is vertically aligned with the position of the small crack; at the same time, the connecting rod system adjusts the height of the support component 20 to ensure that the ultrasonic probe maintains an appropriate distance from the crack area.

[0108] When the support component 20 is positioned at the crack detection position, the spraying device 60 is started, and the oil-based coupling agent is evenly sprayed on the cracks in the detection area through the nozzle. The oil-based coupling agent optimizes the contact between the ultrasonic probe and the bridge surface to ensure that the ultrasonic signal can be effectively transmitted, avoid signal attenuation or incomplete reflection, and thus improve the accuracy of crack depth detection. In some embodiments, in order to ensure that the oil-based coupling agent can be sprayed on the detection area and avoid the contamination of the ultrasonic probe by the coupling agent spraying, a push rod 50 is provided between the ultrasonic probe and the support component 20. The main function of the push rod 50 is to adjust the length and position of the push rod 50 when spraying the coupling agent to avoid the ultrasonic probe being directly exposed to the sprayed oil-based coupling agent. After the spraying is completed, the push rod 50 will push the ultrasonic probe back to the bridge pavement for normal crack depth detection.

[0109] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. The patent protection scope of the present invention shall be based on the claims. All equivalent structural changes made using the contents of the description and drawings of the present invention should also be included in the protection scope of the present invention.

Claims

1. A bridge defect detection and early warning system, suitable for detecting cracks in bridge pavement, characterized in that: include: A first detection unit is used to obtain a grayscale image of the bridge pavement within the detection area, identify cracks after edge processing of the grayscale image, extract size information of each crack and mark its position information relative to the bridge pavement, and then output the size information and position information of each crack; wherein a first size threshold is set in the first detection unit, and cracks whose size information is smaller than the first size threshold are classified as small cracks, and the remaining cracks are classified as large cracks; A second detection unit is used to receive the size information and position information of each small crack output by the first detection unit, obtain the echo data of each small crack on the bridge pavement after applying ultrasonic waves, and obtain the crack depth information of the small crack at the corresponding position after analyzing the echo data; an early warning unit, configured to receive the size information and position information of the large crack and the small crack output by the first detection unit, and receive the crack depth information of the small crack output by the second detection unit; wherein a second size threshold and a first depth threshold are set in the early warning unit, For a large crack whose size information is greater than the second size threshold, assigning a first priority; For a large crack whose size information is smaller than the second size threshold, a third priority is assigned; For small cracks whose crack depth information is greater than the first depth threshold, assign a second priority; For small cracks whose crack depth information is less than the first depth threshold, a fourth priority is assigned; Among them, the levels to be repaired of the first priority, the second priority, the third priority and the fourth priority decrease in sequence.

2. A bridge defect detection and early warning system according to claim 1, characterized in that: The first detection unit performs the edge processing on the grayscale image after removing noise and enhancing contrast; wherein: The noise removal operation includes smoothing the image and suppressing the interference of local light fluctuations; The contrast enhancement operation includes applying histogram equalization to increase the contrast between the cracks and the background.

3. A bridge defect detection and early warning system according to claim 2, characterized in that: The first detection unit identifies cracks after performing canny edge detection on the grayscale image, specifically: Calculate the gradient amplitude and direction of the grayscale image to find the area in the grayscale image where the intensity changes dramatically; Use non-maximum suppression to eliminate false edges in the gradient magnitude map; A double threshold method is used to distinguish strong edges from weak edges, and the weak edges are connected to extract a complete contour, which characterizes the crack.

4. A bridge defect detection and early warning system according to claim 3, characterized in that: The first detection unit generates a binary edge map after performing canny edge detection on the grayscale image, extracts the size information of each crack after performing skeleton extraction and geometric analysis on the binary edge map, wherein the size information includes length information and width information of the crack; and A spatial coordinate system is constructed based on the bridge pavement, and after extracting the plane pixel coordinates of the crack in the binary edge map, the plane pixel coordinates of the crack are mapped to the spatial coordinate system of the bridge pavement to generate the position information of the crack relative to the bridge pavement.

5. A bridge defect detection and early warning system according to claim 4, characterized in that: The first size threshold set in the first detection unit specifically includes a first length threshold and a first width threshold. For the length information and width information of the crack identified and calculated from the binary edge map, if the maximum length value in the crack length information is less than the preset first length threshold, and the maximum width value in the crack width information is less than the preset first width threshold, the corresponding crack is classified as a small crack; Otherwise, it is classified as a large crack.

6. A bridge defect detection and early warning system according to claim 5, characterized in that: After identifying the small cracks, the first detection unit sends the size information and position information corresponding to each small crack to the second detection unit, and the second inspection unit is used to detect and obtain the crack depth information of each small crack; specifically, At each of the small cracks, an ultrasonic emission point is selected and an ultrasonic wave with fixed specification parameters is emitted, and the second detection unit receives echo data at the corresponding ultrasonic emission point, and calculates the crack depth value of the small crack at the ultrasonic emission point by analyzing the echo flight time difference and echo intensity characteristics in the echo data; A plurality of ultrasonic emission points are selected at different positions of the small crack, the echo data are received and calculated point by point, and the crack depth value corresponding to each ultrasonic emission point is integrated into the crack depth information of the small crack.

7. A bridge defect detection and early warning system according to claim 6, characterized in that: The second size threshold value set in the early warning unit specifically includes a second length threshold value and a second width threshold value. If the maximum length value in the crack length information is greater than the second length threshold value, and the maximum width value in the crack width information is greater than the second width threshold value, the corresponding large crack is assigned the first priority, and the position information of the large crack is recorded; If the maximum length value in the crack length information is less than the second length threshold, or the maximum width value in the crack width information is less than the second width threshold, the corresponding large crack is assigned the third priority and the position information of the large crack is recorded.

8. A bridge defect detection and early warning system according to claim 6, characterized in that: After the early warning unit receives the crack depth information of each small crack outputted from the second detection unit, the early warning unit compares the information with the first depth threshold. If at least one crack depth value in the crack depth information is greater than the first depth threshold, the corresponding small crack is assigned the second priority and the position information of the small crack is recorded. If all the crack depth values ​​in the crack depth information are smaller than the first depth threshold, the corresponding small cracks are assigned the fourth priority and the position information of the small cracks is recorded.

9. A defect detection and early warning device for a bridge, characterized in that: include: Moving parts (10); A supporting component (20) is located on the moving component (10), and the supporting component (20) can move in a direction perpendicular to the bridge pavement; A visual detection device (30), an ultrasonic detection device (40), an alarm device (70) and a memory are located on the supporting component (20), and the visual detection device (30), the ultrasonic detection device (40), the alarm device (70) and the memory are configured to be able to jointly perform the functions of the defect detection and early warning system described in any one of claims 1 to 8; wherein the ultrasonic detection device (40) also includes a spraying device (60), and the spraying device (60) is used to spray an oil-based coupling agent onto the bridge pavement before performing ultrasonic detection on cracks on the bridge pavement, so as to enhance the coupling effect of the ultrasonic signal when it propagates on the bridge pavement.

Images