A construction bridge pavement crack detection system

Abstract

The present invention relates to the field of bridge crack detection, and is used to solve the problem that bridge pavement is easily disturbed during detection and there is a lack of effective regularity and time axis analysis of bridge cracks. The present invention is specifically a construction bridge pavement crack detection system. In the present invention, when detecting bridge cracks, the cracks are superimposed and analyzed by combining visual image detection and ultrasonic image detection, and secondary confirmation of the cracks is achieved based on the superimposed analysis results to avoid missed detection and wrong detection during crack detection of the bridge pavement. After obtaining crack data, interval analysis is performed based on the distribution positions of the bridge cracks to obtain the distribution dispersion of the bridge pavement cracks, and the crack concentration area is judged based on the dispersion of the crack distribution. At the same time, the crack data is automatically numbered and stored, and the cracks are analyzed based on the time axis according to the results of the automatic numbering and storage, so as to determine the growth and expansion of the cracks and achieve dynamic analysis of crack changes.

Description

Technical Field

[0001] The invention relates to the field of bridge crack detection, in particular to a construction bridge pavement crack detection system. Background Art

[0002] Cracks are one of the common bridge diseases. According to the causes, they can be divided into seven categories: cracks caused by load, cracks caused by temperature changes, cracks caused by shrinkage, cracks caused by foundation deformation, cracks caused by steel corrosion, cracks caused by frost heave and cracks caused by construction.

[0003] Bridge cracks have a unique linear topological structure with irregular distribution, which will cause changes in the bridge structure, accelerate the aging of components, reduce the bearing capacity, and thus shorten the service life of the bridge, while increasing the danger during use;

[0004] At present, the bridge crack detection system in the existing technology still has shortcomings. The traditional bridge crack detection system generally uses manual naked eye detection for surface cracks, which has low detection efficiency and is easy to miss small cracks. Although the ultrasonic crack detection method can monitor internal cracks and small cracks, it is easy to cause noise in the monitoring results due to environmental factors, thereby affecting the detection results. At the same time, there is a lack of effective regularization and time axis analysis for bridge cracks, which affects the detection effect of bridge cracks.

[0005] In view of the above technical problems, this application proposes a solution. Summary of the invention

[0006] In the present invention, when detecting cracks in a bridge, a superposition analysis is performed on the cracks by combining visual image detection and ultrasonic image detection, and a secondary confirmation of the cracks is achieved based on the superposition analysis results, so as to avoid missed detection and wrong detection when detecting cracks in the bridge pavement. After obtaining the crack data, an interval analysis is performed based on the distribution position of the bridge cracks to obtain the distribution dispersion of the cracks in the bridge pavement, and the crack concentration area is judged based on the dispersion of the crack distribution. At the same time, the crack data detected multiple times are automatically numbered and stored, and the cracks are analyzed on the time axis based on the results of the automatic numbering storage, so as to determine the growth and expansion of the cracks, realize dynamic analysis of crack changes, improve the supervision level of crack growth risks, solve the problem that the bridge pavement is easily disturbed when detecting and lacks effective regularization and time axis analysis of the cracks in the bridge, and propose a construction bridge pavement crack detection system.

[0007] The object of the present invention can be achieved by the following technical solutions: A construction bridge pavement crack detection system, comprising an image acquisition unit, the image acquisition unit is used to acquire images of pavement cracks, and analyze the acquired images, and obtain an analysis image and a spare image according to the analysis results;

[0008] An ultrasonic acquisition unit, wherein the ultrasonic acquisition unit performs ultrasonic detection on the bridge pavement and generates an ultrasonic crack image according to the detection result;

[0009] A data fusion unit, wherein the data fusion unit marks cracks in the ultrasonic image and the analysis image, compares and analyzes the cracks at the marked locations, and generates a crack confirmation signal and a suspected crack signal according to the analysis result;

[0010] A regional analysis unit, wherein the regional analysis unit obtains crack information through the data fusion unit, performs position spacing analysis on the crack information, and determines the occurrence position of the crack concentration area according to the analysis result;

[0011] A crack monitoring unit monitors cracks by number, performs historical data analysis on each numbered crack, obtains the development and change of the cracks, and generates corresponding early warning signals according to the development and change of the cracks.

[0012] As a preferred embodiment of the present invention, the image acquisition unit uses each frame of the acquired picture as the original image, judges the image clarity through a gradient function, uses images with clarity that meet the requirements as valid images, and uses images with clarity that do not meet the requirements as invalid images. The image acquisition unit deletes the invalid images, retains the valid images, and performs overlap comparison on the valid images. One of the images with similarity greater than a preset requirement in the overlap comparison is retained and recorded as an analysis image, and the other images are used as backup images. The image acquisition unit sends the analysis image and the backup image to the data fusion unit at the same time.

[0013] As a preferred embodiment of the present invention, the ultrasonic acquisition unit transmits ultrasonic waves to the road surface through an ultrasonic transmitter, and obtains returned ultrasonic waves through an ultrasonic receiver, collects the depth and size of the cracks based on the returned ultrasonic waves, obtains an ultrasonic crack image, and sends the ultrasonic crack image to the data fusion unit.

[0014] As a preferred embodiment of the present invention, after the data fusion unit acquires the analysis image and the ultrasonic crack image, the cracks in the analysis image are marked, and the cracks existing in the image are marked. The data fusion analysis unit extracts the size and position of the cracks and records them as image cracks;

[0015] The data fusion unit extracts cracks in the ultrasonic crack image, obtains the size and position of the cracks in the ultrasonic crack image, and records the cracks as surface cracks when the cracks in the ultrasonic crack image exist on the road surface, and records the cracks as deep cracks when the cracks in the ultrasonic crack image exist in the deep layer of the road surface;

[0016] The data fusion unit compares the surface cracks and image cracks at the same position, and after the overlap, records the pixel points where the surface cracks and image cracks overlap as overlap points, and records the pixel points where no overlap occurs as difference points. The data fusion unit calculates the proportion of overlap points in all pixel points and records it as overlap degree. If the overlap degree is greater than a preset ratio, a crack confirmation signal is generated, and if the overlap degree is less than the preset ratio, a crack abnormality signal is generated.

[0017] After generating the abnormal crack signal, the data fusion unit extracts the corresponding crack position, searches for an image containing the crack position in the backup image, extracts the size and position of the crack from the backup image, obtains the backup image crack, and compares the backup image crack with the surface crack again. If the overlap is greater than or equal to a preset ratio, a crack confirmation signal is generated. If the overlap is less than the preset ratio, a suspicious crack signal is generated.

[0018] As a preferred embodiment of the present invention, the data fusion unit records the cracks that generate crack confirmation signals and deep cracks as verified cracks, and sends the verified cracks to the regional analysis unit. The regional analysis unit obtains the position of the verified cracks, and records the shortest distance between the two verified cracks as the crack spacing. If the crack spacing between the two verified cracks is less than a preset spacing threshold, the two verified cracks are recorded as interference cracks. The regional analysis unit then takes the two interference cracks as starting points, selects a verified crack and calculates the minimum distance between the interference crack and the starting point. If the minimum distance is less than the preset spacing threshold, the selected verified crack is also recorded as an interference crack. The data fusion unit records all the interference cracks as starting points again, selects another verified crack for analysis again, and if the other verified crack is not recorded as an interference crack, replaces the verified crack for continued analysis. If the other verified crack is also recorded as an interference crack, the number of interference cracks as the starting point is updated again, and other verified cracks are selected for analysis again until all verified cracks are calculated.

[0019] The regional analysis unit calculates the proportion of interference cracks in the verified cracks. If the proportion of interference cracks in the verified cracks is greater than or equal to a preset value, the location of the interference cracks is recorded as a crack concentration area. If the proportion of interference cracks in the verified cracks is less than the preset value, no response is made.

[0020] As a preferred embodiment of the present invention, the crack monitoring unit assigns a number i, i=1, 2, 3..., n to the verified crack each time the data fusion unit obtains the verified crack, stores the size and position of the verified crack, and assigns the same number i when subsequently obtaining cracks at the same position. The size of the verified crack includes the length of the center line of the crack and the maximum width of the crack. The crack monitoring unit compares cracks with the same number in multiple crack detection results, and calculates the difference in the center line length of the crack in the two comparisons. If the difference in the center line length is greater than a preset value, a crack extension signal is generated, and the difference in the maximum width of the crack in the two comparisons is calculated. If the difference in the maximum width of the crack is greater than a preset value, a crack expansion signal is generated. After generating either the crack expansion signal or the crack extension signal, the crack monitoring unit records the crack as a developing crack.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] 1. In the present invention, when detecting cracks in a bridge, a superimposed analysis of the cracks is performed by combining visual image detection and ultrasonic image detection, and a secondary confirmation of the cracks is achieved based on the superimposed analysis results, thereby improving the accuracy of crack detection and avoiding missed detection and wrong detection when detecting cracks in the bridge pavement.

[0023] 2. In the present invention, when detecting cracks in a bridge, interval analysis is performed based on the distribution positions of the cracks in the bridge to obtain the distribution dispersion of cracks in the bridge pavement. Based on the dispersion of the crack distribution, the crack concentration area is judged, thereby increasing the functionality of the crack detection system when in use.

[0024] 3. In the present invention, the crack data detected multiple times are automatically numbered and stored, and the time axis analysis of the cracks is carried out according to the results of the automatic numbering and storage, so as to determine the growth and expansion of the cracks, realize the dynamic analysis of the crack changes, and improve the supervision level of the crack growth risk. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0026] Figure 1 is a system block diagram of the present invention;

[0027] Figure 2 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0028] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] Example 1: Please refer to Figure 1 - Figure 2 As shown, a crack detection system for a building bridge pavement includes an image acquisition unit, an ultrasonic acquisition unit, a data fusion unit, a regional analysis unit and a crack supervision unit, wherein the image acquisition unit acquires images of the bridge pavement, acquires images through a short shutter, and uses each acquired frame of the image as an original image, the image acquisition unit regards the image as a two-dimensional discrete matrix, judges the image clarity through the Laplacian gradient function, regards the image clarity that meets the requirements as a valid image, and regards the image clarity that does not meet the requirements as an invalid image, the image acquisition unit deletes the invalid image, retains the valid image, and performs overlap comparison on the valid images, selects one of the images whose similarity in the overlap comparison is greater than the preset requirement and retains it, records it as an analysis image, and uses the other images as spare images, and the image acquisition unit sends the analysis image and the spare image to the data fusion unit at the same time;

[0030] The ultrasonic acquisition unit transmits ultrasonic waves to the road surface through an ultrasonic transmitter, and obtains the returned ultrasonic waves through an ultrasonic receiver, collects the depth and size of the cracks based on the returned ultrasonic waves, obtains an ultrasonic crack image, and sends the ultrasonic crack image to the data fusion unit;

[0031] After the data fusion unit obtains the analysis image and the ultrasonic crack image, it marks the cracks in the analysis image through the deep learning image segmentation algorithm, marks the cracks existing in the image, and the data fusion analysis unit extracts the size and position of the cracks and records them as image cracks;

[0032] The data fusion unit extracts cracks in the ultrasonic crack image, obtains the size and position of the cracks in the ultrasonic crack image, and records the cracks as surface cracks when the cracks in the ultrasonic crack image exist on the road surface, and records the cracks as deep cracks when the cracks in the ultrasonic crack image exist in the deep layer of the road surface;

[0033] The data fusion unit compares the surface cracks and image cracks at the same position, and after the overlap, the pixel points where the surface cracks and image cracks overlap are recorded as overlap points, and the pixel points where no overlap occurs are recorded as difference points. The data fusion unit calculates the proportion of overlap points in all pixel points and records it as the overlap degree. If the overlap degree is greater than a preset ratio, a crack confirmation signal is generated, and if the overlap degree is less than the preset ratio, a crack abnormality signal is generated.

[0034] After generating the abnormal crack signal, the data fusion unit extracts the corresponding crack position, searches for the image containing the crack position in the backup image, extracts the size and position of the crack in the backup image, obtains the backup image crack, and compares the backup image crack with the surface crack again. If the overlap is greater than or equal to the preset ratio, a crack confirmation signal is generated. If the overlap is less than the preset ratio, a suspicious crack signal is generated.

[0035] Example 2: Please refer to Figure 1 - Figure 2 As shown, the data fusion unit records the cracks that generate crack confirmation signals and deep cracks as verified cracks, and sends the verified cracks to the regional analysis unit. The regional analysis unit obtains the position of the verified cracks, and records the shortest distance between the two verified cracks as the crack spacing. If the crack spacing between the two verified cracks is less than the preset spacing threshold, the two verified cracks are recorded as interference cracks. The regional analysis unit then takes the two interference cracks as starting points, selects a verified crack and calculates the minimum distance between the interference crack and the starting point. If the minimum distance is less than the preset spacing threshold, the selected verified crack is also recorded as an interference crack. The data fusion unit records all the interference cracks as starting points again, selects another verified crack for analysis again, and if the other verified crack is not recorded as an interference crack, replaces the verified crack for continued analysis. If the other verified crack is also recorded as an interference crack, the number of interference cracks as the starting point is updated again, and other verified cracks are selected for analysis again until all verified cracks are calculated.

[0036] The regional analysis unit calculates the proportion of interference cracks in the verified cracks. If the proportion of interference cracks in the verified cracks is greater than or equal to a preset value, the location of the interference cracks is recorded as a crack concentration area. If the proportion of interference cracks in the verified cracks is less than the preset value, no response is made.

[0037] Example 3: Please refer to Figure 1 - Figure 2As shown, the crack monitoring unit assigns a number i, i=1, 2, 3..., n to the verified crack each time the crack is obtained through the data fusion unit, and stores the size and position of the verified crack. When a crack at the same position is obtained subsequently, the same number i is assigned. The size of the verified crack includes the length of the center line of the crack and the maximum width of the crack. The crack monitoring unit compares cracks with the same number in multiple crack detection results, and calculates the difference in the center line length of the crack in the two comparisons. If the difference in the center line length is greater than a preset value, a crack extension signal is generated, and the difference in the maximum width of the crack in the two comparisons is calculated. If the difference in the maximum width of the crack is greater than a preset value, a crack extension signal is generated. After generating either the crack extension signal or the crack extension signal, the crack monitoring unit records the crack as a developing crack and sends the developing crack to the management personnel through the network.

[0038] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A construction bridge pavement crack detection system, characterized in that: It includes an image acquisition unit, which is used to acquire images of road cracks, analyze the acquired images, and obtain an analysis image and a backup image according to the analysis results; An ultrasonic acquisition unit, which performs ultrasonic detection on the bridge pavement by ultrasound and generates an ultrasonic crack image according to the detection result; A data fusion unit, wherein the data fusion unit marks cracks in the ultrasonic image and the analysis image, compares and analyzes the cracks at the marked locations, and generates a crack confirmation signal and a suspected crack signal according to the analysis result; A regional analysis unit, wherein the regional analysis unit obtains crack information through the data fusion unit, performs position spacing analysis on the crack information, and determines the occurrence position of the crack concentration area according to the analysis result; A crack monitoring unit, which monitors cracks by number, analyzes historical data of each crack with a number, obtains the development and change of the cracks, and generates corresponding early warning signals according to the development and change of the cracks; After the data fusion unit acquires the analysis image and the ultrasonic crack image, the cracks in the analysis image are marked, and the cracks existing in the image are marked. The data fusion unit extracts the size and position of the cracks and records them as image cracks; The data fusion unit extracts cracks in the ultrasonic crack image, obtains the size and position of the cracks in the ultrasonic crack image, and records the cracks as surface cracks when the cracks in the ultrasonic crack image exist on the road surface, and records the cracks as deep cracks when the cracks in the ultrasonic crack image exist in the deep layer of the road surface; The data fusion unit compares the surface cracks and image cracks at the same position, and after the overlap, records the pixel points where the surface cracks and image cracks overlap as overlap points, and records the pixel points where no overlap occurs as difference points. The data fusion unit calculates the proportion of overlap points in all pixel points and records it as overlap degree. If the overlap degree is greater than a preset ratio, a crack confirmation signal is generated, and if the overlap degree is less than the preset ratio, a crack abnormality signal is generated. After generating the abnormal crack signal, the data fusion unit extracts the corresponding crack position, searches for an image containing the crack position in the backup image, extracts the size and position of the crack from the backup image, obtains the backup image crack, and compares the backup image crack with the surface crack again. If the overlap is greater than or equal to a preset ratio, a crack confirmation signal is generated; if the overlap is less than the preset ratio, a suspicious crack signal is generated. The data fusion unit records the cracks that generate crack confirmation signals and deep cracks as verified cracks, and sends the verified cracks to the regional analysis unit. The regional analysis unit obtains the position of the verified cracks, and records the shortest distance between the two verified cracks as the crack spacing. If the crack spacing between the two verified cracks is less than a preset spacing threshold, the two verified cracks are recorded as interference cracks. The regional analysis unit then uses the two interference cracks as starting points, selects a verified crack and calculates the minimum distance between the interference crack and the starting point. If the minimum distance is less than the preset spacing threshold, the selected verified crack is also recorded as an interference crack. The data fusion unit records all the interference cracks as starting points again, selects another verified crack for analysis again, and if the other verified crack is not recorded as an interference crack, replaces the verified crack for continued analysis. If the other verified crack is also recorded as an interference crack, the number of interference cracks as the starting point is updated again, and other verified cracks are selected for analysis again until all verified cracks are calculated. The regional analysis unit calculates the proportion of interference cracks in the verified cracks, and if the proportion of interference cracks in the verified cracks is greater than or equal to a preset value, the location of the interference cracks is recorded as a crack concentration area; if the proportion of interference cracks in the verified cracks is less than the preset value, no response is made; Each time the crack monitoring unit obtains a verified crack through the data fusion unit, it assigns a number i to the verified crack, i=1, 2, 3..., n, stores the size and position of the verified crack, and assigns the same number i when subsequently obtaining cracks at the same position. The size of the verified crack includes the length of the center line of the crack and the maximum width of the crack. The crack monitoring unit compares cracks with the same number in multiple crack detection results, and calculates the difference in the center line length of the crack in the two comparisons. If the difference in the center line length is greater than a preset value, a crack extension signal is generated. The difference in the maximum width of the crack in the two comparisons is calculated. If the difference in the maximum width of the crack is greater than a preset value, a crack expansion signal is generated. After generating either the crack expansion signal or the crack extension signal, the crack monitoring unit records the crack as a developing crack.

2. A construction bridge pavement crack detection system according to claim 1, characterized in that: The image acquisition unit uses each frame of the acquired picture as the original image, judges the image clarity through the gradient function, uses the image clarity that meets the requirements as a valid image, and uses the image clarity that does not meet the requirements as an invalid image. The image acquisition unit deletes the invalid image, retains the valid image, and performs overlap comparison on the valid images. One of the images with similarity greater than the preset requirement in the overlap comparison is retained and recorded as the analysis image, and the other images are used as spare images. The image acquisition unit sends the analysis image and the spare image to the data fusion unit at the same time.

3. A construction bridge pavement crack detection system according to claim 1, characterized in that: The ultrasonic acquisition unit transmits ultrasonic waves to the road surface through an ultrasonic transmitter, and obtains returned ultrasonic waves through an ultrasonic receiver, collects the depth and size of the cracks based on the returned ultrasonic waves, obtains an ultrasonic crack image, and sends the ultrasonic crack image to the data fusion unit.

Images