A concrete crack detection method for bridge engineering

Abstract

The present invention relates to the field of image processing technology, and more specifically to a concrete crack detection method for bridge engineering, comprising: obtaining a crack region and a principal tensile stress direction; segmenting the crack region into a plurality of crack subregions; segmenting the crack subregion into a plurality of local crack subregions; obtaining the degree of harmfulness of the crack subregion based on the extension direction and principal tensile stress direction of the local crack subregion; obtaining the degree of splicing between adjacent crack subregions based on the extension direction and harmfulness of adjacent crack subregions, thereby obtaining a plurality of harmful cracks; and assessing the quality of the bridge based on the distribution, harmfulness, and area of all harmful cracks in the bridge. The present invention analyzes the causes of harmful cracks in the bridge, extracts the characteristics of the harmful cracks, obtains the harmful cracks based on the characteristics of the harmful cracks, and further improves the accuracy of bridge quality assessment based on the distribution and size of the harmful cracks.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a concrete crack detection method for bridge engineering. Background Art

[0002] In order to ensure the safety of the overall beam structure and extend the service life of the bridge, it is necessary to detect cracks in the concrete in the bridge to evaluate the quality of the bridge. When the poured concrete has not yet solidified, as the moisture in the concrete gradually evaporates, small cracks will inevitably form on the concrete surface as the surface shrinks. However, such cracks are a normal phenomenon when pouring concrete and will not endanger the structural safety of the bridge. Traditional concrete crack detection methods will identify all cracks, but harmless cracks will always appear during the concrete solidification process. Harmless cracks do not harm the structure of the bridge. Therefore, the traditional method of only evaluating the size of cracks on the bridge surface cannot accurately assess the quality of the bridge. Summary of the Invention

[0003] The present invention provides a concrete crack detection method for bridge engineering to solve the existing problem that the traditional method cannot accurately evaluate the quality of a bridge by only measuring the size of cracks on the bridge surface.

[0004] The present invention provides a concrete crack detection method for bridge engineering using the following technical solutions:

[0005] The following steps are involved:

[0006] Obtain several crack areas in the bridge and the direction of the principal tensile stress at each location;

[0007] Dividing the crack region into a number of crack sub-regions; dividing the crack sub-region into a number of local crack sub-regions; obtaining a harmfulness factor of the local crack sub-region based on the extension direction of the local crack sub-region and the direction of the principal tensile stress at the corresponding position, wherein the harmfulness factor represents the perpendicularity between the extension direction and the direction of the principal tensile stress; obtaining the degree of harmfulness of the crack sub-region based on the difference between the harmfulness factors of each local crack sub-region and the adjacent local crack sub-region in the crack sub-region;

[0008] Determining the extension consistency of adjacent crack sub-regions based on the difference in the inclination angles of adjacent crack sub-regions; determining the parameter consistency of adjacent crack sub-regions based on the relative deviation of the harmfulness of adjacent crack sub-regions; determining the degree of splicing between adjacent crack sub-regions based on the extension consistency and the parameter consistency, wherein both the extension consistency and the parameter consistency are positively correlated with the splicing degree; and obtaining a number of harmful cracks based on the splicing degree between adjacent crack sub-regions.

[0009] The harmfulness of the harmful cracks is determined based on the harmfulness of all crack sub-areas among all harmful cracks. The degree of harm of the harmful cracks to the bridge is obtained based on the distance between the harmful crack and all other harmful cracks, combined with the harmfulness of all harmful cracks, where the harmfulness is positively correlated with the degree of harm. The bridge quality is evaluated based on the harmfulness of all harmful cracks and the area of all harmful cracks.

[0010] Preferably, the method of dividing the crack region into a plurality of crack sub-regions includes:

[0011] For any crack region, the Guo-Hall algorithm is used to obtain the skeleton of the crack region; the pixels on the skeleton are recorded as skeleton pixels, and for any skeleton pixel, if there are three or more skeleton pixels in its eight-neighborhood, the skeleton pixel is used as a segmentation point; the skeleton is divided into several segments using all the segmentation points in the skeleton to obtain several skeleton segments;

[0012] For any pixel point in the crack area, obtain the distance between the pixel point and each skeleton segment, and use the skeleton segment closest to the pixel point as the skeleton segment corresponding to the pixel point;

[0013] The skeleton segment corresponding to each pixel point in the crack region is obtained, and several pixel points corresponding to the same skeleton segment are classified into the same crack sub-region to obtain several crack sub-regions.

[0014] Preferably, the method of dividing the crack sub-region into several local crack sub-regions includes:

[0015] For any crack sub-region, the skeleton segment of the crack sub-region is divided into several segments of length The sub-skeleton segment, is the preset sub-skeleton segment length;

[0016] For any pixel point in the crack sub-region, obtain the distance between the pixel point and each skeleton sub-segment, and use the skeleton sub-segment closest to the pixel point as the skeleton sub-segment corresponding to the pixel point;

[0017] A skeleton subsegment corresponding to each pixel point in the crack sub-region is obtained, and several pixel points corresponding to the same skeleton subsegment are classified into the same local crack sub-region to obtain several local crack sub-regions.

[0018] Preferably, the method of obtaining the harmful factor of the local crack sub-region according to the extension direction of the local crack sub-region and the direction of the principal tensile stress at the corresponding position includes:

[0019] For any local crack sub-region, a linear fitting is performed based on the coordinate positions of all pixel points in the local crack sub-region using the least squares method to obtain a fitting straight line for the local crack sub-region, and the direction of the fitting straight line for the local crack sub-region is used as the extension direction of the local crack sub-region;

[0020] The harmfulness factor of the local crack sub-region is obtained according to the extension direction of the local crack sub-region and the principal tensile stress direction at each pixel point position in the local crack sub-region.

[0021] Preferably, the specific calculation formula for obtaining the harmfulness factor of the local crack sub-region is:

[0022]

[0023] Where, A harmful factor representing the local crack sub-region; represents the number of pixels in the local crack sub-region; An extension direction of the local crack sub-region; Indicates the first The principal tensile stress direction at each pixel position; represents the sine trigonometric function; Represents the absolute value function.

[0024] Preferably, the method of obtaining the harmfulness of a crack sub-region according to the difference between the harmfulness factors of each local crack sub-region and the adjacent local crack sub-region in the crack sub-region includes the following specific methods:

[0025] For any crack sub-region, the harmfulness of the crack sub-region is obtained based on the harmfulness factor of each local crack sub-region in the crack sub-region and the difference in harmfulness factors between adjacent local crack sub-regions in the crack sub-region. The specific calculation formula is:

[0026]

[0027] Where, Indicates the harmfulness of the crack sub-area; represents the number of local crack sub-regions in the crack sub-region; Indicates the first Harmful factors of local crack sub-regions; Indicates the crack sub-area with the The number of local crack sub-regions adjacent to a local crack sub-region; Indicates the crack sub-area with the The adjacent local crack sub-region Harmful factors of local crack sub-regions; Indicates the absolute value function; Represents an exponential function with a natural constant as its base.

[0028] Preferably, the combining of the extension consistency and the parameter consistency to determine the degree of integration between adjacent crack sub-regions includes the following specific methods:

[0029] For any crack sub-region, the crack sub-region is recorded as a reference region, and the crack sub-region adjacent to the reference region is recorded as a target region;

[0030] The calculation formula for the degree of integration between the reference area and the target area is:

[0031]

[0032] Where, Indicates the degree of fit between the reference area and the target area; represents the slope of the fitted straight line of the target area; represents the slope of the fitted line in the reference area; Indicates the harmfulness of the target area; Indicates the degree of harmfulness of the reference area; represents the inverse tangent function; Represents the preset hyperparameters; represents the linear normalization function; The extension consistency of the reference area and the target area; is the tilt angle of the target area; is the tilt angle of the reference area; The consistency of parameters between the reference area and the target area;

[0033] Among them, the method for obtaining the slope of the fitting line of each crack sub-region is as follows: for each crack sub-region, according to the position coordinates of each pixel point in the crack sub-region, the least squares method is used to fit the crack sub-region separately to obtain the fitting line of the crack sub-region; the slope of the fitting line of the crack sub-region is used as the slope of the fitting line of the crack sub-region.

[0034] Preferably, the method of obtaining a number of harmful cracks according to the degree of splicing between adjacent crack sub-regions includes:

[0035] Preset a stitching threshold If the degree of fit between the reference area and the target area is greater than or equal to , classifying the benchmark area and the target area as the same harmful crack;

[0036] The degree of splicing of all crack sub-regions and their adjacent crack sub-regions is judged to obtain several harmful cracks.

[0037] Preferably, the calculation formula for the degree of harm caused by the harmful cracks to the bridge is:

[0038]

[0039] Where, Indicates the The degree of damage of each harmful crack; Indicates the The harmfulness of each harmful crack; Indicates the The harmfulness of each harmful crack; Indicates the number of harmful cracks; Indicates the The harmful cracks and The distance between harmful cracks; Represents the sigmoid function.

[0040] Preferably, the bridge quality is assessed based on the degree of damage of all harmful cracks and the area of all harmful cracks, including the following specific methods:

[0041] The bridge quality is obtained based on the degree of damage of all harmful cracks and the area of all harmful cracks. The specific calculation formula is:

[0042]

[0043] Where, Indicates the quality of the bridge; Indicates the number of harmful cracks; Indicates the The degree of damage of each harmful crack; Indicates the The area of harmful cracks; Represents an exponential function with a natural constant as its base.

[0044] The beneficial effects of the technical solution of the present invention are as follows: the present application divides the crack area into several crack sub-areas; divides the crack sub-areas into several local crack sub-areas. Since the crack area in the bridge may be composed of the intersection of multiple cracks, and when the crack area is composed of the intersection of harmful cracks and harmless cracks, the harmful cracks and harmless cracks in the crack area will interfere with each other, resulting in the inability to accurately quantify the degree of harm of the crack area to the bridge, it is necessary to divide the crack area into several crack sub-areas; and since harmful cracks are caused by the concrete at the crack being subjected to stress exceeding its load upper limit, the harmful cracks are always distributed in the direction perpendicular to its principal tensile stress. Therefore, the harmful factor of the local crack sub-area is obtained according to the extension direction of the local crack sub-area and the degree of perpendicularity to the principal tensile stress direction at the corresponding position; the harmful degree of the crack sub-area is obtained according to the harmful factor of each local crack sub-area in the crack sub-area.

[0045] Because when assessing the degree of damage to a bridge caused by cracks, it is not possible to split a complete crack into several crack segments and assess the degree of damage to the bridge based on the several crack segments, the degree of splicing between adjacent crack sub-regions is obtained based on the differences in the extension direction of adjacent crack sub-regions and the degree of damage of each adjacent crack sub-region; based on the degree of splicing between adjacent crack sub-regions, several harmful cracks are obtained; based on the degree of damage of harmful cracks, the degree of damage of harmful cracks is obtained based on the distribution position of all harmful cracks in the bridge; based on the degree of damage of all harmful cracks, the area of all harmful cracks is combined to assess the quality of the bridge. This application analyzes the causes of harmful cracks in the bridge, extracts the characteristics of harmful cracks, obtains harmful cracks based on the characteristics of harmful cracks, and further improves the accuracy of bridge quality assessment based on the distribution of harmful cracks and the size of harmful cracks. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0047] Figure 1 This is a flow chart of the steps of a concrete crack detection method for bridge engineering according to the present invention;

[0048] Figure 2 This is the grayscale legend of the bridge;

[0049] Figure 3 This is an illustration of the edge of a bridge crack. DETAILED DESCRIPTION

[0050] To further illustrate the technical means and effectiveness of the present invention in achieving its intended objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effectiveness of a concrete crack detection method for bridge engineering proposed by the present invention. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0051] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0052] The specific scheme of the concrete crack detection method for bridge engineering provided by the present invention is described in detail below with reference to the accompanying drawings.

[0053] See also Figure 1 , which shows a flowchart of a concrete crack detection method for bridge engineering provided by one embodiment of the present invention, the method comprising the following steps:

[0054] Step S001: Obtain several crack areas in the bridge and the principal tensile stress directions at various locations.

[0055] It should be noted that to ensure the safety of the overall bridge structure and extend the service life of the bridge, it is necessary to evaluate the quality of the bridge. Cracks in the bridge can seriously affect the quality of the bridge. To accurately evaluate the quality of the bridge, it is necessary to obtain the cracks in the bridge. However, traditional methods for evaluating bridge quality only evaluate based on the size of the cracks in the bridge, without considering the different degrees of damage to the bridge caused by cracks in different bridges. In other words, traditional methods based on bridge cracks cannot accurately evaluate bridge quality. Therefore, this embodiment proposes a concrete crack detection method for bridge engineering. Specifically, by analyzing the morphological characteristics of each crack in the bridge, the degree of damage to the bridge caused by each crack is evaluated, thereby improving the accuracy of bridge quality assessment. To this end, it is first necessary to capture an image of the bridge surface and obtain the crack area therein. Since cracks that seriously endanger bridge safety are caused by concrete being subjected to stress exceeding its upper load limit, they are always distributed perpendicular to the principal tensile stress. To assess the degree of damage to the bridge caused by cracks, it is necessary to further obtain the principal tensile stress direction at each location in the bridge.

[0056] Specifically, the bridge surface image is captured by a high-definition camera, and the bridge surface image is grayscaled and denoised by Gaussian filtering to obtain a bridge grayscale image, such as Figure 2 As shown, Figure 2 For the bridge grayscale image, we further use the edge detection algorithm to obtain the edges in the bridge grayscale image, such as Figure 3 As shown, Figure 3 This is a legend image of the bridge crack edge, and the edge of the bridge grayscale image and the area surrounded by the edge are regarded as the crack area;

[0057] Furthermore, based on the structure of the bridge, the finite element analysis method (Finite Element Analysis) is used to obtain the direction of the principal tensile stress at various positions in the bridge.

[0058] It should be noted that grayscale conversion, Gaussian filtering, finite element analysis, and edge detection algorithms are all well-known existing technologies and are not described in detail in this embodiment. This embodiment does not impose strict requirements on the edge detection algorithm used to obtain edges in the bridge grayscale image. In this embodiment, the Sobel edge detection algorithm is used to obtain edges in the bridge grayscale image. The edges in the bridge grayscale image represent crack edges in the bridge.

[0059] At this point, the crack area in the bridge and the direction of the principal tensile stress at each position are obtained.

[0060] Step S002: Divide the crack area into several crack sub-areas; divide the crack sub-areas into several local crack sub-areas; obtain the harmfulness factor of the local crack sub-area according to the extension direction of the local crack sub-area and the principal tensile stress direction at the corresponding position, wherein the harmfulness factor represents the perpendicularity between the extension direction and the principal tensile stress direction; obtain the harmfulness degree of the crack sub-area according to the difference between the harmfulness factors of each local crack sub-area and the adjacent local crack sub-area in the crack sub-area.

[0061] It should be noted that cracks in bridges can be divided into harmful cracks and harmless cracks. Harmful cracks are caused by the concrete at the cracks being subjected to stress exceeding its load limit, which usually seriously endangers the safety of the bridge structure; while harmless cracks are caused by the evaporation of water and volume shrinkage of the concrete during solidification, which usually do not endanger the safety of the bridge structure; but the traditional evaluation of bridge quality through cracks in bridges is only based on the size of the cracks in the bridge, without considering the degree of harm of different cracks to the bridge. Therefore, the traditional evaluation of bridge quality through cracks in bridges cannot accurately evaluate the quality of bridges; therefore, this embodiment proposes a concrete crack detection method for bridge engineering, specifically by analyzing the morphological characteristics of each crack to obtain the degree of harm of each crack to the bridge, so as to accurately evaluate the quality of the bridge.

[0062] It should be further explained that since the crack area in the bridge may be composed of the intersection of multiple cracks, and when the crack area is composed of the intersection of harmful cracks and harmless cracks, the harmful cracks and harmless cracks in the crack area will interfere with each other, resulting in the inability to accurately quantify the degree of damage the crack area causes to the bridge. Therefore, it is necessary to divide the crack area into several crack sub-areas; and since harmful cracks are caused by the concrete at the crack being subjected to stress exceeding its load limit, harmful cracks are always distributed in the direction perpendicular to its principal tensile stress. Therefore, this can be used as a basis to obtain the harmfulness factor of the crack sub-area.

[0063] Preferably, in a specific embodiment of the present invention, for any crack region, the Guo-Hall algorithm is used to obtain the skeleton of the crack region. Since the Guo-Hall algorithm is a well-known prior art, it will not be described in detail in this embodiment; the pixel points on the skeleton are recorded as skeleton pixel points, and for any skeleton pixel point, if there are three or more skeleton pixel points in its eight-neighborhood area, the skeleton pixel point is used as a segmentation point; the skeleton is divided into a plurality of segments using all the segmentation points in the skeleton to obtain a plurality of skeleton segments;

[0064] Furthermore, for any pixel point in the crack area, the distance between the pixel point and each skeleton segment is obtained, and the skeleton segment closest to the pixel point is used as the skeleton segment corresponding to the pixel point;

[0065] The skeleton segment corresponding to each pixel point in the crack region is obtained, and several pixel points corresponding to the same skeleton segment are classified into the same crack sub-region to obtain several crack sub-regions.

[0066] It should be noted that the pixel points in the crack sub-area are pixel points in the same crack. The cracks in the bridge are not distributed vertically along the main tensile stress direction at their corresponding positions. Therefore, in order to better capture the harmful crack characteristics of the crack sub-area, the crack sub-area needs to be further segmented.

[0067] Preferably, in a specific embodiment of the present invention, for any crack sub-region, the skeleton segment of the crack sub-region is equally divided into several segments of length The sub-skeleton segment, is the preset sub-skeleton segment length, The specific value of can be set according to the actual situation. This embodiment does not make a hard requirement. Take this as an example (if the length of the last sub-skeleton segment in the crack sub-area is less than , the remaining skeleton segment is taken as the last sub-skeleton segment);

[0068] Furthermore, for any pixel point in the crack sub-region, the distance between the pixel point and each skeleton sub-segment is obtained, and the skeleton sub-segment closest to the pixel point is used as the skeleton sub-segment corresponding to the pixel point;

[0069] A skeleton subsegment corresponding to each pixel point in the crack sub-region is obtained, and several pixel points corresponding to the same skeleton subsegment are classified into the same local crack sub-region to obtain several local crack sub-regions.

[0070] It should be noted that the more the extension direction of the local crack sub-region is perpendicular to the principal tensile stress direction at its position, the more the local crack sub-region has the characteristics of a harmful crack. Therefore, the extension direction of the local crack sub-region is obtained first, and the harmful factor of the local crack sub-region is obtained according to the degree of perpendicularity between the extension direction of the local crack sub-region and the principal tensile stress direction of the local crack sub-region.

[0071] Preferably, in a specific embodiment of the present invention, for any local crack sub-region, a linear fitting is performed on it according to the coordinate positions of all pixel points in the local crack sub-region by the least squares method. Since the least squares method is a well-known prior art, it will not be described in detail in this embodiment. A fitting straight line of the local crack sub-region is obtained, and the direction of the fitting straight line of the local crack sub-region is used as the extension direction of the local crack sub-region, wherein the extension direction is obtained based on from top to bottom and from left to right.

[0072] Furthermore, according to the extension direction of the local crack sub-region and the principal tensile stress direction at each pixel position in the local crack sub-region, the harmfulness factor of the local crack sub-region is obtained, and the specific calculation formula is:

[0073]

[0074] Where, A harmful factor representing the local crack sub-region; represents the number of pixels in the local crack sub-region; An extension direction of the local crack sub-region; Indicates the first The principal tensile stress direction at each pixel position; represents the sine trigonometric function; Represents the absolute value function.

[0075] It should be noted that represents the angle between the extension direction of the local crack sub-region and the principal tensile stress direction at each pixel position of the local crack sub-region; therefore, when The larger the value of , the more likely it is that the extension direction of the local crack sub-region is perpendicular to the principal tensile stress direction at each pixel point in the local crack sub-region. In other words, the local crack sub-region is more likely to be a harmful crack, and is more likely to damage the structure of the bridge.

[0076] It should be further explained that the harmfulness factor of a local crack sub-region represents the degree of perpendicularity between the extension direction of the local crack sub-region and the direction of the principal tensile stress at its location. The more and more continuous the local crack sub-regions with high harmfulness factors are in the crack sub-region, the more likely it is that the crack sub-region has been subjected to a load exceeding its upper limit. Therefore, the degree of harmfulness of the crack sub-region can be obtained based on this.

[0077] Preferably, in a specific embodiment of the present invention, for any crack sub-region, the harmfulness of the crack sub-region is obtained according to the harmfulness factor of each local crack sub-region in the crack sub-region and the difference in harmfulness factors between adjacent local crack sub-regions in the crack sub-region. The specific calculation formula is:

[0078]

[0079] Where, Indicates the harmfulness of the crack sub-area; represents the number of local crack sub-regions in the crack sub-region; Indicates the first Harmful factors of local crack sub-regions; Indicates the crack sub-area with the The number of local crack sub-regions adjacent to a local crack sub-region; Indicates the crack sub-area with the The adjacent local crack sub-region Harmful factors of local crack sub-regions; Indicates the absolute value function; Represents an exponential function with a natural constant as the base. In this embodiment, Model to present inverse proportional relationship and normalization processing, As the input of the model, the implementer can set the inverse proportional function and normalization function according to the actual situation.

[0080] It should be noted that It represents the difference in harmful factors between adjacent local crack sub-regions. Since the harmful cracks are distributed along the direction of the principal tensile stress at their corresponding positions, the difference in harmful factors between adjacent local sub-regions in the crack sub-region is small. The larger the value of is, the more and more continuous the number of local crack sub-regions with high harmful factors is. The larger the value of is, the more the crack sub-region has the characteristics of a harmful crack and the greater its harmfulness is.

[0081] Thus, the harmfulness of the crack sub-region is obtained.

[0082] Step S003: Determine the extension consistency of adjacent crack sub-regions based on the difference in the inclination angles of adjacent crack sub-regions; determine the parameter consistency of adjacent crack sub-regions based on the relative deviation of the harmfulness of adjacent crack sub-regions; determine the degree of splicing between adjacent crack sub-regions based on the extension consistency and the parameter consistency, wherein the extension consistency and the parameter consistency are both positively correlated with the splicing degree; obtain a number of harmful cracks based on the splicing degree between adjacent crack sub-regions.

[0083] It should be noted that harmful cracks are the outward manifestation of damage to the bridge structure, and generally, the stress concentration at the tip of a long crack is more significant, making it easier for the tip of a long crack to approach the critical fracture size of the material. At the same time, long cracks are more likely to directly penetrate the key stress-bearing area of the bridge, significantly weakening the bearing capacity of the bridge structure and causing overall instability, while scattered short cracks have less impact on the bearing capacity of the remaining material. For example, the damage caused by a 10-centimeter-long harmful crack to a bridge is greater than the damage caused by two 5-centimeter harmful cracks. Therefore, in order to accurately assess the degree of damage caused by cracks to the bridge, it is not possible to split a complete crack into several crack segments to assess its degree of damage to the bridge. To accurately assess the quality of the bridge, it is necessary to calculate the degree of splicing of the several crack sub-regions decomposed in step S002 and splice them to obtain several complete harmful cracks; the complete harmful cracks as a whole can be used to accurately assess the quality of the bridge.

[0084] Preferably, in a specific embodiment of the present invention, for any crack sub-region, the crack sub-region is recorded as a reference region, and the crack sub-region adjacent to the reference region is recorded as a target region;

[0085] The calculation formula for the degree of integration between the reference area and the target area is:

[0086]

[0087] Where, Indicates the degree of fit between the reference area and the target area; represents the slope of the fitted straight line of the target area; represents the slope of the fitted line in the reference area; Indicates the harmfulness of the target area; Indicates the degree of harmfulness of the reference area; represents the inverse tangent function; Represents the preset hyperparameter, the purpose of which is to avoid the situation where the denominator is 0 during the fractional operation. The specific value of can be set according to the actual situation. This embodiment does not make a hard requirement. Take this as an example to describe; Represents a linear normalization function, whose specific normalization range is the reference area and all target areas ; The extension consistency of the reference area and the target area; is the tilt angle of the target area; is the tilt angle of the reference area; The parameter consistency between the reference area and the target area.

[0088] In a specific embodiment of the present invention, the extension consistency is determined based on the difference in the inclination angles of adjacent crack sub-regions. More specifically, the difference in the inclination angles of adjacent crack sub-regions is negatively correlated with the sum of preset hyperparameters, and the result value of the negative correlation mapping is the extension consistency of adjacent crack sub-regions.

[0089] In a specific embodiment of the present invention, parameter consistency is determined based on the relative deviation in the harmfulness of adjacent crack sub-regions. More specifically, the sum of the harmfulness of adjacent crack sub-regions is used as the numerator, and the absolute value of the difference in the harmfulness of adjacent crack sub-regions and the sum of a preset hyperparameter are used as the denominator. The ratio of the numerator and denominator is used as the parameter consistency of adjacent crack sub-regions. Here, the ratio of the sum to the difference in the harmfulness of adjacent crack sub-regions reflects the relative deviation in the harmfulness of adjacent crack sub-regions.

[0090] It should be noted that the degree of fit between the reference area and the target area indicates the possibility that the reference area and the target area belong to the same harmful crack. When the reference area and the target area belong to the same harmful crack, the reference area and the target area are similar in the extension direction. It represents the difference between the reference area and the target area in the extension direction. The smaller its value is, the more likely the reference area and the target area are to belong to the same crack. If the reference area and the target area belong to the same harmful crack, then the reason why the harmful crack is divided into the reference area and the target area is that there are other cracks at the intersection of the reference area and the target area, thereby dividing the harmful crack into the reference area and the target area. Therefore, if the reference area and the target area belong to the same harmful crack, then the reference area and the target area are similar in degree of harm, and the degree of harm between the reference area and the target area is greater. The larger the value of , the more likely it is that the reference area and the target area belong to the same harmful crack.

[0091] In a specific embodiment of the present invention, the method for obtaining the slope of the fitting straight line of each crack sub-region is as follows: for each crack sub-region, the least squares method is used to fit the crack sub-region according to the position coordinates of each pixel point in the crack sub-region to obtain the fitting straight line of the crack sub-region; the slope of the fitting straight line of the crack sub-region is used as the slope of the fitting straight line of the crack sub-region.

[0092] Preferably, in a specific embodiment of the present invention, a threshold value of the degree of integration is preset. , The specific value of can be set according to the actual situation. This embodiment does not make a hard requirement. Describe; if the degree of fit between the reference area and the target area is greater than or equal to , then the benchmark area and the target area are classified as the same harmful crack;

[0093] The degree of splicing of all crack sub-regions and their adjacent crack sub-regions is judged to obtain several harmful cracks.

[0094] It should be noted that when evaluating the degree of damage of the crack area to the bridge, a complete crack cannot be split into several crack segments to evaluate its degree of damage to the bridge; in order to accurately evaluate the degree of damage of the crack to the bridge, it is necessary to obtain the part composed of complete harmful cracks in the crack area; and the harmful cracks in the crack area are composed of several adjacent crack sub-areas; it represents the part composed of complete harmful cracks in the crack area, which prepares for the subsequent accurate evaluation of the degree of damage of the crack area to the bridge.

[0095] At this point, several harmful cracks are obtained.

[0096] Step S004: Determine the harmfulness of the harmful cracks based on the harmfulness of all crack sub-areas in all harmful cracks; obtain the degree of harm of the harmful cracks to the bridge based on the distance between the harmful crack and all other harmful cracks and the harmfulness of all harmful cracks, wherein the harmfulness is positively correlated with the degree of harm; and evaluate the bridge quality based on the harmfulness of all harmful cracks and the area of all harmful cracks.

[0097] The harmfulness of harmful cracks is determined based on the harmfulness of the crack sub-areas in the harmful cracks and the distribution of all harmful cracks in the bridge. The bridge quality is then evaluated based on the harmfulness of all harmful cracks and the area of all harmful cracks.

[0098] It should be noted that the multiple harmful cracks obtained in step S003 represent multiple complete harmful cracks, and the degree of damage to the bridge can be assessed by using the harmful cracks. However, since the stress fields between the harmful cracks in the bridge may affect each other, the closer the distance between the harmful cracks in the bridge, the more their stress fields will affect each other, resulting in the superposition of stress concentration areas in the bridge. When this happens, the concrete between the harmful cracks will be subjected to greater stress, that is, the closer the distance between the harmful cracks, the easier it is for the harmful cracks to expand and form larger cracks. Therefore, when assessing the quality of a bridge based on multiple harmful cracks, it is necessary not only to conduct a comprehensive assessment based on each harmful crack itself, but also to combine the positional distribution relationship between the various harmful cracks.

[0099] Specifically, for any harmful crack, the average of the harmfulness of all crack sub-regions in all harmful cracks is used as the harmfulness of the harmful crack;

[0100] According to the distance between the harmful crack and all other harmful cracks, combined with the harmfulness of all harmful cracks, the harmfulness of the harmful cracks is obtained. The specific calculation formula is:

[0101]

[0102] Where, Indicates the The degree of damage of each harmful crack; Indicates the The harmfulness of each harmful crack; Indicates the The harmfulness of each harmful crack; Indicates the number of harmful cracks; Indicates the The harmful cracks and The distance between harmful cracks; represents a sigmoid function, which is used to perform a normalization operation in this embodiment.

[0103] It should be noted that the degree of harmfulness of harmful cracks refers to the harmful crack characteristics of the harmful cracks. The more harmful crack characteristics the cracks have, the more likely the concrete structure is to be damaged, and the greater the damage to the bridge. When the harmful cracks with high degree of harmfulness are distributed closer, the stress on the concrete between the harmful cracks will be greater, and the subsequent harmful cracks will be more likely to expand into larger cracks, endangering the safety of the bridge structure. The larger the value, the The more harmful cracks there are near the first harmful crack, the greater the degree of its harmfulness. The more likely the harmful crack is to expand, the The more harmful cracks there are, the greater the damage to the bridge. After obtaining the damage degrees of all harmful cracks, we can further combine the areas of all harmful cracks to accurately evaluate the quality of the bridge.

[0104] Specifically, the bridge quality is obtained based on the degree of damage of all harmful cracks and the area of all harmful cracks. The specific calculation formula is:

[0105]

[0106] Where, Indicates the quality of the bridge; Indicates the number of harmful cracks; Indicates the The degree of damage of each harmful crack; Indicates the The area of harmful cracks; Represents an exponential function with a natural constant as the base. In this embodiment, Model to present inverse proportional relationship and normalization processing, As the input of the model, the implementer can set the inverse proportional function and normalization function according to the actual situation.

[0107] It should be noted that this embodiment analyzes the causes of harmful cracks in bridges, extracts the characteristics of harmful cracks, obtains harmful cracks based on the characteristics of harmful cracks, and further improves the accuracy of bridge quality assessment based on the distribution of harmful cracks and the size of harmful cracks.

[0108] At this point, this embodiment is completed.

[0109] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A concrete crack detection method for bridge engineering, characterized in that: The method comprises the following steps: Obtain several crack areas in the bridge and the direction of the principal tensile stress at each location; Dividing the crack region into a number of crack sub-regions; dividing the crack sub-region into a number of local crack sub-regions; obtaining a harmfulness factor of the local crack sub-region based on the extension direction of the local crack sub-region and the direction of the principal tensile stress at the corresponding position, wherein the harmfulness factor represents the perpendicularity between the extension direction and the direction of the principal tensile stress; obtaining the degree of harmfulness of the crack sub-region based on the difference between the harmfulness factors of each local crack sub-region and the adjacent local crack sub-region in the crack sub-region; Determining the extension consistency of adjacent crack sub-regions based on the difference in the inclination angles of adjacent crack sub-regions; determining the parameter consistency of adjacent crack sub-regions based on the relative deviation of the harmfulness of adjacent crack sub-regions; determining the degree of splicing between adjacent crack sub-regions based on the extension consistency and the parameter consistency, wherein both the extension consistency and the parameter consistency are positively correlated with the splicing degree; and obtaining a number of harmful cracks based on the splicing degree between adjacent crack sub-regions. The harmfulness of the harmful cracks is determined based on the harmfulness of all crack sub-areas among all harmful cracks. The degree of harm of the harmful cracks to the bridge is obtained based on the distance between the harmful crack and all other harmful cracks, combined with the harmfulness of all harmful cracks, where the harmfulness is positively correlated with the degree of harm. The bridge quality is evaluated based on the harmfulness of all harmful cracks and the area of all harmful cracks.

2. A concrete crack detection method for bridge engineering according to claim 1, characterized in that: The specific method of dividing the crack region into a plurality of crack sub-regions includes: For any crack region, the Guo-Hall algorithm is used to obtain the skeleton of the crack region; the pixels on the skeleton are recorded as skeleton pixels, and for any skeleton pixel, if there are three or more skeleton pixels in its eight-neighborhood, the skeleton pixel is used as a segmentation point; the skeleton is divided into several segments using all the segmentation points in the skeleton to obtain several skeleton segments; For any pixel point in the crack area, obtain the distance between the pixel point and each skeleton segment, and use the skeleton segment closest to the pixel point as the skeleton segment corresponding to the pixel point; The skeleton segment corresponding to each pixel point in the crack region is obtained, and several pixel points corresponding to the same skeleton segment are classified into the same crack sub-region to obtain several crack sub-regions.

3. The concrete crack detection method for bridge engineering according to claim 1, characterized in that: The specific method of dividing the crack sub-region into a number of local crack sub-regions includes: For any crack sub-region, the skeleton segment of the crack sub-region is divided into several segments of length The sub-skeleton segment, is the preset sub-skeleton segment length; For any pixel point in the crack sub-region, obtain the distance between the pixel point and each skeleton sub-segment, and use the skeleton sub-segment closest to the pixel point as the skeleton sub-segment corresponding to the pixel point; A skeleton subsegment corresponding to each pixel point in the crack sub-region is obtained, and several pixel points corresponding to the same skeleton subsegment are classified into the same local crack sub-region to obtain several local crack sub-regions.

4. The concrete crack detection method for bridge engineering according to claim 1, characterized in that: The method of obtaining the harmful factor of the local crack sub-region according to the extension direction of the local crack sub-region and the direction of the principal tensile stress at the corresponding position includes the following specific methods: For any local crack sub-region, a linear fitting is performed based on the coordinate positions of all pixel points in the local crack sub-region using the least squares method to obtain a fitting straight line for the local crack sub-region, and the direction of the fitting straight line for the local crack sub-region is used as the extension direction of the local crack sub-region; The harmfulness factor of the local crack sub-region is obtained according to the extension direction of the local crack sub-region and the principal tensile stress direction at each pixel point position in the local crack sub-region.

5. A concrete crack detection method for bridge engineering according to claim 4, characterized in that: The specific calculation formula for obtaining the harmfulness factor of the local crack sub-region is: Where, A harmful factor representing the local crack sub-region; represents the number of pixels in the local crack sub-region; An extension direction of the local crack sub-region; Indicates the first The principal tensile stress direction at each pixel position; represents the sine trigonometric function; Represents the absolute value function.

6. The concrete crack detection method for bridge engineering according to claim 1, characterized in that: The method of obtaining the harmfulness of a crack sub-region based on the difference between the harmfulness factors of each local crack sub-region and the adjacent local crack sub-regions in the crack sub-region includes the following specific methods: For any crack sub-region, the harmfulness of the crack sub-region is obtained based on the harmfulness factor of each local crack sub-region in the crack sub-region and the difference in harmfulness factors between adjacent local crack sub-regions in the crack sub-region. The specific calculation formula is: Where, Indicates the harmfulness of the crack sub-area; represents the number of local crack sub-regions in the crack sub-region; Indicates the first Harmful factors of local crack sub-regions; Indicates the crack sub-area with the The number of local crack sub-regions adjacent to a local crack sub-region; Indicates the crack sub-area with the The adjacent local crack sub-region Harmful factors of local crack sub-regions; Indicates the absolute value function; Represents an exponential function with a natural constant as its base.

7. The concrete crack detection method for bridge engineering according to claim 1, characterized in that: The specific method of determining the degree of splicing between adjacent crack sub-regions by combining the extension consistency and the parameter consistency includes: For any crack sub-region, the crack sub-region is recorded as a reference region, and the crack sub-region adjacent to the reference region is recorded as a target region; The calculation formula for the degree of integration between the reference area and the target area is: Where, Indicates the degree of fit between the reference area and the target area; represents the slope of the fitted straight line of the target area; represents the slope of the fitted line in the reference area; Indicates the harmfulness of the target area; Indicates the degree of harmfulness of the reference area; represents the inverse tangent function; Represents the preset hyperparameters; represents the linear normalization function; The extension consistency of the reference area and the target area; is the tilt angle of the target area; is the tilt angle of the reference area; The consistency of parameters between the reference area and the target area; Among them, the method for obtaining the slope of the fitting line of each crack sub-region is as follows: for each crack sub-region, according to the position coordinates of each pixel point in the crack sub-region, the least squares method is used to fit the crack sub-region separately to obtain the fitting line of the crack sub-region; the slope of the fitting line of the crack sub-region is used as the slope of the fitting line of the crack sub-region.

8. A concrete crack detection method for bridge engineering according to claim 7, characterized in that: The specific method of obtaining a number of harmful cracks according to the degree of splicing between adjacent crack sub-regions includes: Preset a stitching threshold If the degree of fit between the reference area and the target area is greater than or equal to , classifying the benchmark area and the target area as the same harmful crack; The degree of splicing of all crack sub-regions and their adjacent crack sub-regions is judged to obtain several harmful cracks.

9. The concrete crack detection method for bridge engineering according to claim 1, characterized in that: The calculation formula for the degree of damage caused by the harmful cracks to the bridge is: Where, Indicates the The degree of damage of each harmful crack; Indicates the The harmfulness of each harmful crack; Indicates the The harmfulness of each harmful crack; Indicates the number of harmful cracks; Indicates the The harmful cracks and The distance between harmful cracks; Represents the sigmoid function.

10. The concrete crack detection method for bridge engineering according to claim 1, characterized in that: The bridge quality is assessed based on the degree of damage of all harmful cracks and the area of all harmful cracks, including the following specific methods: The bridge quality is obtained based on the degree of damage of all harmful cracks and the area of all harmful cracks. The specific calculation formula is: Where, Indicates the quality of the bridge; Indicates the number of harmful cracks; Indicates the The degree of damage of each harmful crack; Indicates the The area of harmful cracks; Represents an exponential function with a natural constant as its base.

Images