A bridge corrosion and fatigue monitoring system and method based on computer vision

Abstract

The present invention provides a bridge corrosion and fatigue monitoring system and method based on computer vision, and relates to the technical field of bridge corrosion and fatigue monitoring. The method comprises: obtaining an initial image and a current image of a bridge, as well as a current construction time; determining the partition positions of the initial image and the current image based on the current construction time and a preset image partition position mapping table; dividing the initial image and the current image into a plurality of sub-initial image regions and a plurality of sub-current image regions according to the partition positions; determining a deformed sub-current image region and an undeformed sub-current image region; determining a pixel matching relationship; comparing the undeformed sub-current image region with the corresponding sub-initial image region, and comparing the deformed sub-current image region with the corresponding sub-initial image region based on the pixel matching relationship to obtain a corrosion degree; determining the fatigue degree of the bridge according to the corrosion degree; the method of the present invention can improve the accuracy of monitoring the corrosion degree and fatigue degree of the bridge.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge corrosion and fatigue monitoring, and in particular to a bridge corrosion and fatigue monitoring system and method based on computer vision. Background Art

[0002] As an important transportation infrastructure, the structural safety of bridges is directly related to the smoothness of transportation and public safety. Corrosion and fatigue are one of the main causes of bridge structural damage. They will cause the strength and stiffness of bridge components to decrease, and in severe cases may cause safety accidents such as bridge collapse; therefore, corrosion and fatigue monitoring of bridges can timely discover and deal with potential damage to ensure the structural safety of bridges. In the prior art, one of the methods for monitoring the corrosion and fatigue of bridges is through image comparison, that is, by comparing the current image of the bridge with the initial image of the bridge to determine the corrosion and fatigue of the bridge; however, as time goes by, the bridge will undergo a certain deformation, and the deformation degree of different areas of the bridge is different, and the deformation degree of the same area of ​​the bridge will also change over time; if the current image of the bridge is simply compared with the pixels with the same coordinates in the initial image of the bridge, the pixels compared will not correspond to the same position of the bridge, which will have a greater impact on the corrosion and fatigue monitoring of the bridge, resulting in low accuracy of the corrosion and fatigue monitoring of the bridge. Summary of the invention

[0003] The object of the present invention is to provide a bridge corrosion and fatigue monitoring system and method based on computer vision to improve the accuracy of determining the corrosion and fatigue of the bridge.

[0004] According to a first aspect of the present application, a bridge corrosion and fatigue monitoring method based on computer vision is provided, the method comprising the following steps:

[0005] Get the initial image and current image of the bridge, as well as the current construction time;

[0006] Determine the partition positions of the initial image and the current image based on the current construction time and a preset image partition position mapping table; in the mapping table, each row corresponds to a construction time range and a number of image partition positions;

[0007] The initial image and the current image are divided into a plurality of sub-initial image regions and a plurality of sub-current image regions according to the partition positions; the actual positions of the same partition positions on the bridge are the same;

[0008] According to the current construction time, the deformed sub-image area and the undeformed sub-image area are determined;

[0009] Determine the pixel matching relationship between each deformed sub-current image region and its corresponding sub-initial image region;

[0010] Compare the undeformed sub-current image area with the corresponding sub-initial image area, and compare the deformed sub-current image area with the corresponding sub-initial image area based on the pixel matching relationship to obtain the corrosion degree;

[0011] Determine the fatigue level of the bridge based on the degree of corrosion.

[0012] Furthermore, the step of determining the deformed sub-current image region and the undeformed sub-current image region according to the current construction time includes the following steps:

[0013] If the current construction time is less than or equal to the first preset time, each sub-current image region is determined as an undeformed sub-current image region;

[0014] If the current construction time is greater than the first preset time and less than the second preset time, the sub-current image area whose distance from the nearest bridge support point is greater than the preset distance corresponding to the bridge at the current moment is determined as the deformed sub-current image area;

[0015] If the current construction time length is greater than or equal to the second preset time length, each sub-current image region is determined as a deformed sub-current image region.

[0016] Furthermore, the step of determining the pixel matching relationship between each deformed sub-current image region and its corresponding sub-initial image region comprises the following steps:

[0017] Acquire several feature points of the bridge in the current image region of the target deformable sub-image, and several feature points of the bridge in the target sub-initial image region corresponding to the current image region of the target deformable sub-image; wherein the current image region of the target deformable sub-image is any current image region of the deformable sub-image;

[0018] The coordinates of several feature points of the bridge in the current image area of ​​the target deformable sub-image and the coordinates of several feature points of the bridge in the target sub-initial image area corresponding to the current image area of ​​the target deformable sub-image are fitted to obtain the pixel matching relationship between the current image area of ​​the target deformable sub-image and the corresponding initial image area of ​​the target sub-image.

[0019] Furthermore, the undeformed sub-current image region is compared with the corresponding sub-initial image region, and the deformed sub-current image region is compared with the corresponding sub-initial image region based on a pixel matching relationship to obtain a corrosion degree, including the following steps:

[0020] Obtaining a first difference in pixel value between each pixel point in each undeformed sub-current image region and a pixel point at the same coordinate position in the corresponding sub-initial image region;

[0021] Determine a pixel point in each undeformed sub-current image region whose first difference value is greater than a first preset pixel value and less than a second preset pixel value as a corroded pixel point;

[0022] Get the first coordinate of each pixel point in the current image area of ​​each deformer;

[0023] Determine the second coordinate of the first coordinate of each pixel point in each deformable sub-current image region in the corresponding sub-initial image region according to the pixel matching relationship corresponding to each deformable sub-current image region;

[0024] Obtain a second difference in pixel values ​​between each pixel point in each deformed sub-current image region and a pixel point at a second coordinate position in the corresponding sub-initial image region;

[0025] Determine a pixel point in the current image region of each deformer whose second difference value is greater than the first preset pixel value and less than the second preset pixel value as a corroded pixel point;

[0026] The corrosion degree of the bridge is determined according to the number of corroded pixels and the total number of pixels corresponding to the bridge in the current image.

[0027] Furthermore, the method of determining the corrosion degree of the bridge according to the number of corroded pixels and the total number of pixels corresponding to the bridge in the current image comprises the following steps:

[0028] Get the number of corroded pixels and the total number of pixels corresponding to the bridge in the current image;

[0029] The ratio of the number of corroded pixels to the total number of pixels corresponding to the bridge in the current image is determined as the corrosion degree of the bridge.

[0030] Furthermore, determining the fatigue degree of the bridge according to the corrosion degree comprises the following steps:

[0031] Compare the initial image and current image of the bridge to determine the deformation corresponding to the bridge at the current moment;

[0032] The fatigue degree of the bridge is obtained by multiplying the deformation amount corresponding to the bridge at the current moment by the first preset weight and the corrosion degree of the bridge by the second preset weight plus the preset initial value.

[0033] Furthermore, the method further comprises the following steps:

[0034] If the corrosion degree of the bridge is greater than a preset corrosion degree threshold and the fatigue degree of the bridge is greater than a preset fatigue degree threshold, a preset first prompt information is generated; wherein the first prompt information is used to prompt the user to perform corrosion and fatigue inspections on the bridge;

[0035] If the corrosion degree of the bridge is greater than a preset corrosion degree threshold and the fatigue degree of the bridge is less than a preset fatigue degree threshold, a preset second prompt information is generated; wherein the second prompt information is used to prompt the user to perform corrosion inspection on the bridge;

[0036] If the corrosion degree of the bridge is less than a preset corrosion threshold and the fatigue degree of the bridge is greater than a preset fatigue threshold, a preset third prompt information is generated; wherein the third prompt information is used to prompt the user to perform fatigue inspection on the bridge.

[0037] Furthermore, the number of partition positions is positively correlated with the current construction time of the bridge.

[0038] Furthermore, the initial image of the bridge is taken at the same position, angle and lighting conditions as the current image.

[0039] According to a second aspect of the present invention, a bridge corrosion and fatigue monitoring system based on computer vision is provided, the system comprising: a computer-readable storage medium and a processor; wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the bridge corrosion and fatigue monitoring method based on computer vision as described in the first aspect is implemented.

[0040] The present invention has at least the following beneficial effects:

[0041] The computer vision-based bridge corrosion and fatigue monitoring method of the present invention obtains an initial image, a current image and a current construction time of the bridge; divides the current image and the initial image corresponding to the bridge into a plurality of sub-initial image areas and sub-current image areas according to the current construction time, establishes a corresponding pixel matching relationship for a deformed sub-current image area with deformation in the sub-current image area, and accurately matches each pixel point in the deformed sub-current image area to a pixel point in the corresponding sub-initial image area through the pixel matching relationship, thereby avoiding the situation that the compared pixels do not correspond to the same position of the bridge, reducing the impact on the corrosion and fatigue monitoring of the bridge, and improving the accuracy of the corrosion and fatigue monitoring of the bridge.

[0042] In addition, when dividing the initial image and the current image corresponding to the bridge, a relationship is established between the current construction time of the bridge and the specific partition position by establishing a bridge image partition position mapping table. The current construction time corresponding to the bridge can be used to quickly determine several partition positions corresponding to the bridge. At the same time, the construction time corresponding to the bridge is different, and the setting method of the corresponding partition position is also different. When the bridge is divided into regions according to the corresponding partition positions, the deformation degree of the actual area of ​​the bridge corresponding to each region is relatively similar. In the subsequent image comparison, the correspondence relationship of the pixel comparison can be improved, thereby further improving the accuracy of bridge corrosion and fatigue monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0044] Figure 1 A flowchart of a bridge corrosion and fatigue monitoring method based on computer vision provided in Embodiment 1 of the present invention;

[0045] Figure 2 A flow chart of determining a deformed sub-image area and an undeformed sub-image area according to the first embodiment of the present invention;

[0046] Figure 3 A flowchart for determining a matching relationship provided in Embodiment 1 of the present invention;

[0047] Figure 4 A flow chart of a first method of determining the corrosion degree of a bridge provided in the first embodiment of the present invention;

[0048] Figure 5 A second flow chart for determining the corrosion degree of a bridge provided in the first embodiment of the present invention;

[0049] Figure 6 A flow chart for determining fatigue of a bridge provided in the first embodiment of the present invention;

[0050] Figure 7 This is a flow chart for monitoring the corrosion and fatigue of a bridge provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0052] It should be noted that, based on the present disclosure, those skilled in the art should understand that an aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects described herein can be used to implement a device and / or practice a method. In addition, other structures and / or functionalities other than one or more of the aspects described herein can be used to implement this device and / or practice this method.

[0053] Embodiment 1:

[0054] The following will refer to Figure 1 The flowchart of the bridge corrosion and fatigue monitoring method based on computer vision is shown, which introduces a bridge corrosion and fatigue monitoring method based on computer vision.

[0055] The bridge corrosion and fatigue monitoring method based on computer vision may include the following steps:

[0056] S100, obtaining an initial image and a current image of the bridge, as well as the current construction time.

[0057] In this embodiment, the initial image of the bridge is the image when the bridge was built, and the current image is the image of the bridge taken at the current moment; by obtaining the construction time point of the bridge and the current time point, the current construction duration corresponding to the current moment of the bridge can be obtained.

[0058] Furthermore, the initial image of the bridge is taken at the same position, angle and lighting conditions as the current image.

[0059] In this embodiment, when taking the initial image and the current image of the bridge, it is necessary to ensure that the shooting position, angle and lighting conditions are the same; specifically, when the bridge is built, the camera can be fixed at a fixed position, and the initial image of the bridge can be taken at a preset time on a clear day. Similarly, at the same preset time on the same day one or two years later, ensuring that the weather conditions are the same, the current image of the bridge can be taken to avoid inaccurate monitoring results due to different environmental factors.

[0060] S200, determining the partition positions of the initial image and the current image based on the current construction time and a preset image partition position mapping table; in the mapping table, each row corresponds to a construction time range and a number of image partition positions.

[0061] In this embodiment, the structural form of the bridge and the parameters such as concrete and steel bars used are known. Therefore, a finite element model of the bridge can be established, and a finite element simulation analysis can be performed on the bridge using a preset simulation software to obtain the deformation position of the bridge and the corresponding relationship between the deformation amount and time. Then, based on the relationship between the deformation amount of each deformation area and the preset deformation amount threshold and time, the corresponding preset bridge image partition position is determined for different construction durations, thereby obtaining a preset bridge image partition position mapping table. It should be noted that those skilled in the art can use existing finite element simulation analysis methods to perform finite element simulation analysis on the bridge according to actual needs to obtain the deformation position of the bridge and the corresponding relationship between the deformation amount and time, which will not be elaborated here.

[0062] In this embodiment, several partition positions corresponding to the initial image and the current image of the bridge are determined to obtain a partition position list A=(A1, A2, ..., A i , …, A n ), i=1, 2,...,n; among them, A i is the determined i-th partition position, and n is the number of determined partition positions; it can be understood that the deformation of the bridge is along the Y-axis, so the partition positions are divided along the X-axis direction. As time goes by, the farther away from the bridge support point, the greater the degree of deformation.

[0063] Furthermore, the number of partition positions is positively correlated with the current construction time; that is, the longer the current construction time, the greater the number of partition positions, and this relationship is consistent with the characteristics of bridge deformation.

[0064] S300, dividing the initial image and the current image into a plurality of sub-initial image regions and a plurality of sub-current image regions according to the partition positions; the actual positions of the same partition positions on the bridge are the same.

[0065] In this embodiment, according to the initial image of the bridge and the corresponding partition positions of the current image, the initial image is divided into a plurality of sub-initial image regions, and the current image is divided into a plurality of sub-current image regions, so as to obtain a sub-initial image region list B = (B1, B2, ..., B j , …, B n+1 ), j = 1, 2, ..., n + 1; and the sub-current image area list C = (C1, C2, ..., C j , …, C n+1 ); where B j is the jth sub-initial image area obtained, C j is the jth sub-current image area obtained; B j and C j Corresponds to the same area of ​​the bridge.

[0066] S400: Determine a deformed sub-current image region and an undeformed sub-current image region according to the current construction time.

[0067] In this embodiment, the deformation of the bridge is small in the early stage of construction and can be considered as no deformation. As time goes by, the deformation will increase. Therefore, it is necessary to determine each deformed sub-current image area and undeformed sub-current image area from several sub-current image areas according to the current construction time corresponding to the current moment of the bridge.

[0068] Further, such as Figure 2 As shown, step S400 may include the following steps:

[0069] S410: If the current construction time is less than or equal to the first preset time, each sub-current image region is determined as an undeformed sub-current image region.

[0070] In this embodiment, the first preset time length can be set based on experience or obtained through finite element simulation analysis; for example: the first preset time length is 6 months to 12 months; if the current construction time corresponding to the current moment of the bridge is less than or equal to the first preset time length, it means that the current bridge has not been deformed, then each sub-current image area in C is determined as an undeformed sub-current image area.

[0071] S420, if the current construction time is greater than the first preset time and less than the second preset time, the sub-current image area whose distance to the nearest bridge support point is greater than the preset distance corresponding to the bridge at the current moment is determined as the deformed sub-current image area.

[0072] In this embodiment, the second preset time length can be set based on experience or obtained through finite element simulation analysis; for example: the second preset time length is 12 months to 36 months; if the construction time length corresponding to the current moment of the bridge is greater than the first preset time length and less than the second preset time length, it means that the sub-current image area whose distance from the nearest support point of the current bridge is greater than the preset distance corresponding to the bridge is determined as the deformed sub-current image area, and the support point can be a bridge pier.

[0073] In this embodiment, the preset distance corresponding to the bridge is obtained by the following steps:

[0074] S421, obtain the maximum design duration T of the bridge max ;

[0075] S422, based on the construction time T corresponding to the current time of the bridge now and T max , determine the preset distance L corresponding to the bridge = (T now / T max )×(W / 2); where W is the distance between adjacent supporting points of the bridge.

[0076] The L determined by the above steps is equal to the construction time T corresponding to the current moment of the bridge. now As increases, will gradually increase, so that the number of the determined deformable sub-image regions in the current image area will also gradually increase, thereby improving the accuracy of determining the corrosion degree.

[0077] S430: If the current construction duration is greater than or equal to the second preset duration, each sub-current image region is determined as a deformed sub-current image region.

[0078] In this embodiment, if the current construction duration corresponding to the current moment of the bridge is greater than or equal to the second preset duration, it means that all areas of the bridge have been deformed. At this time, each sub-current image area is determined as a deformed sub-current image area.

[0079] S500, determining a pixel matching relationship between each deformed sub-current image region and its corresponding sub-initial image region.

[0080] In the present embodiment, it should be noted that this step will be executed only when the current construction duration corresponding to the current moment of the bridge is greater than the first preset duration; since the bridge at the current moment has been deformed, if the current image corresponding to the current moment is directly compared with the initial image for pixel points, then the two pixel points for comparison will not be in the same position of the corresponding bridge, which will result in a large error in the comparison result; therefore, it is necessary to match the pixels.

[0081] Further, such as Figure 3 As shown, step S500 may include the following steps:

[0082] S510, obtaining several feature points of the bridge in the target deformable sub-current image area, and several feature points of the bridge in the target sub-initial image area corresponding to the target deformable sub-current image area; wherein the target deformable sub-current image area is any deformable sub-current image area.

[0083] In this embodiment, feature point 1 of the bridge, for example: feature point 1 is a vertex of the bridge; feature 1 corresponds to a pixel point in the initial image, and also corresponds to a pixel point in the current image, and the coordinates of the two in the same coordinate system may be different because the bridge has been deformed at the current moment.

[0084] It should be noted that those skilled in the art can use existing feature point acquisition methods to obtain several feature points of the bridge in the current image area of ​​the target deformable sub-image, as well as several feature points of the bridge in the target sub-initial image area corresponding to the current image area of ​​the target deformable sub-image, according to actual needs, which will not be elaborated here.

[0085] S520, fitting the coordinates of several feature points of the bridge in the current image area of ​​the target deformable sub-image and the coordinates of several feature points of the bridge in the target sub-initial image area corresponding to the current image area of ​​the target deformable sub-image to obtain a pixel matching relationship between the current image area of ​​the target deformable sub-image and the corresponding initial image area of ​​the target sub-image.

[0086] In this embodiment, after obtaining the coordinates of several feature points in the initial image and the coordinates of several feature points in the current image, the pixel matching relationship between the current image area of ​​the target deformed sub-image and the corresponding initial image area of ​​the target sub-image can be obtained through the existing fitting method; the matching relationship can be understood as a coordinate mapping function.

[0087] S600, comparing the undeformed sub-current image region with the corresponding sub-initial image region, and comparing the deformed sub-current image region with the corresponding sub-initial image region based on a pixel matching relationship to obtain a corrosion degree.

[0088] Further, such as Figure 4 As shown, step S600 may include the following steps:

[0089] S610, obtaining a first difference in pixel value between each pixel point in each undeformed sub-current image region and a pixel point at the same coordinate position in the corresponding sub-initial image region.

[0090] In this embodiment, for any undeformed sub-current image area, the first difference in pixel value between each pixel point in the undeformed sub-current image area and the pixel point at the same coordinate position in the corresponding sub-initial image area can be obtained; the pixel value may be a grayscale value.

[0091] S620: Determine a pixel point in each undeformed sub-current image region whose first difference value is greater than a first preset pixel value and less than a second preset pixel value as a corroded pixel point.

[0092] In this embodiment, each first difference can be traversed. If the first difference is greater than the first preset pixel value and less than the second preset pixel value, it means that the pixel value corresponding to the first difference has undergone a significant change during the construction period, and it can be considered that the position corresponding to the pixel point has been corroded. It should be noted that after the bridge is corroded, the change in its pixel value is usually within a certain range. Therefore, setting the first preset pixel value and the second preset pixel value can avoid the occurrence of misjudgment due to foreign body occlusion, and further improve the accuracy of judgment.

[0093] S630, obtaining the first coordinate of each pixel point in the current image area of ​​each deformer.

[0094] S640: Determine, according to the pixel matching relationship corresponding to each deformed sub-current image region, the second coordinate of the first coordinate of each pixel point in each deformed sub-current image region in the corresponding sub-initial image region.

[0095] In this embodiment, after the matching relationship is determined in the above steps, the corresponding second coordinate in the sub-initial image can be determined based on the first coordinate and the corresponding matching relationship; it can be understood that the position corresponding to the first coordinate on the bridge and the position corresponding to the second coordinate on the bridge are the same.

[0096] S650, obtaining a second difference in pixel value between each pixel point in each deformed sub-current image region and a pixel point at a second coordinate position in the corresponding sub-initial image region.

[0097] S660: Determine a pixel point in the current image region of each deformer whose second difference value is greater than the first preset pixel value and less than the second preset pixel value as a corroded pixel point.

[0098] In this embodiment, after matching each pixel point in the current image area of ​​the deformer, the difference in pixel values ​​obtained will correspond to the pixel at the same position of the bridge, thereby ensuring the accuracy of image comparison.

[0099] S670: Determine the corrosion degree of the bridge according to the number of corroded pixels and the total number of pixels corresponding to the bridge in the current image.

[0100] Further, such as Figure 5 As shown, step S670 may include the following steps:

[0101] S671, obtaining the number of corroded pixels and the total number of pixels corresponding to the bridge in the current image.

[0102] S672: Determine the ratio of the number of corroded pixel points to the total number of pixel points corresponding to the bridge in the current image as the corrosion degree of the bridge.

[0103] In this embodiment, the corrosion degree of the bridge can be determined through the above steps; when determining the corrosion degree of the bridge, the deformation of the bridge is taken into account, and the pixels in the deformed area are re-matched, so that the final result is more accurate.

[0104] S700, Determine the fatigue of the bridge based on the corrosion degree.

[0105] In this embodiment, Figure 6 As shown, step S700 may include the following steps:

[0106] S710, comparing the initial image and the current image of the bridge to determine the deformation amount corresponding to the bridge at the current moment.

[0107] It should be noted that those skilled in the art can use existing image comparison methods to compare the initial image and the current image of the bridge according to actual needs to determine the deformation amount corresponding to the bridge at the current moment, which will not be elaborated here.

[0108] S720, multiplying the deformation amount corresponding to the bridge at the current moment by the first preset weight and the corrosion degree of the bridge by the second preset weight plus the preset initial value to obtain the fatigue degree of the bridge.

[0109] In this embodiment, the fatigue degree of the bridge is affected by many factors, and corrosion and deformation are relatively important factors. Therefore, the fatigue degree of the bridge can be obtained by weighting the two and adding other factors by presetting initial values. The preset initial values ​​can be empirical values ​​or obtained by analyzing a large amount of bridge data.

[0110] Further, such as Figure 7 As shown, the method may further include the following steps:

[0111] S810, if the corrosion degree of the bridge is greater than a preset corrosion threshold and the fatigue degree of the bridge is greater than a preset fatigue threshold, a preset first prompt information is generated; wherein the first prompt information is used to prompt the user to perform corrosion and fatigue inspections on the bridge.

[0112] S820: If the corrosion degree of the bridge is greater than a preset corrosion degree threshold and the fatigue degree of the bridge is less than a preset fatigue degree threshold, a preset second prompt information is generated; wherein the second prompt information is used to prompt the user to perform corrosion inspection on the bridge.

[0113] S830: If the corrosion degree of the bridge is less than a preset corrosion degree threshold and the fatigue degree of the bridge is greater than a preset fatigue degree threshold, a preset third prompt message is generated; wherein the third prompt message is used to prompt the user to perform fatigue maintenance on the bridge.

[0114] In this embodiment, through the above steps, different prompt information can be generated for different situations, so that the user can carry out maintenance on the bridge in a targeted manner to improve the maintenance efficiency.

[0115] The computer vision-based bridge corrosion and fatigue monitoring method of this embodiment divides the current image and initial image corresponding to the bridge into several sub-initial image areas and sub-current image areas according to the current construction time corresponding to the bridge, and establishes corresponding pixel matching relationships for deformed sub-current image areas with deformation in the sub-current image areas. Through the pixel matching relationship, each pixel point in the deformed sub-current image area is accurately matched to the pixel point in the corresponding sub-initial image area, thereby avoiding the situation where the compared pixels do not correspond to the same position of the bridge, reducing the impact on the corrosion and fatigue monitoring of the bridge, and improving the accuracy of the corrosion and fatigue monitoring of the bridge.

[0116] In addition, when dividing the initial image and the current image corresponding to the bridge, a relationship is established between the construction time of the bridge and the specific partition position by establishing a bridge image partition position mapping table. The construction time corresponding to the bridge can be used to quickly determine several partition positions corresponding to the bridge. At the same time, the construction time corresponding to the bridge is different, and the setting method of the corresponding partition position is also different. When the bridge is divided into regions according to the corresponding partition positions, the deformation degree of the actual area of ​​the bridge corresponding to each region is relatively similar. In the subsequent image comparison, the correspondence relationship of the pixel comparison can be improved, thereby further improving the accuracy of bridge corrosion and fatigue monitoring.

[0117] Furthermore, the current image area of ​​the deformer is determined according to the construction duration corresponding to the current moment of the bridge, and pixel matching is performed only on the current image area of ​​the deformer instead of matching all pixels, thereby improving computational efficiency while ensuring accuracy.

[0118] Embodiment 2:

[0119] An embodiment of the present invention also provides a bridge corrosion and fatigue monitoring system based on computer vision, the system comprising: a computer-readable storage medium and a processor; wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, it implements the bridge corrosion and fatigue monitoring method based on computer vision as described in Example 1.

[0120] Although some specific embodiments of the present invention have been described in detail by way of example, it should be understood by those skilled in the art that the above examples are only for illustration, not for limiting the scope of the present invention. It should also be understood by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention.

Claims

1. A bridge corrosion and fatigue monitoring method based on computer vision, characterized in that: The method comprises the following steps: Get the initial image and current image of the bridge, as well as the current construction time; Determine the partition positions of the initial image and the current image based on the current construction time and a preset image partition position mapping table; in the mapping table, each row corresponds to a construction time range and a number of image partition positions; The initial image and the current image are divided into a plurality of sub-initial image regions and a plurality of sub-current image regions according to the partition positions; the actual positions of the same partition positions on the bridge are the same; According to the current construction time, the deformed sub-image area and the undeformed sub-image area are determined; Determine the pixel matching relationship between each deformed sub-current image region and its corresponding sub-initial image region; Compare the undeformed sub-current image area with the corresponding sub-initial image area, and compare the deformed sub-current image area with the corresponding sub-initial image area based on the pixel matching relationship to obtain the corrosion degree; Determine the fatigue of the bridge based on the degree of corrosion; The step of comparing the undeformed sub-current image region with the corresponding sub-initial image region, and comparing the deformed sub-current image region with the corresponding sub-initial image region based on a pixel matching relationship to obtain a corrosion degree includes the following steps: Obtaining a first difference in pixel value between each pixel point in each undeformed sub-current image region and a pixel point at the same coordinate position in the corresponding sub-initial image region; Determine a pixel point in each undeformed sub-current image region whose first difference value is greater than a first preset pixel value and less than a second preset pixel value as a corroded pixel point; Get the first coordinate of each pixel point in the current image area of ​​each deformer; Determine the second coordinate of the first coordinate of each pixel point in each deformable sub-current image region in the corresponding sub-initial image region according to the pixel matching relationship corresponding to each deformable sub-current image region; Obtain a second difference in pixel values ​​between each pixel point in each deformed sub-current image region and a pixel point at a second coordinate position in the corresponding sub-initial image region; Determine a pixel point in the current image region of each deformer whose second difference value is greater than the first preset pixel value and less than the second preset pixel value as a corroded pixel point; The corrosion degree of the bridge is determined according to the number of corroded pixels and the total number of pixels corresponding to the bridge in the current image.

2. The bridge corrosion and fatigue monitoring method based on computer vision according to claim 1 is characterized in that: The step of determining the deformed sub-current image region and the undeformed sub-current image region according to the current construction time includes the following steps: If the current construction time is less than or equal to the first preset time, each sub-current image region is determined as an undeformed sub-current image region; If the current construction time is greater than the first preset time and less than the second preset time, the sub-current image area whose distance from the nearest bridge support point is greater than the preset distance corresponding to the bridge at the current moment is determined as the deformed sub-current image area; If the current construction time length is greater than or equal to the second preset time length, each sub-current image region is determined as a deformed sub-current image region.

3. The bridge corrosion and fatigue monitoring method based on computer vision according to claim 1 is characterized in that: The step of determining the pixel matching relationship between each deformed sub-current image region and its corresponding sub-initial image region comprises the following steps: Acquire several feature points of the bridge in the current image region of the target deformable sub-image, and several feature points of the bridge in the target sub-initial image region corresponding to the current image region of the target deformable sub-image; wherein the current image region of the target deformable sub-image is any current image region of the deformable sub-image; The coordinates of several feature points of the bridge in the current image area of ​​the target deformable sub-image and the coordinates of several feature points of the bridge in the target sub-initial image area corresponding to the current image area of ​​the target deformable sub-image are fitted to obtain the pixel matching relationship between the current image area of ​​the target deformable sub-image and the corresponding initial image area of ​​the target sub-image.

4. The bridge corrosion and fatigue monitoring method based on computer vision according to claim 1 is characterized in that: Determining the corrosion degree of the bridge according to the number of corroded pixels and the total number of pixels corresponding to the bridge in the current image comprises the following steps: Get the number of corroded pixels and the total number of pixels corresponding to the bridge in the current image; The ratio of the number of corroded pixels to the total number of pixels corresponding to the bridge in the current image is determined as the corrosion degree of the bridge.

5. The bridge corrosion and fatigue monitoring method based on computer vision according to claim 1 is characterized in that: Determining the fatigue degree of the bridge according to the corrosion degree comprises the following steps: Compare the initial image and current image of the bridge to determine the deformation corresponding to the bridge at the current moment; The fatigue degree of the bridge is obtained by multiplying the deformation amount corresponding to the bridge at the current moment by the first preset weight and the corrosion degree of the bridge by the second preset weight plus the preset initial value.

6. The bridge corrosion and fatigue monitoring method based on computer vision according to claim 1 is characterized in that: The method further comprises the following steps: If the corrosion degree of the bridge is greater than a preset corrosion degree threshold and the fatigue degree of the bridge is greater than a preset fatigue degree threshold, a preset first prompt information is generated; wherein the first prompt information is used to prompt the user to perform corrosion and fatigue inspections on the bridge; If the corrosion degree of the bridge is greater than a preset corrosion degree threshold and the fatigue degree of the bridge is less than a preset fatigue degree threshold, a preset second prompt information is generated; wherein the second prompt information is used to prompt the user to perform corrosion inspection on the bridge; If the corrosion degree of the bridge is less than a preset corrosion threshold and the fatigue degree of the bridge is greater than a preset fatigue threshold, a preset third prompt information is generated; wherein the third prompt information is used to prompt the user to perform fatigue inspection on the bridge.

7. The bridge corrosion and fatigue monitoring method based on computer vision according to claim 1 is characterized in that: The number of partition locations is positively correlated to the current construction age of the bridge.

8. The bridge corrosion and fatigue monitoring method based on computer vision according to claim 1 is characterized in that: The initial image of the bridge is taken at the same location, angle, and lighting conditions as the current image.

9. A bridge corrosion and fatigue monitoring system based on computer vision, characterized in that: The system includes: a computer-readable storage medium and a processor; wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the computer vision-based bridge corrosion and fatigue monitoring method as described in any one of claims 1 to 8 is implemented.

Images