A method and system for three-dimensional visualization of bridge cracks

Abstract

The present invention provides a method and system for three-dimensional visualization of bridge cracks, the method comprising: obtaining a plurality of local images with overlapping areas, performing grayscale processing and local equalization processing on them to obtain an adjustment histogram; converting the adjustment histogram into an enhanced image and then performing smoothing processing, threshold segmentation and denoising processing to form a standby image including a reserved area; obtaining the minimum bounding rectangle of the reserved area, judging whether the reserved area is longitudinally extended or transversely extended based on the minimum bounding rectangle; if it is longitudinally extended, forming a first crack curve; if the reserved area is transversely extended, forming a second crack curve; mapping the first crack curve and / or the second crack curve into the local image, and obtaining a bridge model after splicing the local images. The bridge and cracks are three-dimensionalized, and there is no need for manual on-site monitoring, which shortens the monitoring cycle, reduces monitoring errors, and improves monitoring efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method and system for three-dimensional visualization of bridge cracks. Background Art

[0002] With the rapid development of society, the number of bridges, as one of the important infrastructures, is also increasing. Bridges are artificial structures that cross rivers, lakes, seas, deep canyons and other obstacles. They are key nodes and hub projects in transportation projects.

[0003] At present, most bridges are reinforced concrete structures. Under the influence of negative factors such as environmental erosion and vehicle squeezing on the bridge for a long time, cracks are prone to occur. Cracks not only affect the appearance of the bridge, but also easily cause safety problems such as steel corrosion, reduced durability of components, and reduced impermeability, which in turn affects the service life of the bridge. Therefore, crack monitoring has become one of the important indicators for bridge health status monitoring.

[0004] Currently, crack monitoring of bridges is generally carried out on-site by monitoring personnel, who use monitoring equipment such as crack width gauges to get close to the surface of the bridge structure for manual monitoring and recording. This not only has a long monitoring cycle, low efficiency, and large monitoring errors, but it is also difficult to intuitively grasp the location of cracks on the bridge, affecting the efficiency of subsequent bridge maintenance. Summary of the invention

[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a method and system for three-dimensional visualization of bridge cracks, aiming to solve the technical problem that the method of recording cracks through manual monitoring in the prior art not only has a long monitoring cycle, low efficiency, and large monitoring error, but also makes it difficult to intuitively grasp the location of cracks on the bridge, affecting the efficiency of subsequent bridge maintenance.

[0006] In order to achieve the above objectives, in a first aspect, an embodiment of the present application provides a method for three-dimensional visualization of bridge cracks, comprising the following steps:

[0007] Acquire a plurality of local images corresponding to the bridge, wherein there are overlapping areas between the local images, perform grayscale processing on the local images to obtain grayscale images, acquire a histogram image corresponding to the grayscale image, and perform local equalization processing on the histogram image to obtain an adjusted histogram;

[0008] Converting the adjusted histogram into an enhanced image, performing smoothing and threshold segmentation on the enhanced image to obtain a binary image, and performing denoising on the binary image to obtain a standby image including a plurality of reserved areas;

[0009] Constructing a coordinate system on the stand-by image, assigning pixel coordinates to all pixel points in the reserved area, obtaining a minimum bounding rectangle of the reserved area, and determining whether the reserved area extends vertically or horizontally based on the minimum bounding rectangle;

[0010] If the reserved area extends longitudinally, pixel connection judgment is performed on adjacent pixel rows in the reserved area to form a first crack curve; if the reserved area extends transversely, pixel connection judgment is performed on adjacent pixel columns in the reserved area to form a second crack curve;

[0011] The first crack curve and / or the second crack curve are mapped into the local image, and the local image is spliced ​​based on the overlapping area to obtain a bridge model with cracks.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: after obtaining the grayscale image, only the corresponding histogram image is locally equalized, avoiding the simultaneous enhancement of noise points and background outside the cracks in the grayscale image, and strengthening the contrast between the cracks and other areas; by judging the extension direction of the reserved area and combining its extension direction, the crack point judgment is performed on each pixel point in the reserved area respectively, thereby ensuring the continuity of the cracks and avoiding the generation of breakpoints, and the crack points are judged to exist inside the extracted reserved area, which can maximize the restoration of the actual cracks; by splicing several of the local images into the bridge model, the overall restoration of the bridge is completed, and at the same time, combined with the above steps, the cracks are restored in the local image, and the position of the cracks and the specific structure of the cracks can be intuitively grasped through the bridge model, replacing manual monitoring in an automated and intelligent manner, without the need for manual field monitoring, shortening the monitoring cycle, reducing monitoring errors, and improving monitoring efficiency.

[0013] Furthermore, the histogram includes a plurality of initial grayscale values ​​and the number of pixels corresponding to the initial grayscale values, and the step of performing local equalization processing on the histogram to obtain an enhanced image includes:

[0014] Obtaining a quantity change value of the number of pixels corresponding to the adjacent initial grayscale values, and determining a starting grayscale value and an end grayscale value from a plurality of the initial grayscale values ​​based on the quantity change value to obtain an adjustment interval;

[0015] Obtain the total number of pixels in the adjustment interval, determine the occurrence probability of the initial grayscale value in the adjustment interval through the number of pixels and the total number of pixels, obtain the cumulative probability through the occurrence probability, and convert the initial grayscale value in the adjustment interval into an adjustment grayscale value based on the occurrence probability and the cumulative probability.

[0016] Furthermore, the step of performing denoising on the binary image to obtain a standby image including a plurality of reserved areas includes:

[0017] Selecting a plurality of connected regions from the binary image based on pixel values ​​of pixels in the binary image;

[0018] The area of ​​the connected region is compared with an area threshold to separate the connected regions into noise regions and reserved regions, and the noise regions are removed to form a standby image including the reserved regions.

[0019] Furthermore, the step of determining whether the reserved area is extended vertically or horizontally based on the minimum circumscribed rectangle is specifically as follows:

[0020] The aspect ratio of the minimum circumscribed rectangle is obtained. If the aspect ratio of the minimum circumscribed rectangle is greater than 1, it is determined that the reserved area is extended vertically. If the aspect ratio of the minimum circumscribed rectangle is less than 1, it is determined that the reserved area is extended horizontally.

[0021] Furthermore, the step of performing pixel connection judgment on adjacent pixel rows in the reserved area to form a first crack curve is specifically:

[0022] Selecting the pixel row located at the bottom as a reference row, and selecting the pixel row located at the top as an extended row;

[0023] Respectively obtain a first central pixel point of the reference row and a second central pixel point of the extended row, and compare the x-axis coordinates of the first central pixel point and the second central pixel point to determine whether the crack direction is left or right;

[0024] If the crack runs to the right, the rightmost pixel in the extended row is obtained, a first x-axis coordinate difference between the first central pixel and the rightmost pixel is determined, and a crack curve point is determined in the reference row based on the first x-axis coordinate difference;

[0025] If the crack runs in a left direction, obtaining the leftmost pixel point in the extended row, determining the second x-axis coordinate difference between the first central pixel point and the leftmost pixel point, and determining the crack curve point in the reference row based on the second x-axis coordinate difference;

[0026] The crack curve points are connected to form a first crack curve.

[0027] Furthermore, the step of determining the crack curve point in the reference row based on the first x-axis coordinate difference includes:

[0028] Determine whether the first x-axis coordinate difference is greater than 1;

[0029] If the first x-axis coordinate difference is less than or equal to 1, the first central pixel point is determined as a crack curve point;

[0030] If the first x-axis coordinate difference is greater than 1, obtaining a plurality of left pixel points of the first central pixel point in the reference row, obtaining a third x-axis coordinate difference between the plurality of left pixel points and the first central pixel point, selecting a first marked pixel point from the plurality of left pixel points based on the third x-axis coordinate difference, and determining the first marked pixel point and the first central pixel point as crack curve points;

[0031] The step of determining the crack curve point in the reference row based on the second x-axis coordinate difference comprises:

[0032] Determine whether the second x-axis coordinate difference is greater than 1;

[0033] If the second x-axis coordinate difference is less than or equal to 1, the first central pixel point is determined as a crack curve point;

[0034] If the second x-axis coordinate difference is greater than 1, obtain several right side pixel points of the first center pixel point in the reference row, obtain the fourth x-axis coordinate difference between the several right side pixel points and the first center pixel point, select a second marked pixel point from the several right side pixel points based on the fourth x-axis coordinate difference, and determine the second marked pixel point and the first center pixel point as crack curve points.

[0035] Furthermore, the step of performing stitching processing on the local images based on the overlapping areas includes:

[0036] Performing feature extraction on the two partial images respectively to obtain a first geometric feature point and a second geometric feature point respectively;

[0037] Performing a similarity comparison between the first geometric feature point and the second geometric feature point to determine whether there is an overlapping area between the two partial images;

[0038] If there is an overlapping area between the two partial images, the position transformation relationship between the first geometric feature point and the second geometric feature point is obtained, and the two partial images are spliced ​​according to the position transformation relationship.

[0039] In a second aspect, an embodiment of the present application provides a system for three-dimensional visualization of bridge cracks, which is applied to the method for three-dimensional visualization of bridge cracks as described in the first aspect above, and the system includes:

[0040] A first processing module is used to obtain a plurality of local images corresponding to the bridge, wherein there are overlapping areas between the local images, perform grayscale processing on the local images to obtain grayscale images, obtain a histogram image corresponding to the grayscale image, and perform local equalization processing on the histogram image to obtain an adjusted histogram;

[0041] A second processing module is used to convert the adjusted histogram into an enhanced image, perform smoothing and threshold segmentation on the enhanced image to obtain a binary image, and perform denoising on the binary image to obtain a stand-by image including a plurality of reserved areas;

[0042] A judgment module is used to construct a coordinate system on the standby image to assign pixel coordinates to all pixel points in the reserved area, obtain a minimum bounding rectangle of the reserved area, and judge whether the reserved area extends vertically or horizontally based on the minimum bounding rectangle;

[0043] A connection module, configured to, if the reserved area extends longitudinally, perform pixel connection judgment on adjacent pixel rows in the reserved area to form a first crack curve; if the reserved area extends transversely, perform pixel connection judgment on adjacent pixel columns in the reserved area to form a second crack curve;

[0044] A construction module is used to map the first crack curve and / or the second crack curve into the local image, and to perform splicing processing on the local image based on the overlapping area to obtain a bridge model with cracks.

[0045] In a third aspect, an embodiment of the present application provides a computer, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for three-dimensional visualization of bridge cracks as described in the first aspect above is implemented.

[0046] In a fourth aspect, an embodiment of the present application provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for three-dimensional visualization of bridge cracks as described in the first aspect above. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flow chart of the method for three-dimensional visualization of bridge cracks in the first embodiment of the present invention;

[0048] Figure 2 It is a structural block diagram of a system for three-dimensional visualization of bridge cracks in a second embodiment of the present invention;

[0049] The following specific implementation manner will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0050] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0051] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0052] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0053] See also Figure 1 The method for three-dimensional visualization of bridge cracks provided by the first embodiment of the present invention comprises the following steps:

[0054] S10: obtaining a plurality of local images corresponding to the bridge, wherein there are overlapping areas between the local images, performing grayscale processing on the local images to obtain a grayscale image, obtaining a histogram image corresponding to the grayscale image, and performing local equalization processing on the histogram image to obtain an adjusted histogram;

[0055] In this embodiment, the drone travels along a preset route and captures the local images. Preferably, the overlapping area between the two local images is greater than 30%. The histogram includes a plurality of initial grayscale values ​​and the number of pixels corresponding to the initial grayscale values, that is, the histogram is a corresponding curve graph of the initial grayscale value and the number of pixels, the horizontal axis of which is the different initial grayscale values, and the vertical axis is the number of pixels corresponding to the initial grayscale value.

[0056] Specifically, the step S10 includes:

[0057] S110: Obtaining a quantity change value of the number of pixels corresponding to the adjacent initial grayscale values, and determining a starting grayscale value and an end grayscale value from a plurality of the initial grayscale values ​​based on the quantity change value to obtain an adjustment interval;

[0058] After the grayscale processing, there is a certain difference in the grayscale values ​​between the crack area and the background area. After being converted into a histogram, there is a sudden change in the number of pixels. For example, before the initial grayscale value is 100, the number of pixels corresponding to different initial grayscale values ​​is 2-5. When the initial grayscale value is 100-300, the number of pixels corresponding to different initial grayscale values ​​suddenly changes to 100-1000, and increases with the increase of the initial grayscale value. When the initial grayscale value is 300-500, the number of pixels corresponding to different initial grayscale values ​​suddenly changes to 100-1000, and decreases with the increase of the initial grayscale value. When the initial grayscale value is greater than 500, the number of pixels corresponding to different initial grayscale values ​​drops back to 2-5. If the histogram is directly equalized, the non-mutation area will be enhanced too much and the mutation area will decrease more, which reduces the contrast between the crack area and the background area and affects the accuracy of crack extraction. With the above distance, the grayscale value of the starting point is 100, the grayscale value of the end point is 500, and the adjustment interval is 100-500.

[0059] S120: Acquire the total number of pixels in the adjustment interval, determine the occurrence probability of the initial grayscale value in the adjustment interval according to the number of pixels and the total number of pixels, acquire the cumulative probability according to the occurrence probability, and convert the initial grayscale value in the adjustment interval into an adjustment grayscale value based on the occurrence probability and the cumulative probability;

[0060] The calculation formula of the occurrence probability is:

[0061] ,

[0062] in, represents the probability of occurrence of the i-th initial gray value, represents the number of pixels corresponding to the i-th initial gray value, Indicates the total number of pixels.

[0063] The calculation formula of the cumulative probability is:

[0064] ,

[0065] in, represents the cumulative probability, Indicates the number of initial grayscale values.

[0066] The calculation formula for adjusting the gray value is:

[0067] ,

[0068] in, represents the i-th adjusted gray value, Represents the i-th initial grayscale value.

[0069] S20: converting the adjusted histogram into an enhanced image, performing smoothing and threshold segmentation on the enhanced image to obtain a binary image, and performing denoising on the binary image to obtain a standby image including a plurality of reserved areas;

[0070] Based on the conversion relationship between the grayscale image and the histogram, the adjusted histogram is converted into the enhanced image. In this embodiment, the smoothing process is a bilateral mean filter process, and threshold segmentation of an adaptive threshold is performed by the Otus algorithm.

[0071] The step S20 comprises:

[0072] S210: Selecting a plurality of connected regions from the binary image based on pixel values ​​of pixels in the binary image;

[0073] After acquiring the binary image, the pixel values ​​of the pixels therein are 0 or 1. A certain pixel is selected as the initial pixel, and the pixel values ​​of the pixels adjacent to the initial pixel are retrieved to see whether they are the same, so as to form several initial connected areas. Generally, the pixel value of the background area in the binary image is 1. Therefore, the pixel values ​​of the pixels within the initial connected area are determined. If the pixel values ​​of the pixels within the initial connected area are 0, the initial connected area is determined as the connected area.

[0074] S220: Compare the area of ​​the connected region with an area threshold to separate the connected regions into noise regions and reserved regions, and remove the noise regions to form a standby image including the reserved regions.

[0075] After obtaining the connected area, there is still the noise area formed by the discrete noise points that have not been completely removed. In order to further reduce the impact of noise on crack acquisition and improve the accuracy of cracks, it is necessary to judge the area of ​​the connected area. If the area of ​​the connected area is less than the area threshold, the connected area is determined as the noise area and removed.

[0076] S30: constructing a coordinate system on the standby image, assigning pixel coordinates to all pixel points in the reserved area, obtaining a minimum bounding rectangle of the reserved area, and determining whether the reserved area extends vertically or horizontally based on the minimum bounding rectangle;

[0077] The contour area corresponding to the reserved area in the standby image can be extracted by an edge detection algorithm, and the minimum bounding rectangle of the contour area can be obtained. The edge detection algorithm and the acquisition of the minimum bounding rectangle have been widely used and will not be described here. It can be understood that after the coordinate system is constructed, each pixel point has an x-axis coordinate value and a y-axis coordinate value.

[0078] After obtaining the aspect ratio of the minimum circumscribed rectangle, if the aspect ratio of the minimum circumscribed rectangle is greater than 1, the reserved area is determined to be longitudinally extended, and if the aspect ratio of the minimum circumscribed rectangle is less than 1, the reserved area is determined to be horizontally extended.

[0079] S40: If the reserved area extends longitudinally, performing pixel connection judgment on adjacent pixel rows in the reserved area to form a first crack curve; if the reserved area extends transversely, performing pixel connection judgment on adjacent pixel columns in the reserved area to form a second crack curve;

[0080] The step S40 comprises:

[0081] S410: selecting the pixel row located at the bottom as a reference row, and selecting the pixel row located at the top as an extended row;

[0082] S420: respectively obtaining a first central pixel point of the reference row and a second central pixel point of the extended row, and comparing the x-axis coordinates of the first central pixel point and the second central pixel point to determine whether the crack direction is leftward or rightward;

[0083] S430: If the crack is in the right direction, obtain the rightmost pixel in the extended row, determine the first x-axis coordinate difference between the first central pixel and the rightmost pixel, and determine the crack curve point in the reference row based on the first x-axis coordinate difference;

[0084] Specifically, determine whether the first x-axis coordinate difference is greater than 1; if the first x-axis coordinate difference is less than or equal to 1, determine the first center pixel point as a crack curve point; if the first x-axis coordinate difference is greater than 1, obtain a number of left pixel points of the first center pixel point in the reference row, obtain a third x-axis coordinate difference between the several left pixel points and the first center pixel point, select a first marked pixel point from the several left pixel points based on the third x-axis coordinate difference, and determine the first marked pixel point and the first center pixel point as crack curve points.

[0085] S440: If the crack is in a left direction, obtain the leftmost pixel in the extended row, determine the second x-axis coordinate difference between the first central pixel and the leftmost pixel, and determine the crack curve point in the reference row based on the second x-axis coordinate difference;

[0086] Specifically, determine whether the second x-axis coordinate difference is greater than 1; if the second x-axis coordinate difference is less than or equal to 1, determine the first center pixel point as a crack curve point; if the second x-axis coordinate difference is greater than 1, obtain a number of right side pixel points of the first center pixel point in the reference row, obtain a fourth x-axis coordinate difference between the several right side pixel points and the first center pixel point, select a second marked pixel point from the several right side pixel points based on the fourth x-axis coordinate difference, and determine the second marked pixel point and the first center pixel point as crack curve points.

[0087] S450: Connect the crack curve points to form a first crack curve.

[0088] It should be noted that the steps for obtaining the second crack curve are basically the same as those for obtaining the first crack curve. The second crack curve obtains the crack curve points through the reference column and the extended column and the y-axis coordinates therebetween, which will not be repeated here.

[0089] S50: mapping the first crack curve and / or the second crack curve into the local image, and performing splicing processing on the local image based on the overlapping area to obtain a bridge model with cracks;

[0090] The step S50 comprises:

[0091] S510: performing feature extraction on the two partial images respectively to obtain a first geometric feature point and a second geometric feature point respectively;

[0092] In this embodiment, the feature extraction is performed using the SIFT algorithm.

[0093] S520: performing a similarity comparison between the first geometric feature point and the second geometric feature point to determine whether there is an overlapping area between the two partial images;

[0094] S530: If there is an overlapping area between the two partial images, obtain a position transformation relationship between the first geometric feature point and the second geometric feature point, and stitch the two partial images together according to the position transformation relationship.

[0095] After completing the stitching process, a bridge mapping model is obtained. In this embodiment, after obtaining the bridge mapping model, a first point cloud model corresponding to the bridge is obtained, the first point cloud model is projected from a three-dimensional coordinate system to a plane coordinate system to obtain a plane point cloud model, a plurality of interconnected triangular meshes are constructed in the plane point cloud model, and the triangular meshes are projected from the plane coordinate system to a three-dimensional coordinate system to obtain a second point cloud model, the boundary of the second point cloud model is optimized, and the second point cloud model after boundary optimization is filled into a solid model, the bridge attachment model is structurally compared with the solid model to determine the alignment relationship between the bridge attachment model and the solid model, and the bridge mapping model and the solid model are merged into the bridge model based on the alignment relationship.

[0096] After acquiring the grayscale image, only the corresponding histogram image is locally equalized to avoid enhancing the noise points and background outside the cracks in the grayscale image simultaneously, thereby enhancing the contrast between the cracks and other areas. By judging the extension direction of the reserved area and combining it with the extension direction, the crack point judgment is performed on each pixel point in the reserved area respectively, thereby ensuring the continuity of the cracks and avoiding breakpoints. The crack points are judged to exist inside the extracted reserved area, thereby restoring the actual cracks to the greatest extent. By splicing several local images into the bridge model, the overall restoration of the bridge is completed. At the same time, combined with the above steps, the cracks are restored in the local image. The position of the cracks and the specific structure of the cracks can be intuitively grasped through the bridge model, thereby replacing manual monitoring in an automated and intelligent manner. There is no need for manual field monitoring, thus shortening the monitoring cycle, reducing monitoring errors, and improving monitoring efficiency.

[0097] See also Figure 2 The second embodiment of the present invention provides a system for three-dimensional visualization of bridge cracks. The system is applied to the method for three-dimensional visualization of bridge cracks in the above-mentioned embodiment. The description has been made and will not be repeated here. As used below, the terms "module", "unit", "subunit", etc. can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0098] The system comprises:

[0099] The first processing module 10 is used to obtain a plurality of local images corresponding to the bridge, wherein there are overlapping areas between the local images, perform grayscale processing on the local images to obtain grayscale images, obtain a histogram image corresponding to the grayscale image, and perform local equalization processing on the histogram image to obtain an adjusted histogram;

[0100] The first processing module 10 comprises:

[0101] The first unit is used to obtain a quantity change value of the number of pixels corresponding to the adjacent initial grayscale values, and determine a starting grayscale value and an end grayscale value from a plurality of the initial grayscale values ​​based on the quantity change value to obtain an adjustment interval;

[0102] A second unit is used to obtain the total number of pixels in the adjustment interval, determine the occurrence probability of the initial grayscale value in the adjustment interval according to the number of pixels and the total number of pixels, obtain the cumulative probability according to the occurrence probability, and convert the initial grayscale value in the adjustment interval into an adjustment grayscale value based on the occurrence probability and the cumulative probability;

[0103] A second processing module 20 is used to convert the adjusted histogram into an enhanced image, perform smoothing and threshold segmentation on the enhanced image to obtain a binary image, and perform denoising on the binary image to obtain a standby image including a plurality of reserved areas;

[0104] The second processing module 20 includes:

[0105] A third unit is used to select a plurality of connected regions from the binary image based on the pixel values ​​of the pixels in the binary image;

[0106] The fourth unit is used to compare the area of ​​the connected region with the area threshold to separate the connected regions into noise regions and reserved regions, and remove the noise regions to form a standby image including the reserved regions;

[0107] A determination module 30 is used to construct a coordinate system on the standby image to assign pixel coordinates to all pixel points in the reserved area, obtain a minimum bounding rectangle of the reserved area, and determine whether the reserved area is extended vertically or horizontally based on the minimum bounding rectangle;

[0108] The judging module 30 comprises:

[0109] A fifth unit is used to obtain the aspect ratio of the minimum circumscribed rectangle, and if the aspect ratio of the minimum circumscribed rectangle is greater than 1, it is determined that the reserved area is extended vertically; if the aspect ratio of the minimum circumscribed rectangle is less than 1, it is determined that the reserved area is extended horizontally;

[0110] A connection module 40 is used for performing pixel connection judgment on adjacent pixel rows in the reserved area to form a first crack curve if the reserved area extends longitudinally; and performing pixel connection judgment on adjacent pixel columns in the reserved area to form a second crack curve if the reserved area extends transversely;

[0111] The connection module 40 includes:

[0112] A sixth unit, configured to select the pixel row located below as a reference row, and select the pixel row located above as an extended row;

[0113] A seventh unit is used to obtain the first central pixel point of the reference line and the second central pixel point of the extended line respectively, and compare the x-axis coordinates of the first central pixel point and the second central pixel point to determine whether the crack direction is left or right;

[0114] An eighth unit is used for obtaining the rightmost pixel point in the extended row if the crack is in the right direction, determining a first x-axis coordinate difference between the first central pixel point and the rightmost pixel point, and determining a crack curve point in the reference row based on the first x-axis coordinate difference;

[0115] The eighth unit is specifically used to determine whether the first x-axis coordinate difference is greater than 1; if the first x-axis coordinate difference is less than or equal to 1, the first center pixel point is determined as a crack curve point; if the first x-axis coordinate difference is greater than 1, a plurality of left pixel points of the first center pixel point in the reference row are obtained, a third x-axis coordinate difference between the plurality of left pixel points and the first center pixel point is obtained, a first marked pixel point is selected from the plurality of left pixel points based on the third x-axis coordinate difference, and the first marked pixel point and the first center pixel point are determined as crack curve points;

[0116] A ninth unit is configured to obtain the leftmost pixel point in the extended row if the crack is in a left direction, determine a second x-axis coordinate difference between the first central pixel point and the leftmost pixel point, and determine a crack curve point in the reference row based on the second x-axis coordinate difference;

[0117] The ninth unit is specifically used to determine whether the second x-axis coordinate difference is greater than 1; if the second x-axis coordinate difference is less than or equal to 1, the first center pixel point is determined as a crack curve point; if the second x-axis coordinate difference is greater than 1, a plurality of right side pixel points of the first center pixel point in the reference row are obtained, a fourth x-axis coordinate difference between the plurality of right side pixel points and the first center pixel point is obtained, a second marked pixel point is selected from the plurality of right side pixel points based on the fourth x-axis coordinate difference, and the second marked pixel point and the first center pixel point are determined as crack curve points;

[0118] a tenth unit, used for connecting the crack curve points to form a first crack curve;

[0119] A construction module 50 is used to map the first crack curve and / or the second crack curve into the local image, and perform splicing processing on the local image based on the overlapping area to obtain a bridge model with cracks;

[0120] The building block 50 comprises:

[0121] An eleventh unit is used to perform feature extraction on the two partial images respectively to obtain a first geometric feature point and a second geometric feature point respectively;

[0122] A twelfth unit is used to perform a similarity comparison between the first geometric feature point and the second geometric feature point to determine whether there is an overlapping area between the two partial images;

[0123] The thirteenth unit is used for obtaining the position transformation relationship between the first geometric feature point and the second geometric feature point if there is an overlapping area between the two local images, and splicing the two local images according to the position transformation relationship.

[0124] The present invention also provides a computer, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for three-dimensional visualization of bridge cracks as described in the above technical solution is implemented.

[0125] The present invention also provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the method for three-dimensional visualization of bridge cracks as described in the above technical solution is implemented.

[0126] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0127] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for three-dimensional visualization of bridge cracks, characterized in that: The following steps are involved: Acquire a plurality of local images corresponding to the bridge, wherein there are overlapping areas between the local images, perform grayscale processing on the local images to obtain grayscale images, acquire a histogram image corresponding to the grayscale image, and perform local equalization processing on the histogram image to obtain an adjusted histogram; Converting the adjusted histogram into an enhanced image, performing smoothing and threshold segmentation on the enhanced image to obtain a binary image, and performing denoising on the binary image to obtain a standby image including a plurality of reserved areas; Constructing a coordinate system on the stand-by image, assigning pixel coordinates to all pixel points in the reserved area, obtaining a minimum bounding rectangle of the reserved area, and determining whether the reserved area extends vertically or horizontally based on the minimum bounding rectangle; If the reserved area extends longitudinally, pixel connection judgment is performed on adjacent pixel rows in the reserved area to form a first crack curve; if the reserved area extends transversely, pixel connection judgment is performed on adjacent pixel columns in the reserved area to form a second crack curve; The step of performing pixel connection judgment on adjacent pixel rows in the reserved area to form a first crack curve is specifically as follows: Selecting the pixel row located at the bottom as a reference row, and selecting the pixel row located at the top as an extended row; Respectively obtain a first central pixel point of the reference row and a second central pixel point of the extended row, and compare the x-axis coordinates of the first central pixel point and the second central pixel point to determine whether the crack direction is left or right; If the crack runs to the right, the rightmost pixel in the extended row is obtained, a first x-axis coordinate difference between the first central pixel and the rightmost pixel is determined, and a crack curve point is determined in the reference row based on the first x-axis coordinate difference; If the crack runs in a left direction, obtaining the leftmost pixel point in the extended row, determining the second x-axis coordinate difference between the first central pixel point and the leftmost pixel point, and determining the crack curve point in the reference row based on the second x-axis coordinate difference; connecting the crack curve points to form a first crack curve; The first crack curve and / or the second crack curve are mapped into the local image, and the local image is spliced ​​based on the overlapping area to obtain a bridge model with cracks.

2. The method for three-dimensional visualization of bridge cracks according to claim 1, characterized in that: The histogram includes a plurality of initial grayscale values ​​and the number of pixels corresponding to the initial grayscale values. The step of performing local equalization processing on the histogram to obtain an enhanced image includes: Obtaining a quantity change value of the number of pixels corresponding to the adjacent initial grayscale values, and determining a starting grayscale value and an end grayscale value from a plurality of the initial grayscale values ​​based on the quantity change value to obtain an adjustment interval; Obtain the total number of pixels in the adjustment interval, determine the occurrence probability of the initial grayscale value in the adjustment interval through the number of pixels and the total number of pixels, obtain the cumulative probability through the occurrence probability, and convert the initial grayscale value in the adjustment interval into an adjustment grayscale value based on the occurrence probability and the cumulative probability.

3. The method for three-dimensional visualization of bridge cracks according to claim 1, characterized in that: The step of performing denoising on the binary image to obtain a standby image including a plurality of reserved areas comprises: Selecting a plurality of connected regions from the binary image based on pixel values ​​of pixels in the binary image; The area of ​​the connected region is compared with an area threshold to separate the connected regions into noise regions and reserved regions, and the noise regions are removed to form a standby image including the reserved regions.

4. The method for three-dimensional visualization of bridge cracks according to claim 1, characterized in that: The step of determining whether the reserved area is extended vertically or horizontally based on the minimum circumscribed rectangle is specifically as follows: The aspect ratio of the minimum circumscribed rectangle is obtained. If the aspect ratio of the minimum circumscribed rectangle is greater than 1, it is determined that the reserved area is extended vertically. If the aspect ratio of the minimum circumscribed rectangle is less than 1, it is determined that the reserved area is extended horizontally.

5. The method for three-dimensional visualization of bridge cracks according to claim 1, characterized in that: The step of determining a crack curve point in the reference row based on the first x-axis coordinate difference comprises: Determine whether the first x-axis coordinate difference is greater than 1; If the first x-axis coordinate difference is less than or equal to 1, the first central pixel point is determined as a crack curve point; If the first x-axis coordinate difference is greater than 1, obtaining a plurality of left pixel points of the first central pixel point in the reference row, obtaining a third x-axis coordinate difference between the plurality of left pixel points and the first central pixel point, selecting a first marked pixel point from the plurality of left pixel points based on the third x-axis coordinate difference, and determining the first marked pixel point and the first central pixel point as crack curve points; The step of determining the crack curve point in the reference row based on the second x-axis coordinate difference comprises: Determine whether the second x-axis coordinate difference is greater than 1; If the second x-axis coordinate difference is less than or equal to 1, the first central pixel point is determined as a crack curve point; If the second x-axis coordinate difference is greater than 1, obtain several right side pixel points of the first center pixel point in the reference row, obtain the fourth x-axis coordinate difference between the several right side pixel points and the first center pixel point, select a second marked pixel point from the several right side pixel points based on the fourth x-axis coordinate difference, and determine the second marked pixel point and the first center pixel point as crack curve points.

6. The method for three-dimensional visualization of bridge cracks according to claim 1, characterized in that: The step of performing stitching processing on the local images based on the overlapping areas comprises: Performing feature extraction on the two partial images respectively to obtain a first geometric feature point and a second geometric feature point respectively; Performing a similarity comparison between the first geometric feature point and the second geometric feature point to determine whether there is an overlapping area between the two partial images; If there is an overlapping area between the two partial images, the position transformation relationship between the first geometric feature point and the second geometric feature point is obtained, and the two partial images are spliced ​​according to the position transformation relationship.

7. A system for three-dimensional visualization of bridge cracks, applied to the method for three-dimensional visualization of bridge cracks as claimed in any one of claims 1 to 6, characterized in that: The system comprises: A first processing module is used to obtain a plurality of local images corresponding to the bridge, wherein there are overlapping areas between the local images, perform grayscale processing on the local images to obtain grayscale images, obtain a histogram image corresponding to the grayscale image, and perform local equalization processing on the histogram image to obtain an adjusted histogram; A second processing module is used to convert the adjusted histogram into an enhanced image, perform smoothing and threshold segmentation on the enhanced image to obtain a binary image, and perform denoising on the binary image to obtain a stand-by image including a plurality of reserved areas; A judgment module is used to construct a coordinate system on the standby image to assign pixel coordinates to all pixel points in the reserved area, obtain a minimum bounding rectangle of the reserved area, and judge whether the reserved area extends vertically or horizontally based on the minimum bounding rectangle; A connection module, configured to, if the reserved area extends longitudinally, perform pixel connection judgment on adjacent pixel rows in the reserved area to form a first crack curve; if the reserved area extends transversely, perform pixel connection judgment on adjacent pixel columns in the reserved area to form a second crack curve; The connection module comprises: A sixth unit, configured to select the pixel row located below as a reference row, and select the pixel row located above as an extended row; A seventh unit is used to obtain the first central pixel point of the reference line and the second central pixel point of the extended line respectively, and compare the x-axis coordinates of the first central pixel point and the second central pixel point to determine whether the crack direction is left or right; An eighth unit is used for obtaining the rightmost pixel point in the extended row if the crack is in the right direction, determining a first x-axis coordinate difference between the first central pixel point and the rightmost pixel point, and determining a crack curve point in the reference row based on the first x-axis coordinate difference; A ninth unit is configured to obtain the leftmost pixel point in the extended row if the crack is in a left direction, determine a second x-axis coordinate difference between the first central pixel point and the leftmost pixel point, and determine a crack curve point in the reference row based on the second x-axis coordinate difference; a tenth unit, used for connecting the crack curve points to form a first crack curve; A construction module is used to map the first crack curve and / or the second crack curve into the local image, and to perform splicing processing on the local image based on the overlapping area to obtain a bridge model with cracks.

8. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for three-dimensional visualization of bridge cracks according to any one of claims 1 to 6 is implemented.

9. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for three-dimensional visualization of bridge cracks according to any one of claims 1 to 6 is implemented.

Images