A road crack morphology fitting, analysis and complexity assessment method

Abstract

The present invention discloses a method for fitting, analyzing, and assessing the complexity of road crack morphology. The method uses ground-penetrating radar to scan the road surface to obtain a grayscale image of the horizontal cross-section of the crack, and then uses a semantic segmentation method to obtain a binary image of the crack. The binary image of the horizontal cross-section of the road is obtained, and then an opening operation is used to filter noise, and a closing operation is used to connect the crack areas. Contour detection is also used to separate different crack areas and process them separately. After obtaining the preprocessed image, the crack area is downsampled to extract the crack area skeleton, and feature points are identified on the skeleton. The feature points are connected using a minimum spanning tree, and finally further simplified to obtain an undirected graph that can represent the road crack morphology. The edge orientation and node degree of the undirected graph are counted to determine the shape of the crack, and the complexity of the crack morphology is assessed by the complexity of the undirected graph. The present invention can be used to guide road maintenance personnel to handle crack diseases according to complexity priority.

Description

Technical Field

[0001] The present invention relates to the field of computer graphics, and in particular to a method for road crack morphology fitting, analysis and complexity evaluation. Background Art

[0002] During road maintenance, road cracks are a significant traffic safety hazard that can lead to serious accidents. Therefore, assessing road crack damage is crucial. With the advancement of radar detection and digital image processing technologies, video-based road crack assessment technology offers promising economic benefits and development potential. Traditional methods for evaluating damage image processing, both domestically and internationally, rely primarily on traditional manual methods. These methods determine crack information through on-site or vehicle-mounted imagery. While this method requires minimal technical expertise, it suffers from high labor costs, low information yield, and unstable information acquisition, making quantitative analysis of road surface damage impossible.

[0003] Currently, the main approaches for analyzing road crack morphology are to use radar detection and semantic segmentation techniques to quickly and effectively extract binary images of crack areas. However, these methods can only determine whether cracks exist in the area, but cannot assess the crack's morphology or complexity. Summary of the Invention

[0004] The purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art and to provide a method for fitting, analyzing and evaluating the complexity of road crack morphology, so as to solve the problem that traditional detection methods are unable to analyze and evaluate the complexity of crack morphology.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] A road crack morphology fitting, analysis and complexity assessment method includes the following steps:

[0007] Step 1: Use opening operation on the road crack binary image to filter out the noise area;

[0008] Step 2: Use a closing operation on the road crack binary map, and the connected areas with a splicing distance less than a set threshold are regarded as the same crack area of ​​the binary map;

[0009] Step 3: Detect the contours of all connected regions as crack regions and separate each connected region to ensure that the subsequent algorithm only processes a single connected region;

[0010] Step 4: For a single connected region, downsample the pixels in the region and use the medial axis transform to refine the connected region to obtain the internal skeleton of the connected region.

[0011] Step 5: Detect the feature points of the connected region skeleton;

[0012] Step 6: Determine whether all feature points are interconnected. If so, use the minimum spanning tree to connect the feature points. If there are any disconnected feature points, use the minimum spanning tree to connect the feature points for each connected component to obtain a spanning forest. Finally, merge the spanning forests to form a connected graph and obtain a new minimum spanning tree.

[0013] Step 7: Depth-traverse the new minimum spanning tree and simplify it to obtain an undirected graph that can describe the road crack morphology analysis and complexity;

[0014] Step 8: Count the degree of each node and the direction of each edge in the undirected graph to determine the type of road cracks.

[0015] In step 9, the complexity of the road cracks can be quantified by counting the number of edges, the number of joints, and the weighted average of the orientations of all edges in the undirected graph.

[0016] In step 4, the medial axis transformation is a process of continuously eliminating pixels. The specific process is as follows:

[0017] First, define the following:

[0018] Traverse the binary graph P of a single connected area, where all the pixels px of the binary graph P i,j The value is 0, 1; traverse to px i,j When judging px i,j Do you need to delete it? i,j The 8 neighborhoods of are marked in clockwise order as follows:

[0019] Seq={px i-1,j-1 ,px i,j-1 ,px i+1,j-1 ,px i+1,j ,px i+1,j+1 ,px i,j+1 ,px i-1,j+1 ,px i-1,j};

[0020] Let neighbor(px i,j ) is px i,j The sum of the pixel values ​​in the 8-neighborhood of curcle(px i,j ) is px i,j The 8-neighborhood of circle(px) is the number of sequence 0-1 in the order of the mark. The number of sequence 0-1 is 2, so circle(px i,j )=2, the summary is:

[0021] neighbor(px i,j )=∑ m,n＝-1,0,1 px i+m,j+n -px i,j ;

[0022]

[0023] Traverse each point of the image, for each pixel px is 1 i,j Make the following judgment:

[0024] 1)2≤neighbor(px i,j )≤6;

[0025] 2)circle(px i,j )=1;

[0026] 3) When it is an odd number of iterations, judge px i,j-1 ×px i+1,j ×px i,j+1 =0, px i+1,j ×px i,j+1 ×px i-1,j = 0 are both true, if so, condition 3 is satisfied);

[0027] When it is an even number of iterations, judge px i,j-1 ×px i+1,j ×px i-1,j =0, px i,j-1 ×px i,j+1 ×px i-1,j = 0 are both true, if so, condition 3 is satisfied);

[0028] If conditions 1), 2), and 3) are met at the same time, the pixel is deleted; the image is traversed until no pixel is deleted, and the medial axis transformation process is completed, and finally a skeleton binary image of the refined crack structure is obtained.

[0029] Step 5: detecting the feature points of the connected region skeleton in the following manner:

[0030] A fixed window is used to slide in any direction on the skeleton binary image of the crack structure, and the degree of change of the pixel grayscale in the window before and after sliding is compared: if there is sliding in any direction and the grayscale change value is greater than the set value, it is considered that there is a corner feature point in the window.

[0031] In step 6, assume that the fracture structure graph G has n connected components, SG = (V, E) is a connected component of G, and T = (U, TE) is the minimum spanning tree of SG, where V and E represent the vertex set and edge set of the connected component, respectively, and U and TE represent the vertex set and edge set of the corresponding minimum spanning tree, respectively; then the prim algorithm steps for constructing the minimum spanning tree T starting from vertex v using SG are as follows:

[0032] (1) Initialize U = {v}, V = VU;

[0033] (2) Repeat the following steps to add all nodes in V to U:

[0034] ①Select node u from U * , select node v from V * , making In U and V, select one node from each set and connect them with the edge with the smallest weight;

[0035]

[0036] ② General Join TE:

[0037]

[0038] ③V * Join U:

[0039] U=U∪{v *};

[0040] ④ V * Remove V:

[0041] V=V-{v *};

[0042] The minimum spanning tree T of the connected component SG can be obtained; using the prim algorithm for each connected component, the minimum spanning forest F = {T i |1≤i≤n}.

[0043] In step 6, the forests are merged to form a connected graph, and a new minimum spanning tree is obtained, as follows:

[0044] First, all edges of the connected components are regarded as pixel sets, and the shortest distance between all pixel sets is calculated;

[0045] Secondly, the endpoints of the connecting line segments corresponding to the shortest distance are added as new nodes to the connected components;

[0046] Finally, the connecting line segments are used as new edges to reorganize multiple connected components into a connected graph.

[0047] In step 7, the new minimum spanning tree is deeply traversed and simplified, specifically:

[0048] (1) For the crack structure graph G, select the node v with degree 1 start Perform a depth-first traversal for the starting node;

[0049] (2) If the depth-first traversal reaches the current node v currentThe degree d(v current )≠2, then keep v current ;

[0050] (3) If the depth-first traversal reaches the current node v current The degree d(v current )=2, indicating that v current There are 2 adjacent edges e pre ,e next , and two adjacent nodes v pre ,v next , where v pre is the node that has been visited by the depth-first traversal before, v next is the next node to be visited, and

[0051] e pre =(v pre ,v current )#(3-27)

[0052] e next =(v current ,v next )#(3-28)

[0053] Determine the two edges e pre ,e next If the angle between That is e pre ,e next Angle Then retain v current and continue traversing; otherwise delete v current , while connecting v pre ,v next , continue to traverse;

[0054] After depth-first traversal of all nodes, the final simplified result is obtained.

[0055] At the same time, the present invention provides:

[0056] A server includes a processor and a memory, wherein the memory stores at least one program, and the program is loaded and executed by the processor to implement the above-mentioned road crack morphology fitting, analysis and complexity assessment method.

[0057] A computer-readable storage medium stores at least one program, which is loaded and executed by a processor to implement the above-mentioned road crack morphology fitting, analysis and complexity assessment method.

[0058] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0059] This invention provides an algorithm that can quickly and effectively assess crack morphology and complexity. To quantitatively analyze crack development within roads and determine their morphology and complexity, the invention uses segmented straight lines to fit the crack morphology. The length and angle of the lines represent the size and orientation of the crack area, and thus the structural morphology of these cracks. This can guide road maintenance personnel in addressing crack damage based on priority of complexity. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is the preprocessing flow chart of the road crack binary image.

[0061] Figure 2 Flowchart for crack structure fitting of road cracks.

[0062] Figure 3 px i,j Schematic diagram of the 8-neighborhood.

[0063] Figure 4 px i,j Example graph of an 8-neighborhood 0-1 sequence.

[0064] Figure 5 Schematic diagram of feature point evaluation.

[0065] Figure 6 Schematic diagram of sliding window feature point detection.

[0066] Figure 7 Schematic diagram of feature points after connected area refinement.

[0067] Figure 8 G is a schematic diagram of a connected graph.

[0068] Figure 9 G is a schematic diagram of a non-connected graph.

[0069] Figure 10 Schematic diagram of connecting a non-connected graph into a connected graph.

[0070] Figure 11 is the histogram of crack complexity C. DETAILED DESCRIPTION

[0071] The present invention will be described in further detail below with reference to the embodiments and drawings, but the embodiments of the present invention are not limited thereto.

[0072] A method for fitting, analyzing, and evaluating the complexity of road crack morphology uses ground-penetrating radar to scan the road surface to obtain a grayscale image of the crack's horizontal cross-section, and then uses a semantic segmentation method to obtain a binary image of the crack. The binary image of the road's horizontal cross-section is then filtered using an opening operation, connected using a closing operation, and contour detection is used to separate and process different crack regions. After obtaining the preprocessed image, the crack region is downsampled to extract the crack region skeleton, feature points are identified on the skeleton, and the feature points are connected using a minimum spanning tree. Finally, further simplification is performed to obtain an undirected graph that can represent the road crack morphology. The edge directions and node degrees of the undirected graph are counted to determine the crack shape. The complexity of the undirected graph can also be used to evaluate the complexity of the crack morphology.

[0073] Specifically, if Figure 1 、 2 , a road crack morphology fitting, analysis and complexity evaluation method, comprising the following steps:

[0074] Step 1: Use opening operation on the road crack binary image to filter out the noise area;

[0075] Step 2: Use a closing operation on the road crack binary map, and the connected areas with a splicing distance less than a set threshold are regarded as the same crack area of ​​the binary map;

[0076] Step 3: Detect the contours of all connected regions as crack regions and separate each connected region to ensure that the subsequent algorithm only processes a single connected region;

[0077] Step 4: For a single connected region, downsample the pixels in the region and use the medial axis transform to refine the connected region to obtain the internal skeleton of the connected region.

[0078] Step 5: Detect the feature points of the connected region skeleton;

[0079] Step 6: Determine whether all feature points are interconnected. If so, use the minimum spanning tree to connect the feature points. If there are any disconnected feature points, use the minimum spanning tree to connect the feature points for each connected component to obtain a spanning forest. Finally, merge the spanning forests to form a connected graph and obtain a new minimum spanning tree.

[0080] Step 7: Depth-traverse the new minimum spanning tree and simplify it to obtain an undirected graph that can describe the road crack morphology analysis and complexity;

[0081] Step 8: Count the degree of each node and the direction of each edge in the undirected graph to determine the type of road cracks.

[0082] In step 9, the complexity of the road cracks can be quantified by counting the number of edges, the number of joints, and the weighted average of the orientations of all edges in the undirected graph.

[0083] In step 4, the medial axis transformation is a process of continuously eliminating pixels. The specific process is as follows:

[0084] First, define the following:

[0085] Traverse the binary graph P of a single connected area, where all the pixels px of the binary graph P i,j The value is 0, 1; traverse to px i,j When judging px i,j Do you need to delete it? i,j The 8 neighborhoods are respectively Figure 3 The clockwise order is:

[0086] Seq={px i-1,j-1 ,px i,j-1 ,px i+1,j-1 ,px i+1,j ,px i+1,j+1 ,px i,j+1 ,px i-1,j+1 ,px i-1,j};

[0087] Let neighbor(px i,j ) is px i,j The sum of the pixel values ​​in the 8-neighborhood of circle(px i,j ) is px i,j The number of 8-neighborhoods in the order of marking 0-1; Figure 4 As shown, the number of sequential sequences 0-1 is 2, so circle(px i,j )=2, the summary is:

[0088] neighbor(px i,j )=∑ m,n＝-1,0,1 px i+m,j+n -px i,j ;

[0089]

[0090] Traverse each point of the image, for each pixel px is 1 i,j Make the following judgment:

[0091] 1)2≤neighbor(px i,j )≤6;

[0092] 2)circle(px i,j )=1;

[0093] 3) When it is an odd number of iterations, judge px i,j-1 ×px i+1,j ×px i,j+1 =0, px i+1,j ×px i,j+1 ×px i-1,j = 0 are both true, if so, condition 3 is satisfied);

[0094] When it is an even number of iterations, judge px i,j-1 ×px i+1,j ×px i-1,j =0, px i,j-1 ×px i,j+1 ×px i-1,j = 0 are both true, if so, condition 3 is satisfied);

[0095] If conditions 1), 2), and 3) are met at the same time, the pixel is deleted; the image is traversed until no pixel is deleted, and the medial axis transformation process is completed, and finally a skeleton binary image of the refined crack structure is obtained.

[0096] However, the internal skeleton obtained by medial axis transformation only contains pixel information and cannot quantitatively analyze the shape and complexity of the crack based on the skeleton. Therefore, it is necessary to find the characteristic points of the skeleton to reflect the spatial structural information of the crack.

[0097] Step 5: detecting the feature points of the connected region skeleton in the following manner:

[0098] A fixed window is used to slide in any direction on the skeleton binary image of the crack structure, and the degree of change of the pixel grayscale in the window before and after sliding is compared: if there is sliding in any direction and the grayscale change value is greater than the set value, it is considered that there is a corner feature point in the window.

[0099] For the window to be detected, its strength is defined as:

[0100] E(u,v)=∑ x,y W(x,y)[P(x+u,y+v)-P(x,y)] 2 ;

[0101] Where (x, y) is the pixel point in the window, W(x, y) is the window function, and (u, v) is the minimum displacement; the window function is a uniform weight window or a Gaussian weight window, which provides weights for the pixels in the window; the larger E(u, v), the greater the difference in pixel value between the point (x, y) and its neighboring point (x+u, y+v), that is, it is necessary to detect corner points by maximizing this error function;

[0102] Applying Taylor expansion, we can get:

[0103]

[0104] Where, P x Represents the partial derivative of the image matrix P in the x direction;

[0105] The Taylor expansion formula can be obtained:

[0106]

[0107] in

[0108]

[0109] By simply analyzing the eigenvalues ​​of the matrix M, the change in pixel value can be scored; let the score

[0110] R = min(λ1,λ2);

[0111] In the formula, λ1,λ2 are the eigenvalues ​​of M, and the derivative of M in the (x,y) direction is measured by the size of λ1,λ2; Figure 5 As shown, when λ1,λ2 are higher than the minimum value λ min When λ1 or λ2 is below the minimum value λ min When λ1 and λ2 are both lower than the minimum value λ min When , it is considered as a flat area (blue area); select λ1,λ2 higher than the minimum value λ min The points at which are found can be used to obtain the feature points.

[0112] Correspondingly, a sliding window is used to detect feature points, and λ1 and λ2 after sliding are compared to determine whether the center of the window is a feature point; Figure 6 shown.

[0113] Compare the refined images to identify feature points, such as Figure 7 As shown, the green points are the identified feature points;

[0114] After detecting these feature points, a set of line segments passing through the feature points can be determined to represent the shape of the connected area.

[0115] In step 6, based on the feature points, an undirected graph G = (V, E) is constructed to represent all possible connection modes of these feature points. Each feature point corresponds to a node v in the crack structure graph G. i ∈V. Since there are n(n-1) / 2 ways to connect n nodes, it is more appropriate to use the adjacency matrix to represent the undirected graph G. For each pair of nodes v i ,v j , check the undirected edge e connecting the node pair i,j =(vi ,v j ) corresponds to the line segment pixel set line(v i ,v j ) is inside the connected area D, that is, to judge Among them, line(v i ,v j )={pt|cos(v j -pt,pt-v i )=1}.

[0116] If e i,j Inside the connected region,

[0117] dist(v i ,v j )=||v j -v i ||2;

[0118] weight(e i,j )=dist(v i ,v j );

[0119] otherwise

[0120] weight(e i,j )=INF;

[0121] Among them, dist(v i ,v j ) represents node v i ,v j Euclidean distance, weight(e i,j ) represents the edge e in the adjacency matrix i,j The weight of e i,j =(v i ,v j ) has a weight of INF, which means v i ,v j There are no connecting edges between them. Finally, all possible connection modes G of the feature points are obtained. However, since the feature point recognition cannot be guaranteed to be completely correct, even when the crack area is a single connected area, it is still possible that some feature points are completely disconnected from other feature points, that is, G is a non-connected graph. Figure 8 、 9 shown.

[0122] In order to determine the most appropriate connection method required between feature points, a minimum spanning tree (MST) is found from the constructed undirected graph G.

[0123] If G is a disconnected graph, the MST is found independently for each connected component to obtain a spanning forest. The MST is the spanning tree with the minimum total weight, so the edges of the MST can connect feature points with the minimum redundancy of the same-direction line segments.

[0124] Assume that the fracture structure graph G has n connected components, SG = (V, E) is a connected component of G, and T = (U, TE) is the minimum spanning tree of SG, where V and E represent the vertex set and edge set of the connected component, respectively, and U and TE represent the vertex set and edge set of the corresponding minimum spanning tree, respectively. The steps of the prim algorithm for constructing the minimum spanning tree T starting from vertex v using SG are as follows:

[0125] (1) Initialize U = {v}, V = VU;

[0126] (2) Repeat the following steps to add all nodes in V to U:

[0127] ①Select node u from U * , select node v from V * , making In U and V, select one node from each set and connect them with the edge with the smallest weight;

[0128]

[0129] ② General Join TE:

[0130]

[0131] ③V * Join U:

[0132] U=U∪{v *};

[0133] ④ V * Remove V:

[0134] V=V-{v *};

[0135] The minimum spanning tree T of the connected component SG can be obtained; using the prim algorithm for each connected component, the minimum spanning forest F = {T i |1≤i≤n}.

[0136] In step 6, the forest is merged to form a connected graph and a new minimum spanning tree is obtained, specifically as follows: since each crack corresponds to a connected area, its structural graph should not be a disconnected graph, so it is necessary to connect each connected component again to form a connected graph.

[0137] First, all edges of the connected components are regarded as pixel sets, and the shortest distance between all pixel sets is calculated;

[0138] Secondly, the endpoints of the connecting line segments corresponding to the shortest distance are added as new nodes to the connected components;

[0139] Finally, the connected components are reorganized into a connected graph, using the connecting line segments as new edges. Essentially, this is equivalent to treating each connected component of the spanning forest as a node and using the prim algorithm again to find the minimum spanning tree for each connected component.

[0140] The result of connecting multiple connected components into a connected graph is as follows Figure 10 shown.

[0141] In step 7, the new minimum spanning tree is deeply traversed and simplified, specifically:

[0142] Since the crack structure graph G obtained by simply connecting feature points using the minimum spanning tree is rather complex, and many edges actually represent the same extension direction, the following algorithm is used to simplify the crack structure graph:

[0143] (1) For the crack structure graph G, select the node v with degree 1 start Perform a depth-first traversal (DFS) for the starting node;

[0144] (2) If the depth-first traversal reaches the current node v current The degree d(v current )≠2, then keep v current ;

[0145] (3) If the depth-first traversal reaches the current node v current The degree d(v current )=2, indicating that v current There are 2 adjacent edges e pre ,e next , and two adjacent nodes v pre ,v next , where v pre is the node that has been visited by the depth-first traversal before, v next is the next node to be visited, and

[0146] e pre =(v pre ,v current )#(3-27)

[0147] e next =(v current ,v next )#(3-28)

[0148] Determine the two edges epre , e next The included angle between, if That is, e pre , e next The included angle of Then retain v current And continue to traverse; otherwise, delete v current , and at the same time connect v pre , v next , and continue to traverse;

[0149] After depth - first traversing all nodes, the final simplified result is obtained.

[0150] After the above operations, the finally simplified crack structure diagram is obtained.

[0151] Through the crack structure diagram G(V, E), the characteristics of the crack structure such as the orientation, the number of joint points, the number of segments, and the total length of the edges can be obtained. Among them, the joint point refers to the node with a degree greater than 2. The joint point connects multiple edges, making G have a more complex structure and being more important than other nodes. These characteristics objectively reflect the complexity of the crack. Intuitively, the greater the difference in the orientation of the crack structure diagram, the more joint points, the more segments, and the greater the total length of the edges, the more complex the crack morphology is.

[0152] Let |E| represent the number of edges of G, J represent the number of joint points of G, and L represent the total length of all edges. The orientation of the crack structure can be represented by the set of included angles Θ = {θ i |0 < i < |E|} of all edges with the positive x - axis. Calculate the standard deviation σ(Θ) of Θ, which objectively reflects the orientation complexity of G. Use the vector to measure the complexity of the crack morphology. Since the weights of the components of are not equal, the components of are standardized according to the formula to eliminate the dimensional gap of the components. Suppose there are n crack samples in total, i,j represents the complexity vector of the i - th sample, and the components are c

[0153]

[0154] To obtain the standardized vector The L2 norm of is used to evaluate the crack complexity C. The histogram of the crack complexity of the obtained test data set is as Figure 11 shown, and the mean and variance of each component are shown in Table 1.

[0155] Table 1 Mean and variance table of edge orientation standard deviation, number of joint points, number of edges, and total length of edges

[0156] Number of pictures 2000 Total number of cracks 2897 <![CDATA[The mean μ1 of σ(Θ)]]> 0.1854 <![CDATA[Standard deviation σ(Θ), σ1]]> 0.1628 <![CDATA[J mean μ2]]> 0.1346 <![CDATA[J standard deviation σ2]]> 0.4068 <![CDATA[L mean μ3]]> 149.9493 <![CDATA[L standard deviation σ3]]> 145.4438 <![CDATA[|E| Mean μ4]]> 13.2506 <![CDATA[|E| Standard Deviation σ4]]> 11.4275

[0157] Figure 11 In , the horizontal axis represents the crack complexity, and the vertical axis represents the number of corresponding complexity intervals. Figure 11 It can be observed that most of the cracks have low complexity, with 80.98% of the cracks having complexity C∈[0,2]. This indicates that most of the cracks in this dataset are not very complex.

[0158] Based on the resulting undirected graph G, a parameterized fracture structure morphology model can be easily derived, allowing the fracture shape category and complexity to be determined. Crack shapes are categorized as linear, curved, and tree-shaped. If the crack's development direction angle remains constant and there are no bifurcations, it is classified as a linear crack. If the crack's development direction angle changes and there are no bifurcations, it is classified as a curved crack. If there is a bifurcation in the crack's development direction, it is classified as a tree-shaped crack. Cracks that cannot be categorized as any of the above are classified as other types.

[0159] First, a depth-first traversal is performed on the structural graph G corresponding to the crack, counting the degrees of each node in G. If the degree of all nodes is less than or equal to 2, the crack is either linear or curved, meaning that G is a unary tree. If G contains a unique node with a degree greater than or equal to 3, the crack is a tree-shaped crack; otherwise, it is a general crack. Furthermore, a depth-first traversal counts the angles between adjacent edges of each node in the unary tree G to further distinguish between linear and curved cracks. If the angle between any two edges does not exceed a specified angle, the crack is linear. Otherwise, it is curved.

[0160] Furthermore, linear cracks can be divided into transverse cracks and longitudinal cracks.

[31] , where the transverse cracks are basically perpendicular to the road centerline, and the longitudinal cracks are basically parallel to the route direction; according to the number of nodes with a degree of 2 and adjacent edges with a large angle, the curved cracks can be further subdivided into L-shaped cracks, C-shaped cracks and other curved cracks; similarly, according to the number of nodes with a degree of 3 or 4, the tree-shaped cracks can be further subdivided into T-shaped cracks, cross-shaped cracks and other tree-shaped cracks.

[0161] In real-world scenarios, because the crack structure maps generated by MSTF can quantitatively analyze the crack morphology, the number of crack morphology categories can be customized based on needs by simply specifying the geometric properties corresponding to the desired morphology. This eliminates the need for relabeling crack shape categories and retraining machine learning methods like neural networks and support vector machines.

[0162] The crack shapes of the test dataset are classified, and the results are shown in Table 2.

[0163] Table 2 Crack shape quantity table

[0164]

[0165] As shown in Table 2, simple transverse and longitudinal cracks account for 28.53% and 48.49% of the total, respectively, far exceeding the remaining crack types. This is consistent with the conclusion of this example's crack complexity assessment: most cracks are of low complexity. In other words, from a purely morphological and structural perspective, the majority of cracks in this dataset are of low severity.

[0166] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be considered as equivalent replacement methods and are included in the scope of protection of the present invention.

Claims

1. A road crack morphology fitting, analysis and complexity assessment method, characterized by: The following steps are involved: Step 1: Use opening operation on the road crack binary image to filter out the noise area; Step 2: Use a closing operation on the road crack binary map, and the connected areas with a splicing distance less than a set threshold are regarded as the same crack area of ​​the binary map; Step 3: Detect the contours of all connected regions as crack regions and separate each connected region to ensure that the subsequent algorithm only processes a single connected region; Step 4: For a single connected region, downsample the pixels in the region and use the medial axis transform to refine the connected region to obtain the internal skeleton of the connected region. Step 5: Detect the feature points of the connected region skeleton; Step 6: Determine whether all feature points are interconnected. If so, use the minimum spanning tree to connect the feature points. If there are any disconnected feature points, use the minimum spanning tree to connect the feature points for each connected component to obtain a spanning forest. Finally, merge the spanning forests to form a connected graph and obtain a new minimum spanning tree. Step 7: Depth-traverse the new minimum spanning tree and simplify it to obtain an undirected graph that can describe the road crack morphology analysis and complexity; Step 8: Count the degree of each node and the direction of each edge in the undirected graph to determine the type of road cracks. In step 9, the complexity of the road cracks can be quantified by counting the standard deviation of the edge orientation, the number of joints, the number of edges, and the total length of the edges in the undirected graph.

2. The method for road crack morphology fitting, analysis and complexity assessment according to claim 1, characterized in that: In step 4, the medial axis transformation is a process of continuously eliminating pixels. The specific process is as follows: First, define the following: Traverse the binary graph P of a single connected area, where all the pixels px of the binary graph P i,j The value is 0, 1; traverse to px i,j When judging px i,j Do you need to delete it? i,j The 8 neighborhoods of are marked in clockwise order as follows: Seq={px i-1,j-1 ,px i,j-1 ,px i+1,j-1 ,px i+1,j ,px i+1,j+1 ,px i,j+1 ,px i-1,j+1 ,px i-1,j }; Let neighbor(px i,j ) is px i,j The sum of the pixel values ​​in the 8-neighborhood of circle(px i,j ) is px i,j The number of 8-neighborhood sequences 0-1 in the order of marking; Traverse each point of the image, for each pixel px is 1 i,j Make the following judgment: 1)2≤neighbor(px i,j )≤6; 2)circle(px i,j )=1; 3) When it is an odd number of iterations, judge px i,j-1 ×px i+1,j ×px i,j+1 =0, px i+1,j ×px i,j+1 ×px i-1,j = 0 are both true, if so, condition 3 is satisfied); When it is an even number of iterations, judge px i,j-1 ×px i+1,j ×px i-1,j =0, px i,j-1 ×px i,j+1 ×px i-1,j = 0 are both true, if so, condition 3 is satisfied); If conditions 1), 2), and 3) are met at the same time, the pixel is deleted; the image is traversed until no pixel is deleted, and the medial axis transformation process is completed, and finally a skeleton binary image of the refined crack structure is obtained.

3. The road crack morphology fitting, analysis and complexity assessment method according to claim 1, characterized in that: Step 5: detecting the feature points of the connected region skeleton in the following manner: A fixed window is used to slide in any direction on the skeleton binary image of the crack structure, and the degree of change of the pixel grayscale in the window before and after sliding is compared: if there is sliding in any direction and the grayscale change value is greater than the set value, it is considered that there is a corner feature point in the window.

4. The method for road crack morphology fitting, analysis and complexity assessment according to claim 1, characterized in that: In step 6, assume that the fracture structure graph G has n connected components, SG = (V, E) is a connected component of G, and T = (U, TE) is the minimum spanning tree of SG, where V and E represent the vertex set and edge set of the connected component, respectively, and U and TE represent the vertex set and edge set of the corresponding minimum spanning tree, respectively; then the prim algorithm steps for constructing the minimum spanning tree T starting from vertex v using SG are as follows: (1) Initialize U = {v}, V = VU; (2) Repeat the following steps to add all nodes in V to U: ①Select node u from U * , select node v from V * , making In U and V, select one node from each set and connect them with the edge with the smallest weight; ② General Join TE: ③V * Join U: U=U∪{v * }; ④ V * Remove V: V=V-{v * }; The minimum spanning tree T of the connected component SG can be obtained; using the prim algorithm for each connected component, the minimum spanning forest F = {T i |1≤i≤n}.

5. The method for road crack morphology fitting, analysis and complexity assessment according to claim 1, characterized in that: In step 6, the forests are merged to form a connected graph, and a new minimum spanning tree is obtained, as follows: First, all edges of the connected components are regarded as pixel sets, and the shortest distance between all pixel sets is calculated; Secondly, the endpoints of the connecting line segments corresponding to the shortest distance are added as new nodes to the connected components; Finally, the connecting line segments are used as new edges to reorganize multiple connected components into a connected graph.

6. The method for road crack morphology fitting, analysis and complexity assessment according to claim 1, characterized in that: In step 7, the new minimum spanning tree is deeply traversed and simplified, specifically: (1) For the crack structure graph G, select the node v with degree 1 start Perform a depth-first traversal for the starting node; (2) If the depth-first traversal reaches the current node v current The degree d(v current )≠2, then keep v current ; (3) If the depth-first traversal reaches the current node v current The degree d(v current )=2, indicating that v current There are 2 adjacent edges e pre ,e next , and two adjacent nodes v pre ,v next , where v pre is the node that has been visited by the depth-first traversal before, v next is the next node to be visited, and yes pre =(v pre ,v current ); yes next =(v current ,v next ); Determine the two edges e pre ,e next If the angle between That is e pre ,e next Angle Then retain v current and continue traversing; otherwise delete v current , while connecting v pre ,v next , continue to traverse; After depth-first traversal of all nodes, the final simplified result is obtained.

7. A server comprising a processor and a memory, wherein the memory stores at least one program, characterized in that: The program is loaded and executed by the processor to implement the road crack morphology fitting, analysis and complexity assessment method according to any one of claims 1 to 6.

8. A computer-readable storage medium storing at least one program, characterized in that: The program is loaded and executed by a processor to implement the road crack morphology fitting, analysis and complexity assessment method according to any one of claims 1 to 6.

Images