Method and system for constructing three-dimensional model of corrosion morphological characteristics of port engineering component based on image detection

Abstract

The invention discloses a port engineering component corrosion morphology feature three-dimensional model construction method and system based on image detection, and the method comprises the steps: carrying out the classification and arrangement of a corrosion morphology image collected in real time through an image clustering method, segmenting and extracting the local information of a corrosion pit, and completing the fine reconstruction of the morphology of a metal corrosion pit; according to the method, the morphology features of the corrosion pit at the overlook angle and the depth features of the corrosion pit at the section view angle are fused together, so that the three-dimensional corrosion pit model is obtained, the model can accurately describe the local corrosion features of a metal structure, metal corrosion feature detection can be efficiently and conveniently achieved, and evaluation of metal service safety is completed. In order to realize detection of the corrosion degree of the steel structure with enriched corrosion products, a corrosion degree generalization method is provided, and comprehensive evaluation of the metal corrosion degree is completed through the method.

Description

Technical Field

[0001] The present invention relates to the technical field of steel component corrosion detection, and in particular to a method and system for constructing a three-dimensional model of corrosion morphology characteristics of a port component based on image detection. Background Art

[0002] Metal structures are widely used in today's buildings, but their corrosion susceptibility will closely affect the service safety of metal structures. Especially for local corrosion of metals, there are still obvious local differences in the degree of corrosion of metal structures in the overall environment. And because of the differences in the load conditions of different structures, the impact of local uneven corrosion of metals will be magnified. Therefore, real-time monitoring of the local corrosion characteristics of metal structures, such as corrosion area, corrosion depth and other indicators, can provide real-time feedback on the safety status of metal structures in service.

[0003] Most of the corrosion characteristics assessments of metal structures in service currently assume uniform corrosion. This assumption has limitations for the study of the load response of metal structures and cannot accurately assess the impact of metal defects. In addition, metal corrosion characteristics are extremely complex, and the large amount of impurities present on the surface will affect the assessment results of metal corrosion characteristics, which cannot be generalized in one way. Furthermore, in view of the complex on-site service environment, the corrosion characteristics assessment technology of metal structures still remains in the form of manual detection, lacking real-time, efficient and convenient technical means.

[0004] Moreover, most monitoring and detection methods for metal structure corrosion rely on manual detection. Therefore, a method that can identify corrosion images through image analysis and processing technology is needed. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide a method for constructing a three-dimensional model of the corrosion morphology characteristics of port construction components based on image detection. The method utilizes image analysis to analyze the corrosion morphology characteristics to construct a three-dimensional corrosion pit model.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] The method for constructing a three-dimensional model of corrosion morphology characteristics of port construction components based on image detection provided by the present invention comprises the following steps:

[0008] Acquire a top-view corrosion image of the object to be detected;

[0009] The topographical features of the rust pits are obtained by image processing on the topographical rust image.

[0010] Obtaining depth characteristics of the rust pits of the object to be detected;

[0011] A three-dimensional rust pit model is constructed according to the morphological characteristics and depth characteristics of the rust pit, and the three-dimensional rust pit model is used to describe the three-dimensional morphological characteristics of the rust pit.

[0012] Further, the rust pit depth characteristics are obtained by the following steps:

[0013] Acquire a cross-sectional corrosion image of the object to be inspected;

[0014] The depth characteristics of the rust pits are obtained by image processing on the cross-sectional rust image;

[0015] A depth feature database is constructed according to the depth features and morphological features of the rust pits, and the depth feature database is used to fit the corresponding relationship between different rust pit morphological features and depth features.

[0016] Furthermore, the morphological characteristics of the rust pits include shape characteristics and / or area characteristics of the rust pits.

[0017] Further, the morphological characteristics of the rust pits are obtained according to the following steps, and the specific process is as follows:

[0018] The shape characteristics of the rust pits are obtained through image segmentation; the center and radius of the circle are determined according to the shape characteristics; the position and area of ​​each rust pit are determined according to the center and radius of the circle; and the effective superposition area of ​​the rust pits is calculated by the superposition state of each circle.

[0019] Further, the depth characteristics of the rust pits are obtained in the following manner:

[0020] A cross-sectional corrosion image is obtained according to the corrosion image processing at the cross-sectional view angle;

[0021] The depth h of the rust pit is calculated according to the following formula:

[0022] h=R′(1―sinα)

[0023] Where α is the angle between the radius of the rust pit and the radius of the generalized space sphere, and R′ represents the radius of the generalized space sphere.

[0024] Furthermore, the effective overlapping area of ​​the rust pits is merged in the following manner:

[0025]

[0026] Among them, S1 is the effective overlapping area of ​​the left circle, and S2 is the effective overlapping area of ​​the right circle; x0, y0 represent the coordinates of the intersection of the two circles; x1, y1 represent the coordinates of the center of the left circle; x2, y2 represent the coordinates of the center of the right circle; R1 represents the radius of the left circle; and L is the distance between the centers of the circles.

[0027] Furthermore, the volume calculation of the rust pit during the fusion process is performed according to the following steps:

[0028] Establish a unified coordinate system;

[0029] Divide the computational grid;

[0030] Determine whether the corrosion pits have merged. When the center distance satisfies d≥R1+R2, it indicates that no merging phenomenon has occurred. Then, calculate the volume of a single corrosion pit and then calculate the total volume of each corrosion pit;

[0031] When the center distance satisfies d<R1+R2, it indicates that a merging phenomenon has occurred. Then, calculate the depth of the intersecting grid and take the maximum depth max as the grid depth; sum up the grid depths for solution; and then calculate the total volume of each corrosion pit.

[0032] The three-dimensional model construction system for the corrosion morphology characteristics of port engineering components based on image detection provided by the present invention includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the above method is implemented.

[0033] The beneficial effects of the present invention are as follows:

[0034] The three-dimensional model construction method for the corrosion morphology characteristics of port engineering components based on image detection provided by the present invention classifies and arranges the corrosion morphology images collected in real time through an image clustering method, and segments and extracts the local information of the corrosion pits, completing the fine reconstruction of the metal corrosion pit morphology; this method integrates the corrosion pit morphology characteristics from the top view angle and the corrosion pit depth characteristics from the cross-sectional view angle, thereby obtaining a three-dimensional corrosion pit model. This model can accurately depict the local corrosion characteristics of the metal structure, can efficiently and conveniently realize the detection of metal corrosion characteristics, and complete the assessment of the metal service safety.

[0035] To solve the problem of detecting the corrosion degree of steel structures with enriched corrosion products, a method for generalizing the corrosion degree is provided, and through this method, a comprehensive assessment (depth - area) of the metal corrosion degree is completed.

[0036] Other advantages, objectives, and features of the present invention will be described to some extent in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the following specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] To make the objectives, technical solutions, and beneficial effects of the present invention clearer, the present invention provides the following drawings for illustration:

[0038] Figure 1 It is a flowchart of the three-dimensional model construction method for the corrosion morphology characteristics of port engineering components based on image detection.

[0039] Figure 2 It is a schematic diagram of the standard circular area of the corrosion pit.

[0040] Figure 3 Schematic diagram of the morphological characteristics of rust pits.

[0041] Figure 4 This is a schematic diagram of the corrosion area calculation from a bird's-eye view.

[0042] Figure 5 Schematic diagram of the rust pit morphology after image segmentation.

[0043] Figure 6 This is a schematic diagram of corrosion depth calculation from a cross-sectional perspective.

[0044] Figure 7 This is a generalized schematic diagram of the rust pit.

[0045] Figure 8 Schematic diagram of the reconstruction of a rust pit into a hemispherical pit.

[0046] Fig. 9 Schematic diagram of the rust pit fusion results.

[0047] Fig.10 This is a three-dimensional schematic diagram of the rust pit fusion results.

[0048] Fig.11 Schematic diagram of the geometric model of the rust pit fusion results.

[0049] Fig.12 Flow chart of the rust pit volume calculation process.

[0050] Fig.13 Create a schematic diagram for the original image coordinates of the rust pit.

[0051] Fig.14 Schematic diagram of the mesh calculated for the rust pit image.

[0052] Fig.15 This is a schematic diagram of non-intersecting rust pits.

[0053] Fig.16 This is a schematic diagram of the calculation of the intersection of rust pit circles.

[0054] Fig.17 This is a schematic diagram of the calculation of the intersection and encryption of rust pit circles. DETAILED DESCRIPTION

[0055] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0056] Example 1

[0057] like Figure 1As shown, the method for constructing a three-dimensional model of the corrosion morphology characteristics of a port construction component based on image detection provided in this embodiment includes the following steps:

[0058] Acquire a top-view corrosion image of the object to be detected;

[0059] The topographical features of the rust pits are obtained by image processing on the topographical rust image.

[0060] Obtaining depth characteristics of the rust pits of the object to be detected;

[0061] Constructing a three-dimensional rust pit model according to the morphological characteristics and depth characteristics of the rust pit, wherein the three-dimensional rust pit model is used to describe the three-dimensional morphological characteristics of the rust pit;

[0062] The three-dimensional rust pit model of this embodiment is obtained by integrating the rust pit morphology characteristics and the rust pit depth characteristics at a top view angle.

[0063] The rust pit depth characteristics in this embodiment can be obtained by the following steps:

[0064] Acquire a cross-sectional corrosion image of the object to be inspected;

[0065] The depth characteristics of the rust pits are obtained by image processing on the cross-sectional rust image;

[0066] A depth feature database is constructed according to the depth features and morphological features of the rust pits, wherein the depth feature database is used to fit the corresponding relationship between different rust pit morphological features and depth features;

[0067] This embodiment detects the corrosion depth by adopting an image recognition algorithm, constructs a database of the morphological features and corrosion depth of large metal components based on existing detection results, and provides real-time feedback of the corrosion depth of metal components through image matching, thereby ensuring the long-term stable tracking and prediction of the metal corrosion status and service life, providing accurate and timely corrosion data for real-time detection of the client, and reducing the volume and cost of experimental work for on-site detection.

[0068] The depth characteristics of the rust pit can be obtained through DR images. The DR image is an image obtained by X-ray acquisition, which is recorded in grayscale by different transmissive radiation intensities to form a digital image, and then a thickness-grayscale mathematical model is constructed, and finally the depth of the rust pit is calculated by an algorithm;

[0069] The top-view rust image and the cross-sectional rust image collected in this embodiment are used to obtain the relationship between the top-view circular radius and the depth of the rust pit in the three-dimensional rust pit model through data fitting. Direct three-dimensional morphology data can also be obtained by on-site paraffin sampling as a sampling test to verify the accuracy of the three-dimensional model.

[0070] In this embodiment, a paraffin wax casting method can be used to construct a three-dimensional model of the rust pit on site:

[0071] Use paraffin wax to mold the on-site rust pit, that is, use soft paraffin wax to completely depict the three-dimensional shape of the on-site rust pit. The first step is to cast the paraffin wax into a 5*5*2cm regular hexahedron as the basic mode; the second step is to evenly heat one side of the paraffin mold until it is slightly softened, and then the wide side can be directly covered on the surface of the on-site steel component, and the temperature is maintained and the extrusion is continued; the third step is to carefully remove the paraffin wax until the thickness is less than 1cm, and then the mold result can be obtained, that is, the convex surface of the paraffin wax is the concave surface of the rust pit.

[0072] According to the bubble theory of rust, the rust pits in the rust pit image can be regarded as spheres, and the area of ​​the rust pits at a top-down angle can be determined according to their depth and radius.

[0073] The morphological characteristics of the rust pits in this embodiment include shape characteristics and area characteristics of the rust pits.

[0074] The morphological characteristics of the rust pits provided in this embodiment are obtained according to the following steps, and the specific process is as follows:

[0075] The shape characteristics of the rust pits are obtained by image segmentation; the center and radius of the circle are determined according to the shape characteristics; the position and area of ​​each rust pit are determined according to the center and radius of the circle; the effective superposition area is calculated by the superposition state of each circle;

[0076] In this embodiment, the rust pit is regarded as a standard circle. After the image processing, the rust pit area of ​​the binary image is obtained, and the rust pit area is fitted into a standard circle. Figure 2 As shown, Figure 2 This is a schematic diagram of the standard circular area of ​​rust pits. Figure 2 The rust pit area in the image is obtained by collecting the morphological image of the steel structure after corrosion, and after image processing, it is standardized as the standard circular area of ​​the rust pit.

[0077] The shapes of rust pits vary from field sampling to indoor test morphology, and various shapes can be used for generalization. This method uses circles for generalization because the parameters of circles are relatively few, and only the radius of the circle needs to be controlled to complete the characterization of rust points. In the process of converting to standard circles, this method uses Boolean operations on circles, and can use standard circles to characterize and depict all uneven and irregular rust pits.

[0078] The shape of the rust pit in this embodiment adopts a standard circular structure, such as Figure 2 As shown, the area characteristics of the rust pit are calculated according to the following formula:

[0079] S=∑S i

[0080] In the formula, Si is the effective superposition area of ​​each standard circle feature. Taking the common circle superposition method as an example, Figure 3 As shown, Figure 3 This is a schematic diagram of the morphological characteristics of the rust pit; Figure 3 In, S i Indicates the small rust pit area on the left side of the figure; S ii Indicates the large rust pit area on the right side of the picture.

[0081] The area of ​​the rust pit is calculated according to the following mathematical model: Figure 4 As shown, Figure 4 This is a schematic diagram of the calculation of the corrosion area from a top-down perspective. The area of ​​the corrosion pit is calculated according to the following formula:

[0082] The expression equation of the circle in the Cartesian coordinate system is:

[0083]

[0084] The coordinates of the intersection point can be obtained:

[0085]

[0086] According to the coordinates of the intersection point, the effective areas of the left and right circles can be obtained:

[0087]

[0088] Among them, S1 is the effective superposition area of ​​the left circle, S2 is the effective superposition area of ​​the right circle; x0, y0 are the coordinates of the intersection of the two circles; x1, y1 are the coordinates of the center of the left circle; x2, y2 are the coordinates of the center of the right circle; R1 is the radius of the left circle; L is the distance between the centers:

[0089] In this embodiment, the rust pit morphology is obtained by image segmentation processing. Figure 5 As shown, Figure 5 Schematic diagram of the rust pit morphology after image segmentation.

[0090] In this embodiment, the corrosion image of the cross-sectional view is processed to obtain the depth characteristics of the corrosion pit, which is specifically performed in the following manner:

[0091] The rust image of the cross-sectional view in this embodiment can be obtained by collecting on-site rubbing, and the cross-sectional rust image can be preliminarily judged to have a corresponding relationship between the rust depth and the radius of the rust circle, which can be generalized as a semicircular or arcuate cross section, such as Figure 6 As shown, Figure 6 The following is a schematic diagram of corrosion depth calculation from a cross-sectional perspective, where the depth of the corrosion pit, h, can be calculated using the following formula:

[0092] h=R′(1―sinα)

[0093] Where α is the angle between the radius of the rust pit and the radius of the generalized space sphere, which is related to the corrosion environment and corrosion rate, and R′ represents the radius of the generalized space sphere;

[0094] This embodiment conducts calibration analysis on the corresponding relationship between the corrosion depth and the corrosion pit radius through experiments to obtain a generalized corrosion pit model, such as Figure 7 As shown, Figure 7 This is a generalized schematic diagram of the rust pit.

[0095] The calibration process in this embodiment is carried out in the following manner:

[0096] The original image of the rust pit morphology is processed into a binary image, and the area where the rust pit area needs to be generalized is selected, and the standard circle image after image processing is obtained;

[0097] Figure 7 In the figure, (a) is the original image of the rust pit, (b) is the image binarization result, (c) is the local enlarged image of the binarization image, and (d) is the standard circle image of the local enlarged image after image processing;

[0098] By combining the shape of the rust pit from the top view with the depth of the rust pit from the cross-sectional view, the rust pit can be reconstructed as a hemispherical pit, such as Figure 8 As shown, Figure 8 This is a schematic diagram of the reconstruction of the rust pit into a hemispherical pit. Fig. 9 Schematic diagram of the rust pit fusion results.

[0099] In this embodiment, the hemispherical pit is reconstructed by image processing, and the circular area of ​​the top view image and the center position of the cross-sectional image are reconstructed, that is, a spherical pit with three-dimensional characteristics is constructed according to the area and depth characteristics;

[0100] In this embodiment, during the development of the rust pit, a spherical pit model with three-dimensional features can be established through rotational molding, that is, the rust circle in the top projection perspective is a spherical rust pit, and the rust semicircle in the cross-sectional perspective is the cross-sectional projection of the spherical rust pit at the circle diameter. Therefore, the three-dimensional spherical pit model can be obtained by rotationally molding the cross-sectional result of the rust pit along the central symmetry axis.

[0101] In this embodiment, based on the rust pit fusion model, a model of two rust pits fused and superimposed is obtained, such as Fig.10 As shown in Figure 2, the area of ​​a single rust pit can be calculated based on the fused model. Fig.10 This is a three-dimensional schematic diagram of the rust pit fusion results.

[0102] According to the three-dimensional rust pit model, the geometric model of the rust pit can be obtained, such as Fig.11 As shown, according to the geometric model, the area of ​​the rust pit can be calculated according to the following formula: Fig.11Schematic diagram of the geometric model of the rust pit fusion results.

[0103]

[0104] In the formula, x1, y1 are the coordinates of the center of the rust pit;

[0105] At the same time, the maximum corrosion depth of the rust pit can also be calculated according to the following formula:

[0106] h=R′(1―sinα)

[0107] Based on the spherical volume formula, the corrosion depth of each point in the corrosion pit can be obtained as:

[0108]

[0109] Where R′ is the radius of the generalized space sphere, which can be calculated by the following formula:

[0110]

[0111] In the formula, the value of α is about 60° to 80°; the value of α in this embodiment is obtained through experimental design;

[0112] The rust volume of the rust pit is:

[0113]

[0114] The total volume of corrosion is expressed as follows:

[0115] V=∑V i

[0116] like Fig.12 As shown, Fig.12 The flowchart of the volume calculation process of the rust pit during the rust pit fusion process is as follows:

[0117] Establish a unified coordinate system; the overall coordinate system can be constructed based on the lower left corner of the sample as the coordinate origin; Fig.13 As shown, Fig.13 Create a schematic diagram for the original image coordinates of the rust pit. Fig.13 The lower left corner of the original image collected is regarded as the origin of the coordinate system;

[0118] Divide the computational grid;

[0119] Divide the computational grid by appropriate spacing; Fig.14 Schematic diagram of the mesh calculated for the rust pit image.

[0120] Judge the fusion of corrosion pits; when the center distance satisfies the following condition: d < R1 + R2, if not, calculate the volume of a single corrosion pit by integration, and then calculate the volume of each corrosion pit; if so, calculate the depth of the intersecting grid; take the maximum depth max as the grid depth; sum up the grid depths for solution; the grid depth in this embodiment represents the average depth of the corrosion pits within each grid.

[0121] In this embodiment, the case where the corrosion pits do not intersect is as Fig.15 shown, Fig.15 which is a schematic diagram of non - intersecting corrosion pits. The corrosion pits can be processed separately, and the area and depth of the corrosion pits can be calculated. Finally, the volume of the corrosion pits can be calculated.

[0122] In this embodiment, the intersection of the corrosion circles is as Fig.16 shown, Fig.16 which is a schematic diagram for calculating the intersection of corrosion pit circles.

[0123] The intersecting circles can use the corrosion depth at each grid as the corrosion pit circles;

[0124] Further refine the grid as Fig.17 shown, Fig.17 which is a schematic diagram for calculating the refined intersection of corrosion pit circles.

[0125] Calculate the volume of each corrosion pit, that is, sum the volumes ∑V;

[0126] In this embodiment, after obtaining the reconstructed three - dimensional corrosion pit model, the detailed features of the corrosion morphology can be obtained, and an accurate overall characterization result of the corrosion degree can also be obtained, which can provide good basic conditions for related research such as corrosion rate calculation and corrosion degree evaluation.

[0127] In this embodiment, the area and depth of a single corrosion pit can be calculated through the top view and sectional view. It can be considered as the initial stage when the pitting corrosion pit just germinates. In the subsequent development stage, the pitting corrosion model can be used for calculation, that is, calculation through a simplified grid model.

[0128] The above - described embodiments are only preferred embodiments given to fully illustrate the present invention, and the protection scope of the present invention is not limited thereto. Equivalent substitutions or transformations made by those skilled in the art on the basis of the present invention are all within the protection scope of the present invention. The protection scope of the present invention is subject to the claims.

Claims

1. A method for constructing a three-dimensional model of corrosion morphology characteristics of port construction components based on image detection, characterized in that: Including the following steps: Obtain the top-view rust image of the object to be detected; Perform image processing on the top-view rust image to obtain the morphological features of the rust pits; Obtain the depth features of the rust pits of the object to be detected; Construct a three-dimensional rust pit model based on the morphological features and depth features of the rust pits, and the three-dimensional rust pit model is used to describe the three-dimensional morphological features of the rust pits.

2. The method for constructing a three-dimensional model of corrosion morphology characteristics of port construction components based on image detection according to claim 1, characterized in that: The depth features of the rust pits are obtained through the following steps: Obtain the sectional view rust image of the object to be detected; Perform image processing on the sectional view rust image to obtain the depth features of the rust pits; Construct a depth feature database based on the depth features and morphological features of the rust pits, and the depth feature database is used to fit the corresponding relationship between different rust pit morphological features and depth features.

3. The method for constructing a three-dimensional model of corrosion morphology characteristics of port construction components based on image detection as claimed in claim 1, characterized in that: The morphological features of the rust pits include the shape features and / or area features of the rust pits.

4. The method for constructing a three-dimensional model of corrosion morphology characteristics of port construction components based on image detection according to claim 1, characterized in that: The morphological features of the rust pits are obtained according to the following steps, and the specific process is as follows: Obtain the shape features of the rust pits through image segmentation; determine the center and radius according to the shape features; determine the positions and areas of each rust pit according to the center and radius; calculate the effective overlapping area of the rust pits through the overlapping state of each circle.

5. The method for constructing a three-dimensional model of corrosion morphology characteristics of port construction components based on image detection according to claim 1, characterized in that: The depth features of the rust pits are obtained in the following manner: Obtain the sectional view rust image according to the rust image processing of the sectional view perspective; Calculate the depth h of the rust pit according to the following formula: h = R′(1―sinα) In the formula, α is the angle between the radius of the rust circular pit and the radius of the generalized space sphere, and R′ represents the radius of the generalized space sphere.

6. The method for constructing a three-dimensional model of corrosion morphology characteristics of port construction components based on image detection as claimed in claim 4, characterized in that: The effective overlapping area of the rust pits is fused in the following manner: Wherein, S1 is the effective overlapping area of the left circle, S2 is the effective overlapping area of the right circle; x0, y0 represent the intersection coordinates of the two circles; x1, y1 represent the center coordinates of the left circle; x2, y2 represent the center coordinates of the right circle; R1 represents the radius of the left circle; L is the center distance.

7. The method for constructing a three-dimensional model of corrosion morphology characteristics of port construction components based on image detection according to claim 6, characterized in that: The volume calculation of the rust pits during the fusion process is carried out according to the following steps: Establish a unified coordinate system; Divide the calculation grid; Judge whether the rust pits are fused. When the center distance satisfies d≥R1+R2, it means that no fusion phenomenon occurs, then calculate the volume of a single rust pit, and then calculate the total volume of each rust pit; When the center distance satisfies d<R1+R2, it means that a fusion phenomenon occurs, then calculate the depth of the intersecting grid, and take the maximum depth max as the grid depth; sum up the grid depths and solve; then calculate the total volume of each rust pit.

8. A system for constructing a three-dimensional model of the corrosion morphology characteristics of port construction components based on image detection, 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 program, it implements the method described in any one of claims 1 to 7 above.

Images