Method, device, equipment and medium for identifying defects of special-shaped steel based on infrared images

Abstract

The present application discloses a method, device, equipment and medium for defect recognition of special-shaped steel based on infrared images, comprising the following steps: acquiring an infrared image of a target special-shaped steel based on a frontal perspective; the target special-shaped steel comprises a rectangular frame, a trapezoidal groove is arranged at the bottom of the rectangular frame, triangular protrusions are arranged on both side walls of the rectangular frame, and two arc-shaped plates are arranged on the top of the rectangular frame; preprocessing the infrared image, and performing contour feature recognition on the preprocessed infrared image to respectively obtain the contour of the trapezoidal groove, the contour of the triangular protrusion and the contour of the arc-shaped plate, thereby respectively obtaining a first defect value F1, a second defect value F2 and a third defect value F3, and finally obtaining a defect evaluation index F of the target special-shaped steel, and judging the qualified status of the target special-shaped steel according to the defect evaluation index F. The present application has the advantages of improving the accuracy of defect detection and high detection efficiency.

Description

Technical Field

[0001] The present application relates to the technical field of image data processing, and in particular to a method, device, equipment and medium for identifying defects in special-shaped steel based on infrared images. Background Art

[0002] Special-shaped steel is the abbreviation of complex and special-shaped cross-section steel. It is a unique and important classification in the steel series. It is significantly different from traditional steel with simple and regular cross-section. This type of steel, with its unique shape and structural design, meets the diverse needs for special steel structures in modern buildings. In certain specific construction projects, especially those steel structures that pursue innovative design and high performance requirements, traditional standard steel often cannot meet their complex structural requirements. At this time, special-shaped steel stands out with its unique advantages.

[0003] Figure 2 and Figure 3 A typical special-shaped steel design is shown in the video. They are presented as a hollow rectangular frame as a whole. This design not only reduces the weight of the material, but also enhances the stability and load-bearing capacity of the structure. A trapezoidal groove is cleverly designed at the bottom of the rectangular frame. This design not only takes into account the aesthetics, but also facilitates precise docking with other components. On the two side walls of the rectangular frame, triangular protrusions are cleverly set. These protrusions not only increase the bonding strength between the steel and the concrete, but also improve the shear strength of the overall structure. What is more unique is that the top of the rectangular frame adopts two symmetrical arc-shaped plate designs. This design not only gives the special-shaped steel an elegant streamlined appearance, but also effectively disperses the top load and enhances the bending resistance of the structure.

[0004] However, it is these complex and delicate design elements, such as trapezoidal grooves, triangular protrusions and curved plates, that become parts that are prone to deformation during the manufacturing process. Even slight deformations in these parts may have a significant impact on the stability and safety of the overall structure. Therefore, it is particularly important to accurately and efficiently detect deformation defects in these key parts.

[0005] Unfortunately, the current defect detection for this type of special-shaped steel mainly relies on manual visual inspection or the use of traditional measuring tools. Although manual visual inspection is intuitive, it is often subject to large subjectivity and errors due to factors such as human visual fatigue and different judgment standards. Although the use of measuring tools can improve accuracy to a certain extent, the detection process is cumbersome, time-consuming, and inefficient, and it is still unable to detect complex shapes and small deformations. These problems not only affect the accuracy of the test results, but also greatly limit the improvement of production efficiency.

[0006] Therefore, exploring and developing a more advanced, accurate and efficient method for detecting defects in special-shaped steel has become an urgent problem to be solved. Summary of the invention

[0007] The main purpose of this application is to provide a method, device, equipment and medium for identifying defects in special-shaped steel based on infrared images, aiming to solve the technical problem of low detection accuracy of existing methods for detecting defects in special-shaped steel.

[0008] To achieve the above object, the present application provides a method for identifying defects in special-shaped steel based on infrared images, comprising the following steps:

[0009] An infrared image of the target special-shaped steel is obtained based on a frontal perspective; wherein the target special-shaped steel comprises a rectangular frame, a trapezoidal groove is provided at the bottom of the rectangular frame, triangular protrusions are provided on both side walls of the rectangular frame, and two arc-shaped plates are provided on the top of the rectangular frame. The frontal perspective is a perspective that simultaneously displays the end contour features of the trapezoidal groove, the triangular protrusion and the arc-shaped plate;

[0010] Preprocessing the infrared image, and performing contour feature recognition on the preprocessed infrared image to obtain the trapezoidal groove contour, the triangular convex contour and the arc plate contour respectively;

[0011] According to the trapezoidal groove profile, a first defect value F1 is obtained; wherein the first defect value F1 represents a deformation degree value of the trapezoidal groove;

[0012] According to the triangular protrusion profile, a second defect value F2 is obtained; wherein the second defect value F2 represents the deformation degree value of the triangular protrusion;

[0013] According to the profile of the curved plate, a third defect value F3 is obtained; wherein the third defect value F3 represents the deformation degree value of the curved plate;

[0014] According to the first defect value F1, the second defect value F2 and the third defect value F3, the defect evaluation index F of the target special-shaped steel is obtained, and according to the defect evaluation index F, the qualification of the target special-shaped steel is judged.

[0015] Optionally, obtaining the first defect value F1 according to the trapezoidal groove profile includes:

[0016] Connect the openings of the trapezoidal groove profile to obtain the target quadrilateral;

[0017] Based on the coincidence of any cusp of the target quadrilateral with the corresponding cusp of a preset standard trapezoid, a fused image is obtained; wherein the standard trapezoid is a trapezoid formed based on a trapezoidal groove that has not been deformed;

[0018] According to the fused image, identify the intersecting area S of the target quadrilateral and the standard trapezoid;

[0019] According to the staggered area S, the first defect value F1 is obtained, F1=S*K1; wherein K1 is the first proportional coefficient.

[0020] Optionally, obtaining the second defect value F2 according to the triangular protrusion profile includes:

[0021] According to the triangular convex contour, a first reference line is constructed; wherein the first reference line is a line connecting the two end points of the triangular convex contour and the rectangular frame;

[0022] Identify the target cusp corresponding to the triangular convex contour; wherein the target cusp is an end point of the triangular convex contour away from the rectangular frame;

[0023] Connect the target point with the midpoint of the first reference line to obtain an offset line;

[0024] Obtaining an angle θ between the offset line and the first reference line;

[0025] According to the angle θ, the second defect value F2 is obtained, F2=(π / 2-θ)*K2; wherein K2 is the second proportional coefficient.

[0026] Optionally, obtaining a third defect value F3 according to the curved plate profile includes:

[0027] Mapping a theoretical reference point at a corresponding position of the infrared image; wherein the theoretical reference point is a center point corresponding to an inner arc line of an arc-shaped plate that has not been deformed;

[0028] Select multiple measuring points on the inner arc of the curved plate, and obtain the distance value between each measuring point and the theoretical reference point respectively;

[0029] The average value X of multiple spacing values ​​is obtained to obtain the third defect value F3, F3=|RX|*K3; wherein R is the arc radius corresponding to the inner arc of the arc plate that has not been deformed, and K3 is the third adjustment coefficient.

[0030] Optionally, mapping the theoretical reference point at the position corresponding to the contour of the curved plate includes:

[0031] Constructing a standard circle based on the standard arc profile, and obtaining the center mark of the standard circle; wherein the standard arc profile is the profile corresponding to the arc plate that has not been deformed;

[0032] Connect the center mark and the target corner point to obtain a second reference line; wherein the target corner point is a corner point adjacent to the center mark in the rectangular frame;

[0033] The end of the second reference line away from the center mark is overlapped with the target corner point corresponding to the target special-shaped steel, so that the center mark is mapped to the corresponding position of the infrared image as a theoretical reference point.

[0034] Optionally, the defect evaluation index F=F1+F2+F3.

[0035] Optionally, judging the qualification of the target special-shaped steel according to the defect evaluation index F includes:

[0036] Compare the defect evaluation index F with a preset evaluation threshold;

[0037] If the defect evaluation index F is greater than the evaluation threshold, the target special-shaped steel is identified as a defective product;

[0038] If the defect evaluation index F is less than or equal to the evaluation threshold, the target special-shaped steel is identified as a qualified product.

[0039] To achieve the above objectives, the present application also provides a device for identifying defects in special-shaped steel based on infrared images, comprising:

[0040] An image acquisition module is used to acquire an infrared image of the target special-shaped steel based on a frontal perspective; wherein the target special-shaped steel comprises a rectangular frame, a trapezoidal groove is provided at the bottom of the rectangular frame, triangular protrusions are provided on both side walls of the rectangular frame, and two arc-shaped plates are provided on the top of the rectangular frame, and the frontal perspective is a perspective that simultaneously displays the contour features of the end of the trapezoidal groove, the triangular protrusion and the arc-shaped plate;

[0041] A feature recognition module is used to preprocess the infrared image and perform contour feature recognition on the preprocessed infrared image to obtain the trapezoidal groove contour, the triangular convex contour and the arc plate contour respectively;

[0042] A first acquisition module is used to acquire a first defect value F1 according to the trapezoidal groove profile; wherein the first defect value F1 represents a deformation degree value of the trapezoidal groove;

[0043] A second acquisition module is used to acquire a second defect value F2 according to the triangular protrusion profile; wherein the second defect value F2 represents a deformation degree value of the triangular protrusion;

[0044] A third acquisition module is used to acquire a third defect value F3 according to the contour of the curved plate; wherein the third defect value F3 represents a deformation degree value of the curved plate;

[0045] The evaluation module is used to obtain the defect evaluation index F of the target special-shaped steel according to the first defect value F1, the second defect value F2 and the third defect value F3, and judge the qualification of the target special-shaped steel according to the defect evaluation index F.

[0046] To achieve the above objectives, the present application also provides a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above method.

[0047] To achieve the above objectives, the present application also provides a computer-readable storage medium, on which a computer program is stored, and a processor executes the computer program to implement the above method.

[0048] The beneficial effects that this application can achieve are as follows:

[0049] The present application obtains an infrared image of the target special-shaped steel based on a frontal perspective and performs preprocessing, so that the end contour features of the trapezoidal groove, the triangular protrusion and the arc-shaped plate can be easily identified and extracted at the same time. Therefore, the first defect value F1 can be calculated according to the trapezoidal groove contour, the second defect value F2 can be calculated according to the triangular protrusion contour, and the third defect value F3 can be calculated according to the arc-shaped plate contour. The contour features of each easily deformed part can be identified in combination with the infrared image and the corresponding defect values ​​can be quantitatively calculated. Finally, the first defect value F1, the second defect value F2 and the third defect value F3 can be combined to obtain the defect evaluation index F of the target special-shaped steel, so as to accurately and effectively judge the qualification of the target special-shaped steel. Compared with manual inspection or tool inspection, human errors are reduced, the accuracy of defect detection and identification is greatly improved, and the inspection efficiency is effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the prior art description. In all drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0051] Figure 1 It is a schematic diagram of the process of the method for identifying defects of special-shaped steel based on infrared images in an embodiment of the present application;

[0052] Figure 2 This is a schematic diagram of the structure of the target special-shaped steel from a front view in the embodiment of the present application;

[0053] Figure 3 for Figure 2 A structural diagram of a left view of FIG.

[0054] Figure 4 This is a schematic diagram of the principle of calculating the staggered area S in the embodiment of the present application;

[0055] Figure 5 This is a schematic diagram of the principle of calculating the angle θ in the embodiment of the present application;

[0056] Figure 6 A schematic diagram of the principle of constructing a second baseline in an embodiment of the present application;

[0057] Figure 7It is a schematic diagram of mapping theoretical reference points to corresponding positions of infrared images in an embodiment of the present application.

[0058] Reference numerals:

[0059] 110 - rectangular frame, 120 - trapezoidal groove, 130 - triangular protrusion, 140 - arc plate, 150 - target quadrilateral, 160 - first reference line, 170 - target cusp, 180 - offset line, 190 - theoretical reference point, 210 - standard circle, 220 - second reference line.

[0060] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0062] It should be noted that all directional indications in the embodiments of the present application (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components in a certain specific posture. If the specific posture changes, the directional indication will also change accordingly.

[0063] In this application, unless otherwise clearly specified and limited, the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "fixation" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0064] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the meaning of "and / or" appearing in the full text includes three parallel schemes. Taking "A and / or B" as an example, it includes scheme A, or scheme B, or a scheme that satisfies both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0065] Embodiment 1:

[0066] Reference Figure 1-Figure 7 This embodiment provides a method for identifying defects in special-shaped steel based on infrared images, comprising the following steps:

[0067] An infrared image of the target special-shaped steel is obtained based on a frontal perspective; wherein the target special-shaped steel includes a rectangular frame 110, a trapezoidal groove 120 is provided at the bottom of the rectangular frame 110, triangular protrusions 130 are provided on both side walls of the rectangular frame 110, and two arc plates 140 are provided on the top of the rectangular frame 110. The frontal perspective is a perspective that simultaneously displays the end contour features of the trapezoidal groove 120, the triangular protrusion 130 and the arc plate 140;

[0068] Preprocessing the infrared image, and performing contour feature recognition on the preprocessed infrared image to obtain the trapezoidal groove contour, the triangular convex contour and the arc plate contour respectively;

[0069] According to the trapezoidal groove profile, a first defect value F1 is obtained; wherein the first defect value F1 represents the deformation degree value of the trapezoidal groove 120;

[0070] According to the triangular protrusion profile, a second defect value F2 is obtained; wherein the second defect value F2 represents the deformation degree value of the triangular protrusion 130;

[0071] According to the curved plate profile, a third defect value F3 is obtained; wherein the third defect value F3 represents the deformation degree value of the curved plate 140;

[0072] According to the first defect value F1, the second defect value F2 and the third defect value F3, the defect evaluation index F of the target special-shaped steel is obtained, and according to the defect evaluation index F, the qualification of the target special-shaped steel is judged.

[0073] In this embodiment, after acquiring the infrared image of the target special-shaped steel based on the frontal perspective and performing preprocessing, it is convenient to simultaneously identify and extract the end contour features of the trapezoidal groove, the triangular protrusion and the arc-shaped plate. Therefore, the first defect value F1 can be calculated according to the trapezoidal groove contour, the second defect value F2 can be calculated according to the triangular protrusion contour, and the third defect value F3 can be calculated according to the arc-shaped plate contour. In this way, the contour features of each easily deformed part can be identified in combination with the infrared image and the corresponding defect values ​​can be quantitatively calculated. Finally, the first defect value F1, the second defect value F2 and the third defect value F3 can be combined to obtain the defect evaluation index F of the target special-shaped steel, so as to accurately and effectively judge the qualification of the target special-shaped steel. Compared with manual inspection or tool inspection, human errors are reduced, the accuracy of defect detection and identification is greatly improved, and the inspection efficiency is effectively improved.

[0074] It should be noted that the trapezoidal groove 120 can be an inverted trapezoidal groove or a regular trapezoidal groove. If there are two or more trapezoidal grooves 120, it is necessary to calculate the deformation degree value corresponding to each trapezoidal groove 120, and then the sum of the deformation degree values ​​of multiple trapezoidal grooves 120 is the corresponding first defect value F1; similarly, since there are two or more triangular protrusions and arc plates, it is necessary to calculate the corresponding deformation degree value of each triangular protrusion and arc plate, and then the sum of multiple deformation degree values ​​is used as the corresponding second defect value F2 and third defect value F3. When preprocessing the infrared image, grayscale processing, noise reduction processing and binarization processing can be performed respectively, so as to obtain a clear target special-shaped steel image, which is convenient for subsequent feature recognition and extraction.

[0075] As an optional implementation, the final defect evaluation index F can be obtained based on the sum of the first defect value F1, the second defect value F2 and the third defect value F3, that is, the defect evaluation index F=F1+F2+F3. Therefore, any high defect value or a low single defect value but a high total defect value will lead to a higher defect evaluation index F. Therefore, it can effectively characterize the overall deformation of the target special-shaped steel and has reference guidance significance.

[0076] As an optional implementation manner, obtaining the first defect value F1 according to the trapezoidal groove profile includes:

[0077] Connect the openings of the trapezoidal groove profile to obtain a target quadrilateral 150;

[0078] Based on the coincidence of any cusp of the target quadrilateral 150 with the corresponding cusp of a preset standard trapezoid, a fused image is obtained; wherein the standard trapezoid is a trapezoid formed based on the trapezoidal groove 120 without deformation;

[0079] According to the fused image, the intersecting area S of the target quadrilateral 150 and the standard trapezoid is identified;

[0080] According to the staggered area S, the first defect value F1 is obtained, F1=S*K1; wherein K1 is the first proportional coefficient.

[0081] In this embodiment, when calculating the first defect value F1, since the deformation of the trapezoidal groove 120 causes the corresponding contour edge to shift, this embodiment obtains the target quadrilateral 150 by connecting the openings of the trapezoidal groove contour, and overlaps the target quadrilateral 150 with the standard trapezoid based on the sharp points at the same position. If the trapezoidal groove 120 is deformed, the constructed target quadrilateral 150 and the standard trapezoid cannot overlap. Therefore, the first defect value F1 can be associated with the intersection area S of the target quadrilateral 150 and the standard trapezoid, that is, the greater the deformation of the trapezoidal groove 120, the larger the intersection area S, and the larger the first defect value F1. The first proportional coefficient K1 can represent the degree of association between the two, so that the first defect value F1 can be quantitatively calculated to accurately characterize the degree of deformation of the trapezoidal groove 120.

[0082] It should be noted that the interleaving area S here mainly refers to the area formed by the non-overlapping part of the contour edge of the target quadrilateral 150 and the standard trapezoid (i.e. Figure 4 If they completely overlap, the interlaced area S=0, that is, the first defect value F1=0, indicating that there is no defect of trapezoidal groove deformation.

[0083] As an optional implementation, obtaining the second defect value F2 according to the triangular protrusion profile includes:

[0084] According to the triangular raised contour, a first reference line 160 is constructed; wherein the first reference line 160 is a line connecting the two end points of the triangular raised contour and the rectangular frame 110;

[0085] Identify the target point 170 corresponding to the triangular convex contour; wherein the target point 170 is an end point of the triangular convex contour away from the rectangular frame 110;

[0086] Connecting the target point 170 with the midpoint of the first reference line 160 to obtain an offset line 180;

[0087] Obtaining an angle θ between the offset line 180 and the first reference line 160;

[0088] According to the angle θ, the second defect value F2 is obtained, F2=(π / 2-θ)*K2; wherein K2 is the second proportional coefficient.

[0089] In this embodiment, when calculating the second defect value F2, since when the triangular protrusion 130 is deformed, the position offset of the corresponding target cusp 170 is the largest, so this embodiment constructs the first reference line 160 through the triangular protrusion contour, where the first reference line 160 is the line between the two end points connecting the triangular protrusion contour and the rectangular frame 110, and the two end points are the coincidence points of the triangular protrusion contour and the rectangular frame 110, which are not easy to deform and are not easy to offset. Therefore, the midpoint of the first reference line 160 can be used as a reference point. If the triangular protrusion 130 is deformed, it will cause the target cusp 170 to offset relative to the midpoint of the first reference line 160. Therefore, the target cusp 170 is offset relative to the midpoint of the first reference line 160. The midpoints of the reference line 160 are connected to obtain the offset line 180, and the angle θ between the offset line 180 and the first reference line 160 is associated with the second defect value F2. Based on the calculation formula of the second defect value F2, if the angle θ is 90°, it means that the offset line 180 is perpendicular to the first reference line 160, that is, the target point 170 and the midpoint of the first reference line 160 are at the same height, which means that the triangular protrusion 130 has not been deformed, and F2=0 at this time; conversely, the smaller the angle θ, the greater the offset, and the greater the value of the second defect value F2, and the second proportional coefficient K2 can represent the degree of association between the two, so that the second defect value F2 can be quantitatively calculated to accurately characterize the degree of deformation of the triangular protrusion 130.

[0090] As an optional implementation manner, obtaining the third defect value F3 according to the curved plate profile includes:

[0091] A theoretical reference point 190 is mapped at a corresponding position of the infrared image; wherein the theoretical reference point 190 is a center point corresponding to an inner arc line of the arc-shaped plate 140 that has not been deformed;

[0092] Select multiple measuring points on the inner arc line of the arc plate 140, and obtain the distance value between each measuring point and the theoretical reference point 190 respectively;

[0093] The average value X of multiple spacing values ​​is obtained to obtain a third defect value F3, F3=|RX|*K3; wherein R is the arc radius corresponding to the inner arc of the undeformed arc plate 140, and K3 is the third adjustment coefficient.

[0094] In this embodiment, when calculating the third defect value F3, a theoretical reference point 190 is mapped at the corresponding position of the infrared image. The theoretical reference point 190 can represent the theoretical position point of the center point of the inner arc line corresponding to the arc plate 140 in the overall contour of the target special-shaped steel in the infrared image when the arc plate 140 is not deformed. Therefore, taking the theoretical reference point 190 as the reference point, if the arc plate 140 is deformed and shifted as a whole in the direction closer to or away from the rectangular frame 110, or the curvature of the arc plate 140 does not meet the requirements, the distance between the measuring point on the inner arc line of the arc plate 140 and the theoretical reference point 190 will change. Here, the measuring point should at least include the end point of the arc plate 140 away from the rectangular frame 110. At the distances d1, d2, d3...dn between the multiple measuring points and the theoretical reference point 190, the average value X of the multiple distance values ​​is calculated, X=(d1+d2+d3+...+dn) / n. Due to the different deformation directions of the curved plate 140, the finally calculated average value X may be greater than or less than the arc radius R corresponding to the inner arc line of the curved plate 140 that has not been deformed. Therefore, based on the calculation formula of the third defect value F3, the absolute value of the difference between the average value X and the arc radius R is used to characterize the deformation degree. At the same time, based on the third proportional coefficient K3, the correlation degree between the two can be represented, so that the third defect value F3 can be quantitatively calculated to achieve accurate quantitative calculation of the deformation degree of the curved plate 140.

[0095] As an optional implementation, after respectively obtaining the spacing value between each measuring point and the theoretical reference point 190 , the spacing values ​​whose difference with the arc radius R is less than a preset value among the multiple spacing values ​​are eliminated.

[0096] In this embodiment, the arc plate 140 may be locally deformed, while the deformation amplitude of other parts of the arc plate 140 is relatively low, and the spacing value between the corresponding measurement point and the theoretical reference point 190 is close to or equal to the arc radius R. Therefore, this spacing value needs to be eliminated, and the spacing value with a large difference from the arc radius R is retained, so as to further accurately characterize the deformation degree of the parts of the arc plate 140 with larger deformation, thereby improving the data accuracy.

[0097] As an optional implementation, mapping the theoretical reference point 190 at the corresponding position of the curved plate contour includes:

[0098] Constructing a standard circle 210 based on the standard arc profile, and obtaining a center mark of the standard circle 210; wherein the standard arc profile is a profile corresponding to the arc plate 140 that has not been deformed;

[0099] Connect the center mark and the target corner point to obtain a second reference line 220; wherein the target corner point is a corner point adjacent to the center mark in the rectangular frame 110;

[0100] The end of the second reference line 220 away from the circle center mark is overlapped with the target corner point corresponding to the target special-shaped steel, so that the circle center mark is mapped as a theoretical reference point 190 to the corresponding position of the infrared image.

[0101] In this embodiment, when mapping the theoretical reference point 190, the standard circle 210 is first constructed based on the standard arc contour, so as to obtain the center of the standard circle 210 and mark it, i.e., the center mark, and then the center mark is connected with the target corner point to obtain the second reference line 220. Since the target corner point is a corner point in the rectangular frame 110 that is not easily deformed, a second reference line 220 with position reference can be obtained. Then, the end of the second reference line 220 away from the center mark is overlapped with the target corner point corresponding to the target special-shaped steel, thereby realizing the mapping of the center mark as the theoretical reference point 190 to the corresponding position of the infrared image, which provides a prerequisite for the subsequent accurate characterization of the deformation of the arc plate 140.

[0102] As an optional implementation, judging the qualification of the target special-shaped steel according to the defect evaluation index F includes:

[0103] Compare the defect evaluation index F with a preset evaluation threshold;

[0104] If the defect evaluation index F is greater than the evaluation threshold, the target special-shaped steel is identified as a defective product;

[0105] If the defect evaluation index F is less than or equal to the evaluation threshold, the target special-shaped steel is identified as a qualified product.

[0106] In this embodiment, after the defect evaluation index F is calculated, an evaluation threshold can be set and the two can be compared. According to the comparison result, the target special-shaped steel can be identified as a defective product or a qualified product. If it is a defective product, the first defect value F1, the second defect value F2 and the third defect value F3 can be output respectively, so that the staff can rework and correct the relevant deformed parts in a targeted manner according to the size of each defect value.

[0107] Embodiment 2:

[0108] Reference Figure 1-Figure 3 Based on the same inventive concept as the above-mentioned embodiment, this embodiment further provides a device for identifying defects of special-shaped steel based on infrared images, comprising:

[0109] An image acquisition module is used to acquire an infrared image of a target special-shaped steel based on a frontal perspective; wherein the target special-shaped steel includes a rectangular frame 110, a trapezoidal groove 120 is provided at the bottom of the rectangular frame 110, triangular protrusions 130 are provided on both side walls of the rectangular frame 110, and two arc plates 140 are provided on the top of the rectangular frame 110, and the frontal perspective is a perspective that simultaneously displays the end contour features of the trapezoidal groove 120, the triangular protrusion 130 and the arc plate 140;

[0110] A feature recognition module is used to preprocess the infrared image and perform contour feature recognition on the preprocessed infrared image to obtain the trapezoidal groove contour, the triangular convex contour and the arc plate contour respectively;

[0111] A first acquisition module is used to acquire a first defect value F1 according to the trapezoidal groove profile; wherein the first defect value F1 represents a deformation degree value of the trapezoidal groove 120;

[0112] A second acquisition module is used to acquire a second defect value F2 according to the triangular protrusion profile; wherein the second defect value F2 represents a deformation degree value of the triangular protrusion 130;

[0113] A third acquisition module is used to acquire a third defect value F3 according to the contour of the curved plate; wherein the third defect value F3 represents a deformation degree value of the curved plate 140;

[0114] The evaluation module is used to obtain the defect evaluation index F of the target special-shaped steel according to the first defect value F1, the second defect value F2 and the third defect value F3, and judge the qualification of the target special-shaped steel according to the defect evaluation index F.

[0115] The relevant explanations and examples of each module in the device of this embodiment can refer to the method of the aforementioned embodiment, and will not be repeated here.

[0116] Embodiment 3:

[0117] Based on the same inventive concept as the above-mentioned embodiment, this embodiment provides a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-mentioned method.

[0118] Embodiment 4:

[0119] Based on the same inventive concept as the aforementioned embodiment, this embodiment provides a computer-readable storage medium, on which a computer program is stored, and a processor executes the computer program to implement the aforementioned method.

[0120] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for identifying defects in special-shaped steel based on infrared images, characterized in that: The following steps are involved: An infrared image of the target special-shaped steel is acquired based on a frontal perspective; wherein the target special-shaped steel comprises a rectangular frame, a trapezoidal groove is provided at the bottom of the rectangular frame, triangular protrusions are provided on both side walls of the rectangular frame, and two arc-shaped plates are provided on the top of the rectangular frame, and the frontal perspective is a perspective that simultaneously displays the contour features of the trapezoidal groove, the triangular protrusion and the end of the arc-shaped plate; Preprocessing the infrared image, and performing contour feature recognition on the preprocessed infrared image to respectively obtain a trapezoidal groove contour, a triangular convex contour, and an arc plate contour; According to the trapezoidal groove profile, a first defect value F1 is obtained; wherein the first defect value F1 represents the deformation degree value of the trapezoidal groove; comprising: connecting the openings of the trapezoidal groove profile to obtain a target quadrilateral; based on the coincidence of any cusp of the target quadrilateral with the corresponding cusp of a preset standard trapezoid, a fused image is obtained; wherein the standard trapezoid is a trapezoid constructed based on the undeformed trapezoidal groove; according to the fused image, an interleaved area S between the target quadrilateral and the standard trapezoid is identified; according to the interleaved area S, a first defect value F1 is obtained, F1=S*K1; wherein K1 is a first proportionality coefficient; According to the triangular convex contour, a second defect value F2 is obtained; wherein the second defect value F2 represents the deformation degree value of the triangular convex; comprising: constructing a first reference line according to the triangular convex contour; wherein the first reference line is a line connecting two end points of the triangular convex contour and the rectangular frame; identifying a target cusp corresponding to the triangular convex contour; wherein the target cusp is an end point of the triangular convex contour away from the rectangular frame; connecting the target cusp with the midpoint of the first reference line to obtain an offset line; obtaining an angle θ between the offset line and the first reference line; according to the angle θ, obtaining a second defect value F2, F2=(π / 2-θ)*K2; wherein K2 is a second proportionality coefficient; According to the contour of the curved plate, a third defect value F3 is obtained; wherein the third defect value F3 represents the deformation degree value of the curved plate; including: mapping a theoretical reference point at a corresponding position of the infrared image; wherein the theoretical reference point is a center point corresponding to the inner arc line of the curved plate that has not been deformed; selecting a plurality of measuring points on the inner arc line of the curved plate, and respectively obtaining the spacing value between each measuring point and the theoretical reference point; obtaining the average value X of the plurality of spacing values ​​to obtain the third defect value F3, F3=|RX|*K3; wherein R is the arc radius corresponding to the inner arc line of the curved plate that has not been deformed, and K3 is a third adjustment coefficient; According to the first defect value F1, the second defect value F2 and the third defect value F3, a defect evaluation index F of the target special-shaped steel is obtained, and according to the defect evaluation index F, the qualification of the target special-shaped steel is judged.

2. The method for identifying defects in special-shaped steel based on infrared images according to claim 1, characterized in that: Mapping theoretical reference points at corresponding positions of the infrared image includes: Constructing a standard circle based on a standard arc profile, and obtaining a center mark of the standard circle; wherein the standard arc profile is a profile corresponding to the arc plate that has not been deformed; Connect the center mark and the target corner point to obtain a second reference line; wherein the target corner point is a corner point in the rectangular frame adjacent to the center mark; The end of the second reference line away from the circle center mark is overlapped with the target corner point corresponding to the target special-shaped steel, so that the circle center mark is mapped as the theoretical reference point to the corresponding position of the infrared image.

3. The method for identifying defects in special-shaped steel based on infrared images according to claim 1 or 2, characterized in that: The defect evaluation index F=F1+F2+F3.

4. The method for identifying defects in special-shaped steel based on infrared images according to claim 1, characterized in that: According to the defect evaluation index F, judging the qualification of the target special-shaped steel includes: Comparing the defect evaluation index F with a preset evaluation threshold; If the defect evaluation index F is greater than the evaluation threshold, the target special-shaped steel is identified as a defective product; If the defect evaluation index F is less than or equal to the evaluation threshold, the target special-shaped steel is identified as a qualified product.

5. The device for identifying defects in special-shaped steel based on infrared images is characterized in that: include: An image acquisition module, used for acquiring an infrared image of a target special-shaped steel based on a frontal viewing angle; wherein the target special-shaped steel comprises a rectangular frame, a trapezoidal groove is provided at the bottom of the rectangular frame, triangular protrusions are provided on both side walls of the rectangular frame, and two arc-shaped plates are provided on the top of the rectangular frame, and the frontal viewing angle is a viewing angle that simultaneously displays the contour features of the trapezoidal groove, the triangular protrusion and the end of the arc-shaped plate; A feature recognition module is used to preprocess the infrared image and perform contour feature recognition on the preprocessed infrared image to respectively obtain a trapezoidal groove contour, a triangular convex contour and an arc plate contour; A first acquisition module is used to acquire a first defect value F1 according to the trapezoidal groove profile; wherein the first defect value F1 represents the deformation degree value of the trapezoidal groove; comprising: connecting the openings of the trapezoidal groove profile to obtain a target quadrilateral; based on the coincidence of any cusp of the target quadrilateral with the corresponding cusp of a preset standard trapezoid, to obtain a fused image; wherein the standard trapezoid is a trapezoid constructed based on the undeformed trapezoidal groove; according to the fused image, identifying the interleaved area S of the target quadrilateral and the standard trapezoid; according to the interleaved area S, acquiring a first defect value F1, F1=S*K1; wherein K1 is a first proportionality coefficient; The second acquisition module is used to acquire a second defect value F2 according to the triangular convex contour; wherein the second defect value F2 represents the deformation degree value of the triangular convex; including: constructing a first reference line according to the triangular convex contour; wherein the first reference line is a line connecting two end points of the triangular convex contour and the rectangular frame; identifying a target cusp corresponding to the triangular convex contour; wherein the target cusp is an end point of the triangular convex contour away from the rectangular frame; connecting the target cusp with the midpoint of the first reference line to obtain an offset line; acquiring an angle θ between the offset line and the first reference line; and acquiring a second defect value F2 according to the angle θ, F2=(π / 2-θ)*K2; wherein K2 is a second proportionality coefficient; A third acquisition module is used to acquire a third defect value F3 according to the contour of the curved plate; wherein the third defect value F3 represents the deformation degree value of the curved plate; including: mapping a theoretical reference point at a corresponding position of the infrared image; wherein the theoretical reference point is a center point corresponding to the inner arc line of the curved plate that has not been deformed; selecting a plurality of measurement points on the inner arc line of the curved plate, and respectively acquiring a spacing value between each measurement point and the theoretical reference point; acquiring an average value X of the plurality of spacing values ​​to obtain a third defect value F3, F3=|RX|*K3; wherein R is an arc radius corresponding to the inner arc line of the curved plate that has not been deformed, and K3 is a third adjustment coefficient; The evaluation module is used to obtain the defect evaluation index F of the target special-shaped steel according to the first defect value F1, the second defect value F2 and the third defect value F3, and judge the qualification of the target special-shaped steel according to the defect evaluation index F.

6. A computer device, characterized in that: The computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method for identifying defects of special-shaped steel based on infrared images as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the processor executes the computer program to implement the method for identifying defects in special-shaped steel based on infrared images as described in any one of claims 1 to 4.

Images