Image Analysis-Based Defect Identification System and Method for Steel Structures

Abstract

This invention relates to the field of steel structure image analysis technology and discloses a steel structure defect identification system and method based on image analysis. The system includes a data acquisition module, an image data analysis module, and a steel structure defect identification and judgment module. The data acquisition module comprises a region division unit and an image data acquisition unit. The system acquires abnormal points in the steel structure after ultrasonic testing and annotates these regions. For clear images, it analyzes and outputs the corresponding regional image influence coefficient and local defect influence coefficient. When the local defect influence coefficient exceeds the local defect threshold, an early warning is issued. When the local defect influence coefficient does not exceed the local defect threshold, the system analyzes the overall image data and determines whether to issue an early warning based on the overall image defect coefficient. This allows for rapid and accurate identification of steel structure surface defects, avoiding the need to consume computing power in areas that do not require inspection due to overall image analysis, thus shortening the inspection time.

Description

Technical Field

[0001] This invention relates to the field of steel structure image analysis technology, specifically to a steel structure defect identification system and method based on image analysis. Background Technology

[0002] Image recognition of steel structures plays a vital role in engineering and manufacturing. It can be used to detect surface defects, cracks, deformations, and other issues, aiding in quality control and timely repair, ensuring the safety and reliability of steel structures. Through image recognition technology, automatic detection and identification of steel structure components can be achieved, improving production efficiency and product quality. It can also be used for positioning and identification during automated assembly processes, enabling real-time monitoring of steel structures, timely detection and identification of potential problems, and thus early maintenance and repair, extending equipment life and reducing maintenance costs.

[0003] However, existing image recognition technologies analyze and judge the overall image of the steel structure. For longer steel structures, the limited width results in lower clarity of the main steel structure in the acquired image, which can lead to misjudgments. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a steel structure defect identification system and method based on image analysis, which has the advantages of rapid and accurate defect identification and solves the aforementioned technical problems.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a steel structure defect identification system based on image analysis, comprising a data acquisition module, an image data analysis module, and a steel structure defect identification and judgment module;

[0006] The data acquisition module includes a region division unit and an image data acquisition unit. The region division unit is used to acquire abnormal points of the steel structure after ultrasonic testing, mark the regions, and transmit the divided regions to the image data acquisition unit. The image data acquisition unit is used to acquire the image data of each region of the steel structure marked in the region division unit and the overall image data of the steel structure.

[0007] The image data analysis module includes an image blur degree analysis unit, a regional image analysis unit, and an overall image analysis unit. The image blur degree analysis unit is used to analyze the image data of each region of the steel structure in the image data acquisition unit, and give the clarity coefficient of each region. Based on the clarity coefficient of each region, it determines whether the corresponding region should be re-shot. The regional image analysis unit is used to perform secondary analysis on the regions that have been judged, and output the corresponding regional image influence coefficient.

[0008] The steel structure defect identification and judgment module performs a comprehensive calculation on the steel structure to be detected based on the influence coefficients of all regional images to obtain the local defect influence coefficient. When the local defect influence coefficient exceeds the local defect threshold, an early warning is issued. When the local defect influence coefficient does not exceed the local defect threshold, the overall image analysis unit in the image data analysis module is called to analyze the overall image data and output the overall image defect coefficient. The steel structure defect identification and judgment module determines whether to issue an early warning based on the overall image defect coefficient.

[0009] As a preferred embodiment of the present invention, the specific expression for the image data acquisition unit to acquire the image data of each region of the steel structure marked in the region division unit is as follows:

[0010]

[0011] in, Represents a region image dataset, These represent the image data of the first region, , No. Image data of each region, , No. Image data for each region, And the first Image data for each area was acquired by mobile shooting equipment.

[0012] As a preferred embodiment of the present invention, the image blur analysis unit is used to analyze the image data of each region of the steel structure in the image data acquisition unit, and to provide the sharpness coefficient of each region. The specific steps for determining whether to re-shoot the corresponding region based on the sharpness coefficient of each region are as follows:

[0013] Step A1: Obtain the first Shooting factors in each area ;

[0014] Step A2: Obtain the first Image factors captured in each area ;

[0015] Step A3: Comprehensive calculation of the first Clarity coefficient of each region The specific expression is as follows:

[0016]

[0017] in, and These represent two weight coefficients that sum to 1;

[0018] Step A4: Determine the first Clarity coefficient of each region If the resolution threshold is exceeded, terminate step A4 and set the resolution to the next step. Image data of each region The image is sent to the regional image analysis unit. If the image does not exceed the clarity threshold, it is retaken, and the settings are adjusted accordingly. Number of retakes in each area , This indicates the number of times step A4 is executed. Number of retakes in each area An error will be reported if the maximum number of retakes is exceeded.

[0019] As a preferred technical solution of the present invention, the data acquisition module also stores the proportion set of black areas after grayscale processing of the original steel structure image of each region.

[0020] In step A1, the first... Shooting factors in each area The specific steps are as follows:

[0021] Step A1.1: Read the first... Image data of each region , will the Image data of each region Perform binarization;

[0022] Step A1.2: Obtain the first The proportion of all black areas in each region And calculate the first Shooting factors in each area The specific expression is as follows:

[0023]

[0024] in, express The absolute value, Indicates the percentage of centralized storage The proportion of black areas in each region Indicates the first The proportion of black areas in each region.

[0025] As a preferred technical solution of the present invention, obtaining the first in step A2 Image factors captured in each area The specific steps are as follows:

[0026] Step A2.1: Read the mobile shooting device during the shooting process. Number of shakes during shooting in each area ;

[0027] Step A2.2: Calculate the first... Image factors captured in each area The specific expression is as follows:

[0028]

[0029] in, Represents the natural constant. This indicates that the mobile shooting equipment is shooting the first... Number of shakes during shooting in each area This represents the number of balance factors that avoid a denominator of 0.

[0030] As a preferred embodiment of the present invention, the specific steps of the region image analysis unit for performing secondary analysis on the judged region and outputting the corresponding region image influence coefficient are as follows:

[0031] Step B1: Read the first... Image data of each region The image was then cropped to retain only the steel structure area.

[0032] Step B2: Identify the surface of the steel structure and... Image data of each region Perform grayscale processing;

[0033] Step B3: Identify all non-black pixels in Step B2, and cluster adjacent non-black pixels to obtain several clustered regions;

[0034] Step B4: Obtain the first The area of ​​the smallest rectangular outline of each cluster region and the area of ​​the smallest closed circle contour ;

[0035] Step B5: Determine the first The area of ​​the smallest rectangular outline of each cluster region and the area of ​​the smallest closed circle contour If the smaller one exceeds the area threshold, then determine the first one. If the number of cluster regions exceeds a certain threshold, the smaller area is stored until all cluster regions have been identified, and the result is calculated. Regional image influence coefficient of each region The specific expression is as follows:

[0036]

[0037] in, Indicates the first The area of ​​the smallest rectangular outline of each cluster region and the area of ​​the smallest closed circle contour The smaller one, Indicates the first The regional image influence coefficient of each region This represents the total area of ​​the image after cropping. Indicates support for the common good The contour areas output by each cluster region are summed. .

[0038] As a preferred technical solution of the present invention, the steel structure defect identification and judgment module performs a comprehensive calculation on the steel structure to be detected based on the influence coefficients of all regional images, and finally obtains the specific expression of the local defect influence coefficient as follows:

[0039]

[0040] in, This represents the influence coefficient of local defects. This represents the summation of the regional image influence coefficients over a total of I regions. .

[0041] As a preferred embodiment of the present invention, the specific steps of the overall image analysis unit in analyzing the overall image data and outputting the overall image defect coefficient are as follows:

[0042] Step C1: Obtain the overall image data of the steel structure, and separately obtain the lengths of both sides of the steel structure;

[0043] Step C2: Calculate the overall image defect coefficient of the steel structure. The specific expression is as follows:

[0044]

[0045] in, and These represent the lengths of different sides of the steel structure. This represents the overall image defect coefficient of the steel structure. express The absolute value of.

[0046] As a preferred embodiment of the present invention, the specific steps of the steel structure defect identification and judgment module in determining whether to issue an early warning based on the overall image defect coefficient are as follows: when the overall image defect coefficient of the steel structure... Exceeding the overall judgment threshold When the overall image defect coefficient of the steel structure is... Not exceeding the overall judgment threshold No warning is issued at that time.

[0047] This invention also provides a steel structure defect identification method based on image analysis. Based on the aforementioned image analysis-based steel structure defect identification system, the method includes the following steps:

[0048] Step 1: Obtain the locations of abnormal points in the steel structure after ultrasonic testing and mark the areas accordingly;

[0049] Step 2: Obtain image data of each area of ​​the labeled steel structure and the overall image data of the steel structure;

[0050] Step 3: Analyze the image data of each area of ​​the steel structure, provide the sharpness coefficient of each area, and determine whether to re-shoot the corresponding area based on the sharpness coefficient of each area;

[0051] Step 4: Perform a secondary analysis on the identified regions and output the corresponding region image influence coefficients;

[0052] Step 5: Based on the influence coefficients of all regions, perform a comprehensive calculation on the steel structure to be inspected to finally obtain the local defect influence coefficient;

[0053] Step Six: When the local defect impact coefficient exceeds the local defect threshold, an early warning is issued. When the local defect impact coefficient does not exceed the local defect threshold, the overall image data is analyzed and the overall image defect coefficient is output. At the same time, it is determined whether to issue an early warning based on the overall image defect coefficient.

[0054] Compared with existing technologies, this invention provides a steel structure defect identification system and method based on image analysis, which has the following advantages:

[0055] This invention acquires abnormal points in steel structures after ultrasonic testing, marks these areas, captures images of the marked areas, and provides a clarity coefficient for each area. Based on the clarity coefficient, it determines whether to re-capture the corresponding area. For clear images, it analyzes and outputs the corresponding regional image influence coefficient and local defect influence coefficient. When the local defect influence coefficient exceeds the local defect threshold, an early warning is issued. When the local defect influence coefficient does not exceed the local defect threshold, the overall image data is analyzed, and an early warning is issued based on the overall image defect coefficient. This allows for rapid and accurate identification of surface defects in steel structures, avoiding the need to analyze the entire image and thus conserving computing power for areas that do not require testing, thereby shortening the testing time. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the system framework of the present invention;

[0057] Figure 2This is a schematic diagram of the process of the present invention. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Please see Figure 1 - Figure 2 A steel structure defect identification system based on image analysis includes a data acquisition module, an image data analysis module, and a steel structure defect identification and judgment module.

[0060] The data acquisition module includes a region segmentation unit and an image data acquisition unit. The region segmentation unit is used to acquire the abnormal points of the steel structure after ultrasonic testing, and to mark the regions. The segmented regions are then transmitted to the image data acquisition unit. The image data acquisition unit is used to acquire the image data of each region of the steel structure marked in the region segmentation unit, as well as the overall image data of the steel structure. The specific expression for acquiring the image data of each region of the steel structure marked in the region segmentation unit is as follows:

[0061]

[0062] in, Represents a region image dataset, These represent the image data of the first region, , No. Image data of each region, , No. Image data for each region, And the first Image data for each area is acquired by a mobile imaging device, which can be a crawling robot or a drone. Since the ultrasonic detection provides the distance to the specific anomaly point, the crawling robot or drone can be moved to the target point and the image is acquired after stabilization.

[0063] The image data analysis module includes an image blur degree analysis unit, a regional image analysis unit, and an overall image analysis unit. The image blur degree analysis unit is used to analyze the image data of each region of the steel structure in the image data acquisition unit, and give the clarity coefficient of each region. Based on the clarity coefficient of each region, it determines whether the corresponding region should be re-shot. The regional image analysis unit is used to perform secondary analysis on the regions that have been judged, and output the corresponding regional image influence coefficient.

[0064] The image blur analysis unit analyzes the image data of each region of the steel structure in the image data acquisition unit, provides the sharpness coefficient for each region, and determines whether to re-photograph the corresponding region based on the sharpness coefficient. The specific steps are as follows:

[0065] Step A1: Obtain the first Shooting factors in each area ;

[0066] In step A1, obtain the first Shooting factors in each area The specific steps are as follows:

[0067] Step A1.1: Read the first... Image data of each region , will the Image data of each region Perform binarization;

[0068] Step A1.2: Obtain the first The proportion of all black areas in each region And calculate the first Shooting factors in each area The specific expression is as follows:

[0069]

[0070] in, express The absolute value, Indicates the percentage of centralized storage The proportion of black areas in each region Indicates the first The proportion of black areas in each region;

[0071] Step A2: Obtain the first Image factors captured in each area ;

[0072] In step A2, obtain the first Image factors captured in each area The specific steps are as follows:

[0073] Step A2.1: Read the mobile shooting device during the shooting process. Number of shakes during shooting in each area ;

[0074] Step A2.2: Calculate the first... Image factors captured in each area The specific expression is as follows:

[0075]

[0076] in, Represents the natural constant. This indicates that the mobile shooting equipment is shooting the first... Number of shakes during shooting in each area This represents a balance factor to avoid a denominator of 0;

[0077] Step A3: Comprehensive calculation of the first Clarity coefficient of each region The specific expression is as follows:

[0078]

[0079] in, and These represent two weight coefficients that sum to 1;

[0080] Step A4: Determine the first Clarity coefficient of each region If the resolution threshold is exceeded, terminate step A4 and set the resolution to the next step. Image data of each region The image is sent to the regional image analysis unit. If the image does not exceed the clarity threshold, it is retaken, and the settings are adjusted accordingly. Number of retakes in each area , This indicates the number of times step A4 is executed. Number of retakes in each area An error will be reported if the maximum number of retakes is exceeded.

[0081] The data acquisition module also stores the proportion set of black areas after grayscale processing of the original steel structure images of each region.

[0082] The specific steps of the region image analysis unit to perform secondary analysis on the identified regions and output the corresponding region image influence coefficients are as follows:

[0083] Step B1: Read the first... Image data of each region The image was then cropped to retain only the steel structure area.

[0084] Step B2: Identify the surface of the steel structure and... Image data of each region Perform grayscale processing;

[0085] Step B3: Identify all non-black pixels in Step B2, and cluster adjacent non-black pixels to obtain several clustered regions;

[0086] Step B4: Obtain the first The area of ​​the smallest rectangular outline of each cluster region and the area of ​​the smallest closed circle contour ;

[0087] Step B5: Determine the first The area of ​​the smallest rectangular outline of each cluster region and the area of ​​the smallest closed circle contour If the smaller one exceeds the area threshold, then determine the first one. If the number of cluster regions exceeds a certain threshold, the smaller area is stored until all cluster regions have been identified, and the result is calculated. Regional image influence coefficient of each region The specific expression is as follows:

[0088]

[0089] in, Indicates the first The area of ​​the smallest rectangular outline of each cluster region and the area of ​​the smallest closed circle contour The smaller one, Indicates the first The regional image influence coefficient of each region This represents the total area of ​​the image after cropping. Indicates support for the common good The contour areas output by each cluster region are summed. .

[0090] The steel structure defect identification and judgment module comprehensively calculates the local defect influence coefficient based on the influence coefficients of all regional images of the steel structure to be detected. When the local defect influence coefficient exceeds the local defect threshold, an early warning is issued. When the local defect influence coefficient does not exceed the local defect threshold, the overall image analysis unit in the image data analysis module is called to analyze the overall image data and output the overall image defect coefficient. The steel structure defect identification and judgment module determines whether to issue an early warning based on the overall image defect coefficient.

[0091] The steel structure defect identification and judgment module comprehensively calculates the local defect influence coefficient based on the influence coefficients of all regional images of the steel structure to be detected, and the specific expression of the local defect influence coefficient is as follows:

[0092]

[0093] in, This represents the influence coefficient of local defects. This represents the summation of the regional image influence coefficients over a total of I regions. The specific steps by which the overall image analysis unit analyzes the overall image data and outputs the overall image defect coefficient are as follows:

[0094] Step C1: Obtain the overall image data of the steel structure, and separately obtain the lengths of both sides of the steel structure;

[0095] Step C2: Calculate the overall image defect coefficient of the steel structure. The specific expression is as follows:

[0096]

[0097] in, and These represent the lengths of different sides of the steel structure. This represents the overall image defect coefficient of the steel structure. express The absolute value of.

[0098] The specific steps of the steel structure defect identification and judgment module in determining whether to issue an early warning based on the overall image defect coefficient are as follows: when the overall image defect coefficient of the steel structure... Exceeding the overall judgment threshold When the overall image defect coefficient of the steel structure is... Not exceeding the overall judgment threshold No warning is issued at that time;

[0099] When issuing an alert, the captured photos are sent to staff for review.

[0100] Example 1:

[0101] Two abnormalities were found during the ultrasound examination conducted during this procedure, and the images were taken using a drone.

[0102] When photographing the first anomaly, the sharpness coefficient of the first area is determined as shown in Table 1 below:

[0103] Table 1

[0104]

[0105] , At this point, the calculation yields >0.90, at this point the judgment is clear, and two clusters are obtained through the regional image analysis unit. See Table 2 below for specific parameters;

[0106] Table 2

[0107]

[0108] At this point, the calculation yields... ;

[0109] When photographing the second anomaly, the sharpness coefficient of the second area is determined as shown in Table 3 below:

[0110] Table 3

[0111]

[0112] , At this point, the calculation yields >0.90, at this point the judgment is clear, and one cluster is obtained through the regional image analysis unit. See Table 4 below for specific parameters;

[0113] Table 4

[0114]

[0115] At this point, the calculation yields... ;

[0116] At this point, a judgment is made. =0.304, an early warning is issued. At this time, the overall image data of the steel structure is not acquired, so that the surface defects of the steel structure can be identified quickly and accurately. This avoids the situation where the equipment's computing power is occupied by areas that do not need to be detected due to the analysis of the overall image, and shortens the detection time.

[0117] Example 2:

[0118] In this embodiment, the local defect influence coefficient does not exceed the local defect threshold. At this time, the overall image data of the steel structure is acquired, and the lengths of both sides of the steel structure are acquired separately (see Table 5 below).

[0119] Table 5

[0120]

[0121] Defect coefficient of overall steel structure image It did not exceed the overall judgment threshold. No warning will be issued at this time;

[0122] The threshold value in the above embodiments is set to facilitate comparison. The value of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each group of sample data. As long as it does not affect the ratio between the parameter and the quantized value, it can be determined by those skilled in the art based on each sample data and multiple rounds of experiments.

[0123] This invention also provides a steel structure defect identification method based on image analysis. Based on the aforementioned image analysis-based steel structure defect identification system, the method includes the following steps:

[0124] Step 1: Obtain the locations of abnormal points in the steel structure after ultrasonic testing and mark the areas accordingly;

[0125] Step 2: Obtain image data of each area of ​​the labeled steel structure and the overall image data of the steel structure;

[0126] Step 3: Analyze the image data of each area of ​​the steel structure, provide the sharpness coefficient of each area, and determine whether to re-shoot the corresponding area based on the sharpness coefficient of each area;

[0127] Step 4: Perform a secondary analysis on the identified regions and output the corresponding region image influence coefficients;

[0128] Step 5: Based on the influence coefficients of all regions, perform a comprehensive calculation on the steel structure to be inspected to finally obtain the local defect influence coefficient;

[0129] Step Six: When the local defect impact coefficient exceeds the local defect threshold, an early warning is issued. When the local defect impact coefficient does not exceed the local defect threshold, the overall image data is analyzed and the overall image defect coefficient is output. At the same time, it is determined whether to issue an early warning based on the overall image defect coefficient.

[0130] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A steel structure defect recognition system based on image analysis, characterized by: The data acquisition module, the image data analysis module and the steel structure defect identification and judgment module are included. The data acquisition module includes a region division unit and an image data acquisition unit, the region division unit is used for acquiring abnormal point positions of the steel structure after ultrasonic detection, and region labeling is performed, and the divided regions are transmitted to the image data acquisition unit, and the image data acquisition unit is used for acquiring image data of each region of the steel structure labeled by the region division unit and overall image data of the steel structure. The image data analysis module includes an image blur degree analysis unit, a regional image analysis unit and an overall image analysis unit, the image blur degree analysis unit is used for analyzing the image data of each region of the steel structure in the image data acquisition unit, giving a clear coefficient of each region, and determining whether to re-shoot the corresponding region based on the clear coefficient of each region, the regional image analysis unit is used for secondary analysis of the judged region, and outputs a corresponding regional image influence coefficient. The steel structure defect identification and judgment module performs comprehensive calculation on the steel structure to be detected based on all regional image influence coefficients, and finally obtains a local defect influence coefficient, when the local defect influence coefficient exceeds a local defect threshold, a warning is given, when the local defect influence coefficient does not exceed the local defect threshold, the overall image analysis unit in the image data analysis module is called to analyze the overall image data, and an overall image defect coefficient is output, and the steel structure defect identification and judgment module judges whether to give a warning based on the overall image defect coefficient.

2. The image analysis based steel structure defect identification system according to claim 1, characterized in that: The specific expression of the image data acquisition unit for acquiring the image data of each region of the steel structure labeled by the region division unit is as follows: in, Represents a region image dataset, These represent the image data of the first region, , No. Image data of each region, , No. Image data for each region, And the first Image data for each area was acquired by mobile shooting equipment.

3. The image analysis based steel structure defect identification system according to claim 2, characterized in that: The specific steps of the image blur degree analysis unit for analyzing the image data of each region of the steel structure in the image data acquisition unit, giving a clear coefficient of each region, and determining whether to re-shoot the corresponding region based on the clear coefficient of each region are as follows: Step A1: Obtain the shooting factor of the first region ​ Step A2: Obtain the image factor of the first region ​ Step A3: Comprehensive calculation of the first Clarity coefficient of each region The specific expression is as follows: wherein and respectively denote two weight coefficients that sum to 1. Step A4: judging whether the distinctness coefficient of the first region exceeds a distinctness threshold value, if exceeding the distinctness threshold value, terminating step A4 and sending the image data of the first region to a region image analysis unit, if not exceeding the distinctness threshold value, re-photographing and setting the re-photographing number of the first region, ​​​​​​​​​ 4. The image analysis based steel structure defect recognition system of claim 3, wherein: The data acquisition module also stores a set of proportions of black regions after gray processing of the original image of the steel structure in each region; The specific steps of acquiring the shooting factor of the first region shot in the step A1 are as follows: The specific steps of acquiring the shooting factor of the first region shot in the step A1 are as follows:​ Step A1.1: Read image data of the first region Step A1.2: Perform binarization processing on the image data of the first region Step A1.3: Perform edge detection on the image data of the first region Step A1.4: Perform contour extraction on the image data of the first region Step A1.5: Perform shape recognition on the image data of the first region Step A1.2: The proportion of all black regions of the first region is obtained , and the shooting factor of the first region is calculated , and the specific expression is as follows: wherein, represents the absolute value of represents a proportion of black regions of the th region stored in the proportion, represents a proportion of black regions of the th region.

5. The image analysis based steel structure defect identification system according to claim 4, characterized in that: The step A2 obtains the image factor of the first region The specific steps are as follows: The specific steps are as follows: Step A2.1: Read the number of times the mobile camera device has shaken while taking pictures of the first region ; Step A2.2: Calculate the image factor for the first region , which is expressed as follows: wherein, represents a natural constant, represents the number of times the mobile photographing device is shaken when photographing the first region, represents a balancing factor to avoid a denominator of 0.

6. The image analysis based steel structure defect identification system of claim 3, wherein: The specific steps of the regional image analysis unit for secondary analysis of the judged region and output of the corresponding regional image influence coefficient are as follows: Step B1: Read the image data of the first region and crop the image to keep only the steel structure region. Step B2: Read the image data of the second Step B2: identify the surface of the steel structure, and perform gray scale processing on the image data of the first region ; Step B3: identify all non-black pixel points in step B2, and cluster adjacent non-black pixel points to obtain a plurality of clustering regions; Step B4: Obtain the area of the minimum rectangular profile of the first cluster region and the area of the minimum closed circular profile ;​ Step B5: Determine the first The area of ​​the smallest rectangular outline of each cluster region and the area of ​​the smallest closed circle contour If the smaller one exceeds the area threshold, then determine the first one. If the number of cluster regions exceeds a certain threshold, the smaller area is stored until all cluster regions have been identified, and the result is calculated. Regional image influence coefficient of each region The specific expression is as follows: wherein, represents the area of the minimum rectangular profile of the th cluster region and the area of the minimum closed circular profile the smaller one of, represents the area image influence coefficient of the th region, represents the total area of the cropped image, represents the summing of the profile areas output for the cluster regions, .

7. The image analysis based steel structure defect recognition system according to claim 6, characterized in that: The specific expression of the steel structure defect identification and judgment module for performing comprehensive calculation on the steel structure to be detected based on all regional image influence coefficients and finally obtaining a local defect influence coefficient is as follows: wherein denotes a local defect influence coefficient, denotes a summation over the region image influence coefficients for the co I regions, .

8. The image analysis based steel structure defect identification system according to claim 7, characterized in that: The specific steps of the overall image analysis unit for analyzing the overall image data and outputting an overall image defect coefficient are as follows: Step C1: acquire overall image data of the steel structure, and acquire the lengths of both sides of the steel structure respectively; Step C2: Calculate the overall image defect coefficient of the steel structure The specific expression is as follows: wherein, and respectively represent the length of different sides of the steel structure, represents the overall image defect coefficient of the steel structure, represents the absolute value of.

9. The image analysis based steel structure defect identification system according to claim 8, characterized in that: The specific steps of the steel structure defect identification and judgment module judging whether to perform early warning based on the overall image defect coefficient are as follows: when the steel structure overall image defect coefficient exceeds the overall judgment threshold , early warning is performed; when the steel structure overall image defect coefficient does not exceed the overall judgment threshold , no early warning is performed.

10. A method of identifying defects in steel structures based on image analysis, according to the system for identifying defects in steel structures based on image analysis according to any one of claims 1 to 9, characterized in that: The method includes the following steps: Step one: acquire abnormal point positions of the steel structure after ultrasonic detection, and perform region labeling; Step two: acquire image data of each region of the labeled steel structure and overall image data of the steel structure; Step three: analyze the image data of each region of the steel structure, and give the clarity coefficient of each region, and determine whether to re-shoot the corresponding region based on the clarity coefficient of each region; Step four: secondary analysis is performed on the region whose judgment is completed, and the corresponding region image influence coefficient is output; Step five: based on all the region image influence coefficients, the steel structure to be detected is comprehensively calculated to finally obtain the local defect influence coefficient; Step six: when the local defect influence coefficient exceeds the local defect threshold, a warning is given; when the local defect influence coefficient does not exceed the local defect threshold, the overall image data is analyzed, and the overall image defect coefficient is output, and whether to give a warning is judged based on the overall image defect coefficient.

Images