A method for determining the treatment method of concrete cracks for bridges using image processing

Abstract

The present invention proposes a method for determining a crack treatment method for concrete bridges by image processing, which mainly includes: image acquisition, grayscale processing of the maximum value of the image, binarization processing of the grayscale processed image by the maximum entropy threshold method, identification of the start and end points of the cracks in the binarized image, identification of the nodes of the cracks, burr removal of the image after the crack nodes are identified, identification of the maximum length and total length of the cracks in the image after the burr removal, calculation of the crack danger level, and determination of the crack treatment method according to the calculated crack danger level. The method avoids the influence of the measurement personnel themselves on the measurement results, the missed detection and multiple detection of cracks, saves manpower and material resources, and ensures the personal safety of the detection personnel.

Description

Technical Field

[0001] The present invention relates to the field of information model technology, and in particular to the application of image processing technology in bridge engineering. The present invention proposes a method for determining a method for processing concrete cracks for bridges using image processing. Background Art

[0002] During the use of bridges, they will experience a lot of damage from vehicles, man-made damage and natural erosion, resulting in different conditions and degrees of damage on the bridge surface. At the same time, due to the degradation of the performance of the materials used in bridge construction, cracks will appear on the bottom plate and sides of the bridge, and even exposed steel bars will appear, which poses a serious safety hazard. If cracks on the bridge surface can be discovered and repaired in time at the early stage of formation, the probability of accidents caused by bridge diseases will be greatly reduced. Therefore, regular crack detection of bridges is of great significance.

[0003] With the rapid development of computer technology and the continuous progress of computer vision algorithm research, the use of non-contact methods such as digital image processing technology to detect bridge cracks has become a research hotspot in recent years. However, due to the complex background of bridge cracks, the detected cracks often contain a large number of noise pixels, resulting in reduced detection accuracy. At the same time, traditional detection methods process image pixels roughly during image recognition, which can easily lead to missed detection or over-detection of cracks.

[0004] Therefore, there is an urgent need for an automated image recognition technology for bridge cracks to further analyze bridge diseases and take corresponding protective measures based on the analysis results. This will greatly improve the quality and speed of crack detection, reduce the cost of crack detection, and lay a theoretical and practical foundation for further real-time AI intelligent detection of building damage and unsupervised classification of remote sensing technology. Summary of the invention

[0005] In order to solve the problem that it is difficult to accurately detect and process concrete cracks in bridges, the present invention proposes a method for determining a processing method for concrete cracks in bridges using image processing.

[0006] A method for determining a treatment method for cracks in concrete used in bridges by image processing comprises the following steps:

[0007] Step 1: Use a drone to take pictures of bridge concrete cracks;

[0008] Step 2: Perform maximum grayscale processing on the photographed bridge concrete crack image;

[0009] Step 3: Binarize the grayscale image using the maximum entropy threshold method;

[0010] Step 4: Identify the starting and ending points of the cracks in the binary processed image;

[0011] Step 5: Identify the nodes of the cracks in the image after the start and end points of the cracks are identified;

[0012] Step 6: Remove burrs from the image after the crack nodes are identified;

[0013] Step 7: Identify the maximum length and total length of the crack in the image after burr removal;

[0014] Step 8: Calculate the crack danger level based on the maximum length and total length of the identified cracks;

[0015] Step 9: Determine the crack treatment method based on the calculated crack danger level.

[0016] Further technology of the present invention:

[0017] Preferably, the starting and ending points in step 4 are points that meet two conditions: the grayscale value of the point itself is 1 and there is only one pixel point with a grayscale value of 1 in the neighborhood a.

[0018] Preferably, the neighborhood a is determined by:

[0019] , take the integer;

[0020] Where:

[0021] k — gray level of the image;

[0022] —The percentage of pixels with gray level i in the image.

[0023] Preferably, the node identification rule in step 5 is 8. There are b points with gray value of 1 in the neighborhood, where the gray value b can be determined by the following formula:

[0024] ;

[0025] Where: M is the total number of pixels in the row, is the serial number of the pixel in the row, N is the total number of pixels in the column, is the serial number of the pixel in the column, Represents the horizontal gradient of the defect image, Represents the vertical gradient of the defect image.

[0026] Preferably, the burr removal in step six is: treating branches with a pixel length less than 3 as burrs caused by errors that need to be removed during crack skeleton generation, and treating branches with a pixel length greater than or equal to 3 as normal branches that need to be retained and marked.

[0027] Preferably, in step seven:

[0028] The total length of the cracks is converted to the total number of pixels contained in each crack, and the maximum length of the cracks is converted to the maximum number of pixels contained in each crack.

[0029] Preferably, the conversion relationship between the crack length and the number of pixels is:

[0030] ;

[0031] Where:

[0032] —crack length, m;

[0033] — number of pixels;

[0034] —Resolution.

[0035] Preferably, step eight is specifically:

[0036] The crack danger level R is calculated using the following formula:

[0037] ;

[0038] Where:

[0039] R—Crack danger level;

[0040] —bridge span, m;

[0041] —Radius of the circular curve of the bridge route, m;

[0042] —bridge height, m;

[0043] —Total length of cracks in bridge concrete, m;

[0044] —Maximum length of cracks in bridge concrete, m;

[0045] —Average speed of vehicles on the bridge, km / h;

[0046] —Maximum speed limit for vehicles on bridges, km / h.

[0047] Preferably, the crack treatment method in step nine is specifically as follows:

[0048] If R is less than 1, the surface treatment method is adopted; if 1≤R≤2, the grouting method is adopted; if R is greater than 2, the structural reinforcement method is adopted.

[0049] The beneficial effects of the present invention are:

[0050] The image processing bridge concrete crack processing method proposed by the present invention is based on the theory of computer image processing. First, image data is collected, and the collected image is further grayed and binarized, and the starting and ending points and nodes of the cracks are identified, and the burrs of the image are further removed. The maximum length and total length of the cracks are calculated by computer and graphic software, and the degree of crack danger is calculated. The crack processing method is finally determined based on this. The following situations are avoided: a large amount of manpower and material resources are required when conducting large-scale measurements over a large range; the safety of the surveyors is ensured, and it is dangerous to approach the bridge for inspection, and it is even difficult to approach, resulting in missed data; the measurement results are greatly affected by the professional quality of the surveyors themselves, resulting in instability of the measurement results. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0052] Figure 1 A schematic diagram of a flow chart of a method for determining a method for processing cracks in concrete used in bridges by image processing;

[0053] Figure 2 This is a schematic diagram of the image binarization processing using the maximum entropy threshold method in step 3;

[0054] Figure 3 This is a schematic diagram of the crack start and end points after identification in step 4;

[0055] Figure 4 A schematic diagram of the nodes identified by the computer in step 5;

[0056] Figure 5 This is the image after removing the burrs in step 6;

[0057] Figure 6 This is the identification diagram of the two cracks in step seven. DETAILED DESCRIPTION

[0058] In order to make the technical means, creative features, purpose and efficacy of the present invention easy to understand, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain this description and are not intended to limit the present invention.

[0059] Example: Detect concrete cracks on a highway bridge in a mountainous area in southern Anhui. Figure 1 , mainly including the following steps:

[0060] Step 1: Use a drone to take pictures of the bridge concrete cracks. The focal length is 4.50mm, the exposure is 1 / 300 seconds, and the color format is RGB.

[0061] Step 2: Perform maximum grayscale processing on the image taken in step 1.

[0062] Step 3: Binarize the image after grayscale processing in step 2 using the maximum entropy threshold method. Figure 2 .

[0063] Step 4: Identify the starting and ending points of the cracks in the image that has been binarized in step 3.

[0064] The computer recognition rule is that the starting and ending points of the cracks in the image are points that meet two conditions: the gray value itself is 1 and there is only one pixel with a gray value of 1 in the neighborhood a. The neighborhood a is determined by the following method:

[0065] , take the integer;

[0066] Where:

[0067] —Process the neighborhood value of each pixel in the image;

[0068] k—the gray level of the image, which is 1, 2, 3, ... 255;

[0069] —The proportion of pixels with gray level i in the image. The computer automatically identifies the proportion of gray level i (1, 2, 3, ... 255).

[0070] The calculation can obtain the neighborhood value of this image Take 3, 5, and 7. The starting and ending points of the identified cracks are as follows Figure 3 As shown, the left mark is the starting point and the left mark is the end point.

[0071] Step 5: Identify the nodes of the cracks in the image after identifying the starting and ending points of the cracks in step 4.

[0072] The computer recognition rule is 8. There are b points with gray value 1 in the neighborhood. The gray value b can be determined by the following formula:

[0073] ;

[0074] Where: M is the total number of pixels in the row, is the serial number of the pixel in the row, N is the total number of pixels in the column, is the serial number of the pixel in the column, Represents the horizontal gradient of the defect image, Represents the vertical gradient of the defect image.

[0075] The nodes identified by the computer are as follows Figure 4 shown.

[0076] Step six, remove burrs from the image after the crack nodes are identified in step five.

[0077] The removal rule is: branches with a pixel length (i.e., the number of pixels contained in the branch) less than 3 are regarded as burrs caused by errors that need to be removed during crack skeleton generation, and branches greater than or equal to 3 are normal branches that need to be retained and marked.

[0078] After deburring Figure 5 shown.

[0079] Step seven, identifying the maximum length and total length of the crack in the image after burr removal in step six.

[0080] The identification and calculation rules are as follows: the total length of the crack is converted to the sum of the number of pixels contained in each crack. The maximum length of the crack is converted to the maximum number of pixels contained in each crack.

[0081] The two cracks in this image are shown in Figures L1 and L2. The total length of the cracks is 14173 pixels, which is the number of pixels contained in L1 and L2, and the maximum length of the cracks is the maximum value of 9921 pixels contained in the cracks of L1 and L2.

[0082] Crack length The conversion relationship with the number of pixels px is:

[0083] ;

[0084] Where:

[0085] —crack length, m;

[0086] — number of pixels;

[0087] - resolution;

[0088] Calculation shows that the total length of the bridge concrete cracks is 1.2m, and the maximum length of the bridge concrete cracks is 0.7m.

[0089] Step 8: Calculate the crack danger level R using the total length and maximum length of the cracks identified in step 7:

[0090] ;

[0091] Where:

[0092] R—Crack danger level;

[0093] - Bridge span, 120m;

[0094] —The radius of the circular curve of the bridge route is 2650m;

[0095] — Bridge height, 2.3m;

[0096] —The total length of the cracks in the bridge concrete is taken as 1.2m;

[0097] —The maximum length of a crack in bridge concrete is 0.7m;

[0098] —The average speed of vehicles on the bridge is 95 km / h;

[0099] —The maximum speed limit for vehicles on bridges is 120km / h;

[0100] Calculation shows that the crack danger level R is 0.007.

[0101] Step 9: Determine the crack treatment method based on the crack danger level R calculated in step 8:

[0102] If R is less than 1, the surface treatment method is adopted; if 1≤R≤2, the grouting method is adopted; if R is greater than 2, the structural reinforcement method is adopted.

[0103] The crack danger level R calculated in step eight is 0.007, so the cracks in this bridge are treated using the surface treatment method.

[0104] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for determining a treatment method for concrete cracks in bridges using image processing, characterized in that: The following steps are involved: Step 1: Use a drone to take pictures of bridge concrete cracks; Step 2: Perform maximum grayscale processing on the photographed bridge concrete crack image; Step 3: Binarize the grayscale image using the maximum entropy threshold method; Step 4: Identify the starting and ending points of the cracks in the binary processed image; Step 5: Identify the nodes of the cracks in the image after the start and end points of the cracks are identified; Step 6: Remove burrs from the image after the crack nodes are identified; Step 7: Identify the maximum length and total length of the crack in the image after burr removal; In step seven: The total length of the crack is converted to the sum of the number of pixels contained in each crack, and the maximum length of the crack is converted to the maximum number of pixels contained in each crack; The conversion relationship between crack length and pixel number is: Where: l—crack length, m; px—number of pixels; t—resolution; Step 8: Calculate the crack danger level based on the maximum length and total length of the identified cracks; Step 8 Specific: The crack danger level R is calculated using the following formula: Where: R—Crack danger level; L—bridge span, m; r—radius of the circular curve of the bridge route, m; h—bridge height, m; l 总 —Total length of cracks in bridge concrete, m; l 最大 —Maximum length of cracks in bridge concrete, m; V 平均 —Average speed of vehicles on the bridge, km / h; V 最大 — Maximum speed limit for vehicles on bridges, km / h; Step 9: Determine the crack treatment method based on the calculated crack danger level; Step 9: The specific method of crack treatment is as follows: If R is less than 1, the surface treatment method is adopted; if 1≤R≤2, the grouting method is adopted; if R is greater than 2, the structural reinforcement method is adopted.

2. The method for determining the treatment method of concrete cracks for bridges by image processing according to claim 1, characterized in that: In step 4, the starting and ending points are points that meet two conditions: the grayscale value of the point itself is 1 and there is only one pixel with a grayscale value of 1 in the neighborhood a.

3. The method for determining the treatment method of concrete cracks for bridges by image processing according to claim 2, characterized in that: The neighborhood a is determined as follows: Integer Where: k — gray level of the image; P i —The percentage of pixels with gray level i in the image.

4. According to the method for determining the treatment method of concrete cracks for bridges by image processing as described in claim 1, it is characterized in that: The node identification rule in step 5 is that there are b points with gray value 1 in the 8-neighborhood; the gray value b can be determined by the following formula: Where: M is the total number of pixels in the row, i is the serial number of the pixel in the row, N is the total number of pixels in the column, j is the serial number of the pixel in the column, Represents the horizontal gradient of the defect image, Represents the vertical gradient of the defect image.

5. According to the method for determining the treatment method of concrete cracks for bridges by image processing as described in claim 1, it is characterized in that: In step 6, burr removal is as follows: branches with a pixel length less than 3 are regarded as burrs caused by errors that need to be removed during crack skeleton generation, and branches with a pixel length greater than or equal to 3 are normal branches that need to be retained and marked.

Images