CN114638805B - Track slab crack detection method, system and storage medium - Google Patents
Track slab crack detection method, system and storage medium Download PDFInfo
- Publication number
- CN114638805B CN114638805B CN202210274970.5A CN202210274970A CN114638805B CN 114638805 B CN114638805 B CN 114638805B CN 202210274970 A CN202210274970 A CN 202210274970A CN 114638805 B CN114638805 B CN 114638805B
- Authority
- CN
- China
- Prior art keywords
- plaque
- point
- image
- local extremum
- crack
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 45
- 238000003384 imaging method Methods 0.000 claims abstract description 60
- 238000000034 method Methods 0.000 claims abstract description 37
- 238000006243 chemical reaction Methods 0.000 claims abstract description 36
- 238000005336 cracking Methods 0.000 claims abstract description 15
- 238000012216 screening Methods 0.000 claims description 15
- 239000003550 marker Substances 0.000 claims description 11
- 238000013507 mapping Methods 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 4
- 230000001788 irregular Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 description 7
- 230000004075 alteration Effects 0.000 description 5
- 238000005259 measurement Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 241001270131 Agaricus moelleri Species 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- 229910001294 Reinforcing steel Inorganic materials 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 201000009310 astigmatism Diseases 0.000 description 1
- 230000005587 bubbling Effects 0.000 description 1
- 238000003763 carbonization Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The application discloses a track slab crack detection method, a system and a storage medium, wherein the method comprises the following steps: acquiring a target acquisition image comprising a track plate crack and a calibration plate, wherein the calibration plate is arranged in a preset range around the track plate crack; under the preset local extremum conditions, respectively identifying plaque points corresponding to the mark patterns from the target acquisition image according to the estimated connected domain characteristics and the connected domain radius; determining an optimal local extremum condition from a plurality of local extremum conditions based on the degree of difference between the plaque point imaging position and the actual position; under the optimal local extremum condition, determining a real plane coordinate string reflecting the actual long width of the crack based on a coordinate conversion mode between the plaque point and the corresponding image imaging plane; and determining the cracking degree of the rail plate crack according to the real plane coordinate string. The implementation of the method can improve the detection accuracy of the rail plate crack.
Description
Technical Field
The invention relates to the technical field of track detection, in particular to a track slab crack detection method, a track slab crack detection system and a storage medium.
Background
The crack of the high-speed rail plate is a phenomenon commonly existing in a concrete structure, and the occurrence of the crack can not only reduce the impervious capacity of a building and influence the service function of the building, but also cause corrosion of reinforcing steel bars, carbonization of the concrete and reduce the durability of materials and influence the bearing capacity of the building, so that the crack of the high-speed rail plate is carefully researched and treated differently and is treated by a reasonable method. However, the high-speed railway track slab cracks have various causes, and in order to ensure the safe operation of railways, accurate detection and control of the track slab cracks are needed. In the prior art, the data investigation is carried out on the road section needing to carry out the crack repair of the track slab, then the severity degree of the crack is manually judged, and the priority and the method of the repair construction are determined according to the severity degree. However, this kind of detection mode based on manual work has the cost of labor height, detects the slow problem of speed, and easily leaks and examines, has reduced track board crack detection accuracy from this.
Disclosure of Invention
The invention aims to solve the technical problem of low detection accuracy of a track slab crack in the prior art, and provides a track slab crack detection method, a track slab crack detection system and a storage medium.
The technical scheme adopted for solving the technical problems is as follows: a method for detecting a crack of a track slab is constructed, comprising:
s1, acquiring a target acquisition image comprising a track plate crack and a calibration plate, wherein the calibration plate is arranged in a preset range around the track plate crack;
s2, under the preset multiple local extremum conditions, respectively identifying plaque points corresponding to the marker graph from the target acquisition image according to the estimated connected domain characteristics and the connected domain radius;
S3, determining an optimal local extremum condition from the local extremum conditions based on the difference degree between the imaging position and the actual position of the plaque point;
S4, under the optimal local extremum condition, determining a real plane coordinate string reflecting the actual long width of the crack based on a coordinate conversion mode between the plaque point and the corresponding image imaging plane;
s5, determining the cracking degree of the rail plate crack according to the real plane coordinate string.
The application discloses a track slab crack detection system, which comprises an image acquisition module, a plaque identification module, a condition screening module, a coordinate conversion module and a degree detection module, wherein:
the image acquisition module is used for acquiring a target acquisition image comprising a track plate crack and a calibration plate, wherein the calibration plate is arranged in a preset range around the track plate crack;
The plaque identification module is used for respectively identifying plaque points corresponding to the mark patterns from the target acquisition image according to the estimated connected domain characteristics and the connected domain radius under the preset multiple local extremum conditions;
the condition screening module is used for determining an optimal local extremum condition from the plurality of local extremum conditions based on the degree of difference between the imaging position and the actual position of the plaque point;
the coordinate conversion module is used for determining a real plane coordinate string reflecting the actual long width of the crack based on a coordinate conversion mode between the plaque point and the corresponding image imaging plane under the optimal local extremum condition;
And the degree detection module is used for determining the cracking degree of the track slab crack according to the real plane coordinate string.
The application discloses a readable storage medium, which comprises a track slab crack detection method program, wherein the track slab crack detection method program realizes the steps of the method when being executed by a processor.
According to the track plate crack detection method, system and storage medium, on one hand, corresponding plaque points are detected from the target acquisition images of the comprehensive calibration plate and the track plate crack based on the preset local extremum conditions, and the detection accuracy of the track plate crack can be ensured under the condition of improving the identification accuracy of the plaque points due to the fact that the fitted center accuracy is high. On the other hand, the optimal local extremum condition is determined based on the difference value between the actual position of the plaque point and the corresponding imaging position, so that the identification error of the plaque point can be reduced, and the measurement accuracy is improved. Finally, because the plane of the calibration plate can be similar to the plane of the track plate, a real plane coordinate string reflecting the actual long width of the crack is determined based on a coordinate conversion mode between the calibration point plate and the corresponding image imaging plane, and the cracking degree of the crack of the track plate is reversely pushed based on the real plane coordinate string, even under the condition that the distance, the angle and the like are unknown, the accurate detection of the cracking degree of the crack of the track plate can be ensured, and the detection efficiency of the system is improved.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a flow chart of a track slab crack detection method in one embodiment of the invention;
fig. 2 is a system block diagram of a track slab crack detection system in one embodiment of the invention.
Detailed Description
The principles and features of the present invention are described below with reference to the drawings, the examples are illustrated for the purpose of illustrating the invention and are not to be construed as limiting the scope of the invention.
In one or more embodiments of the present invention, as shown in fig. 1, a track slab crack detection method is provided, which is applied to a computer device (the computer device may be a terminal or a server, the terminal may be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices, the server may be a stand-alone server or a server cluster composed of a plurality of servers), and includes the following steps:
step S1, acquiring a target acquisition image comprising a track plate crack and a calibration plate, wherein the calibration plate is arranged in a preset range around the track plate crack.
Specifically, the target captured image is captured based on a preset capturing device, such as a mobile phone, a camera, and the like. The computer device will request connection from the photographing device through a preset connection protocol, for example, through tcp connection protocol, udp connection protocol, etc., and upon determining that connection to the photographing device is successful, acquire a target captured image transmitted via the photographing device.
In one embodiment, the calibration plate comprises a plurality of mark points reflecting identification characteristics; the specifications of its corresponding calibration plate are also different for different scale rail plate cracks, wherein: for the identification of the calibration plates with different specifications, the identification is performed based on a preset identification mode, wherein the identification mode comprises at least one of a first identification mode based on color, a second identification mode based on shape and a third identification mode based on size.
When the calibration plates are arranged, the calibration plates with different specifications can be utilized to calibrate the cracks of the track plates with different scales. In addition, in order to improve the accuracy of observing the crack, a calibration plate with smaller size (the specific size of the calibration plate is not limited in the present embodiment, and in different embodiments, the value of the calibration plate can be determined based on the user requirement) can be selected, and in order to ensure that the whole calibration plate can be shot, the shooting lens distance needs to be pulled to be locally close to the crack. In addition, the equipment needs to be stabilized as much as possible in the shooting process, so that the equipment can better shoot and obtain required pictures under the condition of not shaking as much as possible. The specification of the calibration plate can be preconfigured with different colors and/or patterns so that the system can automatically adapt to the calibration plate with the corresponding specification.
And step S2, respectively identifying plaque points corresponding to the marker graph from the target acquisition image according to the estimated connected domain characteristics and the connected domain radius under a plurality of preset local extremum conditions.
Specifically, the computer equipment adopts a watershed detection method based on local extremum, and detects plaque points corresponding to the graph on the calibration plate from the target acquisition image. Of course, in the present embodiment, plaque points corresponding to the marker graphics may be identified from the target acquired image based on other detection methods, for example, centroid method, analysis method, and the like.
In one embodiment, the computer device may determine the target plaque point identified correspondingly based on the detection results obtained by the different methods, so as to ensure the detection accuracy of the plaque point. The computer device may perform comprehensive calculation based on methods such as average calculation and weighted summation calculation, and the embodiment of the present application does not limit the above comprehensive calculation method, and may flexibly adjust based on the difference between the data, the overall data amount, and the corresponding maximum and minimum values.
In one embodiment, the computer device performs connected domain analysis based on coordinates taken by each pixel point in the target acquired image and a connection relationship between different pixel points. Subsequently, when the corresponding connected domain is identified from the target acquired image, the computer equipment screens out the target connected domain with defects from the identified connected domains according to the parameter information, such as size, radius, boundary information and the like, for example, the boundary of the connected domain exceeds the preset boundary range, and the connected domain is too large to meet the preset screening condition and the like.
And step S3, determining an optimal local extremum condition from the local extremum conditions based on the difference degree between the imaging position and the actual position of the plaque point.
Specifically, the computer device calculates the difference value between the imaging position and the actual position of the plaque point under each local extremum condition, and sorts the obtained difference values through a preset sorting algorithm, such as a dichotomy method, an bubbling sorting method and the like. And then, based on the obtained sequencing result, taking the local extremum condition with the minimum corresponding difference value as the optimal local extremum condition. In one embodiment, to ensure the accuracy of the sorting, the computer device may sort the obtained values of each difference multiple times in succession, and when it is determined that the same target sorting result appears multiple times, and the number of times the target sorting result appears is greater than a preset threshold, perform the screening of the optimal local extremum condition based on the target sorting result.
In one embodiment, the computer device may further delete the interference factor from the obtained plaque point based on a preset screening rule, for example, the computer device may match the attribute information with a predetermined feature of the calibration pattern, for example, the roundness, the radius range, the coordinate distribution line feature, and the like, based on the attribute information such as the pixel coordinates, the radius, the roundness, and the like of the plaque point, and identify the interference factor based on the obtained matching result, so as to ensure the accuracy of identifying the plaque point.
And S4, under the optimal local extremum condition, determining a real plane coordinate string reflecting the actual long width of the crack based on a coordinate conversion mode between the plaque point and the corresponding image imaging plane.
Specifically, under the obtained optimal local extremum condition, the computer equipment determines the mapping relation between the imaging position and the actual position of each pixel point in the imaging image based on the coordinate conversion mode between the plaque point and the imaging plane of the corresponding image. And then, based on the mapping relation, converting the determined plane coordinate string reflecting the planned length and width of the crack in the imaging image to obtain a real plane coordinate string reflecting the actual length and width of the crack.
And S5, determining the cracking degree of the rail plate crack according to the real plane coordinate string.
Specifically, after the computer device acquires the target acquisition image containing the track slab, a corresponding track slab graph can be obtained by using a mode of drawing lines along the track slab image. When the computer equipment displays the target acquisition image on a preset display page, the target image can be enlarged and reduced in an interpolation display mode. It should be noted that the track slab graphic screen coordinate string may be obtained by clicking when the target image is enlarged, so that when the pixel coordinates of the track slab graphic are restored, the sub-pixel coordinates are obtained, and thus, the length and the width of the track slab graphic can be measured more accurately.
In the present embodiment, the computer device uses the position of the graphic on the calibration plate on the imaging image, and combines the actual coordinate position of the graphic on the calibration plate to calculate the coordinate conversion mode between the plane of the calibration plate (similar to the plane of the track plate) and the imaging plane. And obtaining a pixel coordinate string of a dimension (length or width) line to be detected of the crack through the measured crack pattern of the track plate. And then, the computer equipment calculates and obtains a real plane coordinate string corresponding to the pixel coordinate string based on the obtained coordinate conversion mode, and calculates the real dimension size of the crack to be detected based on the real plane coordinate string.
According to the track plate crack detection method, on one hand, based on the preset multiple local extremum conditions, corresponding plaque points are detected from the target acquisition images of the comprehensive calibration plate and the track plate crack, and the detection accuracy of the track plate crack can be ensured under the condition of improving the identification accuracy of the plaque points due to the fact that the fitted center accuracy is high. On the other hand, the optimal local extremum condition is determined based on the difference value between the actual position of the plaque point and the corresponding imaging position, so that the identification error of the plaque point can be reduced, and the measurement accuracy is improved. Finally, because the plane of the calibration plate can be similar to the plane of the track plate, a real plane coordinate string reflecting the actual long width of the crack is determined based on a coordinate conversion mode between the calibration point plate and the corresponding image imaging plane, and the cracking degree of the crack of the track plate is reversely pushed based on the real plane coordinate string, even under the condition that the distance, the angle and the like are unknown, the accurate detection of the cracking degree of the crack of the track plate can be ensured, and the detection efficiency of the system is improved.
In one embodiment, in step S2, for a plurality of preset local extremum conditions, identifying plaque points corresponding to the marker graph from the target acquired image according to the estimated connected domain feature and the connected domain radius under the corresponding local extremum conditions, including:
and S21, performing multi-interval region binarization processing on the target acquired image to obtain a corresponding binarized image.
Specifically, the image color image in which the image obtained by the photographing apparatus is color is mainly classified into two types, an RGB type and a CMYK type. Wherein, the RGB type color image is composed of three different color components, one is red, one is green and the other is blue. Whereas a CMYK type color image is composed of four color components: cyan C, magenta M, yellow Y, black K. Note that CMYK type color images are mainly used in the printing industry.
Since the pixels of each image generally correspond to a particular location in two dimensions and there are one or more sample value component values associated with that point. In contrast, a gray-scale image is also called a gray-scale image, and each pixel in the gray-scale image may be represented by a luminance value of 0 (black) to 255 (white), wherein different gray scales are represented between 0 and 255.
In one embodiment, the process of converting a color image into a grayscale image is referred to as the graying process of the image. In the RGB model, if r=g=b, the color will represent a gray color, and the value of r=g=b is also called a gray value. Therefore, only one byte is needed for each pixel in the gray image to store gray values (also called intensity values and brightness values), and the gray scale range is 0-255. Since the color of each pixel in a color image is determined by R, G, B components, each component having 255 values is desirable, such that a pixel will have a range of variation of 1600 tens of thousands (255255255) of colors. And the variation range of one pixel point of the gray level image is 255. Therefore, in digital image processing, images in various formats are generally converted into gray-scale images, so that the subsequent image calculation amount is reduced, and the data calculation efficiency is improved.
Step S22, extracting corresponding connected domains from the binarized image, and calculating the connected center points of the corresponding domains.
Specifically, in a pair of binarized images, the pixel points representing black are marked as 0, the pixel points representing white are marked as 1, and the corresponding area where a plurality of white pixel points or black pixel points are connected together is the required communication area.
In one embodiment, these communication areas may be irregularly shaped. Therefore, the obtained communication center point is not necessarily a standard center position, and may be a position close to the center of the area or a position far from the center of the area, and in different embodiments, the computer device determines the calculation mode of the communication center point based on the shape of the communication area, for example, when determining that the shape of the communication area is square, the computer device may determine the coordinates of the communication center point based on the coordinates taken by four vertexes of the square, which is not limited in the embodiment of the present application.
And S23, fitting and classifying the obtained connected central points to obtain corresponding central point sets, and estimating the characteristics of the connected domain and the radius of the connected domain according to the obtained central point sets.
And step S24, identifying plaque points corresponding to the marker graph from the target acquisition image according to the estimated connected domain characteristics and the connected domain radius.
According to the embodiment, the corresponding plaque points are detected from the target acquisition images of the comprehensive calibration plate and the track plate cracks based on the detection of the local extremum of the watershed, the fitted center is high in precision, the identification accuracy of the plaque points can be improved, and the judgment accuracy of the crack splitting degree of the track plate is ensured.
In one embodiment, in step S3, the determining, based on the degree of difference between the imaging position and the actual position of the plaque point, an optimal local extremum condition from the plurality of local extremum conditions includes:
Step S31, the actual position of the plaque point is obtained.
Specifically, the computer device may store the obtained actual position of the plaque point in a preset storage area, and then, obtain the required data by performing a data request to the storage area.
In one embodiment, the computer device divides the data to be stored into a plurality of classified data sets according to a preset data division strategy, such as classification division, and distributes different database nodes to each classified data set; each database node is mutually independent, each database node respectively stores one type of data correspondingly, and subsequently, when data feedback is needed, the computer equipment performs retrieval and acquisition of the needed data from the database nodes based on the determined retrieval conditions.
And step S32, under each local extremum condition, determining the imaging position of the plaque point in the target acquisition image, and calculating the difference value between the actual position and the imaging position.
Step S33, screening out the target difference value with the minimum value from the obtained difference values, and taking the local extremum condition corresponding to the target difference value as the optimal local extremum condition.
According to the embodiment, the optimal local extremum condition is determined based on the difference value between the actual position of the plaque point and the corresponding imaging position, so that the identification error of the plaque point can be reduced, and the measurement accuracy is improved.
In one embodiment, in step S4, the determining, based on a coordinate conversion manner between the plaque point and the corresponding image imaging plane under the optimal local extremum condition, a real plane coordinate string reflecting an actual long width of the crack includes:
and S41, under the optimal local extremum condition, determining a coordinate conversion mode between the patch point and a corresponding image imaging plane based on the imaging coordinate point of the patch point in the image coordinate system and the actual coordinate point of the patch point in the calibration plate coordinate system.
Specifically, in step S41, the determining a coordinate conversion manner between the patch point and the corresponding image imaging plane based on the imaging coordinate point of the patch point in the image coordinate system and the actual coordinate point of the patch point in the calibration plate coordinate system includes:
In step S411, a one-to-one mapping relationship is established between the imaging coordinate points and the actual coordinate points according to the image imaging principle.
Specifically, the computer device may construct a corresponding mapping table based on the image imaging principle according to a mapping relationship between imaging coordinate points and actual coordinate points. Subsequently, when the required target actual coordinate point needs to be queried, the computer equipment can construct a corresponding target address query instruction based on the attribute information of the target imaging coordinate point, and then screen the target actual coordinate point corresponding to the target address query instruction according to the target address query instruction and the mapping table.
Step S412, according to the established mapping relation, the coordinate conversion mode between the plaque point and the corresponding image imaging plane is determined.
Step S42, acquiring mouse movement data, wherein the mouse movement data comprises drawing data which are synchronously generated by drawing a shooting object recorded in a target acquisition image in a mouse clicking mode.
And S43, determining a pixel coordinate string reflecting the planned length and width of the crack according to the drawing data, and converting the pixel coordinate string into a real plane coordinate string reflecting the actual length and width of the crack based on the coordinate conversion mode.
Specifically, the computer device may display the acquired target acquisition image based on a preset display screen, and determine, when determining that the mouse click action is triggered, whether a click area reflecting the crack of the track board exists in the click range, if yes, determine, based on the position of the click area in the image coordinate system, a pixel coordinate string reflecting the planned length and width of the crack. Subsequently, the computer device calculates a corresponding real space coordinate string based on the coordinate conversion mode determined in step S41, thereby obtaining a real dimension of the crack.
In one embodiment, the computer device may also implement machine automatic delineation of the fracture pattern by learning fracture characteristics based on a machine learning algorithm. After the machine automatically draws the crack pattern, the real space coordinate string is calculated by using the coordinate conversion function, so that the real crack dimension is obtained, and of course, the full-automatic identification of the track plate crack can be realized by using a machine learning method, and the automatic execution efficiency of the system is improved.
In one embodiment, when determining that the track slab crack is curved in the target acquisition image, the method further comprises: and determining a target coordinate conversion mode between the plaque point and the corresponding image imaging plane under the optimal local extremum condition by adopting a curved surface fitting method, and determining a real curved surface coordinate string reflecting the actual long width of the crack based on the target coordinate conversion mode.
In particular, the surface fitting method is a data processing method for approximately describing or simulating a functional relationship between coordinates represented by a discrete point group on a plane (curved surface) by using a continuous curved surface. Taking into account the spherical aberration, astigmatism, curvature of field and aberrations equiaxial symmetric aberrations and non-centered aberrations and Bao Lengjing aberrations, and taking into account the CCD photosensitive image plane of the camera, the non-strict perpendicularity of the measured object surface and the optical axis of the camera, in one embodiment, the computer device will establish a corresponding cubic polynomial and determine the transformation relationship between the image point coordinates and the object coordinates based on the cubic polynomial.
In one embodiment, for the resulting cubic polynomial to be built, there will be multiple unknowns in each set of equations. In one embodiment, for solving the unknowns, the computer device may solve each set of equations according to a least squares method, respectively, to obtain a plurality of sets of equations. And then, solving the coefficient attached to each equation set by using a matrix equation solving method, and under the condition of obtaining corresponding processing parameters (such as image point coordinates), converting between an image coordinate system and a real world coordinate system, thereby realizing system calibration.
In one embodiment, the method further comprises:
step S6, aiming at the mark graph which is centrosymmetric, identifying plaque points corresponding to the mark graph from the target acquisition image by adopting a centroid method, wherein:
Step S61, performing binarization segmentation on the target acquired image to obtain a plurality of segmentation areas.
Step S62, screening out required target areas from the plurality of divided areas based on the identification characteristics of the plaque points, and determining the plaque points corresponding to the marker graph based on the target areas.
In one embodiment, the method further comprises:
Step S7, aiming at the mark graph with an irregular shape, identifying plaque points corresponding to the mark graph from the target acquisition image by adopting an analysis method, wherein:
Step S71, based on the identification feature of the plaque point, identifying the rough outline of the plaque point from the target acquisition image.
Step S72, based on the rough outline of the identified plaque point, an analytical equation of the target to be identified is obtained by adopting a fitting method, and the pixel coordinates of the target point are obtained according to the analytical equation.
Step S73, identifying the plaque point corresponding to the marker graph from the target acquisition image according to the pixel coordinates of the target point.
The embodiment is less affected by noise, and when a part of gray scales change due to shadow shielding on the target object, the pixel coordinates of the target point can still be correctly obtained based on an analysis method, so that the stability is higher, and the plaque point identification accuracy and the identification efficiency can be improved.
Referring to fig. 2, the application discloses a track slab crack detection system 200, wherein the system 200 comprises an image acquisition module 201, a plaque identification module 202, a condition screening module 203, a coordinate conversion module 204 and a degree detection module 205, wherein:
the image acquisition module 201 is configured to acquire an object acquisition image including a track slab crack and a calibration slab, where the calibration slab is disposed in a preset range around the track slab crack.
The plaque identification module 202 is configured to identify plaque points corresponding to the marker graphics from the target acquired image according to the estimated connected domain feature and the connected domain radius under a preset plurality of local extremum conditions.
The condition screening module 203 is configured to determine an optimal local extremum condition from the plurality of local extremum conditions based on a degree of difference between the imaging position and the actual position of the plaque point.
The coordinate conversion module 204 is configured to determine, under the optimal local extremum condition, a real plane coordinate string reflecting an actual long width of the crack based on a coordinate conversion manner between the plaque point and the corresponding image imaging plane.
The degree detection module 205 is configured to determine a cracking degree of the track slab crack according to the real plane coordinate string.
According to the track plate crack detection system, on one hand, based on the preset multiple local extremum conditions, corresponding plaque points are detected from the target acquisition images of the comprehensive calibration plate and the track plate cracks, and the detection accuracy of the track plate cracks can be guaranteed under the condition that the identification accuracy of the plaque points is improved due to the fact that the fitted center accuracy is high. On the other hand, the optimal local extremum condition is determined based on the difference value between the actual position of the plaque point and the corresponding imaging position, so that the identification error of the plaque point can be reduced, and the measurement accuracy is improved. Finally, because the plane of the calibration plate can be similar to the plane of the track plate, a real plane coordinate string reflecting the actual long width of the crack is determined based on a coordinate conversion mode between the calibration point plate and the corresponding image imaging plane, and the cracking degree of the crack of the track plate is reversely pushed based on the real plane coordinate string, even under the condition that the distance, the angle and the like are unknown, the accurate detection of the cracking degree of the crack of the track plate can be ensured, and the detection efficiency of the system is improved.
The application discloses a readable storage medium, which comprises a track slab crack detection method program, wherein the track slab crack detection method program realizes the steps of the method disclosed in any one of the method embodiments when being executed by a processor.
According to the storage medium, on one hand, based on the preset various local extremum conditions, corresponding plaque points are detected from the target acquisition images of the comprehensive calibration plate and the track plate cracks, and the detection accuracy of the track plate cracks can be ensured under the condition of improving the identification accuracy of the plaque points due to the fact that the fitted center accuracy is high. On the other hand, the optimal local extremum condition is determined based on the difference value between the actual position of the plaque point and the corresponding imaging position, so that the identification error of the plaque point can be reduced, and the measurement accuracy is improved. Finally, because the plane of the calibration plate can be similar to the plane of the track plate, a real plane coordinate string reflecting the actual long width of the crack is determined based on a coordinate conversion mode between the calibration point plate and the corresponding image imaging plane, and the cracking degree of the crack of the track plate is reversely pushed based on the real plane coordinate string, even under the condition that the distance, the angle and the like are unknown, the accurate detection of the cracking degree of the crack of the track plate can be ensured, and the detection efficiency of the system is improved.
The foregoing is only illustrative of the present invention and is not to be construed as limiting thereof, but rather as various modifications, equivalent arrangements, improvements, etc., within the spirit and principles of the present invention.
Claims (8)
1.A method for detecting a crack in a track slab, the method comprising:
s1, acquiring a target acquisition image comprising a track plate crack and a calibration plate, wherein the calibration plate is arranged in a preset range around the track plate crack;
s2, under the preset multiple local extremum conditions, respectively identifying plaque points corresponding to the marker graph from the target acquisition image according to the estimated connected domain characteristics and the connected domain radius;
S3, determining an optimal local extremum condition from the local extremum conditions based on the difference degree between the imaging position and the actual position of the plaque point;
S4, under the optimal local extremum condition, determining a real plane coordinate string reflecting the actual long width of the crack based on a coordinate conversion mode between the plaque point and the corresponding image imaging plane;
S5, determining the cracking degree of the rail plate crack according to the real plane coordinate string;
in step S2, for a plurality of preset local extremum conditions, under the corresponding local extremum conditions, according to the estimated connected domain features and the connected domain radius, plaque points corresponding to the marker graphics are identified from the target acquired image, including:
S21, performing multi-interval area binarization processing on the target acquired image to obtain a corresponding binarized image;
s22, extracting corresponding connected domains from the binarized image, and calculating the connected center points of the corresponding domains;
s23, fitting and classifying the obtained connected central points to obtain corresponding central point sets, and estimating the characteristics of the connected domain and the radius of the connected domain according to the obtained central point sets;
s24, identifying plaque points corresponding to the mark graph from the target acquisition image according to the estimated connected domain characteristics and the connected domain radius;
In step S3, the determining, based on the degree of difference between the imaging position and the actual position of the plaque point, an optimal local extremum condition from the plurality of local extremum conditions includes:
S31, acquiring the actual position of the plaque point;
S32, under each local extremum condition, respectively determining imaging positions of plaque points in the target acquisition image, and calculating a difference value between the actual positions and the imaging positions;
s33, screening out the target difference value with the minimum value from the obtained difference values, and taking the local extremum condition corresponding to the target difference value as the optimal local extremum condition.
2. The method according to claim 1, wherein in step S4, the determining, under the optimal local extremum condition, a real plane coordinate string reflecting an actual long width of the crack based on a coordinate conversion manner between the plaque point and the corresponding image imaging plane includes:
s41, under the optimal local extremum condition, determining a coordinate conversion mode between the patch point and a corresponding image imaging plane based on an imaging coordinate point of the patch point in an image coordinate system and an actual coordinate point of the patch point in a calibration plate coordinate system;
S42, acquiring mouse movement data, wherein the mouse movement data comprise drawing data which are synchronously generated by drawing a shooting object recorded in a target acquisition image in a mouse clicking mode;
s43, determining a pixel coordinate string reflecting the planned length and width of the crack according to the drawing data, and converting the pixel coordinate string into a real plane coordinate string reflecting the actual length and width of the crack based on the coordinate conversion mode.
3. The method according to claim 2, wherein in step S41, the determining a coordinate conversion manner between the patch point and the corresponding image imaging plane based on the imaging coordinate point of the patch point in the image coordinate system and the actual coordinate point in the calibration plate coordinate system includes:
s411, establishing a one-to-one mapping relation between the imaging coordinate points and the actual coordinate points according to an image imaging principle;
and S412, determining a coordinate conversion mode between the plaque point and the corresponding image imaging plane according to the established mapping relation.
4. The method of claim 1, wherein upon determining that a track slab crack is curved in the target acquisition image, the method further comprises: and determining a target coordinate conversion mode between the plaque point and the corresponding image imaging plane under the optimal local extremum condition by adopting a curved surface fitting method, and determining a real curved surface coordinate string reflecting the actual long width of the crack based on the target coordinate conversion mode.
5. The method according to any one of claims 1-4, further comprising:
s6, aiming at the mark graph which is centrosymmetric, identifying plaque points corresponding to the mark graph from the target acquisition image by adopting a centroid method, wherein:
S61, performing binarization segmentation on the target acquired image to obtain a plurality of segmentation areas;
s62, screening out required target areas from the plurality of divided areas based on the identification characteristics of the plaque points, and determining the plaque points corresponding to the marker graph based on the target areas.
6. The method according to any one of claims 1-4, further comprising:
S7, aiming at the mark graph with an irregular shape, identifying plaque points corresponding to the mark graph from the target acquisition image by adopting an analysis method, wherein:
S71, identifying the rough outline of the plaque point from the target acquisition image based on the identification characteristic of the plaque point;
S72, based on the rough outline of the identified plaque point, obtaining an analytic equation of the target to be identified by adopting a fitting method, and solving pixel coordinates of the target point according to the analytic equation;
S73, identifying plaque points corresponding to the mark graph from the target acquisition image according to the pixel coordinates of the target point.
7. The utility model provides a track slab crack detection system which characterized in that, the system includes image acquisition module, plaque recognition module, condition screening module, coordinate conversion module and degree detection module, wherein:
the image acquisition module is used for acquiring a target acquisition image comprising a track plate crack and a calibration plate, wherein the calibration plate is arranged in a preset range around the track plate crack;
The plaque identification module is used for respectively identifying plaque points corresponding to the mark patterns from the target acquisition image according to the estimated connected domain characteristics and the connected domain radius under the preset multiple local extremum conditions;
the condition screening module is used for determining an optimal local extremum condition from the plurality of local extremum conditions based on the degree of difference between the imaging position and the actual position of the plaque point;
the coordinate conversion module is used for determining a real plane coordinate string reflecting the actual long width of the crack based on a coordinate conversion mode between the plaque point and the corresponding image imaging plane under the optimal local extremum condition;
the degree detection module is used for determining the cracking degree of the track slab crack according to the real plane coordinate string;
Aiming at a plurality of preset local extremum conditions, the plaque identification module identifies the specific implementation of the plaque points corresponding to the mark graph from the target acquisition image according to the estimated connected domain characteristics and the connected domain radius under the corresponding local extremum conditions, wherein the specific implementation is as follows:
performing multi-interval region binarization processing on the target acquisition image to obtain a corresponding binarization image;
extracting corresponding connected domains from the binarized image, and calculating the connected center points of the corresponding domains;
Fitting and classifying the obtained connected central points to obtain corresponding central point sets, and estimating the characteristics of the connected domain and the radius of the connected domain according to the obtained central point sets;
Identifying plaque points corresponding to the mark graph from the target acquisition image according to the estimated connected domain characteristics and the connected domain radius;
The condition screening module determines the specific implementation of the optimal local extremum condition from the plurality of local extremum conditions based on the difference degree between the imaging position and the actual position of the plaque point, wherein the specific implementation is as follows:
Acquiring the actual position of a plaque point;
Under each local extremum condition, determining imaging positions of plaque points in the target acquisition image respectively, and calculating a difference value between the actual positions and the imaging positions;
And screening out the target difference value with the minimum value from the obtained difference values, and taking the local extremum condition corresponding to the target difference value as the optimal local extremum condition.
8. A readable storage medium, characterized in that it comprises a multi-source data card presentation method program, which, when executed by a processor, implements the steps of the method according to any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210274970.5A CN114638805B (en) | 2022-03-21 | 2022-03-21 | Track slab crack detection method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210274970.5A CN114638805B (en) | 2022-03-21 | 2022-03-21 | Track slab crack detection method, system and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114638805A CN114638805A (en) | 2022-06-17 |
| CN114638805B true CN114638805B (en) | 2024-06-14 |
Family
ID=81949146
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210274970.5A Active CN114638805B (en) | 2022-03-21 | 2022-03-21 | Track slab crack detection method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114638805B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116538953B (en) * | 2023-05-08 | 2024-01-30 | 武汉纵横天地空间信息技术有限公司 | Intelligent detection method and system for elevation targets and readable storage medium |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005332119A (en) * | 2004-05-19 | 2005-12-02 | Hitachi Software Eng Co Ltd | Spot image detecting method and spot image detecting program |
| US8913836B1 (en) * | 2013-12-20 | 2014-12-16 | I.R.I.S. | Method and system for correcting projective distortions using eigenpoints |
| CN104008553B (en) * | 2014-06-17 | 2017-03-22 | 武汉武大卓越科技有限责任公司 | Crack detection method with image gradient information and watershed method conflated |
| CN107179322A (en) * | 2017-06-15 | 2017-09-19 | 长安大学 | A kind of bridge bottom crack detection method based on binocular vision |
| CN107369136B (en) * | 2017-06-22 | 2020-06-09 | 福建省万龙新材料科技有限公司 | Visual detection method for surface cracks of polycrystalline diamond compact |
| CN108764345B (en) * | 2018-05-30 | 2022-04-26 | 河海大学常州校区 | Underwater dam crack detection method based on local and global clustering |
| CN109191483A (en) * | 2018-09-19 | 2019-01-11 | 南昌航空大学 | A kind of quick watershed detection method of helicopter blade Circle in Digital Images shape mark |
| CN111798423B (en) * | 2020-07-01 | 2022-08-23 | 上海理工大学 | Concrete crack picture splicing and detecting method |
| CN111915577A (en) * | 2020-07-15 | 2020-11-10 | 杨静 | Method and system for detecting crack proportion in pavement image |
| CN111968079B (en) * | 2020-07-28 | 2023-11-17 | 武汉光谷卓越科技股份有限公司 | Three-dimensional pavement crack extraction method based on local extremum of section and segmentation sparsity |
| CN113393440B (en) * | 2021-06-15 | 2022-06-14 | 电子科技大学 | Method for automatically enhancing and identifying weld defects based on magneto-optical imaging |
-
2022
- 2022-03-21 CN CN202210274970.5A patent/CN114638805B/en active Active
Non-Patent Citations (2)
| Title |
|---|
| 一种基于种子点扩散的隧道裂缝半自动提取方法;张硕;张尤赛;许智勋;孙露霞;周旭;;计算机与数字工程;20170220;第45卷(第02期);第387-391页 * |
| 基于网格激光器的盲人室内避障方法研究;谷勇;中国优秀硕士学位论文全文数据库 (基础科学辑);20200215;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114638805A (en) | 2022-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5699788B2 (en) | Screen area detection method and system | |
| CN112651968B (en) | Wood board deformation and pit detection method based on depth information | |
| JP7292979B2 (en) | Image processing device and image processing method | |
| CN109543665B (en) | Image positioning method and device | |
| CN113592839B (en) | Distribution network line typical defect diagnosis method and system based on improved fast RCNN | |
| CN110909640A (en) | Method and device for determining water level line, storage medium and electronic device | |
| CN111354047B (en) | Computer vision-based camera module positioning method and system | |
| CN108665468B (en) | Device and method for extracting tangent tower insulator string | |
| WO2022206161A1 (en) | Feature point recognition-based block movement real-time detection method | |
| CN110619258A (en) | Road track checking method based on high-resolution remote sensing image | |
| CN113642463B (en) | Heaven and earth multi-view alignment method for video monitoring and remote sensing images | |
| CN112560837A (en) | Reading method, device, equipment and storage medium of pointer instrument | |
| CN114638805B (en) | Track slab crack detection method, system and storage medium | |
| CN115294145A (en) | Method and system for measuring sag of power transmission line | |
| CN113252103A (en) | Method for calculating volume and mass of material pile based on MATLAB image recognition technology | |
| CN110717910B (en) | CT image target detection method based on convolutional neural network and CT scanner | |
| CN112284509A (en) | Bridge structure vibration mode measuring method based on mobile phone video | |
| JPH05215547A (en) | Method for determining corresponding points between stereo images | |
| JP5218849B2 (en) | Defect inspection equipment | |
| CN116205884A (en) | Concrete dam crack identification method and device | |
| CN115761468A (en) | Water level detection system and method based on image segmentation and target detection technology | |
| CN116051808A (en) | YOLOv 5-based lightweight part identification and positioning method | |
| CN112036393B (en) | Shale gas field production single pointer instrument reading based identification method | |
| CN112964192A (en) | Engineering measurement online calibration method and system based on image video | |
| CN110956640A (en) | Heterogeneous image edge point detection and registration method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |