A method and system for automatic measurement of cracks in aircraft skin

Abstract

The present application belongs to the field of image analysis, and particularly relates to a kind of automatic measurement method and system of aircraft skin crack, to solve the problem of the subjectivity of the existing aircraft skin crack detection method, prone to error. The present application comprises: collecting aircraft skin crack binary image, respectively on the binary image distance transformation and morphological skeleton detection, obtain distance transformation picture and skeleton picture, integrate distance transformation picture and skeleton picture to obtain the skeleton distance graph of aircraft skin crack;Search the branch point, endpoint and connection point of the skeleton distance graph, establish the edge matrix E of the connection relationship of branch point and endpoint as the graph structure feature;Based on the graph structure feature and distance transformation picture, by calculating the circumscribed rectangle attribute and crack attribute, and then evaluating the characteristics and measurement results of aircraft skin crack.

Description

Technical Field

[0001] This invention belongs to the field of image analysis, and specifically relates to a method for automatically measuring cracks in aircraft skin. Background Technology

[0002] In aircraft maintenance, aircraft walk-around inspection is a frequent routine task, documented in both routine work orders and scheduled maintenance work cards. Examples include pre-flight, short-stop, and post-flight work orders for routine operations, and work card packages or non-routine work orders for scheduled maintenance. The traditional walk-around procedure is generally the same, with the time varying slightly depending on the aircraft size, typically around 20 minutes. Routine walk-around inspections require adherence to the walk-around route marked on the work card, and deviations are not permitted. Scheduled maintenance, on the other hand, involves a zoned walk-around based on the work card content within the work card package.

[0003] Among all inspection items, surface damage to aircraft skin (such as cracks) is a significant hidden danger affecting flight safety. According to current civil aviation regulations, surface damage to aircraft must be inspected via line checks or routine hangar checks. Routine checks require comprehensive coverage of all parts of the aircraft, including the back, wings, and tail, making the inspection process demanding and complex. Cracks in aircraft skin are a common defect that affects the reliability and integrity of the structure. The detection, assessment, and repair of cracks play a crucial role in equipment maintenance.

[0004] Traditional manual inspection methods require inspectors to carry specialized equipment to closely inspect and measure the cracked area. This consumes significant manpower and is difficult to implement in certain environments, such as cracks located high on the fuselage or at fuselage joints, where manual measurement with tools is challenging. Furthermore, manual measurement is highly subjective and prone to error. Therefore, an automated method for measuring aircraft skin cracks is crucial. Summary of the Invention

[0005] To address the aforementioned problems in existing technologies, namely the high subjectivity and susceptibility to errors in current aircraft skin crack detection methods, this invention provides an automatic method for measuring aircraft skin cracks, the method comprising:

[0006] Step S100: Acquire a binarized image of the aircraft skin crack, perform distance transformation and morphological skeleton detection on the binarized image to obtain a distance transformation image and a skeleton image, and integrate the distance transformation image and the skeleton image to obtain a skeleton distance map of the aircraft skin crack.

[0007] The image transformation is an image processing method used to calculate the distance from each pixel in the image to the nearest background pixel.

[0008] The morphological skeleton detection is an image processing method that represents shapes. It extracts feature lines from a shape image to form an image containing only the feature lines. The skeleton is also commonly referred to as the central axis, ridge line, or skeleton line.

[0009] Step S200: Search for branch points, endpoints, and connection points in the skeleton distance graph, and establish the edge matrix E of the connection relationship between branch points and endpoints as graph structure features;

[0010] The graph structure is an important data structure used to represent relationships between multiple objects. A graph consists of nodes and edges, where nodes represent objects and edges represent relationships between nodes.

[0011] Step S300: Based on the graph structure features and distance transformation image, the characteristics and measurement results of aircraft skin cracks are evaluated by calculating the circumscribed rectangle attribute and crack attribute.

[0012] In some preferred embodiments, step S100 specifically includes:

[0013] Step S110: Acquire a binarized image B of the aircraft skin crack;

[0014] The pixel at (x1, y1) is designated as the first current pixel. The value of the current pixel is b(x1, y1), and the value range of b(x1, y1) is [0, 1]. When b(x1, y1) = 1, it means that the first current pixel is the foreground point of the segmented image. When b(x1, y1) ≠ 1, it means that the first current pixel is the background point.

[0015] Step S120: Perform a distance transformation on the binarized image B of the aircraft skin crack to obtain a distance-transformed image D inside the crack. Each pixel point located at coordinates (x2, y2) on the distance-transformed image D is denoted as the second current pixel point, with a value of d(x2, y2). d(x2, y2)≥0. When d(x2, y2)=0, it represents that the second current pixel point is a background point; when d(x2, y2)>0, the second current pixel point represents the shortest distance from coordinates (x2, y2) to the boundary on the distance-transformed image D inside the crack, where the distance transformation value of points on the boundary is 0.

[0016] Step S130: Morphological skeleton extraction is performed based on the binarized image B of the aircraft skin crack to obtain the skeleton image S of the aircraft skin crack. Each pixel point located at coordinates (x3, y3) on the skeleton image S is denoted as the third current pixel point, with a value of s(x3, y3). When s(x3, y3) = 1, it means that the third current pixel point is the foreground point; when s(x3, y3) = 0, it means that the third current pixel point is the background point.

[0017] In some preferred embodiments, step S200 specifically includes:

[0018] Step S210: Search for connection points, branch points, and endpoints in the skeleton distance graph;

[0019] Step S220: Traverse the third current pixel in the skeleton distance map and check the number of non-zero points in the eight neighborhoods of the third current pixel.

[0020] The third pixel in the eight-neighborhood with only one zero point is designated as the endpoint;

[0021] The third pixel in the eight-neighborhood with more than 2 non-zero points is designated as the branch point;

[0022] The third pixel point that has exactly two non-zero points in its eight neighborhood is recorded as the connection point;

[0023] Step S230: Aggregate adjacent branch points and connection points using the region generation method to obtain a point set P;

[0024] The region generation method is used to aggregate adjacent branch points and connection points. The branch points and endpoints in the skeleton distance image are numbered sequentially, and all endpoints and branch points are stored in the point set P.

[0025] Step S240: Establish the connection relationship between the branch points and endpoints in the point set P to obtain the graph structure features of the aircraft skin cracks;

[0026] The region generation method is used to aggregate all adjacent branch points and connection points in the eight neighborhoods. Each aggregation result corresponds to an edge, and the number of pixels in the edge is the length of the edge. Each pixel in each edge is traversed and recorded as the fourth current pixel. If there are endpoints and branch points in the eight neighborhoods of the fourth current pixel, the endpoints and branch points of the fourth current pixel are used as the endpoints of the edge.

[0027] Based on the endpoint connection relationships of each edge, construct an edge matrix E to store the endpoint connection relationships. The size of matrix E is n. P ×n P ;

[0028] n P Let e(i,j) represent the number of elements in matrix E. Elements e(i,j) in matrix E are integers greater than or equal to 0. When e(i,j) = 0, it represents a point P in matrix E. i and P j Disconnected, when e(i,j)>0, represents a point P in matrix E. i and P j The length of the connecting edge between the two endpoints is the value of e(i,j).

[0029] In some preferred embodiments, the calculation process for the circumscribed rectangle attribute includes:

[0030] Traverse the binary image B of the aircraft skin cracks and construct a set {b} of all foreground points. i (x, y)};

[0031] The formula for calculating the circumscribed rectangle parameter of the crack region is: starting point coordinate x b =min(x i ), y b =min(y i The width of the bounding rectangle is w = max(x). i )-min(x i The width of the circumscribed rectangle is h = max(y). i )-min(y i The area of ​​the crack region is the set {b}. i The number of pixels in (x, y)}.

[0032] In some preferred embodiments, the crack attributes include: the longest length, average length, maximum width, average width attribute, number of crack segments, average length of crack segments, and maximum ring length of crack.

[0033] In some preferred embodiments, the calculation process for the longest length, average length, maximum width, and average width attributes of the crack specifically includes:

[0034] Any pair of points P in the binarized image B of the aircraft skin crack i and P j Calculate the minimum path length for this pair of points in the graph, denoted as {l}. min The formula for calculating the length of the crack in (i, j) is: l = max i，j (1min(i,j));

[0035] The average length of the crack is the set {l min The average value of (i, j)};

[0036] The maximum width of the crack is the set The maximum value;

[0037] The average width of the crack is set The average value.

[0038] In some preferred embodiments, the calculation process for the area ratio of the crack body to the total crack area specifically includes:

[0039] Let the area of ​​the main body of the crack be The total area of ​​the crack is S m Based on the total number of pixels in the crack segmentation region;

[0040] r b The calculation formula is: r b =S b / S m .

[0041] In some preferred embodiments, both the number of crack segments and the length of each segment are calculated, including:

[0042] The count of all non-zero elements in matrix E is denoted as n. nonzero The number of segments in the crack is

[0043] The average of all non-zero elements in matrix E is the average segment length of the crack.

[0044] In some preferred embodiments, the calculation process for the maximum ring length of the crack specifically includes:

[0045] The Dijkstra algorithm is used to identify annular regions within the crack skeleton. The length of each annular region, the ratio of its length to the total length, and the ratio of the area of ​​the main part of the annular region to the total area are then calculated.

[0046] In another aspect, the present invention provides a system for automatically measuring cracks in aircraft skin, the system comprising:

[0047] Image collection module: configured to acquire binary images of aircraft skin cracks, perform distance transformation and morphological skeleton detection on the binary images respectively to obtain distance transformation images and skeleton images, and integrate the distance transformation images and skeleton images to obtain a skeleton distance map of aircraft skin cracks;

[0048] Graph structure feature calculation module: Configured to search for branch points, endpoints and connection points of the skeleton distance graph, and establish the edge matrix E of the connection relationship between branch points and endpoints as graph structure features;

[0049] Measurement Results Module: Configured to evaluate the characteristics and measurement results of aircraft skin cracks by calculating the bounding rectangle attribute and crack attribute based on the graph structure features and distance transformation image.

[0050] The beneficial effects of this invention are:

[0051] (1) This invention proposes an automated measurement method for aircraft skin based on digital image processing technology, which can automatically and in a standardized manner measure the acquired aircraft skin images, thereby improving measurement efficiency, reducing labor costs and minimizing errors.

[0052] (2) This invention quantitatively analyzes skin defects, including the size, shape and location of the defects, which can more objectively assess the degree of harm of the defects. The information of skin defects is presented in the form of images, which makes it easier for maintenance personnel to observe and understand them intuitively.

[0053] (3) The present invention can process a large amount of data in a short time by automatically running the algorithm, thus improving the analysis efficiency.

[0054] (4) This invention eliminates the interference of human subjective factors through a unified result analysis standard, avoids human error, and has high precision, accuracy, reliability and consistency.

[0055] (5) The present invention can realize the visualization of measurement results so that users can quickly detect and judge, thereby improving visualization capabilities and work efficiency. Attached Figure Description

[0056] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0057] Figure 1 This is a flowchart illustrating the method for automatically measuring cracks in aircraft skin according to an embodiment of the present invention.

[0058] Figure 2 This is a schematic diagram of a binarized image in an embodiment of the present invention;

[0059] Figure 3 This is a schematic diagram illustrating the effect of distance transformation images in an embodiment of the present invention;

[0060] Figure 4 This is a schematic diagram of the skeleton image in an embodiment of the present invention;

[0061] Figure 5 This is a schematic diagram illustrating the effect of the structural features of the aircraft skin crack in an embodiment of the present invention.

[0062] Figure 6 This is a schematic diagram of the measurement results in an embodiment of the present invention. Detailed Implementation

[0063] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0064] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0065] To more clearly explain the method for automatic measurement of aircraft skin cracks according to the present invention, the following is in conjunction with... Figure 1 The steps in the embodiments of the present invention will be described in detail below.

[0066] During aircraft maintenance, the manual measurement of surface skin damage is carried out manually. This requires manual measurement and recording of values ​​using handheld equipment. This process is not only cumbersome but also prone to errors due to individual subjective differences, resulting in low efficiency. Furthermore, it is difficult to uniformly summarize and analyze the data, and typical measurements only provide width information.

[0067] Existing crack measurement and description techniques generally only evaluate the crack by analyzing the outer rectangle of the crack and its length and width, or by calculating the distance between pixels to obtain the crack's length and width. This approach fails to provide a detailed and comprehensive description of the crack's morphology, which is detrimental to subsequent comprehensive evaluation.

[0068] To address these issues, this embodiment proposes an automated aircraft skin inspection technology based on digital image processing. This technology can improve measurement efficiency, reduce the time and cost of manual measurement, avoid human error, and provide quantitative analysis of skin defects. It can obtain information such as the size, shape, and location of defects, thereby better assessing the severity of the defects. The information on skin defects is presented in image form, making it easier for maintenance personnel to observe and understand them intuitively.

[0069] The method for automatically measuring aircraft skin cracks according to the first embodiment of the present invention includes steps S100-S300, such as... Figure 1 As shown, each step is described in detail below:

[0070] Step S100: Acquire a binarized image of the aircraft skin crack, perform distance transformation and morphological skeleton detection on the binarized image to obtain a distance transformation image and a skeleton image, and integrate the distance transformation image and the skeleton image to obtain a skeleton distance map of the aircraft skin crack.

[0071] The distance transformation image is as follows Figure 3 As shown, the skeleton image is as follows Figure 4 As shown.

[0072] In this embodiment, step S100 specifically includes:

[0073] Step S110: Acquire a binarized image B of the aircraft skin crack; image B is as follows: Figure 2 As shown;

[0074] The pixel at (x1, y1) is designated as the first current pixel. The value of the current pixel is b(x1, y1), and the range of b(x1, y1) is [0, 1]. When b(x1, y1) = 1, it means that the first current pixel is the foreground point of the segmented image. When b(x1, y1) ≠ 1, it means that the first current pixel is the background point.

[0075] Step S120: Perform a distance transformation on the binarized image B of the aircraft skin crack to obtain a distance-transformed image D inside the crack. Each pixel point located at coordinates (x2, y2) on the distance-transformed image D is denoted as the second current pixel point, with a value of d(x2, y2). d(x2, y2)≥0. When d(x2, y2)=0, it represents that the second current pixel point is a background point; when d(x2, y2)>0, the second current pixel point represents the shortest distance from coordinates (x2, y2) to the boundary on the distance-transformed image D inside the crack, where the distance transformation value of points on the boundary is 0.

[0076] Step S130: Morphological skeleton extraction is performed based on the binarized image B of the aircraft skin crack to obtain the skeleton image S of the aircraft skin crack. Each pixel point located at coordinates (x3, y3) on the skeleton image S is denoted as the third current pixel point, with a value of s(x3, y3). When s(x3, y3) = 1, it means that the third current pixel point is the foreground point; when s(x3, y3) = 0, it means that the third current pixel point is the background point.

[0077] Step S200: Search for branch points, endpoints, and connection points in the skeleton distance graph, and establish the edge matrix E, which represents the connection relationship between branch points and endpoints, as a graph structure feature.

[0078] In this embodiment, step S200 specifically includes:

[0079] Step S210: Search for connection points, branch points, and endpoints in the skeleton distance graph;

[0080] Step S220: Traverse the third current pixel in the skeleton distance map and check the number of non-zero points in the eight neighborhoods of the third current pixel.

[0081] The third pixel in the eight-neighborhood with only one zero point is designated as the endpoint;

[0082] The third pixel in the eight-neighborhood with more than 2 non-zero points is designated as the branch point;

[0083] The third pixel point that has exactly two non-zero points in its eight neighborhood is recorded as the connection point;

[0084] Step S230: Aggregate adjacent branch points and connection points using the region generation method to obtain a point set P;

[0085] The region generation method is used to aggregate adjacent branch points and connection points. The branch points and endpoints in the skeleton distance image are numbered sequentially, and all endpoints and branch points are stored in the point set P. The recorded information includes the coordinates of the points and the type of the endpoints.

[0086] Step S240: Establish the connection relationship between the branch points and endpoints in the point set P to obtain the graph structure features of the aircraft skin cracks;

[0087] The region generation method is used to aggregate all adjacent branch points in the eight neighborhoods. Each aggregation result corresponds to an edge, and the number of pixels in the edge is the length of the edge. Each pixel in each edge is traversed and recorded as the fourth current pixel. If there are endpoints and branch points in the eight neighborhoods of the fourth current pixel, the endpoints and branch points of the fourth current pixel are used as the endpoints of the edge.

[0088] Based on the endpoint connection relationships of each edge, construct an edge matrix E to store the endpoint connection relationships. The size of matrix E is n. P ×n P ;

[0089] n P Let e(i,j) represent the number of elements in matrix E. Elements e(i,j) in matrix E are integers greater than or equal to 0. When e(i,j) = 0, it represents a point P in matrix E. i and P j Disconnected, when e(i,j)>0, represents a point P in matrix E. i and P j The length of the connecting edge between the two endpoints is the value of e(i, j). The graphical structural features of aircraft skin cracks are as follows: Figure 5 As shown.

[0090] Step S300: Based on the graph structure features and distance-transformed image, the characteristics and measurement results of the aircraft skin cracks are evaluated by calculating the bounding rectangle attribute and crack attribute. The measurement results are as follows: Figure 6 As shown, in Figure 6 In the text, only the length and width information of the crack are displayed.

[0091] In this embodiment, the calculation process for the circumscribed rectangle attribute includes:

[0092] Traverse the binary image B of the aircraft skin cracks and construct a set {b} of all foreground points. i (x, y)};

[0093] The formula for calculating the circumscribed rectangle parameter of the crack region is: starting point coordinate x b =min(x i ), y b =min(yi The width of the bounding rectangle is w = max(x). i )-min(x i The width of the circumscribed rectangle is h = max(y). i )-min(y i The area of ​​the crack region is the set {b}. i The number of pixels in (x, y)}.

[0094] In this embodiment, the crack attributes include: the longest length, average length, maximum width, average width attribute, number of crack segments, average length of crack segments, and maximum ring length of crack.

[0095] In this embodiment, the calculation process for the longest length, average length, maximum width, and average width attributes of the crack specifically includes:

[0096] Any pair of points P in the binarized image B of the aircraft skin crack i and P j Calculate the minimum path length for this pair of points in the graph, denoted as {l}. min The formula for calculating the length of the crack in (i, j) is: l = max i，j (l min (i, j));

[0097] The average length of the crack is the set {l min The average value of (i, j)};

[0098] The maximum width of the crack is the set The maximum value;

[0099] The average width of the crack is set The average value.

[0100] In this embodiment, the calculation process for the area ratio of the main crack to the total crack area specifically includes:

[0101] Let the area of ​​the main body of the crack be The total area of ​​the crack is S m Based on the total number of pixels in the crack segmentation region;

[0102] r b The calculation formula is: r b =S b / S m .

[0103] In this embodiment, the number of crack segments and the length of each segment are calculated, and the calculation process includes:

[0104] The count of all non-zero elements in matrix E is denoted as n. nnonzero The number of segments in the crack is

[0105] The average of all non-zero elements in matrix E is the average segment length of the crack.

[0106] This embodiment includes basic morphological feature analysis such as the number of segments, segment length, and average segment length, as well as features related to annular regions, such as the maximum annular length, the ratio of the annular region length to the total length, and the ratio of the main area of ​​the annular region to the total area. These features can be used for further classification of crack images. Analysis can also be performed by calculating factors such as the shape, density, and distribution of cracks in the aircraft skin.

[0107] In this embodiment, the calculation process for the maximum ring length of the crack specifically includes:

[0108] The Dijkstra algorithm is used to identify annular regions within the crack skeleton. The length of each annular region, the ratio of its length to the total length, and the ratio of the area of ​​the main part of the annular region to the total area are then calculated.

[0109] Although the steps in the above embodiments are described in the above order, those skilled in the art will understand that in order to achieve the effect of this embodiment, different steps do not need to be executed in such an order. They can be executed simultaneously (in parallel) or in a reverse order. These simple variations are all within the protection scope of this invention.

[0110] A system for automatically measuring aircraft skin cracks according to a second embodiment of the present invention, the system comprising:

[0111] Image collection module: configured to acquire binary images of aircraft skin cracks, perform distance transformation and morphological skeleton detection on the binary images respectively to obtain distance transformation images and skeleton images, and integrate the distance transformation images and skeleton images to obtain a skeleton distance map of aircraft skin cracks;

[0112] Graph structure feature calculation module: Configured to search for branch points, endpoints and connection points of the skeleton distance graph, and establish the edge matrix E of the connection relationship between branch points and endpoints as graph structure features;

[0113] Measurement Results Module: Configured to evaluate the characteristics and measurement results of aircraft skin cracks by calculating the bounding rectangle attribute and crack attribute based on the graph structure features and distance transformation image.

[0114] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the system described above can be found in the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0115] It should be noted that the automatic aircraft skin crack measurement system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be further decomposed or combined. For example, the modules in the above embodiments can be merged into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the various modules or steps and are not considered as an improper limitation of the present invention.

[0116] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the storage device and processing device described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0117] Those skilled in the art will recognize that the modules and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. The programs corresponding to the software modules and method steps can be placed in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in the art. To clearly illustrate the interchangeability of electronic hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the invention.

[0118] The terms “first”, “second”, etc., are used to distinguish similar objects, not to describe or indicate a specific order or sequence.

[0119] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent in such process, method, article, or apparatus / device.

[0120] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method of automatic measurement of cracks in aircraft skin, characterized in that, The method comprises: Step S100, collecting an aircraft skin crack binary image, performing distance transformation and morphological skeleton detection on the binary image respectively, obtaining a distance transformation picture and a skeleton picture, and integrating the distance transformation picture and the skeleton picture to obtain a skeleton distance graph of the aircraft skin crack; Step S200, searching for branch points, end points and connection points of the skeleton distance graph, and establishing an edge matrix E of the connection relationship between the branch points and the end points as a graph structure feature; Step S300, based on the graph structure feature and the distance transformation picture, calculating the circumscribed rectangle attribute and the crack attribute, and then evaluating the characteristics and measurement results of the aircraft skin crack; The step S100 specifically comprises: Step S110, collecting an aircraft skin crack binary image B; Will The pixel at point b is denoted as the first current pixel; the current pixel value is b. b The value range of b is [0,1]; when b When b = 1, it indicates that the first current pixel is the foreground point of the segmented image. When ≠1, it means that the first current pixel is a background pixel; Step S120, distance transformation is performed on the binary image B of the crack of the aircraft skin to obtain a distance transformation image D. Each pixel point at coordinate on the distance transformation image D is denoted as a second current pixel point, and the value of the second current pixel point is , When =0, the second current pixel point represents a background point. When >0, the second current pixel point represents the shortest distance from the coordinate to the boundary on the distance transformation image D inside the crack, wherein the distance transformation value of the point on the boundary is 0. Step S130, according to the aircraft skin crack binary image B, the morphological skeleton extraction is carried out, and the skeleton picture S of the aircraft skin crack is obtained. Each pixel point located at the coordinate on the skeleton picture S is recorded as a third current pixel point, and the value is When =1, it represents that the third current pixel point is a foreground point. When =0, it represents that the third current pixel point is a background point. The step S200 specifically comprises: Step S210, searching for connection points, branch points and end points in the skeleton distance graph; Step S220, traversing a third current pixel point in the skeleton distance graph, and checking the number of non-zero points in the eight-neighborhood of the third current pixel point; The third pixel point with only one non-zero point in the eight-neighborhood is recorded as an end point; The third pixel point with more than two non-zero points in the eight-neighborhood is recorded as a branch point; The third pixel point with exactly two non-zero points in the eight-neighborhood is recorded as a connection point; Step S230, aggregating adjacent branch points and connection points by a region generation method to obtain a point set P; Adjacent branch points and connection points are aggregated by the region generation method, and the branch points and the end points in the skeleton distance picture are sequentially numbered, and all the end points and the branch points are stored in the point set P; Step S240, establishing the connection relationship between the branch points and the end points in the point set P to obtain a graph structure feature of the aircraft skin crack; All the eight-neighborhood adjacent branch points and connection points are aggregated by the region generation method, each aggregation result corresponds to an edge, and the number of pixel points in the edge is the length of the edge; each pixel point in each edge is traversed and recorded as a fourth current pixel point; if there are end points and branch points in the eight-neighborhood of the fourth current pixel point, the end points and the branch points of the fourth current pixel point are taken as the end points of the edge. According to the end-point connection relationship of each edge, an edge matrix E storing the end-point connection relationship is established, and the size of the matrix E is ; This represents the number of elements in matrix E. medium elements For integers greater than or equal to 0, when When, it represents a point in matrix E. and No connection, when When, it represents a point in matrix E. and The length of the connecting edge between the two endpoints is The value of .

2. The method of claim 1, wherein, The calculation process of the circumscribed rectangle attribute comprises: traversing the binary image B of the crack of the aircraft skin, all the foreground points form a set ; The parameter calculation formula of the circumscribed rectangle of the crack region is: the starting point coordinate , , the circumscribed rectangle width , the circumscribed rectangle width , and the crack region area is the number of pixel points in the set .

3. A method of automatic measurement of cracks in an aircraft skin as claimed in claim 2, wherein, The crack attribute comprises: the longest length, the average length, the maximum width, the average width attribute of the crack, the number of crack segments, the average length of crack segments, and the maximum ring length of the crack.

4. The method of claim 3, wherein, The calculation process of the longest length, the average length, the maximum width and the average width attribute of the crack specifically comprises: Any pair of points in the binary image B of the crack of the aircraft skin and , the minimum path length of the pair of points in the graph is calculated, denoted as The length of the crack is calculated as: ; The average length of the cracks is the average value of the set of average values; The maximum width of the crack is the maximum value of the set of maximum values; The average width of the cracks is the average value of the set of average values.

5. A method of automatic measurement of cracks in an aircraft skin as claimed in claim 4, wherein, The calculation process of the area ratio of the crack main body to the crack total body specifically comprises: Let the area of the crack body be , and the total area of the crack be , based on the total number of pixels in the crack segmentation region; The calculation formula is: .

6. A method of automatic measurement of cracks in an aircraft skin as claimed in claim 5, wherein, The calculation process of the number of crack segments and the average length of crack segments comprises: The number of all non-zero elements in the statistical matrix E is denoted as , and the number of the crack segments is ; The average value of all non-zero elements in the matrix E is the average length of crack segments.

7. A method of automatic measurement of cracks in an aircraft skin as claimed in claim 6, wherein, The calculation process of the maximum ring length of the crack specifically comprises: According to the Dijkstra algorithm, a ring region in the crack skeleton is found out, the length of the ring region is counted, the ratio of the length of the ring region to the total length is calculated, and the ratio of the main body area of the ring region to the total area is calculated.

8. A system for automated measurement of cracks in aircraft skin, characterized by, The system is applied to the method for automatically measuring the aircraft skin crack according to any one of claims 1-7. The image collection module is configured to collect a binary image of the aircraft skin crack, perform distance transformation and morphological skeleton detection on the binary image respectively, obtain a distance transformation picture and a skeleton picture, and integrate the distance transformation picture and the skeleton picture to obtain a skeleton distance graph of the aircraft skin crack. The graph structure feature calculation module is configured to search for branch points, end points and connection points of the skeleton distance graph, and establish an edge matrix E of the connection relationship between the branch points and the end points as a graph structure feature. The measurement result module is configured to calculate the circumscribed rectangle attribute and the crack attribute based on the graph structure feature and the distance transformation picture, and then evaluate the characteristics and measurement results of the aircraft skin crack.

Images