Quality detection method and device for component image to be audited, equipment and medium
By extracting the edge and texture features of the image of the component to be reviewed, combining multiple traversals and multi-mask fusion, the cracks in the image of the component to be reviewed are determined, which solves the complex and time-consuming problem of manual detection and improves the accuracy of detection and management reliability.
Patent Information
- Application Number
- CN202311694645.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-11
- Publication Date
- 2025-06-13
AI Technical Summary
In petroleum engineering construction, the image quality inspection of manually inspected components to be reviewed has complex processes, time-consuming and labor-intensive problems and is prone to favoritism and fraud, resulting in deviations in quality audit recognition.
By obtaining the image of the component to be reviewed, extracting its edge features and texture features, determining the initial crack feature points, and determining the true crack feature points through multiple traversals and multi-mask fusion, and finally determining the cracks in the image of the component to be reviewed.
It improves the accuracy and accuracy of crack detection, avoids conflicts and quality and safety issues caused by manual inspection, and enhances the reliability, safety and effectiveness of the quality management of components to be reviewed in petroleum engineering construction.
Smart Images
Figure CN120147208A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to a method, apparatus, device and medium for quality detection of an image of a component to be audited. Background Art
[0002] In recent years, the oil engineering has entered the era of intelligent management. During the oil engineering construction process, there are many construction matters that need quality inspection, and the construction process and conditions need to be monitored and managed. The process nodes involved are numerous and the correlation relationships are complex, and there are many difficulties in manual monitoring and management. In addition, for quality inspection, manual supervision usually adopts the method of joint inspection by three parties, that is, the construction party reports the construction situation, the third-party supervision unit conducts on-site inspections, and the owner conducts on-site inspections. The process is complex, time-consuming and laborious, and conflicts are also likely to occur. In some cases, there is also prone to favoritism and fraud, which brings great disadvantages to the construction of the construction project. Especially for on-site quality audits, conflicts are extremely likely to arise due to deviations in quality audit determination. Summary of the Invention
[0003] In order to better implement the quality detection of the image of the component to be audited involving welding and prefabrication in oil engineering construction, the embodiments of the present application provide a method, apparatus, device and medium for quality detection of the image of the component to be audited.
[0004] In a first aspect, the embodiments of the present application provide a method for quality detection of an image of a component to be audited, the method comprising:
[0005] Obtain an image of the component to be audited, and extract the edge features and texture features of the image of the component to be audited;
[0006] Determine all preliminary crack feature points in the image of the component to be audited according to the edge features and the texture features;
[0007] Based on the edge features, obtain a first crack connected region of the image of the component to be audited, and based on the texture features, obtain a second crack connected region of the image of the component to be audited;
[0008] Fuse the first crack connected region and the second crack connected region to obtain a preliminary crack connected region;
[0009] Based on the preliminary crack connected region, determine all main crack feature points according to all the preliminary crack feature points;
[0010] Based on a pre-constructed first mask and second mask, traverse each pixel point of the image of the component to be audited to obtain a crack binary image;
[0011] Determine all true crack feature points according to all the main crack feature points and the crack binary image;
[0012] Determine the cracks in the image of the component to be reviewed based on all the true crack feature points.
[0013] In an alternative embodiment of the present application, the determining all the preliminary crack feature points in the image of the component to be reviewed according to the edge feature and the texture feature includes:
[0014] For each pixel point in the image of the component to be reviewed, determine whether the pixel point is both an edge feature point and a texture feature point;
[0015] If so, determine the pixel point as a preliminary crack feature point in the image of the component to be reviewed.
[0016] In an alternative embodiment of the present application, the fusing the first crack connected region and the second crack connected region to obtain a preliminary crack connected region includes:
[0017] For each pixel point in the first crack connected region and the second crack connected region, determine a first pixel difference from the region covered by the preset LBP kernel in the first crack connected region to the preset LBP kernel, and a second pixel difference from the region covered by the preset LBP kernel in the second crack connected region to the preset LBP kernel;
[0018] Determine whether the difference between the first pixel difference and the second pixel difference is less than a first preset difference;
[0019] If so, perform weighted fusion on the region covered by the preset LBP kernel in the first crack connected region and the region covered by the preset LBP kernel in the second crack connected region to obtain a fused region;
[0020] Map all the fused regions into an empty image to obtain a preliminary crack connected region.
[0021] In an alternative embodiment of the present application, the determining all the main crack feature points based on the preliminary crack connected region according to all the preliminary crack feature points includes:
[0022] Perform texture detection on the preliminary crack connected region to obtain connected texture information;
[0023] Determine whether each preliminary crack feature point exists in the connected texture information;
[0024] If so, determine the preliminary crack feature point as a main crack feature point.
[0025] In an optional implementation manner of the embodiment of the present application, traversing each pixel point of the component image to be audited based on the pre-constructed first mask and second mask to obtain a crack binary image includes:
[0026] For each pixel point of the component image to be audited, obtain a third pixel difference based on the pre-constructed first mask;
[0027] Based on the pre-constructed second mask and the third pixel difference, obtain a fourth pixel difference;
[0028] Determine whether the fourth pixel difference is greater than a second preset difference;
[0029] If so, set the pixel value of the pixel point corresponding to the third pixel difference to 1;
[0030] If not, set the pixel value of the pixel point corresponding to the third pixel difference to 0;
[0031] Obtain the crack binary image according to the pixel values of each pixel point of the component image to be audited.
[0032] In an optional implementation manner of the embodiment of the present application, determining all true crack feature points according to all the main crack feature points and the crack binary image includes:
[0033] For each of the main crack feature points, determine whether the pixel value of the pixel point at the corresponding position in the crack binary image is 1;
[0034] If so, determine the main crack feature point as a true crack feature point.
[0035] In an optional implementation manner of the embodiment of the present application, determining the cracks in the component image to be audited according to all the true crack feature points includes:
[0036] Obtain the distances between adjacent true crack feature points among all the true crack feature points to obtain an average distance;
[0037] Randomly select a true crack feature point as the center, set a crack confirmation circle with the average distance as the radius, and connect the true crack feature points covered by the crack confirmation circle;
[0038] Select the true crack feature point closest to the crack confirmation circle from the true crack feature points, and use the true crack feature point as the center of the next crack confirmation circle;
[0039] Repeat the process of setting the crack confirmation circle and connecting each true crack feature point covered by the crack confirmation circle until the center of the next crack confirmation circle cannot be obtained, and a crack in the image of the component to be reviewed is obtained.
[0040] In an optional implementation manner of the embodiment of the present application, it further includes:
[0041] After obtaining any crack in the image of the component to be reviewed, determine whether there are unconnected true crack feature points;
[0042] If so, set the crack confirmation circle with an unconnected true crack feature point as the center, and repeat the process of setting the crack confirmation circle and connecting each true crack feature point covered by the crack confirmation circle to obtain the next crack in the image of the component to be reviewed.
[0043] In a second aspect, an embodiment of the present application provides a quality detection device for an image of a component to be reviewed, and the device includes:
[0044] A feature extraction module, configured to obtain an image of a component to be reviewed and extract the edge features and texture features of the image of the component to be reviewed;
[0045] A first determination module, configured to determine all preliminary crack feature points in the image of the component to be reviewed according to the edge features and the texture features;
[0046] A second determination module, configured to obtain a first crack connected region of the image of the component to be reviewed based on the edge features, and obtain a second crack connected region of the image of the component to be reviewed based on the texture features;
[0047] A fusion module, configured to fuse the first crack connected region and the second crack connected region to obtain a preliminary crack connected region;
[0048] A third determination module, configured to determine all main crack feature points according to all the preliminary crack feature points based on the preliminary crack connected region;
[0049] A mask traversal module, configured to traverse each pixel point of the image of the component to be reviewed based on a pre-constructed first mask and a second mask to obtain a crack binary image;
[0050] A fourth determination module, configured to determine all true crack feature points according to all the main crack feature points and the crack binary image;
[0051] A crack determination module, configured to determine the cracks in the image of the component to be reviewed according to all the true crack feature points.
[0052] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the quality detection method for the image of the component to be reviewed as described above is implemented.
[0053] In a fourth aspect, an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the quality detection method for the image of the component to be reviewed as described above is implemented.
[0054] In a fifth aspect, an embodiment of the present application provides a computer program product containing instructions. When the computer program product runs on a computer device, the computer device is enabled to execute the quality detection method for the image of the component to be reviewed as described above.
[0055] In a sixth aspect, an embodiment of the present application provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor, and the processor is used to run a computer program or instruction to implement the quality detection method for the image of the component to be reviewed as described above.
[0056] The beneficial effects of the above technical solutions provided by the embodiments of the present application at least include:
[0057] For the quality detection method for the image of the component to be reviewed provided by the embodiment of the present application, by first extracting the edge features and texture features of the image to be reviewed to determine the preliminary crack feature points, and then determining the preliminary crack connection region to judge whether the preliminary crack feature points are the main crack feature points. In addition, by traversing the image to be reviewed based on the pre-constructed first mask and second mask, the corresponding crack binary image is obtained. Then, the main crack feature points are corresponded to the crack binary image to determine the true crack feature points. Finally, according to all the true crack feature points, the cracks in the image of the component to be reviewed are determined. This method is based on the source image for detection. Through two rounds of detection, crack detection is carried out from multiple dimensions, multiple traversals, and multi-mask fusion, improving the accuracy and precision of crack detection; using the crack detection result as a reference for quality inspection, avoiding unnecessary conflicts and quality and safety problems caused by manual inspection, and improving the reliability, safety, and effectiveness of the quality management of the components to be reviewed involving welding and prefabrication in oil engineering construction, and improving the accuracy of quality grade determination.
[0058] Other features and advantages of the present application will be described in the subsequent description. Moreover, some of them will become obvious from the description or be understood by implementing the present application. The objectives and other advantages of the present application can be achieved and obtained through the structures specifically pointed out in the written description, claims, and drawings.
[0059] The technical solutions of the present application will be further described in detail below through the drawings and embodiments. Brief Description of the Drawings
[0060] The drawings are used to provide a further understanding of the present application and form a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation to the present application. In the drawings:
[0061] Figure 1 It is a schematic diagram of the steps of the method for detecting the quality of the image of the component to be reviewed provided by the embodiment of the present application;
[0062] Figure 2 It is a schematic diagram of the first mask example provided by the embodiment of the present application;
[0063] Figure 3 It is a schematic diagram of the second mask example provided by the embodiment of the present application;
[0064] Figure 4 It is a schematic diagram of the structure of the device for detecting the quality of the image of the component to be reviewed provided by the embodiment of the present application. Detailed Description of the Embodiments
[0065] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system architectures, technologies, etc. are presented to provide a thorough understanding of the embodiments of the present application. However, those skilled in the art should understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.
[0066] It should be understood that when used in the specification and the appended claims of the present application, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.
[0067] It should also be understood that the term "and / or" as used in the specification and the appended claims of the present application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.
[0068] As used in the specification and the appended claims of the present application, the term "if" can be interpreted as "when" or "once" or "in response to determining" or "in response to detecting" according to the context. Similarly, the phrase "if determined" or "if [the described condition or event] is detected" can be interpreted as meaning "once determined" or "in response to determining" or "once [the described condition or event] is detected" or "in response to detecting [the described condition or event]" according to the context.
[0069] In addition, in the description of the specification and the appended claims of the present application, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be construed as indicating or implying relative importance.
[0070] The reference to "one embodiment" or "some embodiments" etc. described in the specification of the present application means that in one or more embodiments of the present application, specific features, structures or characteristics described in connection with that embodiment are included. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.
[0071] It should be understood that the magnitudes of the sequence numbers of the steps in the following embodiments do not mean the order of execution is prior or subsequent. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.
[0072] In order to illustrate the technical solution of the present application, the following will be described through specific embodiments.
[0073] The inventors found that in the prior art, petrochemical engineering construction projects have a large amount of information, construction management is difficult and it is difficult to conduct quality inspections safely and effectively. For the traditional quality inspection method of images of parts to be reviewed, when detecting cracks, the source image needs to be binarized, and then texture and skeleton information are extracted to obtain crack information. However, in this way, a lot of image information will be lost when the image is binarized, resulting in a decrease in the accuracy and precision of crack detection, which does not meet the expectations of the inventors. Based on this, after further research and development, the inventors made the present application, providing a quality inspection method, device, equipment and medium for images of parts to be reviewed.
[0074] Embodiment 1
[0075] The embodiment of the present application provides a quality inspection method for images of parts to be reviewed. Referring to Figure 1 as shown, the method includes:
[0076] S101: Obtain an image of a part to be reviewed, and extract the edge features and texture features of the image of the part to be reviewed.
[0077] At the construction site of petroleum engineering, construction workers, third-party supervisors or the owner can take pictures at the construction site to obtain the welding interface images and prefabricated part images of the construction site. These images can be collectively referred to as the images of parts to be reviewed. Generally, it is only necessary for construction workers to take complete images of the parts to be reviewed, and there is no need for third-party supervisors or the owner to specifically go to the construction site to take pictures.
[0078] After obtaining the images of the parts to be reviewed, mark the images of the parts to be reviewed, mainly marking the signals and positions of the parts captured in the images of the parts to be reviewed for subsequent processing.
[0079] After completing the above work, perform pre-recognition processing on the images of the parts to be reviewed. It is necessary to extract the edge features and texture features of the images of the parts to be reviewed to obtain each edge feature point and each texture feature point. Among them, the extraction of edge features can be realized by the canny operator; the extraction of texture features can be realized by the Local Binary Pattern (LBP) algorithm. The specific implementation methods of the above-mentioned edge feature and texture feature extraction can refer to the detailed description in the prior art and are not specifically limited here.
[0080] S102: Determine all preliminary crack feature points in the image of the part to be reviewed according to the edge features and the texture features.
[0081] In the embodiment of the present application, the step of determining all preliminary crack feature points in the image of the part to be reviewed according to the edge features and the texture features specifically includes:
[0082] For each pixel point in the image of the part to be reviewed, judge whether the pixel point is both an edge feature point and a texture feature point;
[0083] If so, determine that the pixel point is a preliminary crack feature point in the image of the part to be reviewed.
[0084] In the embodiment of the present application, for each pixel point in the image of the part to be reviewed, if the pixel point is both an edge feature point and a texture feature point, then it is considered that the pixel point is a preliminary crack feature point in the image of the part to be reviewed.
[0085] S103: Obtain the first crack connected region of the image of the part to be reviewed based on the edge features, and obtain the second crack connected region of the image of the part to be reviewed based on the texture features.
[0086] In the embodiments of the present application, based on the extracted edge features, a first crack connected region of the component image to be audited is obtained; based on the extracted texture features, a second crack connected region of the component image to be audited is obtained. Among them, the specific implementation manner of obtaining the crack connected region can refer to the detailed description in the prior art and will not be specifically limited herein.
[0087] S104: Fuse the first crack connected region and the second crack connected region to obtain a preliminary crack connected region.
[0088] In the embodiments of the present application, the fusing the first crack connected region and the second crack connected region to obtain a preliminary crack connected region specifically includes:
[0089] For each pixel point in the first crack connected region and the second crack connected region, determine a first pixel difference between the region covered by the preset LBP kernel in the first crack connected region and the preset LBP kernel, and a second pixel difference between the region covered by the preset LBP kernel in the second crack connected region and the preset LBP kernel;
[0090] Judge whether the difference between the first pixel difference and the second pixel difference is less than a first preset difference;
[0091] If so, perform weighted fusion on the region covered by the preset LBP kernel in the first crack connected region and the region covered by the preset LBP kernel in the second crack connected region to obtain a fused region;
[0092] If not, do not perform weighted fusion on the region covered by the preset LBP kernel in the first crack connected region and the region covered by the preset LBP kernel in the second crack connected region;
[0093] Map all the fused regions into an empty image to obtain a preliminary crack connected region.
[0094] In the embodiment of the present application, the fusion of two crack-connected regions is completed by a preset LBP kernel, where the preset LBP kernel can be set as an 8-neighborhood LBP kernel. For each pixel point in the first crack-connected region, by setting a preset LBP kernel on this pixel point, calculate the first pixel difference between the region covered by the preset LBP kernel in the first crack-connected region and the preset LBP kernel; for each pixel point in the second crack-connected region, by setting a preset LBP kernel on this pixel point, calculate the second pixel difference between the region covered by the preset LBP kernel in the second crack-connected region and the preset LBP kernel. Calculate the difference between the first pixel difference and the second pixel difference. If the difference is within the preset range, that is, the difference is less than the first preset difference, then perform pixel value weighted fusion on the region covered by the preset LBP kernel in the first crack-connected region and the region covered by the preset LBP kernel in the second crack-connected region to obtain a fusion region; if the difference is not within the preset range, that is, the difference is not less than the first preset difference, then no weighted fusion is performed. Among them, the first preset difference is set according to specific circumstances, which can be set to 20 for color images and 5 for black-and-white images.
[0095] Move the preset LBP kernel on the first crack-connected region and the second crack-connected region, and obtain the corresponding fusion region through the above operations at each pixel point until all pixel points in the first crack-connected region and the second crack-connected region are traversed. Then create an empty image and map all the obtained fusion regions to this empty image to obtain a preliminary crack-connected region.
[0096] S105: Based on the preliminary crack-connected region, determine all main crack feature points according to all the preliminary crack feature points.
[0097] In the embodiment of the present application, the determining all main crack feature points based on the preliminary crack-connected region according to all the preliminary crack feature points specifically includes:
[0098] Perform texture detection on the preliminary crack-connected region to obtain connected texture information;
[0099] Judge whether each of the preliminary crack feature points exists in the connected texture information;
[0100] If so, determine the preliminary crack feature point as a main crack feature point;
[0101] If not, determine that the preliminary crack feature point is not a main crack feature point.
[0102] In the embodiments of the present application, after obtaining the preliminary crack connection region, texture detection is performed on the preliminary crack connection region to obtain texture detection information. Further, among all the preliminary crack feature points obtained in S102, the preliminary crack feature points in the connected texture information are determined as the main crack feature points; the preliminary crack feature points not in the connected texture information are not the main crack feature points and will not be processed subsequently. Among them, the specific implementation manner of texture detection can refer to the detailed description in the prior art and will not be specifically limited herein.
[0103] S106: Based on the pre-constructed first mask and second mask, traverse each pixel point of the image of the component to be reviewed to obtain a binary crack image.
[0104] In the embodiments of the present application, the step of traversing each pixel point of the image of the component to be reviewed based on the pre-constructed first mask and second mask to obtain a binary crack image specifically includes:
[0105] For each pixel point of the image of the component to be reviewed, based on the pre-constructed first mask, obtain a third pixel difference;
[0106] Based on the pre-constructed second mask and the third pixel difference, obtain a fourth pixel difference;
[0107] Judge whether the fourth pixel difference is greater than a second preset difference;
[0108] If so, set the pixel value of the pixel point corresponding to the third pixel difference to 1;
[0109] If not, set the pixel value of the pixel point corresponding to the third pixel difference to 0;
[0110] According to the pixel values of each pixel point of the image of the component to be reviewed, obtain the binary crack image.
[0111] In the embodiments of the present application, after pre-identifying the image of the component to be reviewed through the above steps S101 to S105, further, the image of the component to be reviewed is accurately identified in combination with the mask.
[0112] Two different masks are pre-constructed, namely the first mask and the second mask. Among them, the first mask can be as Figure 2 shown, and the second mask can be as Figure 3 shown. The second mask contains two circles with the same center. The smaller circle is hereinafter referred to as the inner circle, and the larger circle is hereinafter referred to as the outer circle.
[0113] For each pixel point of the image of the component to be reviewed, with this pixel point as the center point, the first mask and the second mask are successively covered on this center point. For the first mask, calculate the pixel differences between the pixel point at the center point and the surrounding 8 pixel points to obtain the third pixel difference. For the second mask, calculate the first pixel value gradient between the circle of pixel points enclosed by the inner circle and the circle of pixel points enclosed by the outer circle in the second mask, and further calculate the second pixel value gradient between the circle of pixel points enclosed by the inner circle in the second mask and the pixel point at the center point; for each pixel point between the inner circle and the outer circle, determine whether both its first pixel value gradient and second pixel value gradient are greater than the second preset difference. If so, after performing weighted summation on the first pixel value gradient and the second pixel value gradient of this pixel point, add the third pixel difference to obtain the fourth pixel difference.
[0114] Determine whether the fourth pixel difference is greater than the second preset difference. If so, set the pixel value of the pixel point corresponding to the third pixel difference (i.e., the pixel point that performs difference operation with the pixel point at the center point) to 1; if not, set the pixel value of the pixel point corresponding to the third pixel difference to 0. Among them, the second preset difference is set according to specific circumstances. For a color image, it can be set to 20, and for a black-and-white image, it can be set to 5.
[0115] By means of the first mask and the second mask, traverse each pixel point of the image of the component to be reviewed, complete the setting of the pixel values, and obtain the crack binary image corresponding to the image of the component to be reviewed.
[0116] S107: Determine all true crack feature points according to all the main crack feature points and the crack binary image.
[0117] In the embodiment of the present application, the step of determining all true crack feature points according to all the main crack feature points and the crack binary image specifically includes:
[0118] For each of the main crack feature points, determine whether the pixel value of the pixel point at the corresponding position of the main crack feature point in the crack binary image is 1;
[0119] If so, determine that the main crack feature point is a true crack feature point;
[0120] If not, determine that the main crack feature point is a false crack feature point.
[0121] In the embodiment of the present application, in combination with the crack binary image determined in step S106 and all the main crack feature points determined in step S105, the crack feature points are further accurately identified. Taking the position information of pixels as the corresponding relationship, the pixel points corresponding to each main crack feature point in the crack binary image are found, and it is judged whether the pixel value of the pixel point is 1. If it is 1, it is confirmed that the main crack feature point is a true crack feature point; if it is 0, it is confirmed that the main crack feature point is a false crack feature point.
[0122] All true crack feature points are obtained in the above manner. When extracting cracks subsequently, only true crack feature points are considered, and false crack feature points are no longer processed.
[0123] S108: Determine the cracks in the image of the component to be reviewed according to all the true crack feature points.
[0124] In the embodiment of the present application, the determining the cracks in the image of the component to be reviewed according to all the true crack feature points specifically includes:
[0125] Obtain the distances between adjacent true crack feature points among all the true crack feature points to obtain an average distance;
[0126] Randomly select a true crack feature point as the center, set a crack confirmation circle with the average distance as the radius, and connect the true crack feature points covered by the crack confirmation circle;
[0127] Screen the true crack feature points closest to the crack confirmation circle from the true crack feature points, and use the true crack feature point as the center of the next crack confirmation circle;
[0128] Repeat the process of setting the crack confirmation circle and connecting the true crack feature points covered by the crack confirmation circle until the center of the next crack confirmation circle cannot be obtained, to obtain a crack in the image of the component to be reviewed.
[0129] In an embodiment of the present application, the distance between two adjacent (nearest) true crack feature points among all true crack feature points is calculated, and then the average value of these distances is calculated to obtain the average distance. With this average distance as the radius and a randomly selected true crack feature point as the center, a crack confirmation circle is set. Each true crack feature point covered by the crack confirmation circle is connected (adjacent true crack feature points are connected together), and the connection result is in a broken line shape. Then, the true crack feature point closest to the crack confirmation circle is selected from each true crack feature point covered by the crack confirmation circle and used as the center of the next crack confirmation circle. The crack confirmation circle is set again in the above manner, and each true crack feature point covered by the crack confirmation circle is connected, and then the process of confirming the next crack confirmation circle is repeated until the center of the next crack confirmation circle cannot be obtained. At this time, all the connected true crack feature points are connected to form a crack, that is, a crack in the image of the component to be reviewed is obtained.
[0130] In an embodiment of the present application, after obtaining any crack in the image of the component to be reviewed, it further includes:
[0131] After obtaining any crack in the image of the component to be reviewed, it is determined whether there are unconnected true crack feature points;
[0132] If so, the crack confirmation circle is set with an unconnected true crack feature point as the center, and the process of setting the crack confirmation circle and connecting each true crack feature point covered by the crack confirmation circle is repeatedly executed to obtain the next crack in the image of the component to be reviewed;
[0133] If not, it is determined that the cracks in the image of the component to be reviewed have been obtained.
[0134] In an embodiment of the present application, there may be more than one crack in an image of a component to be reviewed. At this time, after obtaining any crack, there may still be unvisited true crack feature points in the image of the component to be reviewed. Therefore, after obtaining any crack in the image of the component to be reviewed, it is determined whether there are unconnected true crack feature points. If there are, the connection can continue by setting the crack confirmation circle, that is, the crack confirmation circle is set with an unconnected true crack feature point as the center, and the process of setting the crack confirmation circle and connecting each true crack feature point covered by the crack confirmation circle is repeatedly executed to obtain the next crack in the image of the component to be reviewed; if not, that is, all true crack feature points in the image of the component to be reviewed are connected, then the cracks in the image of the component to be reviewed have been obtained.
[0135] In the embodiments of the present application, if no crack is obtained after traversing the image of the component to be reviewed through the processes of the above steps S101 to S108, it is determined that there is no crack in the component to be reviewed corresponding to the image of the component to be reviewed. By the above method, based on the image of the component to be reviewed, it is checked whether there is a crack in the component to be reviewed. If there is a crack, a quality report for inspecting the component to be reviewed is issued; if no crack is detected, the signal and position of the component to be inspected are stored in the inspection queue.
[0136] The quality detection method for the image of the component to be reviewed provided by the embodiments of the present application first extracts the edge features and texture features of the image to be reviewed to determine the preliminary crack feature points, then determines the preliminary crack connection region, and judges whether the preliminary crack feature points are the main crack feature points. In addition, based on the pre-constructed first mask and second mask, the image to be reviewed is traversed to obtain the corresponding binary crack image. Then, the main crack feature points are corresponding to the binary crack image to determine the true crack feature points. Finally, according to all the true crack feature points, the cracks in the image of the component to be reviewed are determined. This method is based on the source image for detection. Through two rounds of detection, crack detection is carried out from multiple dimensions, multiple traversals, and multi-mask fusion, improving the accuracy and precision of crack detection; using the crack detection result as a reference for quality inspection, avoiding unnecessary conflicts and quality and safety problems caused by manual inspection, and improving the reliability, safety, and effectiveness of the management of the quality of the components to be reviewed involved in welding and prefabrication in petroleum engineering construction, and improving the accuracy of quality grade determination.
[0137] Embodiment 2
[0138] Based on the same inventive concept, the embodiments of the present application also provide a quality detection device for the image of the component to be reviewed. Referring to Figure 4 as shown, the device includes:
[0139] A feature extraction module 101, configured to obtain an image of a component to be reviewed and extract the edge features and texture features of the image of the component to be reviewed;
[0140] A first determination module 102, configured to determine all preliminary crack feature points in the image of the component to be reviewed according to the edge features and the texture features;
[0141] A second determination module 103, configured to obtain a first crack connection region of the image of the component to be reviewed based on the edge features, and obtain a second crack connection region of the image of the component to be reviewed based on the texture features;
[0142] A fusion module 104, configured to fuse the first crack connection region and the second crack connection region to obtain a preliminary crack connection region;
[0143] A third determination module 105, configured to determine all main crack feature points based on the preliminary crack connection regions and according to all the preliminary crack feature points;
[0144] A mask traversal module 106, configured to traverse each pixel point of the component image to be audited based on a pre-constructed first mask and a second mask, and obtain a crack binary image;
[0145] A fourth determination module 107, configured to determine all true crack feature points according to all the main crack feature points and the crack binary image;
[0146] A crack determination module 108, configured to determine the cracks in the component image to be audited according to all the true crack feature points.
[0147] Embodiment III
[0148] Based on the same inventive concept, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the quality detection method for the component image to be audited as described in Embodiment I above.
[0149] Embodiment IV
[0150] Based on the same inventive concept, an embodiment of the present application further provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the quality detection method for the component image to be audited as described in Embodiment I above.
[0151] Embodiment V
[0152] Based on the same inventive concept, an embodiment of the present application further provides a computer program product containing instructions. When the computer program product runs on a computer device, it causes the computer device to execute the quality detection method for the component image to be audited as described in Embodiment I above.
[0153] Embodiment VI
[0154] Based on the same inventive concept, an embodiment of the present application further provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor, and the processor is used to run a computer program or instructions to implement the quality detection method for the component image to be audited as described in Embodiment I above.
[0155] Those skilled in the art should understand that the embodiments of this application can be provided as a method, a system, or a computer program product. Therefore, this application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) that contain computer-usable program code.
[0156] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of this application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.
[0157] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.
[0158] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.
[0159] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalent technologies, this application is also intended to include these modifications and variations.
Claims
1. A method for quality inspection of an image of a component to be reviewed, characterized in that, it includes: Obtain an image of the component to be reviewed, and extract the edge features and texture features of the image of the component to be reviewed; Based on the edge features and the texture features, determine all preliminary crack feature points in the image of the component to be reviewed; Based on the edge features, obtain the first crack connected region of the image of the component to be reviewed, and based on the texture features, obtain the second crack connected region of the image of the component to be reviewed; Fuse the first crack connected region and the second crack connected region to obtain a preliminary crack connected region; Based on the preliminary crack connected region, according to all the preliminary crack feature points, determine all the main crack feature points; Based on a pre-constructed first mask and second mask, traverse each pixel point of the image of the component to be reviewed to obtain a crack binary image; Based on all the main crack feature points and the crack binary image, determine all the true crack feature points; Based on all the true crack feature points, determine the cracks in the image of the component to be reviewed.
2. The method according to claim 1, characterized in that, The step of determining all preliminary crack feature points in the image of the component to be reviewed according to the edge features and the texture features includes: For each pixel point of the image of the component to be reviewed, determine whether the pixel point is both an edge feature point and a texture feature point; If so, determine that the pixel point is a preliminary crack feature point in the image of the component to be reviewed.
3. The method according to claim 1, characterized in that, The step of fusing the first crack connected region and the second crack connected region to obtain a preliminary crack connected region includes: For each pixel point in the first crack connected region and the second crack connected region, determine the first pixel difference from the region covered by the preset LBP kernel in the first crack connected region to the preset LBP kernel, and the second pixel difference from the region covered by the preset LBP kernel in the second crack connected region to the preset LBP kernel; Determine whether the difference between the first pixel difference and the second pixel difference is less than a first preset difference; If so, perform weighted fusion on the region covered by the preset LBP kernel in the first crack connected region and the region covered by the preset LBP kernel in the second crack connected region to obtain a fused region; Map all the fused regions into an empty image to obtain a preliminary crack connected region.
4. The method according to claim 1, characterized in that, The step of determining all the main crack feature points based on the preliminary crack connected region and according to all the preliminary crack feature points includes: Perform texture detection on the preliminary crack connected region to obtain connected texture information; Determine whether each preliminary crack feature point exists in the connected texture information; If so, determine that the preliminary crack feature point is a main crack feature point.
5. The method according to claim 1, characterized in that, The step of traversing each pixel point of the image of the component to be reviewed based on a pre-constructed first mask and second mask to obtain a crack binary image includes: For each pixel point of the image of the component to be audited, based on the pre-constructed first mask, obtain the third pixel difference; Based on the pre-constructed second mask and the third pixel difference, obtain the fourth pixel difference; Determine whether the fourth pixel difference is greater than the second preset difference; If so, set the pixel value of the pixel point corresponding to the third pixel difference to 1; If not, set the pixel value of the pixel point corresponding to the third pixel difference to 0; According to the pixel values of each pixel point of the image of the component to be audited, obtain the crack binary image.
6. The method according to claim 1, wherein, The determining all true crack feature points according to all the main crack feature points and the crack binary image includes: For each of the main crack feature points, determine whether the pixel value of the pixel point at the corresponding position of the main crack feature point in the crack binary image is 1; If so, determine the main crack feature point as a true crack feature point.
7. The method according to claim 1, wherein, The determining the cracks in the image of the component to be audited according to all the true crack feature points includes: Obtain the distances between adjacent true crack feature points among all the true crack feature points to obtain an average distance; Randomly select a true crack feature point as the center of a circle, set a crack confirmation circle with the average distance as the radius, and connect the true crack feature points covered by the crack confirmation circle; Screen the true crack feature points that are closest to the crack confirmation circle from the true crack feature points, and use the true crack feature point as the center of the next crack confirmation circle; Repeat the process of setting the crack confirmation circle and connecting the true crack feature points covered by the crack confirmation circle until the center of the next crack confirmation circle cannot be obtained, to obtain a crack in the image of the component to be audited.
8. The method according to claim 7, wherein, further includes: After obtaining any crack in the image of the component to be audited, determine whether there are unconnected true crack feature points; If so, set the crack confirmation circle with an unconnected true crack feature point as the center, and repeat the process of setting the crack confirmation circle and connecting the true crack feature points covered by the crack confirmation circle to obtain the next crack in the image of the component to be audited.
9. A quality detection device for an image of a component to be audited, wherein, includes: A feature extraction module, configured to obtain an image of a component to be audited, and extract edge features and texture features of the image of the component to be audited; A first determination module, configured to determine all preliminary crack feature points in the image of the component to be audited according to the edge features and the texture features; A second determination module, configured to obtain a first crack connected region of the image of the component to be audited based on the edge features, and obtain a second crack connected region of the image of the component to be audited based on the texture features; A fusion module, configured to fuse the first crack connected region and the second crack connected region to obtain a preliminary crack connected region; A third determination module, configured to determine all main crack feature points based on the preliminary crack connected region and according to all the preliminary crack feature points; A mask traversal module, configured to traverse each pixel point of the component image to be reviewed based on a pre-constructed first mask and a second mask, and obtain a crack binary image; A fourth determination module, configured to determine all true crack feature points according to the all main crack feature points and the crack binary image; A crack determination module, configured to determine the cracks in the component image to be reviewed according to the all true crack feature points.
10. A computer-readable storage medium storing a computer program, which when executed by a processor, causes the processor to execute the quality detection method for the component image to be reviewed according to any one of claims 1-8.
11. A computer device, characterized in that, it includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, it implements the quality detection method for the component image to be reviewed according to any one of claims 1-8.
12. A computer program product containing instructions, which when the computer program product runs on a computer device, causes the computer device to execute the quality detection method for the component image to be reviewed according to any one of claims 1-8.
13. A chip, the chip includes a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run a computer program or instruction to implement the quality detection method for the component image to be reviewed according to any one of claims 1-8.