Three-dimensional reconstruction method and system for X-ray detection of power equipment

Abstract

The invention provides a three-dimensional reconstruction method and system for X-ray detection of power equipment, and relates to the technical field of power equipment detection.The method comprises the following steps that S1, a component label of a to-be-detected component in the power equipment is obtained; s2, obtaining a series of two-dimensional detection images through X-ray tomography scanning power equipment; s3, calling a two-dimensional reference region set obtained by X-ray tomography power equipment in advance according to the component label of the to-be-detected component; s4, screening the series of two-dimensional detection images according to the layer number marks in the two-dimensional reference region set, and performing region extraction on the screened two-dimensional detection images according to the reference region images in the two-dimensional reference region set to obtain region images of the part to be detected on the screened two-dimensional detection images; and S5, carrying out image processing on a region image of the to-be-detected part on the two-dimensional detection image obtained by screening, extracting key points in the region image, and carrying out three-dimensional reconstruction on the region image.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment detection, and in particular to a three-dimensional reconstruction method and system for X-ray detection of power equipment. Background Art

[0002] For power equipment detection, the three-dimensional reconstruction method is currently often used. The power equipment is scanned by X-rays to obtain a two-dimensional image, and the two-dimensional image is image recognized to obtain the feature data of the entire area. The three-dimensional reconstruction of the current power equipment can be achieved based on the feature data in multiple two-dimensional images. Problems with the power equipment can be detected relatively easily in the three-dimensional structure. However, this method has certain defects. Because there are many components in the power equipment, the extraction of feature data in the image recognition process will be greatly increased, resulting in slow calculation speed and poor detection efficiency; and incomplete feature data extraction will also lead to poor three-dimensional reconstruction effect, resulting in detection errors. Summary of the invention

[0003] The purpose of the present invention is to provide a three-dimensional reconstruction method and system for X-ray detection of power equipment. By performing three-dimensional reconstruction of regional images, each component in the power equipment can be accurately detected, while reducing the amount of calculation and ensuring the accuracy of the detection.

[0004] In order to solve the above technical problems, the present invention adopts the following solutions:

[0005] A three-dimensional reconstruction method for X-ray detection of electric power equipment, the method comprising the following steps:

[0006] S1. Obtaining component labels of components to be detected in power equipment;

[0007] S2, obtaining a series of two-dimensional detection images of the power equipment by X-ray tomography;

[0008] S3, calling a two-dimensional reference area set obtained in advance by X-ray tomography scanning of the power equipment according to the component label of the component to be detected;

[0009] S4, screening a series of two-dimensional detection images according to the layer number marks in the two-dimensional reference area set, and performing region extraction on the screened two-dimensional detection images according to the reference area images in the two-dimensional reference area set to obtain a region image of the component to be detected on the screened two-dimensional detection image;

[0010] S5. Perform image processing on the region image of the component to be detected on the screened two-dimensional detection image, extract key points in the region image, and perform three-dimensional reconstruction on the region image.

[0011] Furthermore, in S2, the process of obtaining a series of two-dimensional detection images by X-ray tomography of power equipment is as follows: X-rays are used to perform multiple tomography scans on the power equipment, and cross-sections of the power equipment at different layers are scanned to obtain corresponding two-dimensional detection images, thereby obtaining a series of two-dimensional detection images.

[0012] Furthermore, in S3, the process of constructing the two-dimensional reference area set is as follows:

[0013] First, the power equipment is scanned multiple times by X-rays in advance, and the cross sections of all layers of the power equipment are scanned to obtain two-dimensional reference images of different layers. Meanwhile, the two-dimensional reference images of different layers are marked with corresponding layer numbers.

[0014] Then, the components in all the two-dimensional reference images are identified to obtain the locations of the components, and the two-dimensional reference images are cropped according to the locations of the components to obtain reference area images of different components. At the same time, the reference area images of different components are marked with component labels;

[0015] Finally, the reference area images under the same component label are packaged to form a two-dimensional reference area set.

[0016] Furthermore, the process of cropping the two-dimensional reference image according to the location of the component is as follows:

[0017] The position of the center of mass of the component and the position of the outer contour of the component are obtained, and the position of the outer contour of the component is enlarged by a preset multiple with the position of the center of mass of the component as the center point to obtain the reference area of ​​the component; the reference areas of all components on the two-dimensional reference image can be obtained, and the two-dimensional reference image is cropped based on the reference areas of all components to obtain reference area images of different components.

[0018] Furthermore, the S4 includes the following steps:

[0019] S41, obtaining all reference region images in a two-dimensional reference region set and layer number marks corresponding to the reference region images;

[0020] S42, screening a series of two-dimensional detection images according to the layer number marks corresponding to the reference area image, and screening out two-dimensional detection images with the same layer number marks;

[0021] S43, matching the reference area image with the two-dimensional detection image marked with the same number of layers, and cropping the two-dimensional detection image according to the reference area image to obtain an area image of the component to be detected on the screened two-dimensional detection image.

[0022] Furthermore, in S43, the process of cutting the two-dimensional detection image according to the reference area image is as follows:

[0023] The position of the centroid of the part to be detected in the reference area image is obtained, the part to be detected in the two-dimensional detection image is identified, and the position of the centroid of the part to be detected in the two-dimensional detection image is obtained. The position of the centroid of the part to be detected in the reference area image is matched with the position of the centroid of the part to be detected in the two-dimensional detection image. The reference area image is overlapped on the two-dimensional detection image, and the overlapping part is cropped to obtain a regional image of the part to be detected on the screened two-dimensional detection image.

[0024] Furthermore, the image processing includes image data search, image filtering, image enhancement, and grayscale processing.

[0025] Furthermore, the three-dimensional reconstruction refers to inverting the three-dimensional structure of the part to be detected by using key points of the part to be detected in regional images of different layers.

[0026] A three-dimensional reconstruction system for X-ray detection of electric power equipment, using the three-dimensional reconstruction method for X-ray detection of electric power equipment, comprising:

[0027] The component label confirmation module to be detected is used to obtain the component label of the component to be detected in the power equipment;

[0028] Two-dimensional detection image acquisition module: obtain a series of two-dimensional detection images through X-ray tomography of power equipment;

[0029] A two-dimensional reference area set acquisition module: a two-dimensional reference area set obtained in advance by X-ray tomography scanning of the power equipment is called according to the component label of the component to be detected;

[0030] The regional image acquisition module is used to screen and process a series of two-dimensional detection images according to the layer number marks in the two-dimensional reference region set, and to perform region extraction on the screened two-dimensional detection images according to the reference region images in the two-dimensional reference region set, so as to obtain a region image of the component to be detected on the screened two-dimensional detection image;

[0031] Regional image three-dimensional reconstruction module: performs image processing on the regional image of the component to be detected on the screened two-dimensional detection image, extracts key points in the regional image, and reconstructs the regional image in three dimensions.

[0032] Beneficial effects of the present invention:

[0033] The present invention provides a three-dimensional reconstruction method and system for X-ray detection of electric power equipment. The three-dimensional reconstruction method mainly performs three-dimensional reconstruction on the components to be detected, and cuts the redundant parts in the two-dimensional image through the area where the components to be detected are located, so as to achieve accurate detection of the components to be detected in the electric power equipment, while reducing the amount of calculation and ensuring the accuracy of the detection. In addition, by obtaining a two-dimensional reference image in advance, the redundant parts in the two-dimensional image can be quickly and accurately cut, thereby improving the detection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a flow chart of a three-dimensional reconstruction method for X-ray detection of power equipment in Embodiment 1 of the present invention;

[0035] Figure 2 Schematic diagram of the reference area of ​​the bolt in the two-dimensional reference image in Example 1 of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] The relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.

[0038] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0039] Additionally, descriptions of well-known structures, functions, and configurations may be omitted for clarity and conciseness.One of ordinary skill in the art will recognize that various changes and modifications may be made to the examples described herein without departing from the spirit and scope of the present disclosure.

[0040] Technologies, methods, and apparatus known to ordinary technicians in the relevant field may not be discussed in detail, but where appropriate, such technologies, methods, and apparatus should be considered part of the authorization specification.

[0041] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0042] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments:

[0043] Example 1

[0044] At present, a three-dimensional reconstruction method is often used to detect power equipment. The three-dimensional reconstruction method is specifically: to obtain a two-dimensional image by tomography scanning the power equipment through X-rays, and to perform image recognition on the two-dimensional image to obtain the feature data of the entire area. According to the feature data in multiple two-dimensional images, the three-dimensional reconstruction of the current power equipment can be achieved. The detection of power equipment usually involves detecting whether its internal components are loose, broken, etc. When there are a large number of components in the power equipment, the use of the above three-dimensional reconstruction method has certain defects, which will greatly increase the extraction of the feature data of the components during the image recognition process, slow the calculation speed, and lead to poor detection efficiency; it may also lead to poor three-dimensional reconstruction effect and detection errors due to incomplete extraction of the feature data of the components.

[0045] Therefore, in this embodiment, in order to solve the above problems, a three-dimensional reconstruction method for X-ray detection of power equipment is proposed, such as Figure 1 As shown, the method comprises the following steps:

[0046] S1. Obtaining component labels of components to be detected in power equipment;

[0047] The component label is used to mark different components. The component label can be marked by the name of the component. The component label can be used to know the current component to be detected. The component label can be a bolt. In this case, the three-dimensional reconstruction method is used to detect bolts in power equipment.

[0048] S2, obtaining a series of two-dimensional detection images of the power equipment by X-ray tomography;

[0049] S3, calling a two-dimensional reference area set obtained in advance by X-ray tomography scanning of the power equipment according to the component label of the component to be detected;

[0050] S4, screening a series of two-dimensional detection images according to the layer number marks in the two-dimensional reference area set, and performing region extraction on the screened two-dimensional detection images according to the reference area images in the two-dimensional reference area set to obtain a region image of the component to be detected on the screened two-dimensional detection image;

[0051] S5. Perform image processing on the region image of the component to be detected on the screened two-dimensional detection image, extract key points in the region image, and perform three-dimensional reconstruction on the region image.

[0052] Preferably, in S2, the process of obtaining a series of two-dimensional detection images by X-ray tomography of power equipment is as follows: X-rays are used to perform multiple tomography scans on the power equipment, and cross-sections of the power equipment at different layers are scanned to obtain corresponding two-dimensional detection images, thereby obtaining a series of two-dimensional detection images.

[0053] Preferably, in S3, the process of constructing the two-dimensional reference area set is:

[0054] First, the power equipment is scanned multiple times by X-rays in advance, and the cross sections of all layers of the power equipment are scanned to obtain two-dimensional reference images of different layers. Meanwhile, the two-dimensional reference images of different layers are marked with corresponding layer numbers.

[0055] Then, the components in all the two-dimensional reference images are identified to obtain the locations of the components, and the two-dimensional reference images are cropped according to the locations of the components to obtain reference area images of different components. At the same time, the reference area images of different components are marked with component labels;

[0056] Finally, the reference area images under the same component label are packaged to form a two-dimensional reference area set.

[0057] Preferably, the process of cutting the two-dimensional reference image according to the location of the component is:

[0058] The position of the center of mass of the component and the position of the outer contour of the component are obtained, and the position of the outer contour of the component is enlarged by a preset multiple with the position of the center of mass of the component as the center point to obtain the reference area of ​​the component; the reference areas of all components on the two-dimensional reference image can be obtained, and the two-dimensional reference image is cropped based on the reference areas of all components to obtain reference area images of different components.

[0059] If the component to be inspected is a bolt, the number of bolts in the same section of the power equipment may be multiple, as shown in the following example: Figure 2 As shown, if there are two bolts in the current section during the recognition of the two-dimensional reference image, the positions of the two bolt centroids and the outer contours of the bolts can be obtained, and the two bolts are processed separately. With the position of the bolt centroid as the center point, the position of the outer contour of the bolt is enlarged by a preset multiple, and the reference area of ​​the two bolts in the current section can be obtained.

[0060] Preferably, S4 includes the following steps:

[0061] S41, obtaining all reference region images in a two-dimensional reference region set and layer number marks corresponding to the reference region images;

[0062] S42, screening a series of two-dimensional detection images according to the layer number marks corresponding to the reference area image, and screening out two-dimensional detection images with the same layer number marks;

[0063] S43, matching the reference area image with the two-dimensional detection image marked with the same number of layers, and cropping the two-dimensional detection image according to the reference area image to obtain an area image of the component to be detected on the screened two-dimensional detection image.

[0064] Preferably, in S43, the process of cutting the two-dimensional detection image according to the reference area image is:

[0065] The position of the centroid of the part to be detected in the reference area image is obtained, the part to be detected in the two-dimensional detection image is identified, and the position of the centroid of the part to be detected in the two-dimensional detection image is obtained. The position of the centroid of the part to be detected in the reference area image is matched with the position of the centroid of the part to be detected in the two-dimensional detection image. The reference area image is overlapped on the two-dimensional detection image, and the overlapping part is cropped to obtain a regional image of the part to be detected on the screened two-dimensional detection image.

[0066] Preferably, the image processing includes image data search, image filtering, image enhancement, and grayscale processing.

[0067] Preferably, the three-dimensional reconstruction refers to inverting the three-dimensional structure of the component to be detected by using key points of the component to be detected in regional images of different layers.

[0068] Example 2

[0069] A three-dimensional reconstruction system for X-ray detection of electric power equipment, using the three-dimensional reconstruction method for X-ray detection of electric power equipment, comprising:

[0070] The component label confirmation module to be detected is used to obtain the component label of the component to be detected in the power equipment;

[0071] Two-dimensional detection image acquisition module: obtain a series of two-dimensional detection images through X-ray tomography of power equipment;

[0072] A two-dimensional reference area set acquisition module: a two-dimensional reference area set obtained in advance by X-ray tomography scanning of the power equipment is called according to the component label of the component to be detected;

[0073] The regional image acquisition module is used to screen and process a series of two-dimensional detection images according to the layer number marks in the two-dimensional reference region set, and to perform region extraction on the screened two-dimensional detection images according to the reference region images in the two-dimensional reference region set, so as to obtain a region image of the component to be detected on the screened two-dimensional detection image;

[0074] Regional image three-dimensional reconstruction module: performs image processing on the regional image of the component to be detected on the screened two-dimensional detection image, extracts key points in the regional image, and reconstructs the regional image in three dimensions.

[0075] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. According to the technical essence of the present invention, within the spirit and principles of the present invention, any simple modification, equivalent replacement and improvement made to the above embodiment still falls within the protection scope of the technical solution of the present invention.

Claims

1. A three-dimensional reconstruction method for X-ray detection of power equipment, characterized in that: The method comprises the following steps: S1. Obtaining component labels of components to be detected in power equipment; S2, obtaining a series of two-dimensional detection images of the power equipment by X-ray tomography; S3, calling a two-dimensional reference area set obtained in advance by X-ray tomography scanning of the power equipment according to the component label of the component to be detected; S4, screening a series of two-dimensional detection images according to the layer number marks in the two-dimensional reference area set, and performing region extraction on the screened two-dimensional detection images according to the reference area images in the two-dimensional reference area set to obtain a region image of the component to be detected on the screened two-dimensional detection image; S5. Perform image processing on the region image of the component to be detected on the screened two-dimensional detection image, extract key points in the region image, and perform three-dimensional reconstruction on the region image.

2. A three-dimensional reconstruction method for X-ray detection of power equipment according to claim 1, characterized in that: In S2, the process of obtaining a series of two-dimensional detection images by X-ray tomography of power equipment is as follows: X-rays are performed on the power equipment for multiple times, and cross-sections of the power equipment at different layers are scanned to obtain corresponding two-dimensional detection images, thereby obtaining a series of two-dimensional detection images.

3. A three-dimensional reconstruction method for X-ray detection of power equipment according to claim 1, characterized in that: In S3, the process of constructing the two-dimensional reference area set is as follows: First, the power equipment is scanned multiple times by X-rays in advance, and the cross sections of all layers of the power equipment are scanned to obtain two-dimensional reference images of different layers. Meanwhile, the two-dimensional reference images of different layers are marked with corresponding layer numbers. Then, the components in all the two-dimensional reference images are identified to obtain the locations of the components, and the two-dimensional reference images are cropped according to the locations of the components to obtain reference area images of different components. At the same time, the reference area images of different components are marked with component labels; Finally, the reference area images under the same component label are packaged to form a two-dimensional reference area set.

4. A three-dimensional reconstruction method for X-ray detection of power equipment according to claim 3, characterized in that: The process of cropping the 2D reference image according to the location of the component is as follows: The position of the centroid of the component and the position of the outer contour of the component are obtained, and the position of the outer contour of the component is enlarged by a preset multiple with the position of the centroid of the component as the center point to obtain the reference area of ​​the component; The reference areas of all components on the two-dimensional reference image can be obtained, and the two-dimensional reference image is cropped based on the reference areas of all components to obtain reference area images of different components.

5. A three-dimensional reconstruction method for X-ray detection of power equipment according to claim 3, characterized in that: The S4 includes the following steps: S41, obtaining all reference region images in a two-dimensional reference region set and layer number marks corresponding to the reference region images; S42, screening a series of two-dimensional detection images according to the layer number marks corresponding to the reference area image, and screening out two-dimensional detection images with the same layer number marks; S43, matching the reference area image with the two-dimensional detection image marked with the same number of layers, and cropping the two-dimensional detection image according to the reference area image to obtain an area image of the component to be detected on the screened two-dimensional detection image.

6. A three-dimensional reconstruction method for X-ray detection of power equipment according to claim 5, characterized in that: In S43, the process of cutting out the two-dimensional detection image according to the reference area image is as follows: The position of the centroid of the part to be detected in the reference area image is obtained, the part to be detected in the two-dimensional detection image is identified, and the position of the centroid of the part to be detected in the two-dimensional detection image is obtained. The position of the centroid of the part to be detected in the reference area image is matched with the position of the centroid of the part to be detected in the two-dimensional detection image. The reference area image is overlapped on the two-dimensional detection image, and the overlapping part is cropped to obtain a regional image of the part to be detected on the screened two-dimensional detection image.

7. A three-dimensional reconstruction method for X-ray detection of power equipment according to claim 1, characterized in that: The image processing includes image data search, image filtering, image enhancement, and grayscale processing.

8. A three-dimensional reconstruction method for X-ray detection of power equipment according to claim 1, characterized in that: The three-dimensional reconstruction refers to inverting the three-dimensional structure of the part to be detected through key points of the part to be detected in regional images of different layers.

9. A three-dimensional reconstruction system for X-ray detection of power equipment, characterized in that: A three-dimensional reconstruction method for X-ray detection of power equipment according to any one of claims 1 to 8 is applied, comprising: The component label confirmation module to be detected is used to obtain the component label of the component to be detected in the power equipment; Two-dimensional detection image acquisition module: obtain a series of two-dimensional detection images through X-ray tomography of power equipment; A two-dimensional reference area set acquisition module: a two-dimensional reference area set obtained in advance by X-ray tomography scanning of the power equipment is called according to the component label of the component to be detected; The regional image acquisition module is used to screen and process a series of two-dimensional detection images according to the layer number marks in the two-dimensional reference region set, and to perform region extraction on the screened two-dimensional detection images according to the reference region images in the two-dimensional reference region set, so as to obtain a region image of the component to be detected on the screened two-dimensional detection image; Regional image three-dimensional reconstruction module: performs image processing on the regional image of the component to be detected on the screened two-dimensional detection image, extracts key points in the regional image, and reconstructs the regional image in three dimensions.

Images