X-ray weld joint pseudo-defect segmentation method and device based on skeleton extraction, storage medium and electronic equipment

Abstract

The invention relates to the technical field of pseudo defect segmentation, in particular to an X-ray welding seam pseudo defect segmentation method and device based on skeleton extraction, a storage medium and electronic equipment, and the method comprises the steps: extracting the longest edge in a to-be-segmented X-ray welding seam negative film, and expanding the longest edge to obtain a fused connected image; and carrying out skeleton extraction on the fused connected image, and carrying out expansion on the extracted skeleton image. According to the method, interference of character marks of numbers and letters and the like is stripped through two modes of edge extraction and skeleton extraction, accurate positioning of long-strip-shaped pseudo defects such as hair and fibers is achieved, accurate segmentation of the pseudo defects is completed, the problem that segmentation is difficult due to interference overlapping of the pseudo defects and the numbers, the characters and the like is effectively solved, and the segmentation efficiency is improved. And compared with the existing manual pseudo defect segmentation mode, the method has the advantages of high efficiency and high accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of pseudo defect segmentation, and specifically to a skeleton extraction-based X-ray weld pseudo defect segmentation method, device, storage medium and electronic equipment. Background Art

[0002] As one of the most important material forming and processing technologies in the steel industry, welding is a key process for accurately and quickly connecting steel materials into various forms. It is widely used in construction, manufacturing, transportation and other fields, and has promoted the development of the steel industry. In the field of welding inspection, pseudo defects refer to normal weld features or image interference that are mistakenly identified as defects during the inspection process. These pseudo defects usually do not affect the actual performance and safety of the weld, but their misidentification may affect the defect annotation and the accuracy of the model, thereby affecting the inspection process. Therefore, when rating weld defects on film, pseudo defects need to be segmented and identified, which is a key link in the evaluation of film image quality.

[0003] X-ray weld films generally include a brighter weld area and a darker base material area. The base material area may contain some character markings such as numbers and letters. If small objects such as hair and fiber are accidentally mixed in the imaging process, they will appear as very long and thin images on the imaging screen. These images may pass through the characters in the base material area. At the same time, because the brightness of the pseudo-defects is almost the same as that of the characters, it brings great difficulties to the segmentation of pseudo-defects.

[0004] At present, the methods for segmenting weld pseudo defects in X-ray weld films include: (1) Manual segmentation of pseudo-defects. This method relies on the technical quality and experience of personnel, which is prone to misjudgment and omission, and is inefficient and wastes resources.

[0005] (2) Defect segmentation and identification based on artificial intelligence, for example: An existing public patent document 1, with publication number CN118747739A, discloses a weld defect determination method, apparatus, computer equipment and storage medium, wherein the weld defect determination method includes performing standardization processing on a weld defect image to be determined to obtain an input weld image; using a trained weld recognition model to process the input weld image to obtain standard weld defect data corresponding to the input weld image; and performing inverse standardization processing on the standard weld defect data to determine target weld defect data corresponding to the weld defect image to be determined.

[0006] Existing public patent document 2, publication number CN116883345A, discloses a large-scale weld defect detection method and system based on deep learning, including: inputting the preprocessed weld image into a backbone network model for training, the backbone network model includes two or more convolution units, wherein the convolution kernel of the last layer of convolution units is a void convolution kernel with different void rates; inputting the output result of the backbone network model into a feature pyramid network for feature fusion and enhancement; and inputting the fused and enhanced features into a regression network and a classification network respectively to obtain the classification confidence and position of the weld defect.

[0007] The above-mentioned methods of defect segmentation and identification based on artificial intelligence are all aimed at improving the detection rate and detection accuracy of large-scale defects, but they do not explain how to segment pseudo-defects to make defect identification more accurate, nor do they explain how to solve the problem of difficulty in segmentation caused by the overlap of long strip pseudo-defects and characters in the film image. Summary of the invention

[0008] The present invention provides a method that overcomes the above-mentioned deficiencies of the prior art and can effectively solve the problem in the existing pseudo-defect segmentation method that the overlapping parts of the long strip pseudo-defects and the interferences such as character marks in the X-ray weld film cannot be effectively segmented.

[0009] One of the technical solutions of the present invention is achieved by the following measures: an X-ray weld pseudo defect segmentation method based on skeleton extraction, comprising: Extract the longest edge in the X-ray weld film to be segmented, and expand it to obtain a fused connected image; The skeleton of the fused connected image is extracted, and the extracted skeleton image is expanded.

[0010] The following are further optimizations and / or improvements to the above technical solutions: The longest edge in the X-ray weld film to be segmented is extracted and expanded to obtain a fused connected image, including: The edge detection operator is used to extract the edge of the X-ray weld film to be segmented; The mentioned edges are connected as follows: in, is the image after edge connection; and are the structural element and the number of iterations in the expansion, respectively; Select the longest edge from the multiple edges after edge connection and expand it again to obtain a fused connected image, as follows: in, is a fused connected image; , A function for finding the longest edge in the edge connectivity result; and are the structural element and the number of iterations in the expansion respectively.

[0011] The structure element set during the above-mentioned re-expansion is a 4×4 rectangular structure element, and the number of iterations is 11.

[0012] The above edge detection operator is the Canny edge detection operator.

[0013] The above-mentioned skeleton extraction of the fused connected image and expansion of the extracted skeleton image include: based on The function determines the central axis of the image area in the fused connected image and extracts the skeleton image, as follows: in, is a skeleton image; To fuse connected images; The extracted skeleton image is expanded as follows: in, It is the result of dilating the skeleton image; and are the structural element and the number of iterations in the expansion respectively.

[0014] The second technical solution of the present invention is achieved by the following measures: an X-ray weld pseudo-defect segmentation device based on skeleton extraction, comprising: The edge extraction unit extracts the longest edge in the X-ray weld film to be segmented and expands it to obtain a fused connected image; The skeleton extraction unit extracts the skeleton of the fused connected image and dilates the extracted skeleton image.

[0015] The following are further optimizations and / or improvements to the above technical solutions: The above edge extraction unit comprises: The edge detection module uses the edge detection operator to extract the edge of the X-ray weld film to be segmented; The connectivity module connects the mentioned edges as follows: in, is the image behind the edge; and are the structural elements and iteration numbers of the expansion respectively; The fusion module selects the longest edge among the multiple edges after edge connection and expands it again to obtain a fused connected image, as follows: in, is a fused connected image; , To find the longest edge function in the edge connection results; and are the structural element and the number of iterations in the expansion respectively.

[0016] The skeleton extraction unit comprises: Skeleton image extraction module, based on The function determines the central axis of the image area in the fused connected image and extracts the skeleton image, as follows: in, is a skeleton image; To fuse connected images; The skeleton image expansion module expands the extracted skeleton image as follows: in, It is the result of dilating the skeleton image; and are the structural element and the number of iterations in the expansion respectively.

[0017] The third technical solution of the present invention is achieved through the following measures: an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the computer program is loaded and executed by the processor to implement the steps in the X-ray weld pseudo-defect segmentation method based on skeleton extraction.

[0018] The fourth technical solution of the present invention is achieved through the following measures: a storage medium, characterized in that a computer program that can be read by a computer is stored on the storage medium, and the computer program is configured to execute the steps in the X-ray weld pseudo-defect segmentation method based on skeleton extraction when running.

[0019] The present invention uses edge extraction and skeleton extraction to remove interference such as character logos of numbers and letters, accurately locate long strip-shaped pseudo-defects such as hair and fiber strands, and complete precise segmentation of pseudo-defects, effectively solving the problem of segmentation difficulty caused by overlapping of pseudo-defects with interference such as characters, and has high efficiency and accuracy compared to the existing manual segmentation method of pseudo-defects. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Attached Figure 1 A schematic flow chart of a segmentation method provided by an embodiment of the present invention.

[0021] Attached Figure 2 A schematic diagram of a fusion connectivity method flow chart provided by an embodiment of the present invention.

[0022] Attached Figure 3 A schematic diagram of a skeleton extraction method flow chart provided for one embodiment of the present invention.

[0023] Attached Figure 4 A schematic diagram of an X-ray weld film to be segmented provided in one embodiment of the present invention.

[0024] Attached Figure 5 A schematic diagram of edge extraction results provided by an embodiment of the present invention.

[0025] Attached Figure 6 A schematic diagram of edge connectivity results provided by an embodiment of the present invention.

[0026] Attached Figure 7 A schematic diagram of the longest edge provided by an embodiment of the present invention.

[0027] Attached Figure 8 A schematic diagram of a fused connected image provided by an embodiment of the present invention.

[0028] Attached Fig. 9 A schematic diagram of a skeleton image provided by an embodiment of the present invention.

[0029] Attached Fig.10 A schematic diagram of a skeleton image expansion result provided by an embodiment of the present invention.

[0030] Attached Fig.11 A schematic diagram of the structure of a segmentation device provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0031] The present invention is not limited by the following embodiments, and specific implementation methods can be determined based on the technical solution of the present invention and actual conditions.

[0032] Those skilled in the art will understand that, unless otherwise stated, a "module" or "unit" in the embodiments of the present invention refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as a processing circuit or memory) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0033] In addition, the term “plurality” in the embodiments of the present invention refers to two or more than two, and “first” and “second” are used to distinguish descriptions and should not be understood as implying relative importance.

[0034] The present invention provides a skeleton-extraction-based X-ray weld pseudo-defect segmentation method, device, storage medium and electronic device. The skeleton-extraction-based X-ray weld pseudo-defect segmentation device can be integrated in a computer device, which can be a server or a terminal or other device; it can also be executed by a terminal and a server together, and the above examples should not be understood as limiting the present invention.

[0035] The above-mentioned terminal may include a mobile phone, a wearable smart device, a tablet computer, a notebook computer, a personal computer (PC), and a vehicle-mounted computer, etc., and the present invention does not limit this. The present invention does not limit the number of terminal devices.

[0036] The above-mentioned server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, as well as big data and artificial intelligence platforms. The present invention does not impose any restrictions on this.

[0037] For example, the computer equipment extracts the longest edge in the X-ray weld film to be segmented and expands it to obtain a fused connected image, extracts the skeleton of the fused connected image, and expands the extracted skeleton image. Thus, through edge extraction and skeleton extraction, interferences such as character logos of numbers and letters are removed, and long strip-shaped pseudo-defects such as hair and fiber strands are accurately located, and accurate segmentation of pseudo-defects is completed, effectively solving the problem of segmentation difficulties caused by overlapping interferences such as pseudo-defects and characters, and compared with the existing manual segmentation of pseudo-defects, the efficiency and accuracy are high.

[0038] Based on this, the technical solution of the present invention will be introduced and explained in combination with several examples below.

[0039] Embodiment 1: As attached Figure 1 As shown, an embodiment of the present invention discloses an X-ray weld pseudo defect segmentation method based on skeleton extraction, comprising: Step S110, extracting the longest edge in the X-ray weld film to be segmented and expanding it to obtain a fused connected image.

[0040] In addition to pseudo defects, the X-ray weld film to be segmented may also contain interference such as some digital and alphabetic character identifiers in the base material area. In order to concentrate on processing the pseudo defect-related parts, step S110 performs edge extraction on the X-ray weld film to be segmented. Usually, pseudo defects are mostly hairs, fiber strands and other pseudo defects, which are mostly long strips. Such pseudo defects have the longest edges. Therefore, this step extracts the longest edge in the X-ray weld film to be segmented and expands it to obtain a fused connected image.

[0041] Step S120 , extracting the skeleton of the fused connected image and dilating the extracted skeleton image.

[0042] Since pseudo-defects may also pass through interference such as character logos, the extracted main contour may still contain some characters, and the pseudo-defects cannot be separated from the characters by conventional segmentation methods. Therefore, step S120 uses skeleton extraction to determine the skeleton image of the pseudo-defect to achieve X-ray weld pseudo-defect segmentation, wherein the skeleton image of the pseudo-defect is extracted based on the central axis of the image area in the fused connected image, and the edges of the central axis are gradually peeled off until only the central axis remains.

[0043] The embodiment of the present invention discloses an X-ray weld pseudo-defect segmentation method based on skeleton extraction. Through edge extraction and skeleton extraction, interference such as character identification of numbers and letters is removed, and accurate positioning of long strip pseudo-defects such as hair and fiber filaments is achieved, and precise segmentation of pseudo-defects is completed, effectively solving the problem of segmentation difficulty caused by overlapping of pseudo-defects and interference such as characters, and compared with the existing manual segmentation method of pseudo-defects, the method is efficient and accurate.

[0044] Embodiment 2: As attached Figure 2 As shown, the embodiment of the present invention is a further optimization of the above embodiment, wherein the longest edge in the X-ray weld film to be segmented is extracted and expanded to obtain a fused connected image, including: Step S210, using an edge detection operator to extract edges from the X-ray weld film to be segmented.

[0045] The edge detection operator in this step is selected as needed. In this embodiment, the Canny edge detection operator can be selected but is not limited to be used for edge extraction. The Canny edge detection operator has excellent edge positioning accuracy and anti-noise ability. The edge extraction process using the Canny edge detection operator includes: Firstly, the X-ray weld film to be segmented is processed by Gaussian filtering to smooth the X-ray weld film to be segmented while retaining the key information of the edge; Then, the intensity and direction of the edge are determined by calculating the gradients of the X-ray weld film to be segmented in the horizontal and vertical directions; Then, the non-maximum suppression method is used to process the gradient amplitude in the segmented X-ray weld film to retain the pixels with the maximum local gradient and suppress non-edge pixels. Subsequently, the pixels in the X-ray weld film to be segmented are divided into strong edge pixels, weak edge pixels and non-edge pixels according to the set high threshold and low threshold; Finally, the strong edge pixels are used to connect the adjacent weak edge pixels to form a complete edge.

[0046] The calculation formula of the above Canny edge detection operator is shown as follows: in, is the input image, is the edge image output after detection by the Canny edge detection operator. and They are the high threshold and low threshold of the Canny operator respectively. The high threshold and the low threshold are reasonably selected according to the actual situation, and the purpose is to effectively suppress noise while retaining the key edges in the image. In this embodiment, 10 can be selected as the low threshold and 120 as the high threshold. The low threshold is set to 10 to allow weaker edge information to be detected, ensuring that the subtle edge structure in the image is not missed, but a lower threshold may also bring more noise, so the high threshold is set to 120. The high threshold helps the algorithm distinguish strong edges from weak edges caused by noise, thereby retaining only the most significant edges.

[0047] Step S220, edge connectivity is performed on the mentioned edges, specifically as follows: in, is the image after edge connection; and are the structural element and the number of iterations in the expansion respectively.

[0048] Since the extracted edges may not be connected, the edges mentioned need to be connected in this embodiment. The specific connection process is achieved by expanding the edges. The size, type and number of iterations of the structural element during expansion are selected according to the size of the fractures in the edges, and there is no restriction here.

[0049] Step S230, selecting the longest edge from the multiple edges after edge connection and expanding it again to obtain a fused connected image, as follows: in, is a fused connected image; , () is a function for finding the longest edge in the edge connectivity result; and are the structural element and the number of iterations in the expansion respectively.

[0050] In addition to pseudo defects, the X-ray weld film to be segmented may also contain interference such as some numbers and letters in the base material area, so there will be multiple edges after the edges are connected. Usually, pseudo defects are mostly hair, fiber and other pseudo defects, which are mostly long strips. This type of pseudo defect has the longest edge. Therefore, in this embodiment, the longest edge is selected as the target edge to concentrate on processing the longest edge of the pseudo defect related part.

[0051] Furthermore, since the pseudo-defect may also pass through the character logo, the longest edge may include the edge of the pseudo-defect and the edge of the character logo connected to it. The pseudo-defect and the character logo cannot be separated by conventional segmentation means. Therefore, in order to facilitate subsequent accurate segmentation, this embodiment expands the longest edge again, merges the edge of the pseudo-defect and the edge of the character logo connected to it, and forms a single fused connected image. The size of the rectangular structural element and the number of iterations selected during the second expansion are selected according to the actual situation and are not limited here.

[0052] Embodiment 3: As attached Figure 3 As shown, the embodiment of the present invention is a further optimization of the above embodiment, wherein the skeleton of the fused connected image is extracted, and the extracted skeleton image is expanded, including: Step S310, based on The function determines the central axis of the image area in the fused connected image and extracts the skeleton image, as follows: in, is a skeleton image; To fuse connected images.

[0053] In step S310 of this embodiment, skeleton extraction technology is used to locate and segment pseudo-defects in the presence of interference such as character identification. Skeleton extraction technology is a morphological operation that can accurately represent the basic shape of an object without changing its overall structure. It is achieved by gradually peeling off the edges of the object until only the central axis remains.

[0054] Step S320, dilating the extracted skeleton image, specifically as follows: in, It is the result of dilating the skeleton image; and are the structural element and number of iterations of the expansion respectively.

[0055] In order to restore the approximate width and shape of the original pseudo-defect, this embodiment performs a dilation operation on the extracted skeleton image. The pixels around the skeleton (i.e., the central axis) are increased by dilation, and it is gradually expanded to a size close to the original pseudo-defect. The structural element, type, and number of iterations of the dilation are selected according to actual conditions and are not limited here.

[0056] Embodiment 4: This embodiment discloses an example of an X-ray weld film to be segmented, as shown in the attached Figure 4 As shown, the pseudo defect segmentation of the X-ray weld is performed by the skeleton extraction-based X-ray weld pseudo defect segmentation method disclosed in the present invention, as described in detail as follows: (1) The Canny edge detection operator is used to extract the edges of the segmented X-ray weld film. The extraction results are shown in the attached figure. Figure 5 shown.

[0057] (2) Select a 2×2 rectangular structure element to expand and connect the edge extraction results. The connection results are shown in the attached figure. Figure 6 shown.

[0058] (3) Select the longest edge among the multiple edges after the edges are connected (such as the attached Figure 7 As shown in the figure, the fused connected image is expanded again to obtain the fused connected image. Figure 8 As shown, a 4×4 rectangular structure element is selected when expanding again, and the iteration is set to 11 times.

[0059] (4) Based on The function determines the central axis of the image area in the fused connected image and extracts the skeleton image. The skeleton image is shown in the attached figure. Fig. 9 shown.

[0060] (5) Select to dilate the extracted skeleton image. When dilating, you can choose a 4×4 rectangular structure element. The iteration is set to 11 times. The final skeleton image dilation result is as shown in the attached figure. Fig.10 shown.

[0061] Will attach Fig.10 With attached Figure 4 By comparison, it can be seen that the method disclosed in the present invention can eliminate interference such as character identification, locate pseudo defects, and achieve accurate segmentation of pseudo defects.

[0062] Embodiment 5: As attached Fig.11 As shown, an embodiment of the present invention discloses an X-ray weld pseudo defect segmentation device based on skeleton extraction, comprising: The edge extraction unit extracts the longest edge in the X-ray weld film to be segmented and expands it to obtain a fused connected image; The skeleton extraction unit extracts the skeleton of the fused connected image and dilates the extracted skeleton image.

[0063] Wherein, the edge extraction unit includes: The edge detection module uses the edge detection operator to extract the edge of the X-ray weld film to be segmented; The connectivity module connects the mentioned edges as follows: in, is the image behind the edge; and are the structural elements and iteration numbers of the expansion respectively; The fusion module selects the longest edge among the multiple edges after edge connection and expands it again to obtain a fused connected image, as follows: in, is a fused connected image; , A function for finding the longest edge in the edge connectivity result; and are the structural element and the number of iterations in the expansion respectively.

[0064] The skeleton extraction unit includes: Skeleton image extraction module, based on The function determines the central axis of the image area in the fused connected image and extracts the skeleton image, as follows: in, is a skeleton image; To fuse connected images; The skeleton image expansion module expands the extracted skeleton image as follows: in, It is the result of dilating the skeleton image; and are the structural element and the number of iterations in the expansion respectively.

[0065] Embodiment 6: The embodiment of the present invention discloses a storage medium, on which a computer program readable by a computer is stored, and the computer program is configured to execute an X-ray weld pseudo-defect segmentation method based on skeleton extraction when running.

[0066] The above storage medium may include, but is not limited to: a USB flash drive, a read-only memory, a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0067] Embodiment 7: The embodiment of the present invention discloses an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the computer program is loaded and executed by the processor to implement an X-ray weld pseudo-defect segmentation method based on skeleton extraction.

[0068] The processor may be a central processing unit (CPU), a general purpose processor, a digital signal processor (DSP), an ASIC, an FPGA or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of the present invention. It may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like. The memory may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory, a mobile hard disk, a magnetic disk or an optical disk.

[0069] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The solutions in the embodiments of the present invention may be implemented in various computer languages, for example, object-oriented programming language Java and interpreted scripting language JavaScript, etc.

[0070] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes 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 a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0071] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0072] The above content is only a specific implementation mode of the present invention, which has strong adaptability and implementation effect, but the protection scope of the present invention is not limited to this. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered within the protection scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope covered by the present invention.

Claims

1. A skeleton extraction-based X-ray weld pseudo-defect segmentation method, characterized in that: include: The longest edge in the X-ray weld film to be segmented is extracted and expanded to obtain a fused connected image, including: The edge detection operator is used to extract the edge of the X-ray weld film to be segmented; The mentioned edges are connected as follows: in, is the image after edge connection; and are the structural element and the number of iterations in the expansion, respectively; Select the longest edge from the multiple edges after edge connection and expand it again to obtain a fused connected image, as follows: in, is a fused connected image; , A function for finding the longest edge in the edge connectivity result; and are the structural element and the number of iterations in the expansion, respectively; The skeleton of the fused connected image is extracted, and the extracted skeleton image is expanded.

2. The X-ray weld pseudo defect segmentation method based on skeleton extraction according to claim 1 is characterized in that: The structure element set during the second dilation is a 4×4 rectangular structure element, and the number of iterations is 11.

3. The X-ray weld pseudo defect segmentation method based on skeleton extraction according to claim 1 or 2, characterized in that: The edge detection operator is a Canny edge detection operator.

4. The X-ray weld pseudo defect segmentation method based on skeleton extraction according to claim 1 or 2, characterized in that: The step of extracting a skeleton from the fused connected image and dilating the extracted skeleton image includes: based on The function determines the central axis of the image area in the fused connected image and extracts the skeleton image, as follows: in, is a skeleton image; To fuse connected images; The extracted skeleton image is expanded as follows: in, It is the result of dilating the skeleton image; and are the structural element and the number of iterations in the expansion respectively.

5. An X-ray weld pseudo defect segmentation device based on skeleton extraction using the method as described in any one of claims 1 to 4, characterized in that: include: The edge extraction unit extracts the longest edge in the X-ray weld film to be segmented and expands it to obtain a fused connected image, including: The edge detection module uses the edge detection operator to extract the edge of the X-ray weld film to be segmented; The connectivity module connects the mentioned edges as follows: in, is the image behind the edge; and are the structural elements and iteration numbers of the expansion respectively; The fusion module selects the longest edge among the multiple edges after edge connection and expands it again to obtain a fused connected image, as follows: in, is a fused connected image; , A function for finding the longest edge in the edge connectivity result; and are the structural element and the number of iterations in the expansion, respectively; The skeleton extraction unit extracts the skeleton of the fused connected image and dilates the extracted skeleton image.

6. The X-ray weld pseudo defect segmentation device based on skeleton extraction according to claim 5 is characterized in that: The skeleton extraction unit comprises: Skeleton image extraction module, based on The function determines the central axis of the image area in the fused connected image and extracts the skeleton image, as follows: in, is a skeleton image; To fuse connected images; The skeleton image expansion module expands the extracted skeleton image as follows: in, It is the result of dilating the skeleton image; and are the structural element and the number of iterations in the expansion respectively.

7. An electronic device, characterized in that: The method comprises a processor and a memory, wherein a computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the steps in the method according to any one of claims 1 to 4.

8. A storage medium, characterized in that: The storage medium stores a computer program that can be read by a computer, and the computer program is configured to execute the steps of the method according to any one of claims 1 to 4 when running.

Images