Defect identification method and system for copper bar production

Abstract

The invention discloses a defect identification method and system for copper bar production, and the method comprises the steps: carrying out the comparison of all surface subimages in a certain surface subimage set according to a preset image comparison strategy, and selecting a target surface subimage from the certain surface subimage set based on a comparison result; constructing a target association relationship between the target surface sub-image and other surface sub-images in a certain surface sub-image set; and inputting the target surface sub-image into a preset defect recognition model, outputting a target defect recognition result corresponding to the target surface sub-image by the defect recognition model, and determining the target defect recognition result as a defect recognition result of other surface sub-images by adopting the target association relationship. The to-be-detected copper bars of the same type can be automatically divided into the same set as much as possible, the target surface sub-image is selected from the set, and the defects of the target surface sub-image are identified, so that the copper bar detection efficiency can be effectively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of copper rod detection, and particularly relates to a method and system for defect identification in copper rod production. Background Art

[0002] With the rapid development of industry, China has become the world's largest copper consumer. In high-tech fields such as automobiles and aviation, the quality and stability requirements of copper rods are relatively high, and a tiny defect may have a huge impact on product quality. During the production and processing of copper rods, various types of defects will occur, such as defects like bubbles, scratches, abrasions, and skinning. Therefore, defect detection is an essential part of the process.

[0003] In the prior art, it is necessary to identify each copper rod to be detected one by one, which will increase the amount of identification and thus lead to a decrease in the efficiency of defect detection. Summary of the Invention

[0004] The present invention provides a method and system for defect identification in copper rod production to solve the technical problem that each copper rod to be detected needs to be identified one by one, resulting in low identification efficiency.

[0005] In a first aspect, the present invention provides a method for defect identification in copper rod production, including:

[0006] Obtain surface images of at least one copper rod to be detected, and segment the surface images according to a preset image processing strategy to obtain at least one surface sub-image;

[0007] Classify the at least one surface sub-image based on the number of copper rods to be detected in each surface sub-image to obtain at least one surface sub-image set;

[0008] Compare each surface sub-image in a certain surface sub-image set according to a preset image comparison strategy, and select a target surface sub-image from the certain surface sub-image set based on the comparison result;

[0009] Construct a target association relationship between the target surface sub-image and other surface sub-images in the certain surface sub-image set, where the other surface sub-images are at least one surface sub-image in the certain surface sub-image set excluding the target surface sub-image;

[0010] Input the target surface sub-image into a preset defect identification model, the defect identification model outputs a target defect identification result corresponding to the target surface sub-image, and use the target association relationship to determine the defect identification result of the other surface sub-images as the target defect identification result.

[0011] In a second aspect, the present invention provides a defect recognition system for copper rod production, including:

[0012] A segmentation module configured to obtain a surface image of at least one copper rod to be detected, and segment the surface image according to a preset image processing strategy to obtain at least one surface sub-image;

[0013] A classification module configured to classify the at least one surface sub-image based on the number of copper rods to be detected in each surface sub-image to obtain at least one surface sub-image set;

[0014] A comparison module configured to compare each surface sub-image in a certain surface sub-image set according to a preset image comparison strategy, and select a target surface sub-image from the certain surface sub-image set based on the comparison result;

[0015] A construction module configured to construct a target association relationship between the target surface sub-image and other surface sub-images in the certain surface sub-image set, where the other surface sub-images are at least one surface sub-image in the certain surface sub-image set excluding the target surface sub-image;

[0016] A determination module configured to input the target surface sub-image into a preset defect recognition model, the defect recognition model outputs a target defect recognition result corresponding to the target surface sub-image, and uses the target association relationship to determine the target defect recognition result as the defect recognition result of the other surface sub-images.

[0017] In a third aspect, an electronic device is provided, which includes: at least one processor, and a memory communicatively connected to the at least one processor, where the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the steps of the defect recognition method for copper rod production according to any embodiment of the present invention.

[0018] In a fourth aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the program instructions are executed by a processor, the processor is enabled to execute the steps of the defect recognition method for copper rod production according to any embodiment of the present invention.

[0019] The defect recognition method and system for copper bar production of the present application compare each surface sub-image in a certain set of surface sub-images according to a preset image comparison strategy, and select a target surface sub-image from the set of surface sub-images based on the comparison result; construct a target association relationship between the target surface sub-image and other surface sub-images in the set of surface sub-images; input the target surface sub-image into a preset defect recognition model, and the defect recognition model outputs a target defect recognition result corresponding to the target surface sub-image, and uses the target association relationship to determine the target defect recognition result as the defect recognition result of other surface sub-images. In this way, it is possible to automatically divide the copper bars to be detected of the same type into the same set as much as possible, select a target surface sub-image for the set, and identify the defects of the target surface sub-image, which can effectively improve the efficiency of copper bar detection. Brief Description of the Drawings

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0021] Figure 1 It is a flowchart of a defect recognition method for copper bar production provided by an embodiment of the present invention;

[0022] Figure 2 It is a structural block diagram of a defect recognition system for copper bar production provided by an embodiment of the present invention;

[0023] Figure 3 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. Detailed Embodiments

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0025] Please refer to Figure 1 , which shows a flowchart of a defect recognition method for copper bar production of the present application.

[0026] As Figure 1 shown, the defect recognition method for copper bar production specifically includes the following steps:

[0027] Step S101: Obtain the surface images of at least one copper rod to be detected, and segment the surface images according to a preset image processing strategy to obtain at least one surface sub-image.

[0028] In this step, a camera is used to capture the surface images of each copper rod to be detected within a region, and then the surface images are segmented according to a preset image processing strategy to obtain at least one surface sub-image.

[0029] It should be noted that segmenting the surface images according to a preset image processing strategy to obtain at least one surface sub-image includes:

[0030] Input the surface images into a preset image recognition model, label the contours of each copper rod to be detected in the surface images according to the recognition boxes in the image recognition model to obtain at least one labeled box region; slide a sliding window with a preset size on the edge of at least one labeled box region, and define the region covered by the sliding window as the background region; determine whether the first background region coincides with the second background region, where the first background region is the background region corresponding to the first labeled box region and the second background region is the background region corresponding to the second labeled box region; if the first background region coincides with the second background region, then segment the first labeled box region, the first background region, the second labeled box region, and the second background region from the surface images to obtain a surface sub-image.

[0031] Specifically, the image recognition model can be obtained by training a neural network.

[0032] Step S102: Classify the at least one surface sub-image based on the number of copper rods to be detected in each surface sub-image to obtain at least one surface sub-image set.

[0033] In this step, obtain the number of copper rods to be detected in each surface sub-image; divide the surface sub-images with the same number of copper rods to be detected into the same surface sub-image set to obtain at least one surface sub-image set.

[0034] Step S103: Compare each surface sub-image in a certain surface sub-image set according to a preset image comparison strategy, and select a target surface sub-image from the certain surface sub-image set based on the comparison results.

[0035] In this step, since the positions of the copper rods to be detected in the segmented surface sub-images are regular, it is necessary to screen out the copper rods to be detected at the same position, which is convenient for subsequent image comparison. Specifically:

[0036] Input the first surface sub-image and the second surface sub-image into a preset two-dimensional coordinate system, and obtain the coordinate information of the first annotation box corresponding to the first surface sub-image and the coordinate information of the second annotation box corresponding to the second surface sub-image according to the two-dimensional coordinate system, where the first surface sub-image and the second surface sub-image are both surface sub-images in a certain surface sub-image set;

[0037] Determine whether the minimum coordinate difference between the first annotation box coordinate information and the second annotation box coordinate information is greater than a preset threshold, where the minimum coordinate difference is the minimum difference between each first annotation box coordinate in the first annotation box coordinate information and each second annotation box coordinate in the second annotation box coordinate information;

[0038] If the minimum coordinate difference is not greater than the preset threshold, then divide the first surface sub-image and the second surface sub-image into the same surface sub-image subset;

[0039] If the minimum coordinate difference is greater than the preset threshold, then divide the first surface sub-image and the second surface sub-image into different surface sub-image subsets respectively;

[0040] Obtain the similarity between each surface sub-image in a certain surface sub-image subset, and select a certain surface sub-image with the smallest similarity to other surface sub-images as a certain target surface sub-image of a certain surface sub-image subset.

[0041] Step S104, construct a target association relationship between the target surface sub-image and other surface sub-images in the certain surface sub-image set, where the other surface sub-images are at least one surface sub-image in the certain surface sub-image set excluding the target surface sub-image.

[0042] In this step, associate a certain target surface sub-image with other surface sub-images in the certain surface sub-image subset to obtain a certain target association relationship.

[0043] Step S105, input the target surface sub-image into a preset defect recognition model, the defect recognition model outputs a target defect recognition result corresponding to the target surface sub-image, and use the target association relationship to determine the target defect recognition result as the defect recognition result of the other surface sub-images.

[0044] In this step, determine the target defect recognition result as the defect recognition result of the other surface sub-images according to a certain target association relationship. Specifically, the defect recognition model can be a VGGNet network model, and this model is obtained through machine learning using multiple groups of training images and detection images.

[0045] In summary, the method of the present application compares each surface sub-image in a set of surface sub-images according to a preset image comparison strategy, and selects a target surface sub-image from the set of surface sub-images based on the comparison result; constructs a target association relationship between the target surface sub-image and other surface sub-images in the set of surface sub-images; inputs the target surface sub-image into a preset defect recognition model, and the defect recognition model outputs a target defect recognition result corresponding to the target surface sub-image, and uses the target association relationship to determine the target defect recognition result as the defect recognition result of other surface sub-images. In this way, it is possible to automatically divide the copper rods to be detected of the same type into the same set as much as possible, select a target surface sub-image from this set, and identify the defects of the target surface sub-image, which can effectively improve the efficiency of copper rod detection.

[0046] Please refer to Figure 2 , which shows a structural block diagram of a defect recognition system for copper rod production according to the present application.

[0047] As Figure 2 shown, the defect recognition system 200 includes a segmentation module 210, a classification module 220, a comparison module 230, a construction module 240, and a determination module 250.

[0048] Among them, the segmentation module 210 is configured to obtain a surface image of at least one copper rod to be detected, and segment the surface image according to a preset image processing strategy to obtain at least one surface sub-image; the classification module 220 is configured to classify the at least one surface sub-image based on the number of copper rods to be detected in each surface sub-image to obtain at least one set of surface sub-images; the comparison module 230 is configured to compare each surface sub-image in a set of surface sub-images according to a preset image comparison strategy, and select a target surface sub-image from the set of surface sub-images based on the comparison result; the construction module 240 is configured to construct a target association relationship between the target surface sub-image and other surface sub-images in the set of surface sub-images, where the other surface sub-images are at least one surface sub-image in the set of surface sub-images excluding the target surface sub-image; the determination module 250 is configured to input the target surface sub-image into a preset defect recognition model, and the defect recognition model outputs a target defect recognition result corresponding to the target surface sub-image, and uses the target association relationship to determine the target defect recognition result as the defect recognition result of the other surface sub-images.

[0049] It should be understood that Figure 2 the modules described in Figure 1 correspond to the respective steps in the method described in referenceFigure 2 The various modules in

[0050] The energy module, such as the judgment module, can also be implemented by a processor, which will not be elaborated here.

[0051] In some other embodiments, the embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored. When the program instructions are executed by a processor, the processor is made to execute the defect recognition method for copper bar production in any of the above method embodiments;

[0052] As an implementation manner, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are set as:

[0053] Obtain surface images of at least one copper bar to be detected, and segment the surface images according to a preset image processing strategy to obtain at least one surface sub-image;

[0054] Classify the at least one surface sub-image based on the number of copper bars to be detected in each surface sub-image to obtain at least one surface sub-image set;

[0055] Compare each surface sub-image in a certain surface sub-image set according to a preset image comparison strategy, and select a target surface sub-image from the certain surface sub-image set based on the comparison result;

[0056] Construct a target association relationship between the target surface sub-image and other surface sub-images in the certain surface sub-image set, where the other surface sub-images are at least one surface sub-image in the certain surface sub-image set excluding the target surface sub-image;

[0057] Input the target surface sub-image into a preset defect recognition model. The defect recognition model outputs a target defect recognition result corresponding to the target surface sub-image, and uses the target association relationship to determine the target defect recognition result as the defect recognition result of the other surface sub-images.

[0058] A computer-readable storage medium may include a storage program area and a storage data area. Among them, the storage program area may store an operating system and application programs required for at least one function; the storage data area may store data created according to the use of the defect identification system for copper rod production, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some embodiments, the computer-readable storage medium may optionally include a memory remotely set relative to the processor, and these remote memories may be connected to the defect identification system for copper rod production through a network. Examples of the above network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0059] Figure 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention, as Figure 3 shown, the device includes: a processor 310 and a memory 320. The electronic device may further include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330, and the output device 340 may be connected through a bus or other means, Figure 3 and taking the connection through the bus as an example. The memory 320 is the above-mentioned computer-readable storage medium. The processor 310 executes various functional applications and data processing of the server by running non-volatile software programs, instructions, and modules stored in the memory 320, that is, implements the defect identification method for copper rod production in the above method embodiment. The input device 330 may receive input digital or character information, and generate key signal inputs related to user settings and function controls of the defect identification system for copper rod production. The output device 340 may include a display device such as a display screen.

[0060] The above electronic device may execute the method provided by the embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method. For technical details not described in detail in this embodiment, reference may be made to the method provided by the embodiment of the present invention.

[0061] As an implementation manner, the above electronic device is applied to a defect identification system for copper rod production, and is used for a client, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to:

[0062] Obtain surface images of at least one copper rod to be detected, and segment the surface images according to a preset image processing strategy to obtain at least one surface sub-image;

[0063] Classify the at least one surface sub - image based on the number of copper rods to be detected in each surface sub - image, and obtain at least one set of surface sub - images;

[0064] Compare each surface sub - image in a certain set of surface sub - images according to a preset image comparison strategy, and select a target surface sub - image from the certain set of surface sub - images based on the comparison result;

[0065] Construct a target association relationship between the target surface sub - image and other surface sub - images in the certain set of surface sub - images, where the other surface sub - images are at least one surface sub - image in the certain set of surface sub - images excluding the target surface sub - image;

[0066] Input the target surface sub - image into a preset defect recognition model. The defect recognition model outputs a target defect recognition result corresponding to the target surface sub - image, and use the target association relationship to determine the defect recognition result of the other surface sub - images as the defect recognition result of the target surface sub - image.

[0067] Through the description of the above - mentioned embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general - purpose hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above - mentioned technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer - readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiments.

[0068] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present invention.

Claims

1. A defect identification method for copper rod production, characterized in that, Including: Obtain a surface image of at least one copper rod to be detected, and segment the surface image according to a preset image processing strategy to obtain at least one surface sub-image; Classify the at least one surface sub-image based on the number of copper rods to be detected in each surface sub-image to obtain at least one set of surface sub-images; Compare each surface sub-image in a certain set of surface sub-images according to a preset image comparison strategy, and select a target surface sub-image from the certain set of surface sub-images based on the comparison result; Construct a target association relationship between the target surface sub-image and other surface sub-images in the certain set of surface sub-images, where the other surface sub-images are at least one surface sub-image in the certain set of surface sub-images excluding the target surface sub-image; Input the target surface sub-image into a preset defect recognition model, the defect recognition model outputs a target defect recognition result corresponding to the target surface sub-image, and use the target association relationship to determine the target defect recognition result as the defect recognition result of the other surface sub-images.

2. The defect identification method for copper rod production according to claim 1, wherein The segmenting the surface image according to a preset image processing strategy to obtain at least one surface sub-image includes: Input the surface image into a preset image recognition model, and label the contours of each copper rod to be detected in the surface image according to the recognition frame in the image recognition model to obtain at least one labeled frame area; Slide a sliding window with a preset size on the edge of the at least one labeled frame area, and define the area covered by the sliding window as the background area; Judge whether the first background area coincides with the second background area, where the first background area is the background area corresponding to the first labeled frame area, and the second background area is the background area corresponding to the second labeled frame area; If the first background area coincides with the second background area, then segment the first labeled frame area, the first background area, the second labeled frame area, and the second background area from the surface image to obtain a surface sub-image.

3. The defect recognition method for copper rod production according to claim 1, characterized in that, The classifying the at least one surface sub-image based on the number of copper rods to be detected in each surface sub-image to obtain at least one set of surface sub-images includes: Obtain the number of copper rods to be detected in each surface sub-image; Divide the surface sub-images with the same number of copper rods to be detected into the same set of surface sub-images to obtain at least one set of surface sub-images.

4. A defect identification method for copper rod production according to claim 1, characterized in that, The comparing each surface sub-image in a certain set of surface sub-images according to a preset image comparison strategy, and selecting a target surface sub-image from the certain set of surface sub-images based on the comparison result includes: Input the first surface sub-image and the second surface sub-image into a preset two-dimensional coordinate system, and obtain the first labeled frame coordinate information corresponding to the first surface sub-image and the second labeled frame coordinate information corresponding to the second surface sub-image according to the two-dimensional coordinate system, where the first surface sub-image and the second surface sub-image are both surface sub-images in a certain set of surface sub-images; Determine whether the minimum coordinate difference between the first annotation box coordinate information and the second annotation box coordinate information is greater than a preset threshold, where the minimum coordinate difference is the minimum difference between each first annotation box coordinate in the first annotation box coordinate information and each second annotation box coordinate in the second annotation box coordinate information; If the minimum coordinate difference is not greater than the preset threshold, divide the first surface sub-image and the second surface sub-image into the same surface sub-image subset; If the minimum coordinate difference is greater than the preset threshold, divide the first surface sub-image and the second surface sub-image into different surface sub-image subsets respectively; Obtain the similarity between each surface sub-image in a certain surface sub-image subset, and select a certain surface sub-image with the smallest similarity to other surface sub-images as a certain target surface sub-image of the certain surface sub-image subset.

5. A defect identification method for copper rod production according to claim 4, characterized in that The constructing the target association relationship between the target surface sub-image and other surface sub-images in the certain surface sub-image set includes: Associate the certain target surface sub-image with other surface sub-images in the certain surface sub-image subset to obtain a certain target association relationship.

6. A defect identification method for copper rod production according to claim 5, characterized in that, The adopting the target association relationship to determine the target defect recognition result as the defect recognition result of the other surface sub-images includes: Determine the target defect recognition result as the defect recognition result of the other surface sub-images according to the certain target association relationship.

7. A defect identification system for copper rod production, characterized in that, Includes: A segmentation module, configured to obtain the surface image of at least one copper rod to be detected, and segment the surface image according to a preset image processing strategy to obtain at least one surface sub-image; A classification module, configured to classify the at least one surface sub-image based on the number of copper rods to be detected in each surface sub-image to obtain at least one surface sub-image set; A comparison module, configured to compare each surface sub-image in a certain surface sub-image set according to a preset image comparison strategy, and select a target surface sub-image from the certain surface sub-image set based on the comparison result; A construction module, configured to construct a target association relationship between the target surface sub-image and other surface sub-images in the certain surface sub-image set, where the other surface sub-images are at least one surface sub-image in the certain surface sub-image set excluding the target surface sub-image; A determination module, configured to input the target surface sub-image into a preset defect recognition model, the defect recognition model outputs a target defect recognition result corresponding to the target surface sub-image, and adopt the target association relationship to determine the target defect recognition result as the defect recognition result of the other surface sub-images.

8. An electronic device, characterized in that, Includes: At least one processor, and a memory communicatively connected to the at least one processor, where the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method according to any one of claims 1 to 6.

Images