Ingot surface defect detection method, device and electronic equipment

Abstract

The application relates to a material ingot surface defect detection method and device and electronic equipment. The material ingot surface defect detection method comprises the following steps: inputting surface data of a to-be-detected material ingot into a first defect identification model to enable the first defect identification model to output a first defect identification result; when the first defect identification result indicates that the to-be-detected material ingot has a surface defect, inputting part of the surface data of the to-be-detected material ingot into a second defect identification model to enable the second defect identification model to output a second defect identification result; and generating quality grade data of the to-be-detected material ingot according to the first defect identification result or the second defect identification result. The detection method is accurate.

Description

Technical Field

[0001] This application relates to the field of semiconductor processing technology, and in particular to a method, apparatus and electronic equipment for detecting surface defects in ingots. Background Technology

[0002] In the semiconductor ingot processing, defects are inevitable in every step. If defective products flow into subsequent processes, it will waste production resources and costs. Therefore, timely detection of defective products is extremely important for reducing the cost of semiconductor processing.

[0003] In existing technologies, there are methods that determine semiconductor ingots as unqualified as soon as defects are detected on the surface and the ingots are moved to the unqualified area. However, there are cases where defects that are not part of the surface structure on the ingots are misjudged, requiring manual re-judgment from the unqualified area. This method has low accuracy and low efficiency. Summary of the Invention

[0004] This application provides a method, apparatus, and electronic device for detecting surface defects in semiconductor ingots. Based on the detection of semiconductor ingots, the detection process is simple, avoids misjudgment of non-surface structural defect types of ingots, can accurately determine the defect type, and ensures the accuracy and efficiency of ingot defect judgment, thereby at least partially solving the above-mentioned technical problems.

[0005] To achieve the above objectives, according to a first aspect of this application, a method for detecting surface defects in ingots is provided, comprising:

[0006] The surface data of the ingot to be tested is input into the first defect identification model so that the first defect identification model outputs the first defect identification result;

[0007] When the first defect identification result indicates that the test ingot has a surface defect, a portion of the surface data of the test ingot is input into the second defect identification model so that the second defect identification model outputs the second defect identification result;

[0008] The quality grade data of the ingot to be tested is generated based on the first defect identification result or the second defect identification result.

[0009] Optionally, when the first defect identification result indicates that the ingot under test has a surface defect, inputting a portion of the surface data of the ingot under test into the second defect identification model so that the second defect identification model outputs a second defect identification result includes:

[0010] When the first defect identification result indicates that the ingot to be tested has a surface defect, the surface defect and the type of the surface defect are selected.

[0011] Optionally, the detection method further includes:

[0012] Based on the quality grade data of the ingot to be tested, sorting instruction data for sorting the ingot to be tested is generated.

[0013] Optionally, the detection method further includes:

[0014] The original image of the ingot to be tested is acquired by an image acquisition device; the surface data is generated based on the original image of the ingot to be tested.

[0015] The surface data includes RGB images and three-dimensional point clouds.

[0016] Optionally, acquiring the original image of the ingot to be tested via an image acquisition device includes:

[0017] Control the ingot to be tested to move along a preset trajectory;

[0018] Acquire an image point set of several position points on the first surface of the ingot to be tested along the preset trajectory;

[0019] The original image is generated based on the image point set;

[0020] The first surface is positioned opposite to the image acquisition device.

[0021] Optionally, controlling the movement of the test ingot along a preset trajectory includes:

[0022] The test ingot is controlled to move from a preset position along a preset trajectory.

[0023] Optionally, acquiring the original image of the ingot to be tested via an image acquisition device further includes:

[0024] Obtain the height data of the first surface; wherein, the height data is the height information of different points on the first surface;

[0025] Based on the height data, the vertical distance between the first surface and the image acquisition device is adjusted as the ingot to be tested moves along a preset trajectory.

[0026] According to a second aspect of this application, a method for detecting surface defects in ingots is provided, implemented by an ingot surface defect detection system, the ingot surface defect detection system including a controller, the controller comprising:

[0027] The first defect identification module is used to acquire the surface data of the ingot to be tested and output the first defect identification result based on the surface data of the ingot to be tested;

[0028] The second defect identification module is used to acquire a portion of the surface data of the ingot under test when the first defect identification result indicates that the ingot under test has a surface defect, and output the second defect identification result based on the portion of the surface data;

[0029] The generation module is used to generate quality grade data of the ingot to be tested based on the first defect identification result or the second defect identification result.

[0030] Optionally, the ingot surface defect detection system further includes:

[0031] The moving component is communicatively connected to the controller.

[0032] The image acquisition device is communicatively connected to the controller;

[0033] The moving component is used to move the test ingot along a preset trajectory and adjust the vertical distance between the first surface and the image acquisition device during the movement of the test ingot along the preset trajectory.

[0034] The image acquisition device is used to acquire a set of image points at several positions on the first surface of the ingot to be tested during the process of the ingot moving along the preset trajectory, and to generate a surface image.

[0035] Optionally, the image acquisition device includes a vision sensor.

[0036] Optionally, the moving component is constructed as a three-dimensional linear motion module.

[0037] According to a third aspect of this application, a device for detecting surface defects in ingots is provided, comprising:

[0038] A memory on which computer programs are stored;

[0039] A processor is configured to execute the computer program in the memory to implement the ingot surface defect detection method as described above.

[0040] According to a fourth aspect of this application, a non-transitory computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the ingot surface defect detection method as described above.

[0041] According to a fifth aspect of this application, an electronic device is provided, characterized in that it includes the ingot surface defect detection device as described above.

[0042] The advantages of this application are: based on the detection of ingots, the detection process is simple, it avoids misjudging non-surface structural defect types of ingots, and can ensure the accuracy and efficiency of ingot defect judgment based on defect type.

[0043] More specifically, some embodiments of this application may produce the following specific beneficial effects:

[0044] The ingot surface detection method provided in this application involves inputting the surface data of the ingot to be tested into a first defect identification model and having the first defect identification model output a first defect identification result. When the first defect identification result indicates that the ingot to be tested has a surface defect, a portion of the surface data of the ingot to be tested is then input into a second defect identification model, which outputs a second defect identification result. Based on the first or second defect identification result, quality grade data of the ingot to be tested is generated. Based on the first identification result, the defect is further judged. By combining the universality of deep learning algorithms and the effectiveness of small-scale local point cloud processing algorithms, the accuracy of defect detection is improved, and misjudgment of non-surface structural defect types of ingots is avoided. The method can ensure the accuracy and efficiency of ingot defect judgment based on defect type.

[0045] The ingot surface defect detection method provided in this application obtains the height data of the first surface and adjusts the vertical distance between the first surface and the image acquisition device as the ingot to be tested moves along a preset trajectory based on the height data. This makes the original image acquired by the image acquisition device clearer and can adapt to the surface defect detection of ingots of different models and specifications under different lighting conditions, thereby improving manufacturing efficiency and production capacity, and further improving the accuracy of defect identification.

[0046] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] To gain a more complete understanding of this application and its beneficial effects, the following description will be provided in conjunction with the accompanying drawings, wherein the same reference numerals in the following description denote the same parts.

[0049] Figure 1 This is a flowchart of a method for detecting surface defects in ingots provided in an embodiment of this application;

[0050] Figure 2 This is a block diagram of a surface defect detection system for ingots provided in an embodiment of this application;

[0051] Figure 3This is a schematic diagram of the structure of a material ingot surface defect detection system provided in an embodiment of this application;

[0052] Figure 4 This is a schematic diagram of the connection structure between a stage and a moving component provided in an embodiment of this application;

[0053] Figure 5 This is a block diagram of a surface defect detection device for ingots provided in an embodiment of this application.

[0054] Explanation of reference numerals in the attached figures:

[0055] 200. Ingot Surface Defect Detection System;

[0056] 210. Controller;

[0057] 211. First defect identification module; 212. Second defect identification module; 213. Generation module;

[0058] 220. Moving part; 230. Image acquisition device; 240. Stage. Detailed Implementation

[0059] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the protection scope of this application.

[0060] Reference Figure 3 As shown, for ease of explanation, the corresponding figures use the orientations of up, down, left, right, front, and back to illustrate the relative positional relationships between the parts in this application. These should not be construed as limitations on absolute positions.

[0061] Furthermore, in this application, the first direction corresponds to the front-back direction, the second direction corresponds to the left-right direction, and the third direction corresponds to the up-down direction. Similarly, the first direction here indicates the front-back direction only for the convenience of introducing the specific embodiments of this application. There is no absolute correspondence between the first direction and the front-back direction. Likewise, there is no absolute correspondence between the second direction and the left-right direction, and between the third direction and the up-down direction.

[0062] The first, second, and third directions in this application are only for expressing relative positional relationships; they merely indicate approximate locations rather than absolute geometric relationships.

[0063] In the semiconductor ingot processing, defects are inevitable in every step. If defective products flow into subsequent processes, it will waste production resources and costs. Therefore, timely detection of defective products is extremely important for reducing the cost of semiconductor processing.

[0064] According to a first aspect of this application, embodiments of this application provide a method for detecting surface defects in ingots.

[0065] The surface defects in this application are surface structural defects, such as point defects, line defects, and surface defects on the surface of an ingot.

[0066] Please see Figure 1 , Figure 1 This is a flowchart illustrating a method for detecting surface defects in a material ingot, as provided in an embodiment of this application. The method may include the following steps:

[0067] Step S130: Input the surface data of the ingot to be tested into the first defect identification model so that the first defect identification model outputs the first defect identification result;

[0068] Step S140: When the first defect identification result indicates that there is a surface defect in the ingot to be tested, a portion of the surface data of the ingot to be tested is input into the second defect identification model so that the second defect identification model outputs the second defect identification result;

[0069] Step S150: Generate the quality grade data of the ingot to be tested based on the first defect identification result or the second defect identification result.

[0070] The first defect identification result is obtained by the first defect identification model receiving the surface data of the ingot to be tested and processing it accordingly.

[0071] Both the first and second defect identification models are based on deep learning neural network models. The training process of the identification model for surface defects of ingots is existing technology and will not be described in detail here.

[0072] It should be noted that in this application, the quality grade data is mapped to the corresponding data of surface defects of the ingot. The ingot to be tested is classified into grades, which may include qualified and unqualified, or other grade classifications, such as S (perfect), A (some defects, but not affecting the next process), NG (unqualified, unusable scrap). The specific settings are made according to the actual needs, which will not be elaborated here.

[0073] By inputting the surface data of the ingot to be tested into a first defect identification model and having the first defect identification model output a first defect identification result, when the first defect identification result indicates that the ingot to be tested has a surface defect, a portion of the surface data of the ingot to be tested is then input into a second defect identification model, which outputs a second defect identification result. Based on the first or second defect identification result, the quality grade data of the ingot to be tested is generated. Based on the first identification result, the defect is judged again. By combining the versatility of deep learning algorithms and the effectiveness of small-scale local point cloud processing algorithms, the accuracy of defect detection is improved, and misjudgment of non-surface structural defect types of ingots is avoided. The accuracy and efficiency of ingot defect judgment can be ensured based on the defect type.

[0074] Specifically, when the first defect identification result indicates that the ingot under test has no surface defects, the ingot under test is determined to be qualified; when the first defect identification result indicates that the ingot under test has surface defects, a portion of the surface data of the ingot under test with defects is input into the second defect identification model, and the second defect identification model is used to output the second defect identification result. If the second defect identification result indicates that the ingot under test has surface defects, the ingot under test is determined to be unqualified; if the second defect identification result indicates that the ingot under test has no surface defects, the ingot under test is determined to be qualified.

[0075] In some embodiments, step 140 may include the following specific steps:

[0076] Step 141: When the first defect identification result indicates that there is a surface defect in the ingot to be tested, select the surface defect and indicate the type of surface defect.

[0077] Step 142: By selecting surface defects and indicating the type of surface defects, the second defect identification model re-evaluates a portion of the surface data to improve detection efficiency.

[0078] In some embodiments, the detection method further includes:

[0079] Step 160: Generate sorting instruction data for sorting the ingots to be tested based on the quality grade data of the ingots to be tested.

[0080] In some embodiments, reference Figure 1 Before step 130, the detection method further includes:

[0081] Step 110: Acquire the original image of the ingot to be tested using the image acquisition device 230;

[0082] Step 120: Generate surface data based on the original image of the ingot to be tested;

[0083] The surface data in step 120 includes RGB images and 3D point clouds.

[0084] Due to the high reflectivity and high light absorption of the material surface of the ingot to be inspected, general sensors cannot effectively image the surface. Furthermore, surface defects of the ingot are usually in the form of minute cracks, holes, etc., which are difficult to observe and easy to ignore. Based on the principle of white light confocal imaging, this application selects the image acquisition unit 230 as a visual sensor to achieve ultra-high resolution imaging and high-precision detection values ​​of the surface of the ingot to be inspected, with measurement accuracy reaching the micrometer level.

[0085] Simultaneously, the visual sensor is triggered by the encoder pulse signal to acquire images.

[0086] The visual sensor outputs a digital model of the complete surface of the ingot to be inspected, and the output formats include RGB images and three-dimensional point clouds.

[0087] In some embodiments, step 120 specifically includes:

[0088] Step 121: Control the ingot to be tested to move along the preset trajectory;

[0089] Step 122: Acquire an image point set of several position points on the first surface of the ingot to be tested along a preset trajectory;

[0090] Step 123: Generate the original image based on the image point set;

[0091] The first surface is positioned opposite to the image acquisition unit 230.

[0092] In this application, since the size of the surface to be inspected of the ingot is larger than the FOV of the vision sensor, and in order to be applicable to the inspection of ingots of different models and specifications, step 121 controls the ingot to be inspected to move along a preset trajectory. The preset trajectory is a surface formed by the first direction and the second direction, which can comprehensively acquire the image point set of several position points on the first surface of the ingot to be inspected, and is applicable to the inspection of ingots of different models and specifications.

[0093] In some embodiments, step 121 includes the following steps:

[0094] Control the ingot to be tested to move from a preset position along a preset trajectory.

[0095] By controlling the ingot to be tested to move from a preset position along a preset trajectory, the acquisition of all-round surface data of the ingot to be tested is ensured, and loss is avoided.

[0096] In some embodiments, step 120 further includes the following steps:

[0097] Obtain the height data of the first surface.

[0098] Among them, the height data is the height information of different locations on the first surface;

[0099] Based on the height data, the vertical distance between the first surface and the image acquisition device 230 is adjusted as the ingot to be tested moves along the preset trajectory.

[0100] By adjusting the vertical distance between the first surface and the image acquisition unit 230, the first surface of the ingot to be measured is ensured to be within the depth of field of the image acquisition unit 230 to achieve clear imaging.

[0101] According to a second aspect of this application, a method for detecting surface defects in ingots is provided, implemented by an ingot surface defect detection system 200, with reference to... Figures 2 to 4 The ingot surface defect detection system 200 includes a controller 210, a moving part 220, an image acquisition device 230, and a stage 240.

[0102] The stage 240 is used to hold the ingot to be tested. The stage 240 is mounted on the moving part 220 and is equipped with a negative pressure device to adsorb the ingot using negative pressure. To avoid secondary damage such as scratches and compression to the surface of the ingot during loading and unloading caused by contact mechanical clamping devices, this application adopts a non-contact vacuum adsorption design to fix the ingot on the stage 240, avoiding the influence of external vibration on the detection accuracy.

[0103] Among them, reference Figure 2 The controller 210 includes a first defect identification module 211, a second defect identification module 212, and a generation module 213.

[0104] The first defect identification model in the first defect identification module 211 is used to acquire the surface data of the ingot to be tested and output the first defect identification result based on the surface data of the ingot to be tested;

[0105] The second defect identification module 212 is used to acquire a portion of the surface data of the ingot under test when the first defect identification result indicates that there is a surface defect in the ingot under test, and output the second defect identification result based on the portion of the surface data;

[0106] The generation module 213 is used to generate quality grade data of the ingot to be tested based on the first defect identification result or the second defect identification result.

[0107] The moving component 220 is used to move the test ingot along a preset trajectory and adjust the vertical distance between the first surface and the image acquisition device 230 during the movement of the test ingot along the preset trajectory. The moving component 220 is constructed as a three-dimensional moving linear module, which is communicatively connected to the controller 210. The three-dimensional motion platform composed of the three-dimensional moving linear module is connected to the stage 240. The test ingot is moved using the three-dimensional moving linear module, and the motion positioning accuracy reaches the micrometer level, realizing three-dimensional motion relative to the vision sensor and completing image acquisition in a large-scale three-dimensional space.

[0108] The image acquisition unit 230 is used to acquire a set of image points at several positions on the first surface of the ingot to be tested during the process of the ingot to be tested moving along a preset trajectory, and to generate a surface image.

[0109] According to a third aspect of this application, a device for detecting surface defects in ingots is provided, with reference to... Figure 5 As shown, the ingot surface defect detection device 300 may include a processing unit 301 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 302 or a program loaded from storage device 308 into random access memory (RAM) 303. The random access memory (RAM) 303 also stores various programs and data required for the operation of the ingot surface defect detection device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0110] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows the ingot surface defect detection device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although... Figure 5 A surface defect detection device 300 for ingots with various apparatuses is shown; however, it should be understood that it is not required to implement or possess all of the apparatuses shown. More or fewer apparatuses may be implemented alternatively. Figure 5 Each box shown can represent a device or multiple devices as needed.

[0111] In particular, according to some embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 309, or installed from storage device 308, or installed from ROM 302. When the computer program is executed by processing device 301, it performs the functions defined in the methods of some embodiments of this application.

[0112] It should be noted that the computer-readable medium in some embodiments of this application may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. The computer-readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof, and this application does not specifically limit its use. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0113] In some embodiments of this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In some embodiments of this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0114] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol, such as Hypertext Transfer Protocol (HTTP), and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include Local Area Networks (LANs), Wide Area Networks (WANs), the Internet (e.g., the Internet), and Advanced Developers Hands-On Conference (ADHO) networks, as well as any currently known or future-developed networks.

[0115] The aforementioned computer-readable medium may be included in the aforementioned ingot surface defect detection device, or it may exist independently without being assembled into the ingot surface defect detection device. The aforementioned computer-readable medium carries one or more programs that, when executed by the ingot surface defect detection device, cause the ingot surface defect detection device to: input the surface data of the ingot to be tested into a first defect identification model and have the first defect identification model output a first defect identification result; when the first defect identification result indicates that the ingot to be tested has a surface defect, input a portion of the surface data of the ingot to be tested into a second defect identification model, use the second defect identification model to output a second defect identification result, and generate quality grade data of the ingot to be tested based on the first defect identification result or the second defect identification result.

[0116] Computer program code for performing operations of some embodiments of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages. ― Languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages, include C and similar languages. Program code can execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network (including a local area network (LAN) or a wide area network (WAN)), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0117] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function.

[0118] It should also be noted that in some alternative implementations, the functions marked in the box may occur in a different order than those marked in the attached figures.

[0119] For example, two consecutively represented blocks can actually be executed in substantially parallel order, and sometimes they can be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, as well as combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified functions or operations, or using a combination of dedicated hardware and computer instructions.

[0120] The units described in some embodiments of this application can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including a data acquisition module and a driver module. The names of these units do not necessarily limit the specific unit; for example, a first defect identification module may also be described as "for acquiring surface data of the ingot to be tested and outputting a first defect identification result based on the surface data of the ingot to be tested."

[0121] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Array (FPGA), Application Specific Integrated Circuit (ASIC), Application Specific Standard Parts (ASSP), System on Chips (SOC), Complex Programmable Logic Device (CPLD), and so on.

[0122] According to a fourth aspect of this application, a non-transitory computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the ingot surface defect detection method as described above.

[0123] This non-transitory computer-readable storage medium has all the beneficial effects of the above-described method for detecting surface defects in ingots, which will not be elaborated upon here.

[0124] According to a fifth aspect of this application, embodiments of this application also provide an electronic device, including the ingot surface defect detection device described above. This electronic device possesses all the beneficial effects of the above-described ingot surface defect detection method, which will not be repeated here.

[0125] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0126] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0127] The embodiments, implementation methods, and related technical features of this application can be combined and substituted for each other without conflict.

[0128] The above are merely preferred embodiments of this application and are not intended to limit this application in any way. Although the descriptions of each embodiment in this application have different focuses, and parts not described in detail in a certain embodiment can be referred to the relevant descriptions of other embodiments, any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of this application without departing from the content of the technical solution of this application shall still fall within the scope of the technical solution of this application.

Claims

1. A method for detecting surface defects in ingots, characterized in that, include: Control the ingot to be tested to move from a preset position along a preset trajectory; Acquire the height data of the first surface, and adjust the vertical distance between the first surface and the image acquisition device according to the height data as the ingot to be tested moves along the preset trajectory. Using a vision sensor based on the principle of white light confocal focus, an image point set of several position points on the first surface of the ingot to be tested is acquired along the preset trajectory; an original image is generated based on the image point set, and surface data is generated based on the original image of the ingot to be tested. The height data refers to the height information of different points on the first surface, the first surface is set relative to the image acquisition device, and the surface data includes RGB images and three-dimensional point clouds. The surface data of the ingot to be tested is input into the first defect identification model so that the first defect identification model outputs the first defect identification result; When the first defect identification result indicates that the test ingot has a surface defect, a portion of the surface data of the test ingot is input into the second defect identification model so that the second defect identification model outputs the second defect identification result; The quality grade data of the ingot to be tested is generated based on the first defect identification result or the second defect identification result.

2. The detection method according to claim 1, characterized in that, When the first defect identification result indicates that the ingot under test has a surface defect, a portion of the surface data of the ingot under test is input into the second defect identification model so that the second defect identification model outputs a second defect identification result. include: When the first defect identification result indicates that the ingot to be tested has a surface defect, the surface defect and the type of the surface defect are selected.

3. The detection method according to claim 1, characterized in that, The detection method further includes: Based on the quality grade data of the ingot to be tested, sorting instruction data for sorting the ingot to be tested is generated.

4. A method for detecting surface defects in ingots, implemented by an ingot surface defect detection system, characterized in that, The ingot surface defect detection system includes a controller, the controller comprising: The first defect identification module controls the ingot to be tested to move from a preset position along a preset trajectory; Acquire the height data of the first surface, and adjust the vertical distance between the first surface and the image acquisition device according to the height data as the ingot to be tested moves along the preset trajectory. Using a vision sensor based on the principle of white light confocal focus, an image point set of several position points on the first surface of the ingot to be tested is acquired along the preset trajectory; an original image is generated based on the image point set, and surface data is generated based on the original image of the ingot to be tested. The height data refers to the height information of different points on the first surface, the first surface is set relative to the image acquisition device, and the surface data includes RGB images and three-dimensional point clouds. Used to acquire surface data of the ingot to be tested and output a first defect identification result based on the surface data of the ingot to be tested; The second defect identification module is used to acquire a portion of the surface data of the ingot under test when the first defect identification result indicates that the ingot under test has a surface defect, and output the second defect identification result based on the portion of the surface data; The generation module is used to generate quality grade data of the ingot to be tested based on the first defect identification result or the second defect identification result.

5. The detection method according to claim 4, characterized in that, The ingot surface defect detection system also includes: The moving component is communicatively connected to the controller. The image acquisition device is communicatively connected to the controller; The moving component is used to move the test ingot along a preset trajectory and adjust the vertical distance between the first surface and the image acquisition device during the movement of the test ingot along the preset trajectory. The image acquisition device is used to acquire a set of image points at several positions on the first surface of the ingot to be tested during the process of the ingot moving along the preset trajectory, and to generate a surface image.

6. The detection method according to claim 5, characterized in that, The image acquisition device includes a vision sensor.

7. The detection method according to claim 6, characterized in that, The moving component is constructed as a three-dimensional linear moving module.

8. A device for detecting surface defects in ingots, characterized in that, include: A memory on which computer programs are stored; A processor for executing the computer program in the memory to implement the ingot surface defect detection method according to any one of claims 1 to 7.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the ingot surface defect detection method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, Includes the ingot surface defect detection device as described in claim 8.

Images