Silicon carbide wafer surface scratch detection equipment and method

Abstract

The invention discloses silicon carbide wafer surface scratch detection equipment and method, and relates to the field of semiconductor testing, the silicon carbide wafer surface scratch detection equipment comprises a slide holder, the slide holder is provided with regularly arranged support rods, the support rods are hollow structures, the height of the support rods extending out of the surface of the slide holder is adjustable, and one end of each support rod is connected with a vacuum pump through a vacuum pipe; the other end of the supporting rod is used for supporting the silicon carbide wafer. The slide holder is in transmission connection with the motion platform, and the motion platform controls the slide holder to move in two directions which are perpendicular to each other in the same plane. The optical detection module is arranged in the direction perpendicular to the same plane, the optical detection module comprises a light source, a camera and a reflective mirror, the wavelengths of light rays emitted by the light source are at least two, and the reflective mirror is used for reflecting the light rays of the light source to the surface of the silicon carbide wafer. And the light enters the camera after being reflected by the surface of the silicon carbide wafer. Scratches on the surface can be found as soon as possible, and the influence on the product quality is reduced.

Description

Technical Field

[0001] The present application relates to the field of semiconductor testing technology, and in particular to a silicon carbide wafer surface scratch detection device and method. Background Art

[0002] With the continuous development of science and technology, silicon carbide (SiC) wafers are widely used in new energy, aerospace and other fields due to their high efficiency, high temperature performance and superior electrical properties. However, due to various reasons that may occur during the production process, surface scratches have a serious impact on the performance and service life of silicon carbide wafers. How to detect these scratches early to avoid subsequent impact on product quality is particularly important. Summary of the invention

[0003] In view of the above problems, this application provides a silicon carbide wafer surface scratch detection equipment and method to achieve early detection of scratches and ensure product quality. The specific solution is as follows:

[0004] The first aspect of the present application provides a silicon carbide wafer surface scratch detection device, comprising:

[0005] A wafer stage, on which regularly arranged support rods are arranged, the support rods are hollow and extend out of the surface of the wafer stage to an adjustable height, one end of the support rods is connected to a vacuum pump through a vacuum tube, and the other end of the support rods is used to support the silicon carbide wafer;

[0006] A motion platform, the wafer stage and the motion platform are in driving connection, and the motion platform controls the wafer stage to move in two directions perpendicular to each other in the same plane; and

[0007] An optical detection module, wherein the optical detection module is arranged in a direction perpendicular to the same plane, and comprises: a light source, a camera and a reflector, wherein the light emitted by the light source has at least two wavelengths, and the reflector is used to reflect the light from the light source to the surface of the silicon carbide wafer, so that the light enters the camera after being reflected by the surface of the silicon carbide wafer.

[0008] In one possible implementation, the motion platform includes: a first base and a second base, the second base is connected to the first base via a first transmission part, the first transmission part enables the second base to move along a first direction on the first base, the stage is connected to the second base via a second transmission part, the second transmission part enables the stage to move along a second direction on the second base, and the projections of the first direction and the second direction on the first base are perpendicular to each other.

[0009] In a possible implementation, the silicon carbide wafer surface scratch detection equipment further includes: a conveying module, which is used to take out the silicon carbide wafer from the wafer box port and place it on the support rod of the wafer carrier.

[0010] In one possible implementation, the conveying module includes: a manipulator, an edge finder and a code reader, the manipulator is used to take out the silicon carbide wafer from the wafer box port, the edge finder is used to fix the placement position of the silicon carbide wafer on the wafer carrier, and the code reader is used to read the surface information of the silicon carbide wafer.

[0011] In a possible implementation, the light source includes: a first light source and a second light source, the first light source and the second light source are both annular structures, the second light source is nested inside the first light source, and the wavelengths of light emitted by the first light source and the second light source are different.

[0012] In a possible implementation, it also includes: a support frame, one end of which is fixed on the moving platform, and the optical detection module is arranged at the other end of the support frame, wherein the camera is arranged on the support frame at a position perpendicular to the moving platform, and the light source and the reflector are symmetrically arranged on both sides of the camera.

[0013] The second aspect of the present application provides a method for detecting scratches on the surface of a silicon carbide wafer, which is applied to the silicon carbide wafer surface scratch detection equipment as described in the first aspect above, comprising:

[0014] Control the support rod of the first part on the wafer stage to rise to a first height, the support rod of the second part to rise to a second height, and control the vacuum pump to work, wherein the first height is greater than the second height;

[0015] Controlling the motion platform to drive the wafer stage to move along a set trajectory, controlling the light source to emit light of a first wavelength, and controlling the camera to take pictures each time the wafer stage moves to a point on the set trajectory, so as to obtain a first surface image of the silicon carbide wafer;

[0016] After the slide stage has completed its movement according to the set trajectory, controlling the support rod of the first part to descend to the second height, and the support rod of the second part to rise to the first height;

[0017] Controlling the motion platform to drive the wafer stage so that each shielding area on the silicon carbide wafer moves to the position where the light is irradiated in sequence, and controlling the camera to take pictures when each shielding area moves to the position where the light is irradiated, to obtain a second surface image of the silicon carbide wafer, wherein the shielding area is a corresponding position of each support rod of the first part on the silicon carbide wafer;

[0018] The scratch recognition model is called to recognize the first surface image and the second surface image to obtain first scratch recognition result data.

[0019] In a possible implementation, after obtaining the first scratch identification result data, the method further includes:

[0020] Controlling the support rod of the second part to descend to the second height, and controlling the light source to emit light of a second wavelength, wherein the second wavelength is smaller than the first wavelength;

[0021] Controlling the motion platform to drive the wafer stage to move along the set trajectory, and controlling the camera to take pictures each time the wafer stage moves to a point on the set trajectory, so as to obtain a third surface image of the silicon carbide wafer;

[0022] Calling a scratch recognition model to recognize the third surface image to obtain second scratch recognition result data;

[0023] The scratch identification result data on both surfaces of the silicon carbide wafer are obtained according to the first scratch identification result data and the second scratch identification result data.

[0024] In a possible implementation, the scratch recognition model is based on a pre-trained feature extraction model and is trained using sample images of silicon carbide wafers annotated with scratch defects.

[0025] In a possible implementation, the method for detecting scratches on the surface of a silicon carbide wafer further includes:

[0026] During the training process of the feature extraction model, the anti-interference ability of the feature extraction model is tested and improved in combination with an adversarial training strategy.

[0027] The third aspect of the present application provides a silicon carbide wafer surface scratch detection device, comprising:

[0028] A first height adjustment module is used to control the support rod of the first part on the wafer stage to rise to a first height, the support rod of the second part to rise to a second height, and control the vacuum pump to work, wherein the first height is greater than the second height;

[0029] A first image acquisition module is used to control the motion platform to drive the wafer stage to move along a set trajectory, control the light source to emit light of a first wavelength, and control the camera to take a picture every time the wafer stage moves to a point on the set trajectory, so as to obtain a first surface image of the silicon carbide wafer;

[0030] A second height adjustment module is used to control the support rod of the first part to drop to the second height and the support rod of the second part to rise to the first height after the slide stage completes the movement according to the set trajectory;

[0031] a second image acquisition module, for controlling the motion platform to drive the wafer stage so that each shielded area on the silicon carbide wafer moves to the position where the light is irradiated in sequence, and controlling the camera to take pictures when each shielded area moves to the position where the light is irradiated, so as to obtain a second surface image of the silicon carbide wafer, wherein the shielded area is a corresponding position of each support rod of the first part on the silicon carbide wafer; and

[0032] The scratch recognition module is used to call the scratch recognition model to recognize the first surface image and the second surface image to obtain first scratch recognition result data.

[0033] A fourth aspect of the present application provides a computer program product, comprising computer-readable instructions, which, when executed on an electronic device, enables the electronic device to implement the method for detecting scratches on the surface of a silicon carbide wafer according to the second aspect or any implementation of the second aspect.

[0034] A fifth aspect of the present application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0035] The memory is used to store computer programs;

[0036] The processor is used to execute the computer program so that the electronic device can implement the method for detecting scratches on the surface of a silicon carbide wafer according to the second aspect or any implementation of the second aspect.

[0037] A sixth aspect of the present application provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement the method for detecting scratches on the surface of a silicon carbide wafer according to the second aspect or any implementation of the second aspect.

[0038] By means of the above technical scheme, the surface scratch detection equipment of silicon carbide wafer provided by the present application comprises: a wafer stage, on which regularly arranged support rods are arranged, the support rods are hollow structures and the height of the support rods extending out of the surface of the wafer stage is adjustable, one end of the support rods is connected to the vacuum pump through a vacuum tube, and the other end of the support rods is used to support the silicon carbide wafer. A motion platform, the wafer stage and the motion platform are transmission-connected, and the motion platform controls the movement of the wafer stage in two directions perpendicular to each other in the same plane. And, an optical detection module, which is arranged in a direction perpendicular to the same plane, wherein the optical detection module comprises: a light source, a camera and a reflector, the wavelength of the light emitted by the light source is at least two, and the reflector is used to reflect the light of the light source to the surface of the silicon carbide wafer, so that the light enters the camera after being reflected by the surface of the silicon carbide wafer. Subsequently, by identifying the image of the surface of the silicon carbide wafer obtained, it can be determined whether there is a scratch, so that the scratch on the surface can be found early, reducing the impact on product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0040] Figure 1 An internal structure diagram of a silicon carbide wafer surface scratch detection device provided in this application;

[0041] Figure 2 A structural diagram of a wafer stage provided in this application;

[0042] Figure 3 A structural diagram of the support rod provided for this application;

[0043] Figure 4 A flow chart of a method for detecting scratches on a silicon carbide wafer surface provided in this application;

[0044] Figure 5 A structural diagram of a silicon carbide wafer surface scratch detection device provided in this application;

[0045] Figure 6 A structural diagram of an electronic device provided in this application.

[0046] Reference numerals:

[0047] 1- wafer stage; 11- support rod; 12- vacuum hole; 1101- first support frame; 1102- second support frame; 1103- third support frame; 1104- fourth support frame; 1105- fifth support frame; 1106- sixth support frame; 2- vacuum pump; 3- vacuum tube; 4- motion platform; 41- first base; 42- second base; 43- first transmission part; 44- second transmission part; 5- optical detection module; 51- light source; 52- reflector; 53- camera; 6- support frame; 7- silicon carbide wafer. DETAILED DESCRIPTION

[0048] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0049] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0050] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0051] In order to improve the quality of silicon carbide wafers and reduce production costs, the surface scratch detection of traditional silicon carbide wafers is generally carried out on a microscope, which is cumbersome and inefficient, has great limitations, and can no longer meet the current testing requirements.

[0052] In order to solve the above problems, the present application provides a silicon carbide wafer surface scratch detection device. The silicon carbide wafer surface scratch detection device of the present application is described in detail below in conjunction with the accompanying drawings.

[0053] Reference Figure 1 As shown, the silicon carbide wafer surface scratch detection equipment may specifically include:

[0054] The wafer stage 1 and the wafer stage 2 are provided with regularly arranged support rods 11, the support rods 11 are of a hollow structure and the height thereof extending out of the surface of the wafer stage 1 is adjustable, one end of the support rods 11 is connected to the vacuum pump 2 through a vacuum tube 3, and the other end of the support rods 11 is used to support the silicon carbide wafer 7.

[0055] The moving platform 4 is connected to the slide stage 1 through a transmission mechanism, and the moving platform 4 controls the slide stage 1 to move in two directions perpendicular to each other in the same plane. And,

[0056] The optical detection module 5 is arranged in a direction perpendicular to the same plane. The optical detection module 5 includes: a light source 51, a camera 53 and a reflector 52. The light emitted by the light source 51 has at least two wavelengths. The reflector 52 is used to reflect the light from the light source to the surface of the silicon carbide wafer 7 so that the light enters the camera 53 after being reflected by the surface of the silicon carbide wafer 7.

[0057] Specifically, refer to Figure 1 As shown, the motion platform 4 includes: a first base 41 and a second base 42, the second base 42 is connected to the first base 41 through a first transmission part 43, and the first transmission part 43 enables the second base 42 to move on the first base 41 along a first direction (such as Figure 1 As shown in the figure, the first transmission part 43 enables the second base 42 to move left and right along the direction of the X-axis), the carrier stage 1 is connected to the second base 42 through the second transmission part 44, and the second transmission part 44 enables the carrier stage 1 to move along the second direction on the second base 42 (as shown in the figure). Figure 1 As shown in the figure, the second transmission part 44 realizes the up and down movement of the wafer stage 1 along the Y-axis direction), wherein the projections of the first direction and the second direction on the first base 41 are perpendicular to each other. Thus, the wafer stage 1 can move in all directions. The first transmission part 43 and the second transmission part 44 can use a transmission motor to drive a lead screw, a belt, etc. to realize the relative movement between the wafer stage 1 and the second base 42, and the second base 42 and the first base 41, which will not be repeated here.

[0058] Reference Figure 2 and Figure 3 As shown, the support rods 11 can be regularly arranged on the slide table 1, for example, Figure 3 As shown, 6 support rods are provided from 1101 to 1106, each support rod is a hollow structure with a vacuum hole 12, connected to a vacuum pump 2 through a vacuum tube 3, and generates negative pressure when the vacuum pump 2 is working, thereby being able to fix the silicon carbide wafer 7 placed on the support rod on the support rod 11 to prevent the silicon carbide wafer 7 from falling off when the wafer stage moves. It is understandable that those skilled in the art can select and adjust the number and arrangement of the support rods as needed, and no limitation is made here.

[0059] Reference Figure 1As shown, the light emitted by the light source 51 is reflected by the reflector 52 onto the silicon carbide wafer 7 on the stage 1, and then reflected by the silicon carbide wafer 7 to the camera 53, thereby forming an image of an area on the silicon carbide wafer 7 on the camera 53. The camera 53 is used to shoot and form an image of the surface of the silicon carbide wafer 7. In order to collect the entire image of the surface of the silicon carbide wafer 7, the stage 1 can be made to move in an "S" shape by controlling the motion platform 4, and the camera 53 is started to collect an image once after each movement, thereby realizing the image collection of the entire silicon carbide wafer 7.

[0060] In order to realize the judgment of scratches on the front and back surfaces of the silicon carbide wafer 7, the light source 51 can be composed of a first light source and a second light source of two different wavelengths, wherein the first light source and the second light source are both annular structures, and the second light source is nested inside the first light source, wherein the cross-section of the annular structure can be rectangular, circular, etc., and is not limited here. According to the different transmittances of different wavelengths on the silicon carbide wafer, for example, when white light is used to irradiate the silicon carbide wafer, due to the strong penetration of white light, an image containing scratches on the front and back surfaces of the silicon carbide wafer 7 can be obtained. When blue light with weaker penetration is used for irradiation, only an image containing scratches on the front side of the silicon carbide wafer 7 can be obtained. Subsequently, when the scratch recognition model is used to recognize the two images, on the basis of the scratches on the front and back sides of the image obtained by white light, the scratches on the front side identified by the image obtained by blue light are removed, and the scratches on the back side can be obtained, thereby realizing the recognition of scratches on the front and back sides of the silicon carbide wafer 7.

[0061] During the specific operation, after the silicon carbide wafer 7 is placed on the wafer carrier 1, the vacuum pump 2 starts working. When scanning the first side: support rods 1101, 1103, and 1105 are raised, support rods 1102, 1104, and 1106 fall down, the light source 51 turns on the white light irradiation, the motion platform 4 drives the wafer carrier 1 to make an "S" shape movement, and cooperates with the camera 53 to collect images of each area on the silicon carbide wafer 7.

[0062] After the first side scan is completed, the support rods 1101, 1103, and 1105 fall down, and the support rods 1102, 1104, and 1106 rise up, and then the second scan is performed. During the second side scan: the light source 51 starts white light irradiation, and the motion platform 4 drives the wafer stage 1 to move to three positions blocked by the support rods 1101, 1103, and 1105, and cooperates with the camera 53 to collect images of three areas on the silicon carbide wafer 7. Then, after the second scan is completed, the entire area of ​​the front and back surfaces of the silicon carbide wafer 7 is photographed.

[0063] After the second scan, all support rods fall down, the light source 51 starts blue light irradiation, the motion platform 4 drives the wafer stage 1 to do an "S" shape movement, and cooperates with the camera 53 to collect images of each area on the silicon carbide wafer 7, thereby obtaining a complete image of the front side of the silicon carbide wafer 7. By turning on the blue light scanning result for the front side result of the silicon carbide wafer 7, the white light scanning result for the two sides result of the silicon carbide wafer 7, and filtering out the front side result with the white light result, the front and back scratch results can be distinguished. When scratches are identified based on two images, a pre-trained feature extraction model (such as a convolutional neural network) can be used as the basic model and fine-tuned on the new task. This method can save a lot of time and computing resources while retaining some of the structural knowledge of the pre-trained model. In the field of equipment detection, basic models with better performance, such as ResNet, Inception, etc., can be used to more effectively detect new targets. Those skilled in the art can also use other models to identify scratches, which will not be repeated here.

[0064] The lifting and lowering operation of the above-mentioned support rod can be achieved by driving structures such as motors and cylinders. Those skilled in the art can choose a well-known method to achieve lifting control according to needs, which will not be repeated here.

[0065] From the above, it can be seen that the silicon carbide wafer surface scratch detection equipment is based on optical detection and can cooperate with the deep learning algorithm model convolutional neural network to automatically distinguish scratches from dirt. It can quickly identify scratches on the silicon carbide wafer surface, accurately locate the scratch position, and accurately identify the scratch size, which is convenient for wafer quality analysis, reduces labor costs, and improves the efficiency of scratch detection.

[0066] In a specific embodiment, in order to further improve the efficiency of detection and save manual operations, the silicon carbide wafer surface scratch detection equipment also includes: a transfer module, which is used to take out the silicon carbide wafer from the wafer box port and place it on the support rod of the wafer carrier.

[0067] The conveying module may include: a robot, an edge finder and a code reader. The robot is used to take out the silicon carbide wafer from the wafer box port, the edge finder is used to fix the placement of the silicon carbide wafer on the wafer carrier, and the code reader is used to read the surface information of the silicon carbide wafer.

[0068] The specific structure of the transmission module may adopt a structure well known to those skilled in the art, and will not be described in detail here.

[0069] In other specific embodiments, in order to avoid external interference in the detection process, the silicon carbide wafer surface scratch detection equipment also includes: a support frame 6, one end of the support frame 6 is fixed on the moving platform 4, and an optical detection module 5 is arranged at the other end of the support frame 6, wherein a camera 53 is arranged on the support frame 6 at a position perpendicular to the moving platform, and a light source 51 and a reflector 52 are symmetrically arranged on both sides of the camera 53.

[0070] As a result, the optical detection module 5 no longer needs to be fixed by an additional fixing device, and the entire Figure 1 The silicon carbide wafer surface scratch detection equipment shown is protected to avoid external interference.

[0071] Based on the same design idea Figure 4 As shown, the embodiment of the present application also provides a method for detecting scratches on the surface of a silicon carbide wafer, which is applied to the silicon carbide wafer surface scratch detection equipment described in the above embodiment, and may specifically include the following steps:

[0072] 401. Control the support rod of the first part on the wafer stage to rise to a first height, and the support rod of the second part to rise to a second height, and control the vacuum pump to operate, with the first height being greater than the second height.

[0073] 402. Control the motion platform to drive the wafer carrier to move along a set trajectory, control the light source to emit light of a first wavelength, and control the camera to take pictures each time the wafer carrier moves to a point on the set trajectory, so as to obtain a first surface image of the silicon carbide wafer.

[0074] 403. After the wafer stage has completed its movement according to the set trajectory, the support rod of the first part is controlled to descend to the second height, and the support rod of the second part is controlled to rise to the first height.

[0075] 404. Control the motion platform to drive the wafer stage so that each shielding area on the silicon carbide wafer moves to a position where light is irradiated in sequence, and control the camera to take a picture when each shielding area moves to a position where light is irradiated, so as to obtain a second surface image of the silicon carbide wafer, wherein the shielding area is a corresponding position of each support rod in the support rod of the first part on the silicon carbide wafer.

[0076] 405. Call a scratch recognition model to recognize the first surface image and the second surface image to obtain first scratch recognition result data.

[0077] Specifically, the support rods of the first part are the support rods 1101, 1103, and 1105. The support rods of the second part are the support rods 1102, 1104, and 1106. The above steps 401 to 405 implement the above-mentioned first scanning and second scanning processes, and the specific details can be referred to the scanning process described in the above embodiment, which will not be repeated here.

[0078] Further, after obtaining the first scratch identification result data, the silicon carbide wafer surface scratch detection method further includes the following steps:

[0079] 406. Control the support rod of the second part to descend to a second height, and control the light source to emit light of a second wavelength, where the second wavelength is smaller than the first wavelength.

[0080] 407. Control the motion platform to drive the wafer carrier to move along the set trajectory, and control the camera to take a picture each time the wafer carrier moves to a point on the set trajectory, so as to obtain a third surface image of the silicon carbide wafer.

[0081] 408. Calling a scratch recognition model to recognize the third surface image to obtain second scratch recognition result data.

[0082] 409. Obtain scratch identification result data on two surfaces of the silicon carbide wafer according to the first scratch identification result data and the second scratch identification result data.

[0083] Specifically, the above steps 406 to 408 implement the above third scanning process. In step 409, based on the two scratch identification result data, the scratches on the front and back sides of the silicon carbide wafer 7 are distinguished by comparing the sizes and positions of the scratches recorded in the two scratch identification result data.

[0084] In some specific embodiments, the scratch recognition model is based on a pre-trained feature extraction model and is trained using sample images of silicon carbide wafers marked with scratch defects. In specific implementations, convolutional neural networks (CNNs) and other deep learning models can be used for feature extraction and representation. By combining multiple convolutional layers, pooling layers and other network structures, effective and stable polymorphic features are learned and extracted. With the help of attention mechanisms, residual modules, etc., the generalization ability of the algorithm in nonlinear and multi-task scenarios is improved. By introducing technologies such as data enhancement and transfer learning, the learning effect of the model at each stage is improved; combining physical knowledge and experimental experience, adversarial training strategies are designed to improve the robustness of the algorithm to noise interference; the detection results are fused and optimized to improve the accuracy and reliability of the overall detection system.

[0085] When designing an anti-training strategy, an adversarial training framework can be used, which includes components such as adversarial generators and discriminators to generate adversarial samples (such as dirt) to deceive the detector and determine whether the original sample has been disturbed. During the training process, these adversarial samples can be used to train the detector so that it can maintain a high accuracy rate in the face of disturbances and inconsistencies.

[0086] During software programming, a visual user interface can be built based on easy-to-use and cross-platform compatible programming languages ​​such as Python, so that users can intuitively view and adjust parameters and settings. A plug-in interface can also be provided to expand more functions, such as image processing, annotation generation, etc. One-click docking of device drivers is achieved to simplify the use process; real-time statistical reports and analysis functions are provided to facilitate users to track the changing trends of test results.

[0087] A method for detecting scratches on the surface of a silicon carbide wafer provided in an embodiment of the present application is introduced above. A device for executing the above-mentioned method for detecting scratches on the surface of a silicon carbide wafer will be introduced below.

[0088] See also Figure 5 , Figure 5 This is a schematic diagram of the structure of a silicon carbide wafer surface scratch detection device provided in an embodiment of the present application. Figure 5 As shown, the silicon carbide wafer surface scratch detection device comprises:

[0089] The first height adjustment module 501 is used to control the support rods of the first part on the wafer stage to rise to the first height, the support rods of the second part to rise to the second height, and control the vacuum pump to work, the first height is greater than the second height.

[0090] The first image acquisition module 502 is used to control the motion platform to drive the wafer stage to move along the set trajectory, and control the light source to emit light of the first wavelength, and control the camera to take pictures every time the wafer stage moves to a point on the set trajectory, so as to obtain a first surface image of the silicon carbide wafer.

[0091] The second height adjustment module 503 is used to control the support rods of the first part to drop to the second height and the support rods of the second part to rise to the first height after the wafer stage completes the movement according to the set trajectory.

[0092] The second image acquisition module 504 is used to control the motion platform to drive the wafer stage to move each shielded area on the silicon carbide wafer to the position where the light is irradiated in sequence, and control the camera to take pictures when each shielded area moves to the position where the light is irradiated, so as to obtain a second surface image of the silicon carbide wafer, where the shielded area is the corresponding position of each support rod in the first part on the silicon carbide wafer. And,

[0093] The scratch recognition module 505 is used to call the scratch recognition model to recognize the first surface image and the second surface image to obtain first scratch recognition result data.

[0094] In a possible implementation, the method further includes:

[0095] The third height adjustment module is used to control the support rod of the second part to drop to a second height, and control the light source to emit light of a second wavelength, which is smaller than the first wavelength.

[0096] A third image acquisition module is used to control the motion platform to drive the wafer stage to move along a set trajectory, and control the camera to take pictures every time the wafer stage moves to a point on the set trajectory, so as to obtain a third surface image of the silicon carbide wafer;

[0097] The scratch recognition module is also used to call the scratch recognition model to recognize the third surface image to obtain second scratch recognition result data;

[0098] The recognition result comparison module is used to obtain the scratch recognition result data on the two surfaces of the silicon carbide wafer according to the first scratch recognition result data and the second scratch recognition result data.

[0099] In a possible implementation, the scratch recognition model in the scratch recognition module 505 is based on a pre-trained feature extraction model and is trained using a sample image of a silicon carbide wafer annotated with scratch defects.

[0100] In a possible implementation, the scratch recognition module 505 is also used to test and improve the anti-interference ability of the feature extraction model in combination with an adversarial training strategy during the feature extraction model training process.

[0101] The present application also provides an electronic device in an embodiment. Figure 6 As shown, it shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiment of the present application. The electronic device in the embodiment of the present application may include but is not limited to fixed terminals such as mobile phones, laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 6 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0102] like Figure 6 As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 to a random access memory (RAM) 603. When the electronic device is powered on, various programs and data required for the operation of the electronic device are also stored in the RAM 603. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0103] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a memory card, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0104] An embodiment of the present application also provides a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any one of the silicon carbide wafer surface scratch detection methods provided in the embodiments of the present application.

[0105] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any one of the silicon carbide wafer surface scratch detection methods provided in the embodiments of the present application.

[0106] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0107] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0108] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0109] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A silicon carbide wafer surface scratch detection equipment, characterized in that: include: A wafer stage, on which regularly arranged support rods are arranged, the support rods are hollow and extend out of the surface of the wafer stage to an adjustable height, one end of the support rods is connected to a vacuum pump through a vacuum tube, and the other end of the support rods is used to support the silicon carbide wafer; A motion platform, the slide stage and the motion platform are in transmission connection, and the motion platform controls the slide stage to move in two directions perpendicular to each other in the same plane; as well as, An optical detection module, wherein the optical detection module is arranged in a direction perpendicular to the same plane, and comprises: a light source, a camera and a reflector, wherein the light emitted by the light source has at least two wavelengths, and the reflector is used to reflect the light from the light source to the surface of the silicon carbide wafer, so that the light enters the camera after being reflected by the surface of the silicon carbide wafer.

2. The silicon carbide wafer surface scratch detection equipment according to claim 1, characterized in that: The motion platform includes: a first base and a second base, the second base is connected to the first base through a first transmission part, the first transmission part enables the second base to move along a first direction on the first base, the carrier stage is connected to the second base through a second transmission part, the second transmission part enables the carrier stage to move along a second direction on the second base, and the projections of the first direction and the second direction on the first base are perpendicular to each other.

3. The silicon carbide wafer surface scratch detection equipment according to claim 1, characterized in that: Also includes: A conveying module is used to take out the silicon carbide wafer from the wafer box port and place it on the support rod of the wafer carrier.

4. The silicon carbide wafer surface scratch detection equipment according to claim 3, characterized in that: The conveying module includes: a robot, an edge finder and a code reader, the robot is used to take out the silicon carbide wafer from the wafer box port, the edge finder is used to fix the placement position of the silicon carbide wafer on the wafer carrier, and the code reader is used to read the surface information of the silicon carbide wafer.

5. The silicon carbide wafer surface scratch detection equipment according to claim 1, characterized in that: The light source comprises: a first light source and a second light source, wherein the first light source and the second light source are both annular structures, the second light source is nested inside the first light source, and the wavelengths of the light emitted by the first light source and the second light source are different.

6. The silicon carbide wafer surface scratch detection equipment according to any one of claims 1 to 5, characterized in that: Also includes: A support frame, one end of which is fixed on the moving platform, and the optical detection module is arranged at the other end of the support frame, wherein the camera is arranged on the support frame at a position perpendicular to the moving platform, and the light source and the reflector are symmetrically arranged on both sides of the camera.

7. A method for detecting scratches on a silicon carbide wafer surface, applied to the silicon carbide wafer surface scratch detection equipment as claimed in any one of claims 1 to 6, characterized in that: include: Control the support rod of the first part on the wafer stage to rise to a first height, the support rod of the second part to rise to a second height, and control the vacuum pump to work, wherein the first height is greater than the second height; Controlling the motion platform to drive the wafer stage to move along a set trajectory, controlling the light source to emit light of a first wavelength, and controlling the camera to take pictures each time the wafer stage moves to a point on the set trajectory, so as to obtain a first surface image of the silicon carbide wafer; After the slide stage has completed its movement according to the set trajectory, controlling the support rod of the first part to descend to the second height, and the support rod of the second part to rise to the first height; Controlling the motion platform to drive the wafer stage so that each shielding area on the silicon carbide wafer moves to the position where the light is irradiated in sequence, and controlling the camera to take pictures when each shielding area moves to the position where the light is irradiated, to obtain a second surface image of the silicon carbide wafer, wherein the shielding area is a corresponding position of each support rod of the first part on the silicon carbide wafer; The scratch recognition model is called to recognize the first surface image and the second surface image to obtain first scratch recognition result data.

8. The method for detecting scratches on the surface of a silicon carbide wafer according to claim 7, characterized in that: After obtaining the first scratch identification result data, the method further includes: Controlling the support rod of the second part to descend to the second height, and controlling the light source to emit light of a second wavelength, wherein the second wavelength is smaller than the first wavelength; Controlling the motion platform to drive the wafer stage to move along the set trajectory, and controlling the camera to take pictures each time the wafer stage moves to a point on the set trajectory, so as to obtain a third surface image of the silicon carbide wafer; Calling a scratch recognition model to recognize the third surface image to obtain second scratch recognition result data; The scratch identification result data on both surfaces of the silicon carbide wafer are obtained according to the first scratch identification result data and the second scratch identification result data.

9. The method for detecting scratches on the surface of a silicon carbide wafer according to claim 7 or 8, characterized in that: The scratch recognition model is based on a pre-trained feature extraction model and is trained using a sample image of a silicon carbide wafer annotated with scratch defects.

10. The method for detecting scratches on the surface of a silicon carbide wafer according to claim 9, characterized in that: Also includes: During the training process of the feature extraction model, the anti-interference ability of the feature extraction model is tested and improved in combination with an adversarial training strategy.

Images