Silicon carbide facet imaging system and facet recognition method

Abstract

The present invention relates to the field of optical detection technology, and provides a silicon carbide facet imaging system and a facet identification method, wherein the silicon carbide facet imaging system comprises: a data processing module and a data acquisition module; the data acquisition module comprises: a light source, a first filter, a polarizer, a 1 / 2 wave plate, a first lens group, a collection device, a lens, a second filter, a second lens group, and a dichroic mirror; the light source is sequentially provided with the first filter, a polarizer, a 1 / 2 wave plate, a first lens group, and a dichroic mirror in the direction of emission; the dichroic mirror is arranged at a 45° angle to the light source, and the dichroic mirror is sequentially provided with the second lens group, a second filter, a lens, and a collection device in the direction of emission; the lens is mounted on the collection device; the collection device adopts a CCD camera or a single-photon detector. The present invention can improve the accuracy of facet area identification, reduce manual intervention, and reduce the cost of manual detection.

Description

Technical Field

[0001] The present invention relates to the field of optical detection technology, and in particular to a silicon carbide facet imaging system and a facet recognition method. Background Art

[0002] With the development and progress of semiconductor materials, compared with silicon-based semiconductor materials, the third-generation semiconductor materials represented by silicon carbide (SiC) have many advantages such as high breakdown electric field, high saturated electron drift velocity, and high thermal conductivity, which enable silicon carbide to develop and be used rapidly. Silicon carbide is mainly used in new energy vehicles, photovoltaic energy storage, rail transportation and other fields. Especially in the automotive field, it is expected that the application of vehicle-mounted main inverters, charging modules and other applications will continue to grow rapidly in the next few years. With the continuous increase in demand, it is very important to image and analyze the small face to ensure the quality of SiC wafers after cutting. Since silicon carbide crystals will be doped with a certain concentration of N elements, the concentration of N elements in the core position (small face) will be higher than that in other normal positions during its initial growth process. Compared with other normal positions, the small face has a large concentration of N elements, and its resistivity, density, and hardness will have some differences and changes from other normal positions. The absorption and fluorescence intensity of light of a specific wavelength are different, so the small face of silicon carbide can be imaged and identified.

[0003] The optical property of the high refractive index of silicon carbide ingots makes it difficult to detect and accurately identify faceted areas, especially when some areas of the ingot reflect light, resulting in uneven background. Uneven brightness in non-faceted areas may be mistakenly identified as faceted areas, greatly increasing the difficulty of accurate identification. Traditional testing equipment has a single testing method and few testing methods. The spectral intensity information of traditional imaging systems is very close, and changes in optical signals are difficult to distinguish with the naked eye. In addition, traditional test light sources directly illuminate the silicon carbide ingots, and there is mechanical contact, which may cause scratches on the surface of the ingot. Summary of the invention

[0004] The present invention mainly solves the technical problems of difficulty in detecting and identifying small faces of silicon carbide ingots and low detection accuracy in the prior art, and proposes a silicon carbide small face imaging system and a small face identification method. The small face imaging method is a non-contact and non-destructive test, and improves the accuracy of small face area identification, reduces manual intervention, and reduces manual detection costs.

[0005] The present invention provides a silicon carbide facet imaging system, comprising: a data processing module and a data acquisition module;

[0006] The data acquisition module includes: a light source, a first filter, a polarizer, a 1 / 2 wave plate, a first lens group, a collection device, a lens, a second filter, a second lens group, and a dichroic mirror;

[0007] The first filter, polarizing plate, 1 / 2 wave plate, first lens group and dichroic mirror are arranged in sequence in the emission direction of the light source;

[0008] The dichroic mirror is arranged at an angle of 45° to the emission direction of the light source, and a second lens group, a second filter, a lens and a collection device are arranged in sequence in the emission direction of the dichroic mirror; the lens is mounted on the collection device; the collection device adopts a CCD camera or a single photon detector;

[0009] The acquisition device is signal-connected to the data processing module.

[0010] Preferably, it also includes: a signal synchronization module;

[0011] The signal synchronization module is respectively connected with the acquisition device and the data processing module by signal.

[0012] Preferably, an electric turntable is provided in the reflection direction of the dichroic mirror; the electric turntable is used to place the silicon carbide ingot;

[0013] The first lens group includes: a first concave lens and a first convex lens;

[0014] The second lens group includes: a second convex lens and a second concave lens.

[0015] Preferably, the light source emits light at 365 nm.

[0016] Correspondingly, the present invention further provides a facet recognition method of a silicon carbide facet imaging system according to any embodiment of the present invention, comprising the following process:

[0017] Step S1, using a silicon carbide facet imaging system to collect a photo of the original ingot;

[0018] Step S2, in the original ingot photo, identifying the edge of the ingot, retaining only the area within the edge of the ingot, and obtaining an ingot image;

[0019] Step S3, enhancing the ingot image to highlight the small face area;

[0020] The enhancement processing includes but is not limited to: one or more of gray value scaling processing, mean image smoothing processing, contrast enhancement processing, threshold processing, preliminary screening, and secondary recognition;

[0021] Step S4, circle the edge of the small face area to complete the small face recognition.

[0022] Furthermore, in step 1, the collected original ingot photos are first screened: first, multiple different positions are selected on the original ingot photos as judgment areas for the grayscale values ​​of the photos; then, the grayscale values ​​of the judgment areas are determined, and the grayscale values ​​of the judgment areas are compared with the grayscale value threshold, and the original ingot photos whose grayscale values ​​of all judgment areas are within the grayscale value threshold range are retained for subsequent processing.

[0023] Furthermore, the gray value scaling process includes:

[0024] Calculate the average pixel grayscale value of the ingot image, and use the average pixel grayscale value to scale the grayscale value according to the following formula:

[0025]

[0026] in, x is the horizontal coordinate of the pixel point; y is the vertical coordinate of the pixel point, A 1 is the gray value of the image after scaling, A is the gray value of the image before scaling, c is the transformation coefficient, d is an adjustable constant.

[0027] Furthermore, the contrast enhancement process includes:

[0028] The image is smoothed by the mean value, and the new image grayscale value is calculated, and the contrast enhancement is performed according to the following formula:

[0029]

[0030] in, B 1 represents the gray value of the image after enhancement processing. B Represents the gray value of the image before enhancement processing; round Represents a function used to round a floating point number to the nearest integer; h represents the gray value of the image after mean image smoothing, k Indicates the corrected grayscale value.

[0031] Furthermore, the thresholding process includes:

[0032] By comparing the grayscale values ​​of pixels at corresponding positions in the original image and the image after contrast enhancement, the pixels whose grayscale value difference reaches a certain threshold are determined, so that the small face area and the non-small face area can be initially distinguished; the formula is as follows

[0033]

[0034] in, C 1 represents the grayscale value of the original image,C Represents the gray value of the image after enhancement processing; z is the threshold parameter.

[0035] Further, the preliminary screening: after the facet area and the non-facet area are preliminarily distinguished, parameters are set to preliminarily screen the above areas; the parameters include roundness, aspect ratio, area, and hole area ratio;

[0036] The secondary identification: performing secondary identification on the areas after the preliminary screening by performing gradient operation to obtain areas with large pixel value gradient changes;

[0037] The calculation formula is as follows:

[0038]

[0039] in, D Represents the grayscale value of the image before secondary recognition, ∇D Represents the grayscale value of the image after secondary recognition.

[0040] The silicon carbide facet imaging system and facet recognition method provided by the present invention have the following advantages compared with the prior art:

[0041] 1. In the present invention, the collimated beam expansion is achieved through the lens group, which can not only uniformly amplify the light spot, but also avoid the uneven distribution of light caused by the divergence angle, reduce the systematic error of light intensity at different positions, and effectively improve the accuracy and repeatability of small facet imaging compared to the traditional test light source directly irradiating the silicon carbide ingot.

[0042] 2. The present invention overturns the single testing method and few testing methods of traditional testing equipment. The present invention is compatible with ingots of different sizes, and the lens of the collection part can not only adjust the field of view size but also improve the resolution. The femtosecond time resolution achieved by the present invention can eliminate environmental background interference signals, supplement the deficiencies of spectral measurement and intensity measurement, and greatly improve the sensitivity and accuracy of facet recognition.

[0043] 3. The present invention adopts a signal synchronization module for timing control, which enables the CCD camera to collect images at different times, obtain intensity images with different delay times, and provide a lot of information that traditional optical measurement technology cannot provide, providing a high-sensitivity and multi-time-scale imaging method for the detection and identification of small surface areas.

[0044] 4. The present invention adds polarization-related detection, and adjusts the polarization direction of the light source by rotating the 1 / 2 wave plate, and studies the impact of polarized light in different directions on the test of silicon carbide ingots in a more detailed manner, which is a qualitative improvement compared to the single detection mode of traditional equipment. The facet imaging method of the present invention is a non-contact non-destructive test, which overcomes the mechanical contact of traditional imaging methods and avoids scratches on the surface of the ingot.

[0045] 5. The signal synchronization module in the present invention can accurately adjust the synchronization between signals and precisely control the door width of the camera. It can also cooperate with the precision moving table and rotating table to cooperate with the whole test system to adjust the polarization of the light source and the position of the crystal ingot to meet the needs of different occasions.

[0046] 6. The present invention can perform image processing on the small face area of ​​the ingot, effectively improving the accuracy of small face area recognition, reducing manual intervention, reducing manual inspection costs, and improving the quality and recognition efficiency of silicon carbide wafers. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a schematic diagram of a silicon carbide facet imaging system provided by the present invention;

[0048] Figure 2 It is a principle diagram of the data acquisition module provided by the present invention;

[0049] Figure 3 is a schematic diagram of a CCD camera provided by the present invention;

[0050] Figure 4 is a schematic diagram of a lens provided by the present invention;

[0051] Figure 5 is a timing diagram of the silicon carbide facet imaging system provided by the present invention;

[0052] Figure 6 It is a photo of the original ingot collected by the silicon carbide facet imaging system provided by the present invention;

[0053] Figure 7 It is a photo effect of the crystal ingot without the external environment of the crystal ingot;

[0054] Figure 8 This is the effect picture of the ingot photo after grayscale value scaling processing;

[0055] Fig. 9 This is the effect picture of the ingot photo after mean image smoothing;

[0056] Fig.10 This is the effect picture of the ingot photo after contrast enhancement processing;

[0057] Fig.11 This is a photo of the ingot after small face identification.

[0058] Figure numerals: a, data processing module; b, signal synchronization module; c, data acquisition module; 1, light source; 2, first filter; 3, polarizer; 4, 1 / 2 wave plate; 5, first concave lens; 6, first convex lens; 7, acquisition device; 8, lens; 9, second filter; 10, second concave lens; 11, second convex lens; 12, dichroic mirror; 13, silicon carbide ingot; 14, electric turntable. DETAILED DESCRIPTION

[0059] In order to make the technical problems solved by the present invention, the technical solutions adopted and the technical effects achieved clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for the convenience of description, only the parts related to the present invention are shown in the accompanying drawings, rather than all the contents. Embodiment 1

[0060] like Figure 1-4 As shown, a silicon carbide facet imaging system provided by an embodiment of the present invention includes: a data processing module a, a signal synchronization module b and a data acquisition module c.

[0061] The data acquisition module c comprises: a light source 1, a first filter 2, a polarizer 3, a 1 / 2 wave plate 4, a first lens group, a collection device 7, a lens 8, a second filter 9, a second lens group, and a dichroic mirror 12;

[0062] The emission direction of the light source 1 is sequentially provided with a first filter 2, a polarizer 3, a 1 / 2 wave plate 4, a first lens group, and a dichroic mirror 12; wherein the first lens group includes: a first concave lens 5 and a first convex lens 6. The 1 / 2 wave plate 4 can be electrically rotated, and the rotation of the 1 / 2 wave plate 4 can adjust the polarization direction of the light source.

[0063] The dichroic mirror 12 is arranged at an angle of 45° with the emission direction of the light source 1. The emission direction of the dichroic mirror 12 is sequentially provided with a second lens group, a second filter 9, a lens 8 and a collection device 7; wherein the second lens group comprises: a second convex lens 11 and a second concave lens 10. The lens 8 is mounted on the collection device 7; the collection device 7 adopts a CCD camera or a single photon detector; the collection device 7 is signal-connected to the data processing module a.

[0064] An electric turntable 14 is arranged in the reflection direction of the dichroic mirror 12 ; the electric turntable 14 is used for placing the silicon carbide ingot 13 .

[0065] In the above scheme, the light source 1 emits 365nm light. Specifically, the first filter 2 can be a 365-3nm filter, the polarizer 3 can be a 365nm dedicated polarizer, the 1 / 2 wave plate 4 can be a 365nm dedicated wave plate, the first concave lens 5 can be a quartz-225mm concave lens, the first convex lens 6 can be a +365mm convex lens, the lens 8 can be a +5-+20 continuous zoom lens, the second filter 9 can be a 500-10nm filter, the second convex lens 11 can be a +365mm convex lens, the second concave lens 10 can be a quartz-225mm concave lens, and the dichroic mirror 12 can be a dichroic mirror that reflects 365nm and transmits 500nm.

[0066] The working principle of the data acquisition module c is as follows: first, the light source 1 emits 365nm light, the first filter 2 intercepts light of other wavelengths, and only allows a specific wavelength (365nm) to pass through, and then the light is purified by the polarizer 3 to ensure its polarization; after passing through the 1 / 2 wave plate 4 to adjust the polarization direction of the light, the light is amplified by the first lens group, and the light is irradiated onto the silicon carbide ingot 13 through the dichroic mirror 12. The fluorescence generated by the silicon carbide ingot 13 is transmitted to the second lens group through the dichroic mirror 12, and the second filter 9 filters out the stray light and then transmits the image to the acquisition device 7 (CCD camera) through the lens 8, thereby completing the test. The CCD camera has a gating function, which can delay shooting at different times to achieve a CCD time resolution of the femtosecond level.

[0067] In the present invention, light passes through the first filter 2 to filter out other stray light, and only 365nm light is allowed to pass through and irradiate the silicon carbide ingot 13. Since the concentration of N elements near the silicon carbide during the initial growth is denser than the concentration of the later growing part, the difference in N elements will absorb 365nm light to different degrees, so the reflected light will have different intensities, so that the position of the small face can be captured by the CCD camera.

[0068] The signal synchronization module b is connected to the acquisition device 7 and the data processing module a respectively. The signal synchronization module b mainly includes a signal synchronizer and a signal generator, and its function is to control the signal generator to send a signal, and the signal controls the light source 1 to work. At the same time, the CCD camera will also receive the signal and start to control the shutter switch, so that the CCD camera can collect at different times, so as to perform time-resolved related tests.

[0069] The data processing module a mainly includes an industrial control computer PC. The industrial control PC is mainly used as a computer carrier for data collection and analysis, and is responsible for driving and controlling the logical work and data processing of devices such as CCD cameras and light sources 1.

[0070] The workflow of the silicon carbide facet imaging system based on CCD time resolution of the present invention is as follows: first, the signal synchronization module b transmits a start signal, the light source 1 receives the start signal and starts, the light passes through the first filter 2 and the polarizer 3 for purification, and after the polarization direction of the light is adjusted by the 1 / 2 wave plate 4, it is expanded and collimated by the first lens group, and reflected to the surface of the silicon carbide ingot 13 by the dichroic mirror 12. The silicon carbide ingot 13 generates fluorescence, which is reduced by the second lens group and purified by the second filter 9 to reach the lens 8. At this time, the electrical signal controls the CCD camera to start exposure; another signal controls the CCD camera door signal to start switching. After each acquisition, the door switch of the CCD camera will be delayed to open after the set time, so as to achieve accurate time resolution measurement. Since the concentration of N element in each place of the silicon carbide ingot 13 is different, the fluorescence intensity emitted will be different. The CCD camera collects the data and transmits it to the data processing module a, and the data processing module a processes and identifies the facet. After the present invention identifies the small face, the acquisition device 7 can be subsequently switched to a single-photon detector to test the fluorescence lifetime, so as to facilitate subsequent optical characteristic analysis of the small face area.

[0071] On the one hand, the present invention utilizes a fluorescence detection method to eliminate the uneven background caused by the reflection of the ingot surface as much as possible through the selection and arrangement of light sources. Compared with naked eye detection, the fluorescence detection method improves the accuracy and reliability of identification. Embodiment 2

[0072] An embodiment of the present invention provides a facet recognition method of a silicon carbide facet imaging system according to any embodiment of the present invention, comprising the following process:

[0073] Step S1, using a silicon carbide facet imaging system to collect a photo of the original ingot;

[0074] The original ingot photos collected (such as Figure 6 As shown in the figure, screening is performed first: first, multiple different positions are selected on the original ingot photo as the judgment area of ​​the grayscale value of the photo; then, the grayscale value of the judgment area is determined, and the grayscale value of the judgment area is compared with the grayscale value threshold, and the original ingot photos whose grayscale values ​​of all judgment areas are within the grayscale value threshold range are retained for subsequent processing. Among them, the grayscale value threshold range is, for example, 0 to 200.

[0075] If the grayscale value of the judgment area in the original ingot photo does not meet the grayscale value threshold, the CCD camera exposure time of the silicon carbide facet imaging system is adjusted to re-collect the original ingot photo. The exposure time can be adjusted within the range of 1s-6s.

[0076] Step S2, in the original ingot photo, identify the edge of the ingot, retain only the area within the edge of the ingot, and obtain an ingot image.

[0077] Method for identifying the edge of the ingot: Create a ROI ingot edge identification area.

[0078] After the edge of the ingot is identified, the grayscale value of the area outside the edge of the ingot is set to 0, and only the area within the edge of the ingot is retained to obtain the ingot image (such as Figure 7 This step can remove the influence of the external environment of the ingot on the recognition of the small facet area in the ingot. The obtained ingot image can be subjected to grayscale filtering.

[0079] Step S3, enhancing the ingot image to highlight the small face area.

[0080] The enhancement processing includes but is not limited to: one or more of gray value scaling processing, mean image smoothing processing, contrast enhancement processing, threshold processing, preliminary screening, and secondary recognition;

[0081] 1. Gray value scaling processing: Calculate the average gray value of the pixels in the ingot image, and use the average gray value to scale the gray value according to the following formula to achieve enhanced visual effects (such as Figure 8 shown):

[0082]

[0083] in, x is the horizontal coordinate of the pixel point; y is the vertical coordinate of the pixel point, A 1 is the gray value of the image after scaling, A is the gray value of the image before scaling, c is the transformation coefficient, with a value range of 3±0.5, and the value is adjusted according to the brightness of the wafer area in the ingot image. d It is an adjustable constant and its value is -200.

[0084] 2. Mean image smoothing: By using mean filtering technology, the gray value of each pixel in the ingot image is replaced with the gray value average of the pixels in its surrounding neighborhood (within the filter range) to achieve linear smoothing of the image gray value (e.g. Fig. 9 as shown).

[0085] 3. The contrast enhancement process: by smoothing the mean image and calculating the new image grayscale value, the contrast enhancement process is performed to achieve contrast enhancement (such as Fig.10 as shown).

[0086] Adjust according to the following formula

[0087]

[0088] in, x is the horizontal coordinate of the pixel point;y is the vertical coordinate of the pixel point, B 1 represents the gray value of the image after enhancement processing. B Represents the grayscale value of the image before enhancement. round Represents a function that rounds a floating-point number to the nearest integer. h represents the gray value of the image after mean image smoothing, k Indicates the grayscale value of the correction, ranging from 0.7 to 1.

[0089] 4. Thresholding: Thresholding compares the grayscale values ​​of pixels at corresponding positions in the original image and the image after contrast enhancement, and identifies the pixels whose grayscale value difference reaches a certain threshold, so as to achieve a preliminary distinction between small face areas and non-small face areas. The formula is as follows

[0090]

[0091] in, x is the horizontal coordinate of the pixel point; y is the vertical coordinate of the pixel point, C 1 represents the grayscale value of the original image, C Represents the grayscale value of the image after enhancement. z It is the threshold parameter, and its value range is 15-20.

[0092] 5. The preliminary screening: After the small face area and the non-small face area are preliminarily distinguished, it is also necessary to set parameters for preliminary screening of the above areas; the parameters include roundness, aspect ratio, area, and hole area ratio.

[0093] 6. The secondary identification: performing secondary identification on the areas after the preliminary screening by gradient operation to obtain areas with large pixel value gradient changes.

[0094] The calculation formula is as follows:

[0095]

[0096] in, x is the horizontal coordinate of the pixel point; y is the vertical coordinate of the pixel point, D Represents the grayscale value of the image before secondary recognition, ∇D Represents the grayscale value of the image after secondary recognition.

[0097] The area identified by gradient calculation is the edge of the secondary identified small face area. The edge of the white small face area can be circled and the small face can be presented on the original ingot photo using a coil (such as Fig.11 As shown), the small face recognition is completed.

[0098] The present invention has made improvements on the image recognition method, and has tried various image processing methods to eliminate the background unevenness caused by the reflection on the surface of the ingot. In addition, if the camera exposure time is too long, the photo of the ingot will be oversaturated, which will also affect the recognition effect. The present invention adjusts the camera exposure time by the grayscale value of the photo to keep the grayscale value intensity of the ingot photo within a range suitable for processing. At the same time, it solves the technical problem of poor detection effect in the prior art, and significantly reduces the misjudgment rate of small face recognition in the existing CCD visual inspection software algorithm.

[0099] In order to implement the above embodiments, the present invention further proposes an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the facet recognition method proposed in the above embodiments of the present invention is implemented.

[0100] In order to implement the above embodiments, the present invention further proposes a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the facet recognition method proposed in the above embodiments of the present invention is implemented.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that modifying the technical solutions described in the aforementioned embodiments, or replacing some or all of the technical features therein by equivalents, does not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A facet recognition method based on a silicon carbide facet imaging system, characterized in that: The process includes: Step S1, using a silicon carbide facet imaging system to collect a photo of the original ingot; The silicon carbide facet imaging system comprises: a data processing module (a) and a data acquisition module (c); the data acquisition module (c) comprises: a light source (1), a first filter (2), a polarizer (3), a half-wave plate (4), a first lens group, a collection device (7), a lens (8), a second filter (9), a second lens group, and a dichroic mirror (12); the first filter (2), the polarizer (3), the half-wave plate (4), the first lens group, and the dichroic mirror (12) are arranged in sequence in the emission direction of the light source (1); the dichroic mirror (12) is arranged at an angle of 45° to the emission direction of the light source (1); the second lens group, the second filter (9), the lens (8), and the collection device (7) are arranged in sequence in the emission direction of the dichroic mirror (12); the lens (8) is mounted on the collection device (7); the collection device (7) adopts a CCD camera or a single-photon detector; the collection device (7) is signal-connected to the data processing module (a); Step S2, in the original ingot photo, identifying the edge of the ingot, retaining only the area within the edge of the ingot, and obtaining an ingot image; Step S3, enhancing the ingot image to highlight the small face area; The enhancement processing includes but is not limited to: one or more of gray value scaling processing, mean image smoothing processing, contrast enhancement processing, threshold processing, preliminary screening, and secondary recognition; The preliminary screening: after the facet area and the non-facet area are preliminarily distinguished, the parameters are set to preliminarily screen the above areas; the parameters include roundness, aspect ratio, area, and hole area ratio; The secondary identification: performing secondary identification on the areas after the preliminary screening by performing gradient operation to obtain areas with large pixel value gradient changes; The calculation formula is as follows: ； in, D Represents the gray value of the image before secondary recognition, Represents the gray value of the image after secondary recognition; Step S4, circle the edge of the small face area to complete the small face recognition.

2. The facet recognition method based on the silicon carbide facet imaging system according to claim 1, characterized in that: In step 1, the collected original ingot photos are first screened: first, multiple different positions are selected on the original ingot photos as judgment areas for the grayscale values ​​of the photos; then, the grayscale values ​​of the judgment areas are determined, and the grayscale values ​​of the judgment areas are compared with the grayscale value threshold, and the original ingot photos whose grayscale values ​​of all judgment areas are within the grayscale value threshold range are retained for subsequent processing.

3. The facet recognition method based on the silicon carbide facet imaging system according to claim 1, characterized in that: The grayscale value scaling process comprises: Calculate the average pixel grayscale value of the ingot image, and use the average pixel grayscale value to scale the grayscale value according to the following formula: ； in, x is the horizontal coordinate of the pixel point; y is the vertical coordinate of the pixel point, A 1 is the gray value of the image after scaling, A is the gray value of the image before scaling, c is the transformation coefficient, d is an adjustable constant.

4. The facet recognition method based on the silicon carbide facet imaging system according to claim 3 is characterized in that: The contrast enhancement process comprises: The image is smoothed by the mean value, and the new image grayscale value is calculated, and the contrast enhancement is performed according to the following formula: ； in, B 1 represents the gray value of the image after enhancement processing. B Represents the gray value of the image before enhancement processing; round Represents a function used to round a floating point number to the nearest integer; h represents the gray value of the image after mean image smoothing, k Indicates the corrected grayscale value.

5. The facet recognition method based on the silicon carbide facet imaging system according to claim 4, characterized in that: The thresholding process comprises: By comparing the grayscale values ​​of pixels at corresponding positions in the original image and the image after contrast enhancement, the pixels whose grayscale value difference reaches a certain threshold are determined, so that the small face area and the non-small face area can be initially distinguished; the formula is as follows ； in, C 1 represents the gray value of the original image, C Represents the gray value of the image after enhancement processing; z is the threshold parameter.

6. The facet recognition method based on the silicon carbide facet imaging system according to claim 1, characterized in that: Also includes: Signal synchronization module (b); The signal synchronization module (b) is signal-connected to the acquisition device (7) and the data processing module (a) respectively.

7. The facet recognition method based on the silicon carbide facet imaging system according to claim 1, characterized in that: An electric turntable (14) is arranged in the reflection direction of the dichroic mirror (12); the electric turntable (14) is used to place the silicon carbide crystal ingot (13); The first lens group comprises: a first concave lens (5) and a first convex lens (6); The second lens group comprises a second convex lens (11) and a second concave lens (10).

8. The facet recognition method based on the silicon carbide facet imaging system according to claim 7, characterized in that: The light source (1) emits light at 365 nm.

Images