A method and device for determining the pixel diameter of a crystal rod

Abstract

The present invention provides a method, device and electronic device for determining the pixel diameter of a crystal rod, the method comprising: obtaining an image inside a single crystal furnace, wherein the image inside the single crystal furnace displays an outer aperture of a crystal rod, and multiple regions are preset in the image inside the single crystal furnace; determining the number distribution of aperture points of the outer aperture of the crystal rod in each of the multiple regions; selecting a target region from the multiple regions according to the number distribution; fitting the aperture points in the target region to obtain the pixel diameter of the crystal rod. The method solves the technical problem that in the prior art, all aperture points appearing in the image are often used to fit a circle at any time, resulting in a low accuracy of the fitting circle result.

Description

Technical Field

[0001] The present invention relates to the field of intelligent crystal pulling technology, and in particular to a method and device for determining a crystal rod pixel diameter. Background Art

[0002] Climate change is a global problem facing mankind. Countries around the world are reducing greenhouse gas emissions through global agreements. To this end, my country has proposed carbon peak and carbon neutrality goals, and the demand for clean energy is gradually increasing. Solar energy is an important clean energy source, and monocrystalline silicon is an important raw material for solar panels. In recent years, the market demand for high-quality monocrystalline silicon has increased, which has put forward higher requirements for the production of monocrystalline silicon.

[0003] The Czochralski method is the mainstream method for producing single crystal silicon at present. The main steps of the Czochralski method are welding, seeding, shoulder release, shoulder rotation, and equal diameter. The shoulder release is to allow the single crystal silicon rod to grow to the target size, and the equal diameter is to keep the single crystal silicon rod growing at the target size. In order to realize the automatic control of the single crystal furnace pulling, the diameter of the single crystal silicon rod (can be referred to as the crystal rod) is a parameter that needs to be obtained at all times. In the prior art, manufacturers often use cameras to collect images inside the single crystal furnace, such as Figure 1 As shown, an aperture will appear in the image inside the single crystal furnace. The aperture is formed at the junction of the single crystal silicon rod and the liquid silicon surface. The prior art generates the pixel diameter Dp of the crystal rod by fitting the aperture into a circle, and then estimates a coefficient K through human eye observation, multiplies it by Dp, and obtains the estimated diameter De of the current crystal rod. The control system calculates the difference between the estimated diameter and the preset diameter, and adjusts the pulling speed according to the rate of change of the difference and historical conditions so that the estimated diameter remains within a certain range above and below the preset diameter.

[0004] It should be noted that, combined with Figure 3 to Figure 4 During the shoulder release process, the camera position remains fixed. In the image collected by the camera from the fixed position, the aperture will be blocked by the crystal rod itself or the lower edge of the guide tube, and the aperture display is incomplete. Moreover, as the diameter of the crystal rod increases, the shape of the aperture in the image at different times is also different. In the prior art, all aperture points that appear in the image at any time are often used to fit a circle, resulting in low accuracy of the fitting circle result.

[0005] In view of this, the present invention is proposed. Summary of the invention

[0006] The present invention provides a method, device and electronic device for determining the pixel diameter of a crystal rod, so as to solve the technical problem in the prior art that all aperture points appearing in an image are often used to fit a circle at any time, resulting in low accuracy of the fitting circle result.

[0007] According to a first aspect of the present invention, a method for determining a crystal rod pixel diameter is provided, the method comprising: acquiring an image inside a single crystal furnace, wherein the image inside the single crystal furnace displays an outer aperture of a crystal rod, and a plurality of regions are preset in the image inside the single crystal furnace; determining the quantity distribution of aperture points of the outer aperture of the crystal rod in each of the plurality of regions; screening out a target region from the plurality of regions according to the quantity distribution; and fitting the aperture points in the target region to obtain the pixel diameter of the crystal rod.

[0008] Furthermore, the multiple areas include a first fan-shaped area, a second fan-shaped area and a third fan-shaped area with the same center, and the third fan-shaped area is located between the first fan-shaped area and the second fan-shaped area, wherein the radius of the first fan-shaped area is the same as that of the third fan-shaped area, and the radius of the first fan-shaped area and the third fan-shaped area is larger than that of the second fan-shaped area.

[0009] Furthermore, a target area is screened out from the multiple areas based on the quantity distribution, including: when the number of aperture points in the first fan-shaped area and the third fan-shaped area are both greater than the standard number, the first fan-shaped area and the third fan-shaped area are determined as target areas; when the number of aperture points in the first fan-shaped area or the third fan-shaped area is less than or equal to the standard number, the first fan-shaped area, the second fan-shaped area and the third fan-shaped area are determined as target areas.

[0010] Furthermore, the aperture points in the target area are fitted, including: fitting the aperture points in the target area to obtain an initial fitting circle; performing gradient descent processing on the initial fitting circle using the aperture points in the target area to obtain a first fitting circle; determining a first distance range according to the radius of the first fitting circle; screening out a first aperture point combination from the aperture points in the target area whose distance from the center of the first fitting circle is within the first distance range; and performing gradient descent processing on the first fitting circle using the first aperture point combination to obtain a target fitting circle.

[0011] Furthermore, the aperture points in the target area are fitted to obtain an initial fitting circle, including: when the target area is a first fan-shaped area and a third fan-shaped area, directly fitting the aperture points in the target area to obtain an initial fitting circle.

[0012] Furthermore, the aperture points in the target area are fitted to obtain an initial fitting circle, including: when the target area is a first fan-shaped area, a second fan-shaped area, and a third fan-shaped area, the aperture points in the target area are fitted to obtain a second fitting circle; a second distance range is determined according to the radius of the second fitting circle; a second aperture point combination whose distance from the center of the second fitting circle is within the second distance range is screened from the second fan-shaped area; and the second aperture point combination, the aperture points of the first fan-shaped area, and the aperture points of the third fan-shaped area are fitted to obtain the initial fitting circle.

[0013] Furthermore, the aperture points in the target area are obtained by the following steps: converting the coordinate system of each pixel point in the target area to obtain a polar coordinate diagram of the target area; determining a candidate area from the polar coordinate diagram according to the composition of the target area; and traversing the candidate area to obtain all aperture points in the target area.

[0014] Furthermore, determining a candidate area from the polar coordinate diagram according to the composition of the target area includes: when the target area is a first fan-shaped area and a third fan-shaped area, obtaining the coordinates of two vertices of the first fan-shaped area and the coordinates of two vertices of the third fan-shaped area; determining the candidate area according to the coordinates of the two vertices of the first fan-shaped area and the coordinates of the two vertices of the third fan-shaped area.

[0015] Furthermore, the image inside the single crystal furnace also includes a circular opening at the lower end of the guide tube, a melt surface, and a reflection of the circular opening at the lower end of the guide tube in the melt surface, wherein, after obtaining the pixel diameter of the crystal rod, the method further includes: obtaining a first vertical distance from the lower end of the guide tube to the outer aperture of the crystal rod through the circular opening at the lower end of the guide tube, the melt surface, and the reflection of the circular opening at the lower end of the guide tube in the melt surface; determining the pixel diameter of the circular opening at the lower end of the guide tube through the image inside the single crystal furnace; obtaining the actual distance of the circular opening at the lower end of the guide tube; and obtaining the pixel diameter of the circular opening at the lower end of the guide tube through the image inside the single crystal furnace. The first distance between the opening diameter of the lower end of the guide tube and the optical axis is obtained through the image in the single crystal furnace, and the first pixel length of the first distance in the image plane is obtained; the second distance between the diameter of the crystal rod and the optical axis is obtained through the image in the single crystal furnace, and the second pixel length of the second distance in the image plane is obtained; the actual diameter distance of the crystal rod pixel is obtained according to the pixel diameter of the crystal rod, the first pixel length, the second pixel length, the actual distance of the circular opening at the lower end of the guide tube, the pixel diameter of the circular opening at the lower end of the guide tube, the first vertical distance, the camera focal length parameter and the camera pixel parameter.

[0016] According to a second aspect of the present invention, a device for determining a crystal rod pixel diameter is provided, the device comprising: an acquisition unit, used to acquire an image inside a single crystal furnace, wherein the image inside the single crystal furnace displays an outer aperture of the crystal rod, and a plurality of regions are preset in the image inside the single crystal furnace; a determination unit, used to determine the quantity distribution of aperture points of the outer aperture of the crystal rod in each of the plurality of regions; a screening unit, used to screen out a target region from the plurality of regions according to the quantity distribution; and a fitting unit, used to fit the aperture points in the target region to obtain the pixel diameter of the crystal rod.

[0017] The present invention provides a method and device for determining the pixel diameter of a crystal rod, the method comprising: obtaining an image inside a single crystal furnace, wherein the image inside the single crystal furnace displays an outer aperture of a crystal rod, and multiple regions are preset in the image inside the single crystal furnace; determining the number distribution of aperture points of the outer aperture of the crystal rod in each of the multiple regions; selecting a target region from the multiple regions according to the number distribution; fitting the aperture points in the target region to obtain the pixel diameter of the crystal rod. The method solves the technical problem in the prior art that all aperture points appearing in the image are often used to fit a circle at any time, resulting in low accuracy of the fitting circle result. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 It is a schematic diagram of the aperture in the image of the single crystal furnace;

[0020] Figure 2 is a flow chart of a method for determining a crystal rod pixel diameter according to an embodiment of the present invention;

[0021] Figure 3 to Figure 4 It is a schematic diagram of different states of aperture points in different crystal pulling links;

[0022] Figures 5 and 6 is a schematic diagram of multiple sector-shaped areas according to an embodiment of the present invention;

[0023] Figure 7 is a schematic diagram of a polar coordinate candidate region according to an embodiment of the present invention;

[0024] Figure 8 It is a schematic diagram of the structure of each component of the single crystal furnace;

[0025] Fig. 9 It is a schematic diagram of the simplified model inside the single crystal furnace;

[0026] Figure 10 to Figure 11 This is the optical path diagram of the single crystal furnace camera;

[0027] Fig.12 Schematic diagram of a device for determining a crystal rod pixel diameter according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to make the above and other features and advantages of the present invention more clear, the present invention is further described below in conjunction with the accompanying drawings. It should be understood that the specific embodiments given herein are for the purpose of explaining to those skilled in the art and are only exemplary and not restrictive.

[0029] In the following description, many specific details are set forth to provide a thorough understanding of the present invention. However, it is apparent to one of ordinary skill in the art that specific details need not be adopted to practice the present invention. In other cases, well-known steps or operations are not described in detail to avoid obscuring the present invention.

[0030] Embodiment 1

[0031] The present invention provides a method for determining the pixel diameter of a crystal rod, such as Figure 2 As shown, it is characterized in that the method includes:

[0032] Step S21, acquiring an image inside a single crystal furnace, wherein the image inside the single crystal furnace displays an outer aperture of a crystal rod, and a plurality of regions are preset in the image inside the single crystal furnace.

[0033] Specifically, in this solution, the controller of the intelligent single crystal furnace or other equipment with data processing function can be used as the execution subject of the method of this solution. The single crystal furnace is provided with a camera device for collecting images in the single crystal furnace. It should be noted that during the process of pulling crystals in the single crystal furnace, the above-mentioned crystal rod outer aperture is formed at the junction of the crystal rod and the liquid silicon surface.

[0034] Step S23, determining the quantity distribution of the aperture points of the outer aperture of the crystal rod in each of the multiple regions.

[0035] Specifically, in this solution, the diameter of the crystal rod will change during the different processes of crystal pulling. From the perspective of image acquisition by the camera, the aperture will be blocked by the crystal rod or other objects, thus showing different states at different time points. Figure 3 to Figure 4 In different crystal pulling stages, the distribution number of aperture points in multiple areas of the image in the single crystal furnace is different.

[0036] Step S25, selecting a target area from multiple areas according to quantity distribution.

[0037] Step S27, fitting the aperture points in the target area to obtain the pixel diameter of the crystal rod.

[0038] Specifically, the present solution is different from the prior art in that all aperture points in the image are fitted. Instead, the target area of ​​the best aperture point is screened out according to the distribution of the number of aperture points in multiple areas, and the aperture points in the target area are fitted to obtain the pixel diameter of the crystal rod. Through the above method, in the fitting process, the aperture points with poor effects are filtered out and the best aperture points are retained, thereby greatly improving the accuracy of the fitting circle. The present solution uses aperture points in different areas of the image for fitting under different circumstances, which solves the technical problem that in the prior art, all aperture points in the image in the single crystal furnace are often used to fit the circle at any time, resulting in low accuracy of the fitting circle result.

[0039] It should be noted here that when the diameter of the crystal rod is at a critical point, after this critical point, the aperture shape will change significantly. The existing technology uses all the aperture points in the image at the current moment to fit, which will cause the fitting circle to jitter after the current moment. This solution can effectively solve this problem by selecting the aperture point with the best effect to fit the circle at all times.

[0040] Optional, combined Figure 5 , the multiple regions include a first sector region, a second sector region and a third sector region with the same center, the third sector region is located between the first sector region and the second sector region, and the radius of the first sector and the third sector region is the same and greater than the radius of the second sector. The second sector is used to extract the aperture point in the middle region. The middle region is only effective when the crystal rod size is small, so the radius of the middle sector is relatively small. When the crystal rod diameter is relatively large, the aperture point in the middle sector region is not the best aperture point. Therefore, this scheme should try to obtain as few aperture points in the middle sector region as possible, so the sector with a smaller radius can effectively shield the redundant aperture points.

[0041] It should be noted here that, combined with Figure 5, the aperture diameter of the crystal rod is not the size of the equal diameter stage at the beginning, but gradually increases in the shoulder release stage. The diameter of the crystal rod also needs to be calculated in the shoulder release stage. From the perspective of the single crystal furnace camera device, when the aperture diameter of the crystal rod is small, the number of aperture points falling in the first fan-shaped area and the third fan-shaped area is relatively small, and the number of aperture points falling in the second fan-shaped area is relatively large. From the perspective of the single crystal furnace camera device, as the aperture diameter of the crystal rod continues to increase, the number of aperture points falling in the first fan-shaped area and the third fan-shaped area is relatively large, and the number of aperture points falling in the second fan-shaped area is relatively small. Therefore, this scheme screens out reliable fitting aperture points according to the distribution of the number of multiple fan-shaped areas of the aperture in different processes of crystal pulling, thereby improving the accuracy of crystal rod circle fitting.

[0042] Among them, step S25, screening out the target area from multiple areas according to the quantity distribution, includes:

[0043] Step S251, when the number of aperture points in the first sector-shaped area and the third sector-shaped area is greater than the standard number, the first sector-shaped area and the third sector-shaped area are determined as target areas.

[0044] Specifically, in this solution, if there are more first fan-shaped areas and third fan-shaped areas, it means that only using the aperture points of the first fan-shaped areas and the third fan-shaped areas to fit the circle can ensure the accuracy of the fitting circle.

[0045] Step S252: When the number of aperture points in the first sector-shaped area or the third sector-shaped area is less than or equal to the standard number, the first sector-shaped area, the second sector-shaped area and the third sector-shaped area are determined as target areas.

[0046] Specifically, in this solution, at a certain stage of crystal pulling, if the number of aperture points in the first fan-shaped area and the third fan-shaped area is small, it means that the first fan-shaped area and the third fan-shaped area alone are not enough to ensure the accuracy of the fitting circle. Therefore, this solution sets the aperture points in the first fan-shaped area, the second fan-shaped area and the third fan-shaped area as reliable points for circle fitting, thereby ensuring the accuracy of the fitting circle. Optionally, the above standard number can be 25.

[0047] Therefore, through this solution, at any stage of crystal pulling, the best aperture point can be always selected for fitting according to the distribution of aperture points in multiple sector-shaped areas.

[0048] Optionally, step S27, fitting the aperture points in the target area, includes:

[0049] Step S271, fitting the aperture points in the target area to obtain an initial fitting circle.

[0050] Specifically, the target area in the above step S271 can be the first sector + the third sector, or the first sector + the second sector + the third sector. After the target area is determined, the initial fitting circle is obtained by fitting according to the aperture points in the target area.

[0051] Step S272: performing gradient descent processing on the initial fitting circle using the aperture points in the target area to obtain a first fitting circle.

[0052] Specifically, after obtaining the initial fitting circle, the present solution may use the initial fitting circle as an initial value to fine-tune the initial fitting circle to obtain the first fitting circle.

[0053] The following is an explanation of the details of the gradient descent in this scheme:

[0054] The step size of the gradient descent method is 0.2, the iteration is 400 rounds, and the loss function is the distance from each point in the target area to the current circle.

[0055] The i-th point (x i ,y i ) to the current circle center point (C x ,C y ) horizontal axis C x The partial derivative gx i for:

[0056]

[0057] The i-th point (x i ,y i ) to the current circle center point (C x ,C y ) Vertical axis C y The partial derivative of i for:

[0058]

[0059] The i-th point (x i ,y i ) for the current circle radius C r The partial derivative gr i for:

[0060]

[0061] The center of the circle, the radius, and the total gradient are

[0062]

[0063] Each iteration (C x ,C y ,C r) to update

[0064]

[0065] Step S273: determine a first distance range according to the radius of the first fitting circle.

[0066] Step S274 , selecting from the aperture points in the target area a first aperture point combination whose distance from the center of the first fitting circle is within a first distance range.

[0067] Specifically, the first distance range may be 0.99*the radius of the first fitting circle to 1.01 the radius of the first fitting circle. This solution may screen out the first aperture point combination within the first distance range from the target area.

[0068] Step S275: performing gradient descent processing on the first fitting circle using the first aperture point combination to obtain a target fitting circle.

[0069] Specifically, this solution can use the aperture points in the first aperture point combination, take the center and radius of the first fitting circle as initial values, and use the gradient descent method to fine-tune the first fitting circle to obtain the target fitting circle.

[0070] It should be noted that this solution first determines the target area, and then continues to screen the aperture points in the target area for fitting. Compared with the fitting circle technology in the prior art, the accuracy of the fitting circle is greatly improved.

[0071] The following is a description of the technical effects of fitting + gradient descent and multiple fitting in the above steps: First, each fitting is a least square fitting + gradient descent fine-tuning. The least square method is easily disturbed by outliers. The gradient descent fine-tuning process makes the circle better fit the most correct points on the basis of least squares and reduces the interference of outliers. Multiple fitting, after each fitting circle, legal points will be screened according to the fitting circle to remove outliers. In the scheme, each time the outliers are removed before fitting the circle, the fitting result will be more accurate.

[0072] Optionally, step S271, fitting the aperture points in the target area to obtain an initial fitting circle, includes:

[0073] Step S2711, when the target area is the first sector-shaped area and the third sector-shaped area, directly fit the aperture points in the target area to obtain an initial fitting circle.

[0074] Specifically, when the number of the first fan-shaped area and the third fan-shaped area is greater than 25 and the target area is the first fan-shaped area and the third fan-shaped area, this solution directly sets the first fan-shaped area and the third fan-shaped area as fitting credible points, and directly fits the aperture points in the first fan-shaped area and the third fan-shaped area to obtain the initial fitting circle, and then performs the above-mentioned steps S273 to S275.

[0075] Optionally, step S271, fitting the aperture points in the target area to obtain an initial fitting circle, includes:

[0076] Step S2712: When the target area is the first fan-shaped area, the second fan-shaped area, and the third fan-shaped area, the aperture points in the target area are fitted to obtain a second fitting circle.

[0077] Specifically, when the number of aperture points in the first fan-shaped area or the third fan-shaped area is less than or equal to 25, this solution determines the first fan-shaped area, the second fan-shaped area and the third fan-shaped area as credible points of the fitting circle, and then fits to obtain a second fitting circle.

[0078] Step S2713: determine a second distance range according to the radius of the second fitting circle.

[0079] Step S2714: Screening out from the second sector-shaped area a second aperture point combination whose distance from the center of the second fitting circle is within the second distance range.

[0080] The second distance range may be 0.99*the radius of the second fitting circle to 1.01 the radius of the second fitting circle. The present solution may select a second aperture point combination within the second distance range from the second sector.

[0081] Step S2715, fitting the second aperture point combination, the aperture points of the first fan-shaped area, and the aperture points of the third fan-shaped area to obtain the initial fitting circle.

[0082] Optionally, the aperture point in the target area in step S27 is obtained by the following steps:

[0083] Step S2701, converting the coordinate system of each pixel in the target area to obtain a polar coordinate diagram of the target area.

[0084] Step S2702, determining a candidate region from the polar coordinate map according to the composition of the target region.

[0085] Step S2703, traverse the candidate areas to obtain all aperture points in the target area.

[0086] Specifically, after determining the target area, the present scheme can obtain a polar coordinate diagram of the target area by polar coordinate conversion, and then determine the candidate area from the polar coordinate diagram, and finally obtain all the aperture points in the target area from the candidate area. It should be noted that in the present embodiment, since the target area is screened according to the distribution of the number of aperture points, the present scheme then performs polar coordinate conversion in the target area, so as to obtain each aperture point in the target area. It should be noted that, unlike the prior art in which limited point sampling and then fitting are performed, the polar coordinate conversion scheme of the present scheme can perform dense sampling to obtain each aperture point in the target area. Therefore, the combination of the two technical points of "determining the target area according to the distribution of aperture points" and "performing polar coordinate conversion on the target area" in the present technical scheme can achieve that the aperture points found can ensure both the accuracy of the aperture points required for the fitting circle and the comprehensiveness of the aperture points required for the fitting circle, thereby greatly improving the accuracy of the crystal pixel diameter generation.

[0087] Optionally, step S2702 determines the candidate region from the polar coordinate map according to the composition of the target region, including:

[0088] Step S27021: when the target area is the first sector-shaped area and the third sector-shaped area, the coordinates of two vertices of the first sector-shaped area and the coordinates of two vertices of the third sector-shaped area are obtained.

[0089] Step S27022: determining a candidate region according to the coordinates of the two vertices of the first sector-shaped region and the coordinates of the two vertices of the third sector-shaped region.

[0090] Specific, combined Figure 6 , when the target area is the first sector area and the third sector area, this solution obtains the coordinates of the two vertices A and C of the first sector area and the coordinates of the two vertices B and D of the third sector area. Then, the candidate area to be traversed is determined in the polar coordinate diagram according to the coordinates of A, C, B, and D. Figure 7 , the area between AC and BD is the determined candidate area. It should be noted that in this solution, not all parts in the polar coordinate diagram are traversed, but four vertex coordinates are obtained according to the composition of the target area, and then the candidate area is determined in the polar coordinate diagram according to the coordinates of the four vertices, thereby narrowing the traversal range in the candidate area and greatly improving the efficiency of aperture point determination.

[0091] Optionally, the image in the single crystal furnace further includes the circular opening at the lower end of the guide tube, the melt surface, and the reflection of the circular opening at the lower end of the guide tube in the melt surface, wherein after obtaining the pixel diameter of the crystal rod in step S27, the method further includes:

[0092] Step S29, obtaining a first vertical distance D from the lower end of the guide tube to the outer aperture of the crystal rod through the circular opening at the lower end of the guide tube, the melt surface, and the reflection of the circular opening at the lower end of the guide tube in the melt surface.

[0093] It should be noted here that the first vertical distance D from the lower end of the guide tube to the outer aperture of the crystal rod is the liquid mouth distance. This scheme can estimate the liquid mouth distance according to the existing single crystal distance measurement method based on the circular opening at the lower end of the guide tube, the melt surface and the reflection of the circular opening at the lower end of the guide tube in the melt surface, or it can be obtained by the existing laser distance measurement method.

[0094] Step S31, determining the pixel diameter Dbp of the circular opening at the lower end of the guide tube through the image inside the single crystal furnace.

[0095] Step S33, obtaining the actual distance Dbr of the circular opening at the lower end of the guide tube.

[0096] Step S35, obtaining a first distance Za between the opening diameter of the lower end of the guide tube and the optical axis through the image in the single crystal furnace, and obtaining a first pixel length za′ of the first distance on the image plane.

[0097] Step S37, obtaining a second distance Zb between the diameter of the crystal rod and the optical axis through the image in the single crystal furnace, and obtaining a second pixel length zb′ of the second distance on the image plane.

[0098] Step S39, obtaining the actual diameter distance Dr of the crystal rod pixel according to the pixel diameter Dp of the crystal rod, the first pixel length za, the second pixel length zb, the actual distance Dbr of the circular opening at the lower end of the guide tube, the pixel diameter Dbp of the circular opening at the lower end of the guide tube, the first vertical distance D, the camera focal length parameter f, and the camera pixel parameter k.

[0099] Specifically, in this solution, the actual diameter distance Dr of the above crystal rod pixel can be obtained according to the following formula.

[0100]

[0101] Combine the following Figure 8 to Figure 11 , the derivation of the above calculation formula of Dr is explained as follows:

[0102] The overall structure of the camera, guide tube, and crystal rod is as follows: Figure 8 As shown, the image captured by the camera at this position is as follows Figure 4 shown. Figure 8 The simplified model is as follows Fig. 9 As shown, the camera optical path diagram is as follows Figure 10-11 shown.

[0103] DA and DB are the diameter Dbr and the distance from Dr to the focal plane of the camera respectively. f is the focal length parameter of the camera, that is, the distance from the focal plane of the camera to the image plane (unit: mm). They are the lengths of diameter Dbr and diameter Dr on the image plane (in mm) respectively.

[0104] Depend on Fig.10 The triangle in the

[0105]

[0106]

[0107] Then the vertical distance DB-DA between the two diameters is

[0108]

[0109] Where Δd≈D, in order to further improve the accuracy of measurement, correction can be performed in the following way. The distances of diameter Dbr and diameter Dr from the optical axis are Za and Zb respectively, and za and zb are the projection distances of Za and Zb on the image plane.

[0110] Combination Fig.11 , the following relationship can be obtained from the triangle:

[0111]

[0112]

[0113] The height difference between the two diameters in the direction parallel to the image plane is:

[0114]

[0115] The distance between the guide tube plane and its image plane is d, and Fig.11 The Pythagorean theorem for a red right triangle is:

[0116] In summary

[0117]

[0118] It should be noted that in the actual calculation process, it is impossible to directly obtain the image position za, zb, can only obtain Dbp (i.e., the pixel diameter of the circular opening at the lower end of the guide tube), Dp (i.e., the pixel diameter of the crystal rod), za′ (i.e., the first pixel length za′ of the first distance in the image plane), zb′ (i.e., the second pixel length zb′ of the second distance in the image plane) in the image coordinate system. With Dbp, There are the following coefficient relationships with Dp, za and za′, and zb and zb′:

[0119] Dbp=k×Dbp

[0120] Dp=k×Dp

[0121] za=k×za′

[0122] zb=k×zb′

[0123] Where k is the focal length of the camera pixel, in millimeters per pixel (mm / pixel). Substituting the above coefficient relationship into

[0124]

[0125] The calculation formula of Dr can be derived from the above.

[0126] It should be noted here that in the above steps S21 to S39, an embodiment of a method for calculating the actual diameter of a crystal rod is described. Different from the existing scheme for estimating the diameter of a crystal rod, this embodiment first determines the target area, and then continues to screen the aperture points in the target area from the target area for fitting. Compared with the fitting circle technology in the prior art, the accuracy of fitting the circle and generating the pixel diameter of the crystal rod is greatly improved. Then, this embodiment calculates the absolute diameter of the crystal rod based on the highly accurate pixel diameter of the crystal rod, thereby ensuring the accuracy of the crystal rod diameter measurement.

[0127] Embodiment 2

[0128] The present application also provides a device for determining the pixel diameter of a crystal rod. The device can be arranged in an intelligent single crystal furnace and can also be used to execute the method of the first embodiment. Fig.12 The device includes: an acquisition unit 1110, used to acquire an image inside a single crystal furnace, wherein the image inside the single crystal furnace displays an outer aperture of a crystal rod, and a plurality of regions are preset in the image inside the single crystal furnace; a determination unit 1120, used to determine the quantity distribution of aperture points of the outer aperture of the crystal rod in each of the plurality of regions; a screening unit 1130, used to screen out a target region from the plurality of regions according to the quantity distribution; and a fitting unit 1140, used to fit the aperture points in the target region to obtain the pixel diameter of the crystal rod.

[0129] The multiple units of the above device solve the technical problem that in the prior art, all aperture points in the image in the single crystal furnace are often used to fit the circle at any time, resulting in low accuracy of the fitting circle result.

[0130] It should be understood that the specific features, operations and details described hereinabove about the method of the present invention may also be similarly applied to the device and system of the present invention, or, vice versa. In addition, each step of the method of the present invention described above may be performed by the corresponding parts or units of the device or system of the present invention.

[0131] It should be understood that each module / unit of the device of the present invention can be implemented in whole or in part by software, hardware, firmware or a combination thereof. Each of the modules / units can be embedded in the processor of the computer device in the form of hardware or firmware or independent of the processor, or can be stored in the memory of the computer device in the form of software for the processor to call to perform the operation of each module / unit. Each of the modules / units can be implemented as an independent component or module, or two or more modules / units can be implemented as a single component or module.

[0132] In one embodiment, a computer device is provided, which includes a memory and a processor, wherein the memory stores computer instructions executable by the processor, and the computer instructions instruct the processor to perform each step of the method of the embodiment of the present invention when executed by the processor. The computer device can be a server, a terminal, or any other electronic device with necessary computing and / or processing capabilities in a broad sense. In one embodiment, the computer device may include a processor, a memory, a network interface, a communication interface, etc. connected through a system bus. The processor of the computer device can be used to provide necessary computing, processing and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and an internal memory. An operating system, a computer program, etc. may be stored in or on the non-volatile storage medium. The internal memory can provide an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface and the communication interface of the computer device can be used to connect and communicate with external devices through a network. The steps of the method of the present invention are executed by the processor.

[0133] The present invention may be implemented as a computer-readable storage medium having a computer program stored thereon, which causes the steps of the method of an embodiment of the present invention to be executed when executed by a processor. In one embodiment, the computer program is distributed on a plurality of computer devices or processors coupled to a network so that the computer program is stored, accessed and executed in a distributed manner by one or more computer devices or processors. A single method step / operation, or two or more method steps / operations, may be performed by a single computer device or processor or by two or more computer devices or processors. One or more method steps / operations may be performed by one or more computer devices or processors, and one or more other method steps / operations may be performed by one or more other computer devices or processors. One or more computer devices or processors may perform a single method step / operation, or perform two or more method steps / operations.

[0134] It will be appreciated by those skilled in the art that the method steps of the present invention may be performed by instructing related hardware such as a computer device or a processor through a computer program, and the computer program may be stored in a non-temporary computer-readable storage medium, which causes the steps of the present invention to be performed when the computer program is executed. Depending on the circumstances, any reference to memory, storage, database, or other media herein may include non-volatile and / or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0135] The various technical features described above can be combined arbitrarily. Although all possible combinations of these technical features are not described, any combination of these technical features should be considered to be covered by this specification as long as there is no contradiction in such combination.

[0136] 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 they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do 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 method for determining a crystal rod pixel diameter, characterized in that: The method comprises: An image inside a single crystal furnace is acquired, wherein the image inside the single crystal furnace displays an outer aperture of a crystal rod, and a plurality of regions are preset in the image inside the single crystal furnace; wherein the plurality of regions include a first fan-shaped region, a second fan-shaped region, and a third fan-shaped region having the same center, wherein the third fan-shaped region is located between the first fan-shaped region and the second fan-shaped region, wherein the radius of the first fan-shaped region is the same as that of the third fan-shaped region, and the radius of the first fan-shaped region and the third fan-shaped region is greater than that of the second fan-shaped region; Determine the distribution of the number of aperture points of the outer aperture of the crystal rod in each of the multiple regions; Filtering a target area from the multiple areas according to the quantity distribution; Fitting the aperture points in the target area to obtain the pixel diameter of the crystal rod; Wherein, selecting a target area from the multiple areas according to the quantity distribution includes: When the number of aperture points in the first sector-shaped area and the third sector-shaped area is greater than a standard number, the first sector-shaped area and the third sector-shaped area are determined as target areas; When the number of aperture points in the first fan-shaped area or the third fan-shaped area is less than or equal to the standard number, the first fan-shaped area, the second fan-shaped area and the third fan-shaped area are determined as target areas.

2. The method according to claim 1, characterized in that Fitting the aperture points in the target area includes: Fitting the aperture points in the target area to obtain an initial fitting circle; Using the aperture points in the target area to perform gradient descent processing on the initial fitting circle to obtain a first fitting circle; determining a first distance range according to the radius of the first fitting circle; Filtering from the aperture points in the target area to obtain a first aperture point combination whose distance from the center of the first fitting circle falls within the first distance range; The first aperture point combination is used to perform gradient descent processing on the first fitting circle to obtain a target fitting circle.

3. The method according to claim 2, characterized in that Fitting the aperture points in the target area to obtain an initial fitting circle includes: When the target area is the first sector-shaped area and the third sector-shaped area, the aperture points in the target area are directly fitted to obtain an initial fitting circle.

4. The method according to claim 2, characterized in that: Fitting the aperture points in the target area to obtain an initial fitting circle includes: When the target area is a first fan-shaped area, a second fan-shaped area, and a third fan-shaped area, fitting the aperture points in the target area to obtain a second fitting circle; determining a second distance range according to the radius of the second fitting circle; Screening out from the second sector-shaped area a second aperture point combination whose distance from the center of the second fitting circle falls within the second distance range; The initial fitting circle is obtained by fitting the second aperture point combination, the aperture points of the first fan-shaped area, and the aperture points of the third fan-shaped area.

5. The method according to claim 1, characterized in that The aperture point in the target area is obtained by the following steps: Convert each pixel in the target area into a coordinate system to obtain a polar coordinate diagram of the target area; Determining a candidate region from the polar coordinate map according to the composition of the target region; All aperture points in the target area are obtained by traversing the candidate areas.

6. The method according to claim 5, characterized in that Determining a candidate region from the polar coordinate map according to the composition of the target region includes: When the target area is a first fan-shaped area and a third fan-shaped area, coordinates of two vertices of the first fan-shaped area and coordinates of two vertices of the third fan-shaped area are acquired; The candidate area is determined according to the coordinates of the two vertices of the first fan-shaped area and the coordinates of the two vertices of the third fan-shaped area.

7. The method according to claim 1, characterized in that The single crystal furnace image also includes a circular opening at the lower end of the guide tube, a melt surface, and a reflection of the circular opening at the lower end of the guide tube in the melt surface. After obtaining the pixel diameter of the crystal rod, the method further includes: A first vertical distance from the lower end of the guide tube to the outer aperture of the crystal rod is obtained through the circular opening at the lower end of the guide tube, the melt surface, and the reflection of the circular opening at the lower end of the guide tube in the melt surface; Determine the pixel diameter of the circular opening at the lower end of the guide tube through the image inside the single crystal furnace; Obtaining the actual distance of the circular opening at the lower end of the guide tube; Acquire a first distance between the opening diameter of the lower end of the guide tube and the optical axis through the image in the single crystal furnace, and obtain a first pixel length of the first distance on the image plane; Acquire a second distance between the diameter of the crystal rod and the optical axis through the image in the single crystal furnace, and obtain a second pixel length of the second distance on the image plane; The actual diameter distance of the crystal rod pixel is obtained according to the pixel diameter of the crystal rod, the first pixel length, the second pixel length, the actual distance of the circular opening at the lower end of the guide tube, the pixel diameter of the circular opening at the lower end of the guide tube, the first vertical distance, the camera focal length parameter and the camera pixel parameter.

8. A device for determining a crystal rod pixel diameter, characterized in that: The device comprises: An acquisition unit is used to acquire an image inside a single crystal furnace, wherein the image inside the single crystal furnace displays an outer aperture of a crystal rod, and a plurality of regions are preset in the image inside the single crystal furnace; wherein the plurality of regions include a first fan-shaped region, a second fan-shaped region, and a third fan-shaped region with the same center, wherein the third fan-shaped region is located between the first fan-shaped region and the second fan-shaped region, wherein the radius of the first fan-shaped region is the same as that of the third fan-shaped region, and the radius of the first fan-shaped region and the third fan-shaped region is greater than that of the second fan-shaped region; A determination unit, configured to determine the quantity distribution of the aperture points of the outer aperture of the crystal rod in each of the multiple regions; A screening unit is used to screen out a target area from the multiple areas according to the quantity distribution; wherein screening out the target area from the multiple areas according to the quantity distribution includes: When the number of aperture points in the first sector-shaped area and the third sector-shaped area is greater than a standard number, the first sector-shaped area and the third sector-shaped area are determined as target areas; When the number of aperture points in the first sector-shaped area or the third sector-shaped area is less than or equal to the standard number, the first sector-shaped area, the second sector-shaped area and the third sector-shaped area are determined as target areas; The fitting unit is used to fit the aperture points in the target area to obtain the pixel diameter of the crystal rod.

Images