Apparatus and method for separating wafers, and storage medium

Abstract

Provided in one or more embodiments of the present specification is an apparatus for separating wafers. The apparatus comprises: a container, which is configured to accommodate a separation solution and a group of wafers to be separated, said group comprising a plurality of wafers adhered together; an image acquisition unit, which is configured to acquire an image of said group from at least one angle; a processing unit, which is configured to determine a separation parameter on the basis of the image of said group from the at least one angle; and a separation unit, which is configured to act on the container or the separation solution on the basis of the separation parameter, so as to separate the plurality of wafers adhered together.

Description

An apparatus, method, and storage medium for stripping wafers. Technical Field

[0001] This specification relates to the field of wafer stripping, and more particularly to an apparatus, method, and storage medium for stripping wafers. Background Technology

[0002] After surface polishing, smooth wafers may contain water on their surface during cleaning and other processing. When multiple wafers come into contact, the water in the gaps between adjacent wafers is squeezed out, creating a vacuum at the contact point and making it difficult to separate the wafers. Currently, wafer separation is typically achieved in two ways: one is to slide adjacent wafers to separate them by offsetting them, but this method can easily damage the wafer surface; the other method is to soak the wafers for a long time so that water molecules gradually enter the gaps between adjacent wafers, thereby achieving wafer separation, but this method is more time-consuming. Summary of the Invention

[0003] One embodiment of this specification provides an apparatus for stripping wafers, comprising: a container configured to hold a stripping liquid and a wafer group to be stripped, the wafer group comprising a plurality of wafers adsorbed together; an image acquisition unit configured to acquire an image of the wafer group to be stripped from at least one angle; a processing unit configured to determine stripping parameters based on the image of the wafer group to be stripped from at least one angle; and a stripping unit configured to act on the container or the stripping liquid based on the stripping parameters to strip the plurality of wafers adsorbed together.

[0004] One embodiment of this specification also provides a method for stripping wafers, applied to a wafer stripping apparatus, the wafer stripping apparatus comprising: a container configured to hold a stripping liquid and a wafer group to be stripped, the wafer group comprising a plurality of wafers adsorbed together; an image acquisition unit configured to acquire an image of the wafer group to be stripped from at least one angle; a processing unit; and a stripping unit; the method being performed by the processing unit, comprising: determining stripping parameters based on the image of the wafer group to be stripped from at least one angle; and controlling the stripping unit to act on the container or the stripping liquid based on the stripping parameters to strip the plurality of wafers adsorbed together.

[0005] One embodiment of this specification also provides a computer-readable storage medium storing computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes a method for stripping wafers. The method is applied to a wafer stripping apparatus, the wafer stripping apparatus comprising: a container configured to hold a stripping liquid and a wafer group to be stripped, the wafer group comprising multiple wafers adsorbed together; an image acquisition unit configured to acquire an image of the wafer group to be stripped from at least one angle; and a stripping unit; the method comprising: determining stripping parameters based on the image of the wafer group to be stripped from at least one angle; and controlling the stripping unit to act on the container or the stripping liquid based on the stripping parameters to strip the multiple wafers adsorbed together. Attached Figure Description

[0006] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0007] Figure 1 is an exemplary frame diagram of a wafer stripping apparatus according to some embodiments of this specification;

[0008] Figure 2 is an exemplary flowchart of a method for removing a wafer according to some embodiments of this specification;

[0009] Figure 3 is an exemplary flowchart illustrating the determination of stripping parameters according to some embodiments of this specification;

[0010] Figure 4 is an exemplary flowchart illustrating the periodic determination of stripping parameters based on a detection cycle according to some embodiments of this specification;

[0011] Figure 5 is an exemplary flowchart illustrating the determination of image parameters of an image acquisition unit for the current detection cycle according to some embodiments of this specification;

[0012] Figure 6 is an exemplary flowchart illustrating the determination of image quality scores according to some embodiments of this specification;

[0013] Figure 7 is another exemplary flowchart illustrating the determination of image quality scores according to some embodiments of this specification;

[0014] Figure 8 is another exemplary flowchart illustrating the determination of image quality scores according to some embodiments of this specification;

[0015] Figure 9 is another exemplary flowchart illustrating the determination of an image's quality score according to some embodiments of this specification. Detailed Implementation

[0016] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0017] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0018] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0019] This specification provides an apparatus for stripping wafers, comprising a container, an image acquisition unit, a processing unit, and a stripping unit. The container holds a stripping liquid and a group of wafers to be stripped. The image acquisition unit acquires an image of the wafer group from at least one angle. The processing unit determines stripping parameters based on the image from at least one angle. The stripping unit acts on the container or stripping liquid based on the stripping parameters to strip multiple wafers adsorbed together. In some embodiments, the stripping unit may include an ultrasonic vibrator. The stripping parameters include ultrasonic power and / or vibration time. When the stripping parameters are applied, the ultrasonic vibrator acts on the container or stripping liquid, causing the stripping liquid to penetrate the surface of the adsorbed wafers, thereby stripping multiple wafers adsorbed together.

[0020] The wafer stripping device provided in the embodiments of this specification can efficiently strip multiple wafers adsorbed together while protecting the wafer surface from damage; moreover, the stripping process is a non-contact operation, which can keep the wafer surface clean while saving the time cost and auxiliary material consumption of re-cleaning.

[0021] Figure 1 is an exemplary frame diagram of a wafer stripping apparatus according to some embodiments of this specification.

[0022] As shown in Figure 1, the wafer stripping apparatus 100 (hereinafter referred to as apparatus 100) includes a container 110, an image acquisition unit 120, a processing unit 130, and a stripping unit 140. In some embodiments, after the wafer is stripped from the ingot, it undergoes surface polishing. During the cleaning and other processing after surface polishing, water remains on the wafer surface. When multiple wafers come into contact, the water in the gap between adjacent wafers is squeezed out, creating a vacuum at the contact point, making it difficult to separate the wafers. At this time, apparatus 100 can be used to strip multiple wafers that are adsorbed together.

[0023] Container 110 is configured to hold a stripping solution and a wafer assembly to be stripped. The stripping solution is a liquid used to strip the wafer assembly. In some embodiments, the stripping solution may include water or other solutions. Other solutions may include aqueous solutions with a substance that reduces the surface tension of water as a solute. That is, a certain amount of a substance that reduces the surface tension of water, such as alcohol or soap, is added to water. The wafer assembly to be stripped comprises multiple wafers adsorbed together. Container 110 can be used to hold the stripping solution, in which the wafer assembly to be stripped is placed (or immersed).

[0024] Image acquisition unit 120 is configured to acquire images of at least one angle of the wafer assembly to be stripped. In some embodiments, image acquisition unit 120 may include one or more cameras distributed at different locations on container 110, and the cameras may acquire images of at least one angle of the wafer assembly to be stripped. For example, when image acquisition unit 120 includes one camera, the camera is rotatably fixed at a certain position on container 110, and the camera can acquire images of different angles of the wafer assembly to be stripped as it rotates. As another example, when image acquisition unit 120 includes multiple cameras, the multiple cameras may be distributed at different locations on container 110, and each of the multiple cameras may acquire images of at least one angle of the wafer assembly to be stripped.

[0025] In some embodiments, the wafer group to be peeled in the container 110 or stripping fluid may be in various positions and / or angles. The image acquisition unit 120 acquires images of the wafer group to be peeled from multiple angles. Based on the images of the wafer group to be peeled from multiple angles, the parameter characteristics of the wafer group to be peeled (such as the adhesion characteristics described later) can be determined more accurately, thereby improving the accuracy of subsequent determination of peeling parameters, so as to better achieve the peeling of multiple wafers adsorbed together.

[0026] Processing unit 130 is configured to determine stripping parameters based on an image of at least one angle of the wafer assembly to be stripped. Figure 2 is an exemplary flowchart of a wafer stripping method according to some embodiments of this specification. As shown in Figure 2, process 200 may include:

[0027] Step 210: Determine stripping parameters based on an image of at least one angle of the wafer assembly to be stripped. In some embodiments, step 210 may be performed by processing unit 130.

[0028] The peeling parameters refer to the operating parameters of the peeling unit 140 when peeling multiple wafers adsorbed together. In some embodiments, the peeling parameters may include ultrasonic power, vibration time, temperature control parameters, etc.

[0029] In some embodiments, the processing unit 130 may determine the adhesion characteristics (such as the number of wafers adsorbed together) of the wafer group to be peeled based on an image of at least one angle of the wafer group to be peeled, thereby determining the peeling parameters based on the adhesion characteristics, as described in Figure 3.

[0030] Figure 3 is an exemplary flowchart illustrating the determination of stripping parameters according to some embodiments of this specification. As shown in Figure 3, step 210 may include the following sub-steps:

[0031] Sub-step 211: Determine the adhesion characteristics of the wafer group to be peeled based on an image of at least one angle of the wafer group to be peeled.

[0032] Adhesion characteristics may include the number of wafers adsorbed together. In some embodiments, adhesion characteristics may also include wafer size, cracks, bubbles, adhesion thickness, etc. Wafer size refers to the size of an individual wafer in a group of wafers adsorbed together. In some embodiments, the individual wafers in a group of wafers adsorbed together may have the same or different sizes. Adhesion thickness refers to the total thickness of the wafers adsorbed together. Cracks refer to scratches or cracks on the wafer surface. Bubbles refer to bubbles between two adjacent wafers adsorbed together.

[0033] In some embodiments, the processing unit 130 may determine adhesion features based on the similarity between an image of at least one angle of the wafer group to be peeled and a standard image in a standard image set. The standard image set refers to a collection of images of multiple wafer groups to be peeled from multiple angles with different adhesion features. Each standard image in the standard image set has its corresponding angle and corresponding adhesion feature. For example, the adhesion feature corresponding to standard image B (e.g., images B1, B2, ..., Bn) at different angles (e.g., angles α1, α2, ..., αn) of wafer group A1 is C; the adhesion feature corresponding to standard image D (e.g., images D1, D2, ..., Dn) at different angles (e.g., angles α1, α2, ..., αn) of wafer group A2 is E; and the adhesion feature corresponding to standard image X (e.g., images X1, X2, ..., Xn) at different angles (e.g., angles α1, α2, ..., αn) of wafer group Am is Y. The collection of the above standard images constitutes the standard image set. In some embodiments, the standard image set and the adhesion feature corresponding to each standard image in the standard image set can be obtained experimentally.

[0034] In some embodiments, the processing unit 130 can perform pixel-by-pixel comparison between an image of at least one angle of the wafer assembly to be peeled and a standard image in a standard image set to obtain the similarity between the image of the wafer assembly to be peeled and the standard image at the corresponding angle, and use the adhesion feature corresponding to the standard image with the highest similarity as the adhesion feature of the wafer assembly to be peeled. For example, if the image acquisition unit 120 acquires an image of the wafer assembly to be peeled at one angle, the processing unit 130 can perform pixel-by-pixel comparison between this image and each standard image in the standard image set corresponding to the same angle to obtain the corresponding image similarity. In this case, the adhesion feature corresponding to the standard image with the highest similarity can be used as the adhesion feature of the wafer assembly to be peeled. In some embodiments, the similarity between the image of the wafer assembly to be peeled and the standard image can be calculated by performing pixel-by-pixel comparison between the image of the wafer assembly to be peeled and the standard image, and using the average of the differences between each pixel as the similarity between the two images.

[0035] In some embodiments, when the image acquisition unit 120 acquires images of the wafer assembly to be peeled from multiple angles, it can compare the images of the multiple angles with each standard image in the standard image set corresponding to the same angle pixel by pixel to obtain the corresponding image similarity. In some embodiments, the adhesion feature corresponding to the standard image with the highest similarity can be used as the adhesion feature of the wafer assembly to be peeled. Taking the image acquisition unit 120 acquiring images of the wafer assembly to be peeled from two angles (referred to as the first image and the second image for ease of description) as an example, the first image is compared with each standard image in the standard image set corresponding to the same angle pixel by pixel to obtain the corresponding image similarity, and the second image is compared with each standard image in the standard image set corresponding to the same angle pixel by pixel to obtain the corresponding image similarity. At this time, the adhesion feature corresponding to the standard image with the highest image similarity can be used as the adhesion feature of the wafer assembly to be peeled. In some embodiments, the adhesion features corresponding to the standard image with the highest similarity at each angle can be used as candidate adhesion features for the wafer group to be peeled off. If multiple candidate adhesion features are consistent, the candidate adhesion features are determined as the final adhesion features. If the candidate adhesion features are inconsistent, the candidate adhesion features with more consistency are determined as the final adhesion features. Taking the acquisition of images of the wafer group to be peeled from three angles by the image acquisition unit 120 (referred to as the first image, second image, and third image for ease of description) as an example, the first image is compared pixel by pixel with each standard image in the standard image set corresponding to the same angle to obtain the corresponding image similarity. Among them, the adhesion feature corresponding to the standard image 1 with the highest similarity is used as the candidate adhesion feature 1 of the wafer group to be peeled; the second image is compared pixel by pixel with each standard image in the standard image set corresponding to the same angle to obtain the corresponding image similarity. Among them, the adhesion feature corresponding to the standard image 2 with the highest similarity is used as the candidate adhesion feature 2 of the wafer group to be peeled; the third image is compared pixel by pixel with each standard image in the standard image set corresponding to the same angle to obtain the corresponding image similarity. Among them, the adhesion feature corresponding to the standard image 3 with the highest similarity is used as the candidate adhesion feature 3 of the wafer group to be peeled. If the adhesion features corresponding to candidate adhesion feature 1, candidate adhesion feature 2, and candidate adhesion feature 3 are the same, then the candidate adhesion feature can be determined as the final adhesion feature; if the adhesion features corresponding to candidate adhesion feature 1 and candidate adhesion feature 2 are the same, but not the same as candidate adhesion feature 3, then candidate adhesion feature 1 (or candidate adhesion feature 2) can be determined as the final adhesion feature.

[0036] Consistent adhesion characteristics refer to the fact that at least some parameters of the adhesion characteristics (at least one of the following: the number of wafers adsorbed together, wafer size, cracks, bubbles, adhesion thickness, etc.) are the same. For example, an image of the same wafer group to be peeled at angle α1 can determine the number of wafers adsorbed together and the adhesion thickness, while an image of the same wafer group to be peeled at angle α2 can determine the number of wafers adsorbed together, wafer size, cracks, and bubbles. In this case, both images can determine the number of wafers adsorbed together, and the number is the same. Therefore, the adhesion characteristics of the wafer group to be peeled can be said to be consistent at the two angles.

[0037] In some embodiments, adhesion features can be determined using an adhesion feature determination model. The adhesion feature determination model includes a machine learning model. The input to the machine learning model is an image of the wafer assembly to be peeled from at least one angle. The output is the adhesion features of the corresponding wafer assembly to be peeled.

[0038] A machine learning model can be obtained by training an initial machine learning model based on a large number of training samples with training labels. The training samples can include images of the wafer assembly to be peeled from at least one angle and the corresponding adhesion features of the wafer assembly, where the adhesion features of the wafer assembly to be peeled serve as the training labels. Specifically, the labeled training samples can be input into the initial machine learning model to obtain its output. The parameters of the initial machine learning model are updated through training until the intermediate machine learning model meets preset conditions, resulting in a well-trained machine learning model. These preset conditions can be that the loss function is less than a preset threshold, the loss function converges, or the number of training iterations reaches a preset threshold.

[0039] In some embodiments, the input to the machine learning model may also include light source data. The light source data may include the location and intensity of the light source. The quality of the image of the wafer assembly to be stripped acquired by the image acquisition unit 120 is related to the light source data. By also using the light source data as input to the machine learning model, the shooting angle can be determined more accurately, thus eliminating angle errors caused by the light source.

[0040] In some embodiments, training samples and labels can be obtained through various means such as historical data, networks, and human experience. For example, images of at least one angle of the wafer assembly to be peeled off and the corresponding adhesion characteristics of the wafer assembly can be determined by methods such as laser detection and microscopic observation.

[0041] Machine learning models can be used to determine adhesion features more efficiently and accurately.

[0042] Sub-step 212: Determine the peeling parameters based on the adhesion characteristics.

[0043] In some embodiments, the processing unit 130 may determine the peeling parameters based on adhesion characteristics. In some embodiments, the peeling parameters may be determined based on adhesion characteristics through a feature index parameter table.

[0044] In some embodiments, the feature index parameter table includes a preset correspondence between various adhesion features and their corresponding peeling parameters. In some embodiments, the feature index parameter table can be obtained experimentally. For example, based on the adhesion features, various peeling parameters are selected for experiments, and the peeling effect (such as the time taken, whether wafer damage occurs, etc.) is evaluated. The peeling parameter with the better peeling effect (shortest time taken and no wafer damage) is selected, a preset correspondence between adhesion features and peeling parameters is established, and the feature index parameter table is generated.

[0045] Step 220 involves applying a stripping solution or container based on stripping parameters to strip multiple wafers adsorbed together. In some embodiments, step 220 may be performed by a stripping unit 140. The stripping unit 140 is configured to apply a stripping solution or container based on stripping parameters to strip multiple wafers adsorbed together.

[0046] In some embodiments, the stripping unit 140 may include an ultrasonic vibrator, and the stripping parameters may include ultrasonic power and / or vibration time. The ultrasonic vibrator acts on the container 110 or the stripping fluid based on the ultrasonic power and / or vibration time to strip multiple wafers adsorbed together. In some embodiments, a pre-prepared stripping fluid may be stored in the container 110. Then, the container 110 or the stripping fluid within the container 110 is vibrated by the ultrasonic vibrator. As an example, several ultrasonic vibrators can be installed below the container 110 to vibrate the stripping fluid within the container 110, thereby allowing the stripping fluid to penetrate the surface of the adsorbed wafers and thus strip multiple wafers adsorbed together.

[0047] In some embodiments, ultrasonic power can affect the parameters of cavitation bubbles generated during ultrasonic operation, thereby affecting the wafer stripping effect. A cavitation bubble is a tiny bubble formed in a liquid under certain conditions. In some embodiments, cavitation bubbles can be tiny bubbles formed in the stripping fluid when the ultrasonic vibrator is operating. The parameters of the cavitation bubble may include, but are not limited to, the formation time, size, impact intensity, number, and uniformity of the cavitation bubble. In some embodiments, when the ultrasonic vibrator is based on low-frequency ultrasonic vibration, the cavitation bubble formation time in the stripping fluid is longer, the cavitation bubble size is larger, and the impact force generated by the cavitation bubble is stronger, but the number of cavitation bubbles is smaller and the uniformity is not high. As the ultrasonic power increases, the cavitation bubble formation time in the stripping fluid is shorter, the cavitation bubble size is smaller but the number is greater, and the uniformity and penetration of the cavitation bubbles are higher. Therefore, the ultrasonic power of the ultrasonic vibrator can be reasonably determined according to the adhesion characteristics of the wafer group to be stripped (such as the number and thickness of wafers adsorbed together), thereby improving the stripping efficiency.

[0048] In some embodiments, the stripping unit 140 may further include a temperature control device, wherein the stripping parameters include temperature control parameters, and the temperature control device is configured to adjust the temperature of the stripping solution based on the temperature control parameters. In some embodiments, the temperature of the stripping solution can affect the intermolecular forces of water molecules during the stripping process. By setting the temperature control device to reasonably adjust the temperature of the stripping solution, the stripping efficiency and stripping effect can be improved.

[0049] It should be noted that in some embodiments, if there are damaged wafers (especially broken wafers) in the wafer group to be stripped, the stripping can be suspended to prevent the damaged wafers from damaging other good wafers.

[0050] It should be noted that the above description of process 200 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 200 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0051] In some embodiments, the apparatus 100 may further include a communication unit configured to perform data communication between the image acquisition unit 120, the processing unit 130, and the stripping unit 140. For example, the processing unit 130 may send a command to the stripping unit 140 via the communication unit to cause the stripping unit 140 to start operating.

[0052] In some embodiments, when the stripping unit 140 strips multiple wafers adsorbed together, the following may occur: when the stripping unit 140 acts on the container 110 or the stripping liquid, the stripping liquid enters the surfaces of any two adjacent wafers adsorbed together, thereby stripping all the wafers adsorbed together (i.e., stripping the multiple wafers adsorbed together into multiple single wafers), thus completing the stripping operation. During this process, the stripping unit 140 operates under the same stripping parameters. In other embodiments, in addition to the above situation, partial stripping may also occur during the stripping process, that is, stripping multiple wafers adsorbed together into at least one single wafer and one or more new groups of wafers to be stripped. The number of wafers in the newly generated groups of wafers to be stripped is less than the number of wafers in the original group of wafers to be stripped (the group of wafers not stripped). In this case, the adhesion characteristics of the newly generated groups of wafers to be stripped change, and correspondingly, the stripping parameters also change. If the stripping unit 140 still operates according to the stripping parameters corresponding to the original adhesion characteristics, the stripping efficiency and stripping effect will be reduced. Therefore, in order to improve the peeling efficiency and peeling effect, the processing unit 130 can periodically determine the peeling parameters, and the peeling unit 140 operates based on the peeling parameters determined in each cycle. For details, please see below.

[0053] In some embodiments, the processing unit 130 may periodically determine the stripping parameters based on a detection cycle. The detection cycle is related to the initial characteristics of the wafer assembly to be stripped, including the thickness of the wafer assembly. In some embodiments, the detection cycle may be positively correlated with the thickness of the wafer assembly. For example, the thicker the wafer assembly to be stripped, the longer the detection cycle can be, thereby reducing the time cost of detection; the thinner the wafer assembly to be stripped (indicating tighter adhesion between wafers or fewer wafers adsorbed, making stripping more difficult), the shorter the detection cycle can be, thereby determining more reasonable stripping parameters and improving stripping efficiency.

[0054] Figure 4 is an exemplary flowchart illustrating the periodic determination of stripping parameters based on a detection cycle according to some embodiments of this specification. As shown in Figure 4, process 400 includes:

[0055] Step 410: Based on the image of at least one angle of the wafer group to be peeled in the previous detection cycle, determine the image parameters of the image acquisition unit for the current detection cycle. The image parameters of the image acquisition unit include the image shooting angle. In some embodiments, in the previous detection cycle, the image acquisition unit 120 acquires images of at least one angle of the wafer group to be peeled. The quality of the images of the wafer group to be peeled at different angles will certainly be different. If the image quality of the wafer group to be peeled is high, it indicates that its corresponding shooting angle is better (this angle is denoted as the preferred angle); while if the image quality is poor, it indicates that its corresponding shooting angle is poor (this angle is denoted as the inferior angle). The processing unit 130 may retain the preferred angle from the previous detection cycle and determine the preferred angle from the previous detection cycle as the image parameters of the image acquisition unit 120 for the current cycle; discard the inferior angle from the previous detection cycle.

[0056] In some embodiments, the image parameters of the image acquisition unit 120 can also be determined based on light source data. Light source data may include light source position, light source intensity, etc. In some embodiments, the influence of light source position and / or light source intensity on the image quality captured from different shooting angles can be considered to determine the shooting angle for the current period. For example, light source position and / or light source intensity can affect the image quality score (e.g., sharpness score, noise score, and shading score), and the shooting angle corresponding to the image with a better quality score (e.g., when the quality score is a Boolean value, an image with a Boolean value of 1 has better quality) can be determined as the shooting angle for the current period. For more information on the influence of light source data on image quality (e.g., the influence of light source intensity on the sharpness threshold, noise threshold range, and shading threshold), please see Figures 5-9 and their related descriptions. Image quality is related to light source data; by considering light source data, the shooting angle for the current detection period can be determined more accurately to eliminate angle errors caused by the light source.

[0057] In some embodiments, determining the image parameters of the image acquisition unit for the current detection cycle based on an image of at least one angle of the wafer group to be stripped in the previous detection cycle can include the method shown in FIG5:

[0058] Figure 5 is an exemplary flowchart illustrating the determination of image parameters of an image acquisition unit for the current detection period according to some embodiments of this specification. As shown in Figure 5, step 410 may include:

[0059] Sub-step 411: Based on the image of at least one angle of the wafer group to be stripped in the previous detection cycle, determine the quality score of the image at at least one angle.

[0060] In some embodiments, the quality score of an image at at least one angle can be a Boolean value, corresponding to 1 or 0. A Boolean value of 1 indicates good image quality, while a Boolean value of 0 indicates poor image quality. In some embodiments, an image of at least one angle of the wafer assembly to be stripped from the previous detection cycle can be compared pixel-wise with a standard image taken at the same angle in a standard image set. If the similarity between the two images is higher than a first threshold (e.g., 98%, 95%, 90%, 85%, 80%, etc.), the Boolean value of the image at that angle is 1; if the similarity between the two images is lower than the first threshold, the Boolean value of the image at that angle is 0.

[0061] In some embodiments, the quality score of an image can be determined based on one or more of the image's sharpness score, noise score, and lighting score, as shown in Figures 6-9 and their related descriptions.

[0062] Sub-step 412: Determine the image parameters of the image acquisition unit in the current detection period based on the merit score.

[0063] In some embodiments, the processing unit 130 may retain the image with a quality score of 1 in the previous detection cycle and determine the shooting angle corresponding to the image with a quality score of 1 as the image parameter of the image acquisition unit 120 in the current cycle; the processing unit 130 may discard the shooting angle corresponding to the image with a quality score of 0 in the previous detection cycle.

[0064] Step 420: Based on the image parameters of the image acquisition unit, acquire an image of at least one angle of the wafer group to be stripped in the current detection cycle.

[0065] In some embodiments, the preferred angle from the previous detection cycle (i.e., the shooting angle corresponding to an image with a Boolean value of 1) is retained as the image parameter of the image acquisition unit 120 in the current detection cycle. The image acquisition unit 120 acquires an image of the preferred angle of the wafer assembly to be stripped in the current detection cycle. For example, the image processing unit 120 (e.g., a camera) can adjust its own shooting angle to the preferred angle from the previous detection cycle by rotating.

[0066] In some embodiments, substandard angles (i.e., the shooting angles corresponding to images with a Boolean value of 0) from the previous detection cycle are discarded. The image acquisition unit 120 can acquire images of the wafer assembly to be stripped from at angles other than the substandard angles in the current detection cycle.

[0067] Step 430: Determine the stripping parameters for the current detection cycle based on an image of at least one angle of the wafer assembly to be stripped during the current detection cycle.

[0068] In some embodiments, the processing unit 130 can determine the adhesion features of the wafer group to be peeled based on an image of at least one angle of the wafer group to be peeled in the current detection cycle, thereby determining the peeling parameters for the current detection cycle based on the adhesion features. In some embodiments, the processing unit 130 can determine the adhesion features based on the similarity between an image of at least one angle of the wafer group to be peeled in the current detection cycle and a standard image in a standard image set. In some embodiments, the processing unit 130 can determine the adhesion features using an adhesion feature determination model. In some embodiments, the peeling parameters for the current detection cycle can be determined based on the adhesion features using a feature index parameter table. More details on determining adhesion features and determining peeling parameters based on adhesion features can be found in the relevant content above, such as Figures 2-3, and will not be repeated here.

[0069] Considering that the wafer or wafer group to be stripped may shift during the stripping process, by updating the shooting angle of each detection cycle, the image of the wafer group to be stripped can be obtained based on the shooting angle of the current cycle, which can update the state of the wafer group to be stripped in real time, thereby more accurately determining the stripping parameters for more effective stripping.

[0070] It should be noted that the above description of process 400 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 400 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0071] Figure 6 is an exemplary flowchart illustrating the determination of image quality scores according to some embodiments of this specification. As shown in Figure 6, process 600 may include:

[0072] Step 610: Based on the image of at least one angle of the wafer group to be stripped from the previous detection cycle, determine the sharpness score of the image at at least one angle.

[0073] In some embodiments, an image sharpness score can be used to measure the sharpness of an image. A higher sharpness score indicates higher image sharpness, and a lower sharpness score indicates lower image sharpness. In some embodiments, the image sharpness score can be evaluated by edge detection. For example, the processing unit can identify sharp edges and color contrast in at least one image of the wafer assembly to be stripped in the previous detection cycle to determine the sharpness score of the image at at least one angle.

[0074] Step 620: Determine the quality score of the image from at least one angle based on the sharpness score.

[0075] In some embodiments, a sharpness threshold can be preset. If the sharpness score of an image exceeds the sharpness threshold, the image's quality score is 1; if the sharpness score of an image does not exceed the sharpness threshold, the image's quality score is 0.

[0076] In some embodiments, the sharpness threshold may be related to light source parameters (light source intensity, light source position). In some embodiments, the sharpness threshold may be positively correlated with light source intensity. For example, the stronger the light source intensity, the higher the sharpness threshold; the weaker the light source intensity, the lower the sharpness threshold.

[0077] Image sharpness is an important factor in measuring image quality. High-resolution images have clear and distinct edges, which can improve the distinction between objects and backgrounds. Using image sharpness to determine the quality score of an image and setting an appropriate sharpness threshold can effectively eliminate poor shooting angles.

[0078] It should be noted that the above description of process 600 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 600 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0079] Figure 7 is another exemplary flowchart illustrating the determination of image quality scores according to some embodiments of this specification. As shown in Figure 7, process 700 may include:

[0080] Step 710: Based on the image of at least one angle of the wafer group to be stripped in the previous detection cycle, determine the noise score of the image at at least one angle.

[0081] In some embodiments, the noise score can be used to measure the exposure level of an image. A higher noise score indicates a greater exposure level, while a lower noise score indicates a lower exposure level. The exposure level of an image determines its brightness and darkness. Excessive exposure means the image is too bright, and insufficient exposure means the image is too dark; both excessive and insufficient exposure can lead to unclear image details. In some embodiments, the noise score of an image can be determined based on its texture and color consistency. For example, the processing unit can analyze the texture and color consistency of an image from at least one angle of the wafer assembly to be stripped in the previous detection cycle to determine the noise score of the image from at least one angle.

[0082] Step 720: Determine the quality score of the image from at least one angle based on the noise score.

[0083] In some embodiments, a noise threshold range can be preset. If the noise score of an image exceeds the noise threshold range, the quality score of the image is 0; if the noise score of an image is within the noise threshold range, the quality score of the image is 1.

[0084] In some embodiments, the noise threshold is related to the light source parameters (light source intensity, light source position) and the quality of the image acquisition unit. For example, the better the heat dissipation capability of the image acquisition unit, the larger the noise threshold range can be; the worse the heat dissipation capability of the image acquisition unit, the smaller the noise threshold range can be. As another example, the stronger the light source intensity, the larger the noise threshold range; the weaker the light source intensity, the smaller the noise threshold range.

[0085] Considering that excessive noise may interfere with color reproduction, resulting in inaccurate color reproduction and unnatural color blocks or spots, using the image noise score to determine the quality score of the image and reasonably setting the noise threshold range can effectively eliminate bad shooting angles.

[0086] It should be noted that the above description of process 700 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 700 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0087] Figure 8 is another exemplary flowchart illustrating the determination of image quality scores according to some embodiments of this specification. As shown in Figure 8, process 800 may include:

[0088] Step 810: Based on the image of at least one angle of the wafer group to be stripped in the previous detection cycle, determine the light and shadow score of the image at at least one angle.

[0089] In some embodiments, the light and shadow score of an image can be used to measure the light and shadow level of the image. A higher light and shadow score indicates that the image is more prone to exposure problems (overexposure or underexposure); a lower light and shadow score indicates that the image is less prone to exposure problems (i.e., the image is properly exposed). In some embodiments, a histogram can be used to evaluate the exposure distribution of the image of at least one angle of the wafer assembly to be stripped in the previous detection cycle, thereby identifying overexposed and underexposed areas, and determining the light and shadow score of the image at at least one angle based on the total area of ​​the overexposed and underexposed areas.

[0090] Step 820: Determine the quality score of the image from at least one angle based on the light and shadow score.

[0091] In some embodiments, a light and shadow threshold can be preset. If the light and shadow score of an image exceeds the light and shadow threshold, the image's quality score is 0; if the light and shadow score of an image does not exceed the light and shadow threshold, the image's quality score is 1.

[0092] In some embodiments, the light and shadow threshold can be related to light source parameters (light source intensity, light source position / direction). For example, the higher the light source intensity, the larger the light and shadow threshold can be; the lower the light source intensity, the smaller the light and shadow threshold can be. As another example, the smaller the angle between the shooting angle and the light source direction, the larger the light and shadow threshold can be; the larger the angle between the shooting angle and the light source direction, the smaller the light and shadow threshold can be.

[0093] Good lighting and shadow processing can enhance the contrast of an image, making the transition between the bright and dark parts of the image more natural. Using lighting and shadow scores to determine the quality score of an image and setting lighting and shadow thresholds appropriately can effectively eliminate bad shooting angles.

[0094] It should be noted that the above description of process 800 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 800 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0095] Figure 9 is another exemplary flowchart illustrating the determination of image quality scores according to some embodiments of this specification. As shown in Figure 9, process 900 may include:

[0096] Step 910: Based on the image of at least one angle of the wafer group to be stripped from the previous detection cycle, determine the sharpness score, noise score, and light and shadow score of the image at at least one angle.

[0097] Regarding the image of at least one angle of the wafer assembly to be stripped based on the previous detection cycle, the explanation of determining the sharpness score of the image at at least one angle can be found in the relevant content of Figure 6, and will not be repeated here.

[0098] Regarding the image of at least one angle of the wafer group to be stripped based on the previous detection cycle, the explanation of determining the noise score of the image at at least one angle can be found in the relevant content of Figure 7, and will not be repeated here.

[0099] Regarding the image of at least one angle of the wafer group to be stripped based on the previous detection cycle, the explanation of determining the light and shadow score of the image at at least one angle can be found in the relevant content of Figure 8, and will not be repeated here.

[0100] Step 920: Determine the quality score of the image from at least one angle based on the weighted average of the sharpness score, noise score, and light and shadow score of the image from at least one angle.

[0101] In some embodiments, the processing unit 130 can perform a weighted summation of the sharpness score, noise score, and lighting score of an image from at least one angle to obtain a weighted score for the image. The weighted score is then compared with a comprehensive threshold to determine the image's quality score. For example, when the weight of the sharpness score is positive and the weight of the lighting score is negative, if the image's weighted score exceeds the comprehensive threshold, the image's quality score is 1; if the image's weighted score does not exceed the comprehensive threshold, the image's quality score is 0.

[0102] By weighted summing of the image's sharpness score, noise score, and lighting score, and comparing the weighted score with a comprehensive threshold, image quality can be measured from multiple dimensions, avoiding bias caused by any single factor. In some embodiments, the weights of the sharpness score, noise score, and lighting score can be reasonably set according to requirements, and no further limitations are made here.

[0103] It should be noted that the above description of process 900 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 900 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification. For example, based on an image of at least one angle of the wafer assembly to be stripped from the previous detection cycle, at least two of the following scores—sharpness score, noise score, and luminance score—can be determined for the image at at least one angle, and the quality score of the image at at least one angle can be determined based on the weighted average of the at least two scores.

[0104] Some embodiments of this specification also provide a method for peeling wafers, applied to a wafer peeling apparatus 100. In some embodiments, the wafer peeling method is performed by a processing unit (e.g., processing unit 130) and includes: determining peeling parameters based on an image of a wafer group to be peeled from at least one angle; controlling a peeling unit (e.g., peeling unit 140) to act on a container (e.g., container 110) or a peeling liquid based on the peeling parameters to peel multiple wafers adsorbed together. In some embodiments, the peeling unit includes an ultrasonic vibrator, and the peeling parameters include ultrasonic power and / or vibration time. In some embodiments, the peeling unit further includes a temperature control unit, and the peeling parameters include temperature control parameters, the temperature control unit being configured to adjust the temperature of the peeling liquid based on the temperature control parameters. In some embodiments, determining the peeling parameters based on an image of a wafer group to be peeled from at least one angle includes: determining adhesion characteristics of the wafer group to be peeled from the image of the wafer group to be peeled from at least one angle, the adhesion characteristics including the number of wafers adsorbed together; and determining the peeling parameters based on the adhesion characteristics. In some embodiments, determining the adhesion characteristics of the wafer group to be peeled from an image of the wafer group to be peeled from at least one angle includes: determining the adhesion characteristics using an adhesion characteristic determination model, the adhesion characteristic determination model including a machine learning model. In some embodiments, determining stripping parameters based on an image of at least one angle of the wafer assembly to be stripped includes: periodically determining the stripping parameters based on a detection period, the detection period being related to initial features of the wafer assembly to be stripped, the initial features including the thickness of the wafer assembly to be stripped. Further details regarding the wafer stripping method can be found in the description above and will not be repeated here.

[0105] Some embodiments of this specification also provide a computer-readable storage medium storing computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes a method for stripping wafers. The method is applied to a wafer stripping apparatus, which includes: a container configured to hold a stripping liquid and a wafer set to be stripped, the wafer set comprising multiple wafers adsorbed together; an image acquisition unit configured to acquire an image of the wafer set to be stripped from at least one angle; and a stripping unit. The method includes: determining stripping parameters based on the image of the wafer set from at least one angle; and controlling the stripping unit to act on the container or stripping liquid based on the stripping parameters to strip the multiple wafers adsorbed together.

[0106] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0107] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0108] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0109] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0110] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0111] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0112] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. An apparatus for stripping wafers, comprising: A container is configured to hold a stripping solution and a wafer set to be stripped, the wafer set comprising multiple wafers adsorbed together. An image acquisition unit is configured to acquire an image of at least one angle of the wafer assembly to be stripped. The processing unit is configured to determine peeling parameters based on an image of at least one angle of the wafer assembly to be peeled off, and The stripping unit is configured to act on the container or the stripping liquid based on the stripping parameters to strip the plurality of wafers adsorbed together.

2. The apparatus according to claim 1, wherein, The stripping unit includes an ultrasonic vibrator, and the stripping parameters include ultrasonic power and / or vibration time.

3. The apparatus according to claim 2, wherein, The stripping unit further includes a temperature control unit, and the stripping parameters include temperature control parameters. The temperature control unit is configured to adjust the temperature of the stripping liquid based on the temperature control parameters.

4. The apparatus according to any one of claims 1-3, wherein, Determining the stripping parameters based on an image of at least one angle of the wafer group to be stripped includes: The adhesion characteristics of the wafer group to be peeled are determined based on an image from at least one angle, the adhesion characteristics including the number of wafers adsorbed together; and The peeling parameters are determined based on the adhesion characteristics.

5. The apparatus according to claim 4, wherein, Determining the adhesion features of the wafer group to be peeled based on an image of at least one angle of the wafer group to be peeled includes: determining the adhesion features using an adhesion feature determination model, wherein the adhesion feature determination model includes a machine learning model.

6. The apparatus according to any one of claims 1-5, wherein, Determining the stripping parameters based on an image of at least one angle of the wafer group to be stripped includes: The stripping parameters are determined periodically based on a detection cycle, the detection cycle being related to the initial characteristics of the wafer group to be stripped, the initial characteristics including the thickness of the wafer group to be stripped.

7. The apparatus according to claim 6, wherein, The step of periodically determining the stripping parameters based on the detection cycle includes: Based on an image of at least one angle of the wafer group to be stripped in the previous detection cycle, the image parameters of the image acquisition unit in the current detection cycle are determined, and the image parameters of the image acquisition unit include the image shooting angle; Based on the image parameters of the image acquisition unit, an image of at least one angle of the wafer group to be stripped is obtained in the current detection cycle; and The stripping parameters for the current detection cycle are determined based on an image of at least one angle of the wafer group to be stripped during the current detection cycle.

8. The apparatus according to claim 7, wherein, The image parameters of the image acquisition unit are also determined based on the light source data.

9. The apparatus according to claim 7 or 8, wherein, The method of determining the image parameters of the image acquisition unit in the current detection cycle based on the image of at least one angle of the wafer group to be stripped from the previous detection cycle includes: Based on images of at least one angle of the wafer assembly to be stripped from the detection cycle described above, a quality score for the images at at least one angle is determined; and The image parameters of the image acquisition unit in the current detection cycle are determined based on the quality score.

10. The apparatus according to claim 9, wherein, The determination of the quality score of at least one angle of the image of the wafer group to be peeled off based on the image of the previous detection cycle includes: Based on images of at least one angle of the wafer assembly to be stripped from the detection cycle described above, a sharpness score for the image at at least one angle is determined; and The quality score of an image at at least one angle is determined based on the sharpness score.

11. The apparatus according to claim 9 or 10, wherein, The determination of the quality score of at least one angle of the image of the wafer group to be peeled off based on the image of the previous detection cycle includes: Based on images of at least one angle of the wafer assembly to be stripped from the detection cycle described above, a noise score for the image at at least one angle is determined; and The quality score of an image at at least one angle is determined based on the noise score.

12. The apparatus according to any one of claims 9-11, wherein, The determination of the quality score of at least one angle of the image of the wafer group to be peeled off based on the image of the previous detection cycle includes: Based on an image of at least one angle of the wafer group to be stripped from the previous detection cycle, determine the light and shadow score of the image at at least one angle; and The quality score of an image at at least one angle is determined based on the light and shadow score.

13. The apparatus according to any one of claims 9-12, wherein, The determination of the quality score of an image at at least one angle includes: Based on images of at least one angle of the wafer assembly to be stripped from the detection cycle described above, the sharpness score, noise score, and luminance score of the image at at least one angle are determined; and The quality score of the image from at least one angle is determined by weighting the sharpness score, the noise score, and the light and shadow score of the image from at least one angle.

14. A method for stripping a wafer, applied to a wafer stripping apparatus, the wafer stripping apparatus comprising: A container is configured to hold a stripping solution and a wafer set to be stripped, the wafer set comprising multiple wafers adsorbed together. An image acquisition unit is configured to acquire an image of at least one angle of the wafer assembly to be stripped. Processing unit; as well as Stripping unit; The method is executed by the processing unit and includes: Determine the stripping parameters based on an image of the wafer group to be stripped from at least one angle; The stripping unit is controlled to act on the container or the stripping liquid based on the stripping parameters in order to strip the multiple wafers that are adsorbed together.

15. The method according to claim 14, wherein, The stripping unit includes an ultrasonic vibrator, and the stripping parameters include ultrasonic power and / or vibration time.

16. The method according to claim 15, wherein, The stripping unit further includes a temperature control unit, and the stripping parameters include temperature control parameters. The temperature control unit is configured to adjust the temperature of the stripping liquid based on the temperature control parameters.

17. The method according to any one of claims 14-16, wherein, Determining the stripping parameters based on an image of at least one angle of the wafer group to be stripped includes: The adhesion characteristics of the wafer group to be peeled are determined based on an image from at least one angle, the adhesion characteristics including the number of wafers adsorbed together; and The peeling parameters are determined based on the adhesion characteristics.

18. The method according to claim 17, wherein, Determining the adhesion features of the wafer group to be peeled based on an image of at least one angle of the wafer group to be peeled includes: determining the adhesion features using an adhesion feature determination model, wherein the adhesion feature determination model includes a machine learning model.

19. The method according to any one of claims 14-18, wherein, Determining the stripping parameters based on an image of at least one angle of the wafer group to be stripped includes: The stripping parameters are determined periodically based on a detection cycle, the detection cycle being related to the initial characteristics of the wafer group to be stripped, the initial characteristics including the thickness of the wafer group to be stripped.

20. A computer-readable storage medium storing computer instructions, wherein when a computer reads the computer instructions from the storage medium, the computer executes a method for stripping a wafer, the method being applied to a wafer stripping apparatus, the wafer stripping apparatus comprising: A container is configured to hold a stripping solution and a wafer set to be stripped, the wafer set comprising multiple wafers adsorbed together. An image acquisition unit is configured to acquire an image of at least one angle of the wafer assembly to be stripped. as well as Stripping unit; The method includes: Determine the stripping parameters based on an image of the wafer group to be stripped from at least one angle; The stripping unit is controlled to act on the container or the stripping liquid based on the stripping parameters in order to strip the multiple wafers that are adsorbed together.

Images