Cleaning equipment control system and method for semiconductor processing based on image recognition

Abstract

The invention discloses a cleaning equipment control system for semiconductor processing based on image recognition. The cleaning equipment control system comprises a cleaning module, an image acquisition module and a control module, the cleaning module is used for cleaning the wafer step by step, firstly controlling a spray gun to spray liquid to clean the surface of the wafer, then blow-drying the outer surface of the wafer through an air gun, and then cleaning the inner surface of the wafer; the image acquisition module is used for acquiring an image of the outer surface of the wafer, firstly acquiring the image of the outer surface of the wafer for one time through shooting equipment before cleaning, then cleaning the wafer, and after the outer surface of the wafer is blow-dried, carrying out X-ray shooting on the wafer and cleaning the interior of the wafer; the control module is used for adjusting control parameters of the cleaning equipment; the problem that wafer cleaning equipment in the prior art cannot obtain the cleaning state of the surface of the wafer in the cleaning process is solved through the image recognition technology, the internal and external pollution conditions of the wafer can be recognized, cleaning parameters and paths are adjusted in a targeted mode, excessive cleaning or insufficient cleaning is avoided, and the cleaning quality is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor processing technology, and in particular to a semiconductor processing cleaning equipment control system and method based on image recognition. Background Art

[0002] In the semiconductor manufacturing process, the cleaning process is a crucial link. As the integration of semiconductor devices becomes higher and higher, the cleanliness requirements for the surface of semiconductor wafers are becoming increasingly stringent. Traditional semiconductor cleaning equipment often has the following limitations when controlling the cleaning process:

[0003] (1) Traditional cleaning equipment usually operates according to preset fixed parameters (such as cleaning solution concentration, cleaning time, cleaning solution flow rate, etc.). However, the types and degree of contaminants on the surface of semiconductor wafers may vary significantly at different processing stages. For example, after wafer etching, more etching solution and etching by-products may remain, while after the wafer deposition process, it may mainly be some granular impurities. Fixed parameter cleaning methods are difficult to remove these different types and degrees of contaminants in a targeted manner, which can easily lead to over-cleaning or under-cleaning. Over-cleaning may damage the microstructure of the wafer surface and affect the performance and yield of semiconductor devices, while under-cleaning will leave contaminants on the wafer surface, causing defects such as short circuits and leakage in subsequent processing steps, reducing product quality.

[0004] (2) In particular, traditional cleaning equipment cannot obtain the cleaning status of the wafer surface in real time during operation. Operators can only evaluate the cleaning effect through random inspections after the cleaning is completed. This post-test method cannot timely discover problems in the cleaning process and make adjustments.

[0005] (3) At present, the operation of some cleaning equipment needs to rely on the operator's experience to adjust the cleaning parameters. Different operators may make different parameter adjustments based on their own judgment, which leads to a lack of standardization and consistency in the cleaning process. At present, the operation of some cleaning equipment needs to rely on the operator's experience to adjust the cleaning parameters. Different operators may make different parameter adjustments based on their own judgment, which leads to a lack of standardization and consistency in the cleaning process.

[0006] Based on the above problems, there is an urgent need for a semiconductor processing cleaning equipment control system and method that can monitor and accurately control the cleaning process in real time, so as to improve the quality and efficiency of the semiconductor cleaning process and reduce production costs. The development of image recognition technology provides new ideas and technical means to solve these problems. By collecting and analyzing images of the surface of semiconductor wafers and the inside of the cleaning equipment, real-time monitoring of the cleaning effect and equipment status can be achieved, and intelligent cleaning control can be realized based on this. Summary of the invention

[0007] In order to solve the above-mentioned technical problems, the present invention provides a semiconductor processing cleaning equipment control system and method based on image recognition.

[0008] The technical solution of the present invention is achieved in this way:

[0009] A semiconductor processing cleaning equipment control system based on image recognition, comprising a cleaning module, an image acquisition module and a control module;

[0010] The cleaning module is used to clean the wafer in steps. First, the spray gun is controlled to spray liquid to clean the surface of the wafer, then the outer surface of the wafer is blown dry by the air gun, and then the inner surface of the wafer is cleaned;

[0011] The image acquisition module is used to acquire the image of the outer surface of the wafer. Before cleaning, the outer surface of the wafer is first acquired by a photographing device, and then the wafer is cleaned. After the outer surface of the wafer is dried, the wafer is X-rayed and the inside of the wafer is cleaned.

[0012] The control module is used to adjust the control parameters of the cleaning equipment.

[0013] Preferably, the cleaning module specifically comprises:

[0014] The pre-spray unit sprays the wafer with a pre-wash liquid before cleaning the wafer, covering the entire surface of the wafer. The pre-wash liquid is sprayed in the form of a spray, and the duration is set to 15-30 seconds;

[0015] The external surface cleaning unit controls the spray gun to perform high-pressure cleaning on the wafer after spraying;

[0016] The inner surface cleaning unit cleans the inside of the wafer using ultrasonic cleaning equipment;

[0017] The drying unit uses an air gun to dry the cleaned wafer.

[0018] Preferably, the image acquisition module specifically includes:

[0019] The outer surface image acquisition unit uses a camera to acquire images of the wafers in the cleaning tank. The acquired images are classified into three grades of wafers: a, b and c through image recognition. Grade a wafers are wafers that have not been cleaned, grade b wafers are wafers that have failed cleaning, and grade c wafers are wafers that have passed cleaning.

[0020] The inner surface image acquisition unit uses an X-ray device to capture the internal projection of the wafer in the cleaning tank. The captured projection is distinguished into grade D and grade E wafers through image recognition, where grade D is the wafer that fails the secondary cleaning, and grade E is the wafer that passes the secondary cleaning;

[0021] The image processing unit processes the images and projections collected by the outer surface image collection unit and the inner surface image collection unit through an image recognition algorithm, and identifies the cleaning defects on the outer surface and the inner surface of the round wafer.

[0022] Preferably, the control module specifically includes:

[0023] The spray path unit generates a primary cleaning path and a secondary cleaning path according to the processing information of the image acquisition module, and generates a control path for the spray gun and the air gun;

[0024] The parameter adjustment unit adjusts the cleaning parameters, specifically including the pressure of the spray gun of the external surface cleaning unit and the type of cleaning fluid, the jet pressure of the air gun of the blow-drying unit, the atomization pressure and the type of pre-washing fluid of the pre-spray unit, and the ultrasonic parameters of the ultrasonic cleaning equipment of the internal surface cleaning unit.

[0025] A semiconductor processing cleaning method based on image recognition comprises the following steps:

[0026] S1, the wafer enters the cleaning tank, the outer surface image acquisition unit acquires images of the wafer, and marks the wafer cleaning level on the acquired images, and uncleaned wafers are marked as non-grade a;

[0027] S2, generating a spray gun moving path and an air gun moving path for the wafer marked as level a according to step S1;

[0028] S3, before cleaning the wafer, spray the wafer with a pre-washing liquid, covering the entire surface of the wafer, and spraying the pre-washing liquid in a spray form;

[0029] S4, controlling the movement of the spray gun and the air gun according to the spray gun movement path and the air gun movement path generated in step S2 to clean and dry the wafer;

[0030] S5, collecting the internal projection of the round wafer in the cleaning tank by an X-ray photographing device, and marking the cleaning level of the round wafer on the collected projection image;

[0031] S6, performing ultrasonic cleaning according to the projection image obtained in step S5;

[0032] S7, collecting the outer surface image of the wafer for the second time, identifying and processing the image and marking the grade of the wafer, and generating the secondary spray gun movement path and the air gun movement path;

[0033] S8, repeating steps S1-S7 according to the marked wafer grade of the image in step S7, until all the wafer grades are graded as e.

[0034] Preferably, in step S6, when there is no image with a wafer grade of d in the acquired projection image, the ultrasonic cleaning of step S6 is skipped and step S7 is performed directly.

[0035] Preferably, in step S2, based on the marks in the image and the current positions of the spray gun and air gun, the marks of the most recent positions of the spray gun and air gun are used as the first cleaning order, the marks are arranged from left to right, and the paths of the spray gun and air gun are generated according to the marks.

[0036] The present invention utilizes image recognition technology to solve the problem that the wafer cleaning equipment in the prior art cannot obtain the cleaning status of the wafer surface during the cleaning process, and can identify the internal and external contamination conditions of the wafer, adjust the cleaning parameters and paths in a targeted manner, avoid over-cleaning or insufficient cleaning, and significantly improve the cleaning quality. The control module responds to image acquisition feedback in real time, dynamically optimizes the operating parameters of each cleaning unit, adapts to the cleaning requirements of different processes and batches of wafers, ensures the stability of the cleaning effect, and reasonably plans the cleaning path and collaborative operation mechanism, reduces the ineffective movement and waiting time of the spray gun and air gun, improves the cleaning efficiency per unit time, and conforms to the rhythm of large-scale semiconductor production. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of a semiconductor processing cleaning equipment control system based on image recognition according to the present invention.

[0038] Figure 2 It is a schematic diagram of a semiconductor processing cleaning method based on image recognition according to the present invention. DETAILED DESCRIPTION

[0039] To further explain the technical means and effects taken by the present invention to achieve the predetermined invention purpose, exemplary embodiments will be described in detail here, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are only examples of methods and systems consistent with some aspects of the present application as detailed in the attached claims.

[0040] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0041] The specific implementation methods, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0042] Embodiment 1, as Figure 1 and Figure 2As shown, the present invention is a semiconductor processing cleaning equipment control system based on image recognition, including a cleaning module, an image acquisition module and a control module;

[0043] The cleaning module is used to clean the wafer in steps. First, the spray gun is controlled to spray liquid to clean the surface of the wafer, then the outer surface of the wafer is blown dry by the air gun, and then the inner surface of the wafer is cleaned;

[0044] Specifically include:

[0045] The pre-spray unit sprays the wafer with pre-wash liquid before cleaning the wafer, covering the entire surface of the wafer. The pre-wash liquid is sprayed in the form of a spray, and the duration is set to 15-30 seconds; the pre-wash liquid storage tank is connected to a precision metering pump and an atomizing nozzle to form a pre-spray system. Before the formal cleaning of the wafer, the metering pump accurately extracts an appropriate amount of pre-wash liquid and delivers it to the atomizing nozzle. The nozzle fully covers the wafer surface in a fine and uniform spray form within 15-30 seconds. The composition of the pre-wash liquid is formulated according to the types of common pollutants, and a weak alkaline buffer solution is often used to gently remove large loose impurities such as dust and fibers attached to the surface of the wafer, reducing the burden of subsequent high-pressure cleaning.

[0046] The external surface cleaning unit controls the spray gun to perform high-pressure cleaning on the wafer after spraying. It is equipped with a high-pressure pump and a multi-mode spray gun. The high-pressure pump can output cleaning fluids of different pressure levels as needed, and the spray gun has the function of switching spray modes such as fan-shaped and columnar. For the wafer after spraying, the control module adjusts the appropriate pressure and spray mode according to the size of the wafer and the surface contamination condition, achieving efficient and accurate high-pressure cleaning, and powerfully removing stubborn stains such as photoresist residues and metal ions.

[0047] The inner surface cleaning unit uses ultrasonic cleaning equipment to clean the inside of the wafer. The core of the ultrasonic cleaning equipment is a high-frequency ultrasonic transducer that fits tightly against the wall of the cleaning tank. When working, the transducer converts electrical energy into high-frequency mechanical vibrations, which stimulate a large number of tiny cavitation bubbles in the cleaning liquid. The strong impact force generated by the rupture of the cavitation bubbles penetrates into the micropores and gaps on the inner surface of the wafer, peeling off the particulate impurities and organic residues adsorbed therein, ensuring the cleanliness of the inner surface.

[0048] The drying unit uses an air gun to dry the cleaned wafers. According to the preset program, a large flow rate and low pressure airflow is first quickly swept across the outer surface of the wafer to initially remove a large amount of moisture. Then, fine-tuning is performed to use a small flow rate and high pressure airflow to accurately dry the edges of the wafer, pins and other parts prone to water accumulation.

[0049] The image acquisition module is used to acquire the image of the outer surface of the wafer. Before cleaning, the outer surface of the wafer is first acquired by a photographing device, and then the wafer is cleaned. After the outer surface of the wafer is dried, the wafer is X-rayed and the inside of the wafer is cleaned.

[0050] Specifically include:

[0051] The outer surface image acquisition unit uses a camera to capture images of the wafers in the cleaning tank. The captured images are divided into three grades of wafers: a, b, and c through image recognition. Grade a wafers are wafers that have not been cleaned, grade b wafers are wafers that have failed cleaning, and grade c wafers are wafers that have passed cleaning. High-definition optical cameras are equipped with an annular shadowless light source and are symmetrically installed above and on the sides of the cleaning tank to capture images of the outer surface of the wafer from multiple angles. The captured images are transmitted to the image processing unit in real time, and the deep learning image recognition algorithm is used to quickly distinguish the wafers into three grades: a, b, and c: grade a corresponds to wafers that have not been cleaned, with surface stains and scratches in their original state. Grade b wafers have unqualified cleaning characteristics such as obvious stains and unrepaired scratches. Grade c indicates that the outer surface is clean and has no defects visible to the naked eye.

[0052] The inner surface image acquisition unit uses X-ray equipment to collect the internal projection of the round wafer in the cleaning tank. The collected projection is distinguished into D and E grade round wafers through image recognition, among which D grade round wafers are unqualified round wafers after secondary cleaning, and E grade round wafers are qualified round wafers after secondary cleaning. The X-ray equipment is combined with a high-power X-ray generator and a high-sensitivity detector, and is arranged around the cleaning tank. When the outer surface of the round wafer is blown dry, the internal projection is immediately collected, and the obtained image is transmitted to the image processing unit. After algorithm recognition, it is divided into D and E grades: D grade means that there are still internal impurities and structural defects after secondary cleaning; E grade indicates that the interior is clean and meets production standards.

[0053] The image processing unit processes the images and projections collected by the outer surface image acquisition unit and the inner surface image acquisition unit through the image recognition algorithm to identify the cleaning defects on the outer and inner surfaces of the wafer; it runs the image recognition algorithm based on the convolutional neural network (CNN) to conduct in-depth analysis of the collected outer surface images and inner surface projections. For the outer surface, it accurately locates the location, size, and type of stains; for the inner surface, it identifies internal microcracks, impurity distribution and other defects, providing a key basis for the subsequent adjustment of the cleaning strategy.

[0054] The control module is used to adjust the control parameters of the cleaning equipment.

[0055] Specifically include:

[0056] The spray path unit generates the primary cleaning path and the secondary cleaning path according to the processing information of the image acquisition module, generates the control path for the spray gun and the air gun, receives the feedback data of the image processing unit, and has a built-in intelligent path planning algorithm. The primary cleaning path and the secondary cleaning path are generated by comprehensively considering the number and position of the wafers, the current posture of the spray gun and the air gun. The primary path is for the initial cleaning, and quickly plans the shortest and most efficient route for the spray gun and the air gun to traverse all the A-level wafers; the secondary path focuses on the secondary cleaning, giving priority to the B and D-level wafers, optimizing the spray gun and air gun action sequence, and improving the targeted cleaning;

[0057] The parameter adjustment unit adjusts the cleaning parameters, including the pressure of the spray gun of the external surface cleaning unit and the type of cleaning liquid, the jet pressure of the air gun of the blow-drying unit, the atomization pressure and type of pre-washing liquid of the pre-spraying unit, and the ultrasonic parameters of the ultrasonic cleaning equipment of the internal surface cleaning unit. The parameters of each cleaning unit are monitored and adjusted in real time. According to the degree of contamination and material characteristics of the wafer, the pressure of the spray gun of the external surface cleaning unit is accurately adjusted to meet the cleaning requirements of different intensities; the type of cleaning liquid is switched to flexibly adjust the jet pressure of the air gun of the blow-drying unit to match the specific pollutants generated by processes such as lithography and etching to prevent over- or under-drying; the atomization pressure and pre-washing liquid composition of the pre-spraying unit are adjusted as needed; and the frequency, power and other ultrasonic parameters of the ultrasonic cleaning equipment of the internal surface cleaning unit are finely set.

[0058] A semiconductor processing cleaning method based on image recognition comprises the following steps:

[0059] S1, the wafer enters the cleaning tank, the outer surface image acquisition unit acquires images of the wafer, and marks the wafer cleaning level on the acquired images, and uncleaned wafers are marked as non-grade a;

[0060] S2, generating a spray gun moving path and an air gun moving path for the wafer marked as level a according to step S1;

[0061] S3, before cleaning the wafer, spray the wafer with a pre-washing liquid, covering the entire surface of the wafer, and spraying the pre-washing liquid in a spray form;

[0062] S4, controlling the movement of the spray gun and the air gun according to the spray gun movement path and the air gun movement path generated in step S2 to clean and dry the wafer;

[0063] S5, collecting the internal projection of the round wafer in the cleaning tank by an X-ray photographing device, and marking the cleaning level of the round wafer on the collected projection image;

[0064] S6, performing ultrasonic cleaning according to the projection image obtained in step S5;

[0065] S7, collecting the outer surface image of the wafer for the second time, identifying and processing the image and marking the grade of the wafer, and generating the secondary spray gun movement path and the air gun movement path;

[0066] S8, repeating steps S1-S7 according to the marked wafer grade of the image in step S7, until all the wafer grades are graded as e.

[0067] Preferably, in step S6, when there is no image with a wafer grade of d in the acquired projection image, the ultrasonic cleaning of step S6 is skipped and step S7 is performed directly.

[0068] Preferably, in step S2, based on the marks in the image and the current positions of the spray gun and air gun, the marks of the most recent positions of the spray gun and air gun are used as the first cleaning order, the marks are arranged from left to right, and the paths of the spray gun and air gun are generated according to the marks.

[0069] Preferably, in step S2, the marks a1, a2, a3...an are set, then the spray gun path is a0, a1, a2, a3...an, point a0 is the current position of the spray gun, the spray gun moves in a dot-line-dot manner among adjacent marked points, the moving path of the air gun is a0, a1, a2, a3...an, and the moving position of the air gun is always one marked point slower than the moving position of the spray gun. When the spray gun is a0, the air gun is a0, when the spray gun is a1, the air gun is a0, when the spray gun is a2, the spray gun is a1, and so on.

[0070] Preferably, step S1 further includes, whenever a new batch of wafers enters the cleaning process, selecting defect-free wafers, and collecting images of the wafers before cleaning, at key cleaning nodes, and after cleaning is completed. Comparison images are selected as the A-level mark, C-level mark, and E-level mark of the corresponding wafers.

[0071] Preferably, in step S4, the cleaning and drying of the wafer also includes selecting a cleaning mode, a soft cleaning mode, a high pressure mode and a standard mode.

[0072] Preferably, in the selected cleaning mode, the preset spray gun and air gun pressures are K1 and K2, the spray gun pressure in the soft washing mode is K1, and the air gun pressure is K2, in the standard mode, the spray gun pressure is 1.5*K1, and the air gun pressure is 1.5*K2, and in the high pressure mode, the spray gun pressure is 3*K1, and the air gun pressure is 3*K2.

[0073] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A semiconductor processing cleaning equipment control system based on image recognition, characterized in that: It includes a cleaning module, an image acquisition module and a control module; The cleaning module is used to clean the wafer in steps. First, the spray gun is controlled to spray liquid to clean the surface of the wafer, then the outer surface of the wafer is blown dry by the air gun, and then the inner surface of the wafer is cleaned; The image acquisition module is used to acquire the image of the outer surface of the wafer. Before cleaning, the outer surface of the wafer is first acquired by a photographing device, and then the wafer is cleaned. After the outer surface of the wafer is dried, the wafer is X-rayed and the inside of the wafer is cleaned. The control module is used to adjust the control parameters of the cleaning equipment.

2. A semiconductor processing cleaning equipment control system based on image recognition according to claim 1, characterized in that: The cleaning module specifically comprises: The pre-spray unit sprays the wafer with a pre-wash liquid before cleaning the wafer, covering the entire surface of the wafer. The pre-wash liquid is sprayed in the form of a spray, and the duration is set to 15-30 seconds; The external surface cleaning unit controls the spray gun to perform high-pressure cleaning on the wafer after spraying; The inner surface cleaning unit cleans the inside of the wafer using ultrasonic cleaning equipment; The drying unit uses an air gun to dry the cleaned wafer.

3. A semiconductor processing cleaning equipment control system based on image recognition according to claim 1, characterized in that: The image acquisition module specifically includes: The outer surface image acquisition unit uses a camera to acquire images of the wafers in the cleaning tank. The acquired images are classified into three grades of wafers: a, b and c through image recognition. Grade a wafers are wafers that have not been cleaned, grade b wafers are wafers that have failed cleaning, and grade c wafers are wafers that have passed cleaning. The inner surface image acquisition unit uses an X-ray device to capture the internal projection of the wafer in the cleaning tank. The captured projection is distinguished into grade D and grade E wafers through image recognition, where grade D is the wafer that fails the secondary cleaning, and grade E is the wafer that passes the secondary cleaning; The image processing unit processes the images and projections collected by the outer surface image collection unit and the inner surface image collection unit through an image recognition algorithm, and identifies the cleaning defects on the outer surface and the inner surface of the round wafer.

4. The semiconductor processing cleaning equipment control system based on image recognition according to claim 1, characterized in that: The control module specifically includes: The spray path unit generates a primary cleaning path and a secondary cleaning path according to the processing information of the image acquisition module, and generates a control path for the spray gun and the air gun; The parameter adjustment unit adjusts the cleaning parameters, specifically including the pressure of the spray gun of the external surface cleaning unit and the type of cleaning fluid, the jet pressure of the air gun of the blow-drying unit, the atomization pressure and the type of pre-washing fluid of the pre-spray unit, and the ultrasonic parameters of the ultrasonic cleaning equipment of the internal surface cleaning unit.

5. A semiconductor processing cleaning equipment control system based on image recognition, characterized in that: A semiconductor processing cleaning method based on image recognition comprises the following steps: S1, the wafer enters the cleaning tank, the outer surface image acquisition unit acquires images of the wafer, and marks the wafer cleaning level on the acquired images, and uncleaned wafers are marked as non-grade a; S2, generating a spray gun moving path and an air gun moving path for the wafer marked as level a according to step S1; S3, before cleaning the wafer, spray the wafer with a pre-washing liquid, covering the entire surface of the wafer, and spraying the pre-washing liquid in a spray form; S4, controlling the movement of the spray gun and the air gun according to the spray gun movement path and the air gun movement path generated in step S2 to clean and dry the wafer; S5, collecting the internal projection of the round wafer in the cleaning tank by an X-ray photographing device, and marking the cleaning level of the round wafer on the collected projection image; S6, performing ultrasonic cleaning according to the projection image obtained in step S5; S7, collecting the outer surface image of the wafer for the second time, identifying and processing the image and marking the grade of the wafer, and generating the secondary spray gun movement path and the air gun movement path; S8, repeating steps S1-S7 according to the marked wafer grade of the image in step S7, until all the wafer grades are graded as e.

6. A semiconductor processing cleaning equipment control system based on image recognition according to claim 5, characterized in that: In step S6, when there is no image with a wafer grade of d in the acquired projection image, the ultrasonic cleaning of step S6 is skipped and step S7 is performed directly.

7. A semiconductor processing cleaning equipment control system based on image recognition according to claim 5, characterized in that: In the step S2, according to the mark of the image and the current position of the spray gun and the air gun, the mark of the position of the most recent spray gun and the air gun is used as the first cleaning order, and the marks are arranged from left to right, and the paths of the spray gun and the air gun are generated according to the marks. Preferably, in the step S2, the marks a1, a2, a3...an are set, then the spray gun path is a0, a1, a2, a3...an, a0 point is the current position of the spray gun, the spray gun moves in the form of dot-line-dot in the adjacent mark points, the moving path of the air gun is a0, a1, a2, a3...an, and the moving position of the air gun is always one mark point slower than the moving position of the spray gun. When the spray gun is a0, the air gun is a0, when the spray gun is a1, the air gun is a0, when the spray gun is a2, the spray gun is a1, and so on.

8. The semiconductor processing cleaning equipment control system based on image recognition according to claim 5, characterized in that: The step S1 also includes selecting defect-free wafers whenever a new batch of wafers enters the cleaning process, capturing images of the wafers before cleaning, at key cleaning nodes, and after cleaning, and selecting comparison images as A-level, C-level, and E-level marks for the corresponding wafers.

9. The semiconductor processing cleaning equipment control system based on image recognition according to claim 5, characterized in that: In the step S4, the cleaning and drying of the wafer also includes selecting a cleaning mode, a soft cleaning mode, a high pressure mode and a standard mode.

10. The semiconductor processing cleaning equipment control system based on image recognition according to claim 9, characterized in that: In the selected cleaning mode, the preset spray gun and air gun pressures are K1 and K2. The spray gun pressure in the soft washing mode is K1, and the air gun pressure is K2. In the standard mode, the spray gun pressure is 1.5*K1, and the air gun pressure is 1.5*K2. In the high pressure mode, the spray gun pressure is 3*K1, and the air gun pressure is 3*K2.

Images