Semiconductor product dust removal method and dust removal system based on semi-solidified body

Abstract

The invention discloses a semiconductor product dust removal method and system based on a semi-solidified body, and the method comprises the steps: carrying out the difference regionalization processing of a surface image, collected in real time, of a semiconductor product to be subjected to dust removal, and obtaining a dust condensation image; the difference regionalization processing is designed to diffuse differences of off-group pixel points in the surface image so as to merge discretely distributed dotted dust into regions; performing image segmentation on the dust condensation image to obtain a to-be-identified area; obtaining a dust region according to the gray level distribution feature and the contour feature of each region to be identified in the surface image; dynamically determining a dust removal strategy of each dust area based on the use duration of the dust removal rod, the size feature of each dust area and the corresponding gray scale gradient feature; and a functional part matched with the dust removal rod is controlled to execute a corresponding dust removal strategy on each dust area, so that the dust removal effect on the semiconductor device is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of dust removal of semiconductor devices, and in particular to a dust removal method and a dust removal system for semiconductor products based on semi-solidified bodies. Background Art

[0002] During the manufacturing and use of semiconductor devices, various impurities will accumulate on the surface of semiconductor devices. The accumulation of dust will affect the performance stability of semiconductor precision devices, shorten their service life, cause electrical failure of chips, and even cause failures, resulting in the scrapping of semiconductor precision devices. By removing dust from the surface of semiconductor precision devices, these problems can be effectively prevented, ensuring that the devices are in the best working condition and improving the reliability and durability of the overall system.

[0003] However, since the prior art uses a fixed dust removal mode to remove dust from semiconductor devices, secondary pollution may be caused to the semiconductor devices, resulting in poor dust removal effect on the semiconductor devices.

[0004] It can be seen that how to improve the dust removal effect on semiconductor devices has become a technical problem that needs to be urgently solved by those skilled in the art. Summary of the invention

[0005] The present invention provides a dust removal method and a dust removal system for semiconductor products based on semi-solidified bodies, so as to solve the technical problem that the prior art adopts a fixed dust removal mode to remove dust from semiconductor devices, which may cause secondary pollution to the semiconductor devices and lead to poor dust removal effect on the semiconductor devices.

[0006] In order to solve the above technical problems, an embodiment of the present invention provides a dust removal method for semiconductor products based on a semi-solidified body.

[0007] In the dust removal process, a surface image of the semiconductor product to be dusted is collected in real time, and a difference regionalization process is performed on the surface image to obtain a dust condensation image; the difference regionalization process is designed to diffuse the difference of outlier pixel points in the surface image to merge discretely distributed point-shaped dust into regions; Performing image segmentation on the dust condensation image to obtain various areas to be identified; Acquire each dust area according to the contour features of each area to be identified in the surface image; dynamically determine the dust removal strategy for each dust area based on the use time of the semi-solidified dust removal rod, the size features of each dust area and the corresponding grayscale gradient features; Wherein, the dust removal strategy includes a dust removal mode; The dust removal mode includes a contact dust removal mode and a non-contact dust removal mode. The contact dust removal mode is designed so that the dust removal rod contacts the dust on the surface of the semiconductor product, and the dust removal rod does not contact the semiconductor product; the non-contact dust removal mode is designed so that the dust removal rod does not contact the semiconductor product, and the dust on the surface of the semiconductor product does not contact each other. The functional components matched with the dust removal rod are controlled to execute the corresponding dust removal strategy for each dust area.

[0008] As one of the preferred solutions, performing difference regionalization processing on the surface image to obtain the dust condensation image includes: Based on the grayscale difference between each pixel in the surface image and the overall image, and the grayscale difference between each pixel in the surface image and the neighborhood range, a two-stage screening process is performed on all the pixels in the surface image to obtain each outlier pixel and each background pixel; Calculating the similarity of belonging to the same region between the two outlier pixels based on the position proximity and the corresponding grayscale similarity between the two outlier pixels; The same region belonging similarity is used as a similarity measurement criterion of a clustering algorithm, and the clustering algorithm is used to cluster all the outlier pixels to obtain various point clusters; The background pixels are screened based on the distance between each background pixel and each point-shaped cluster to obtain each pixel to be fused; Performing a fusion update process on the grayscale value of each pixel to be fused according to the grayscale values ​​of all the outlier pixels, so as to obtain a fusion grayscale weighted value of each pixel to be fused; The grayscale value of each pixel to be fused in the surface image is replaced with the fused grayscale weighted value to obtain a dust condensation image.

[0009] As one of the preferred solutions, the two-stage screening process is designed as follows: The average value of the grayscale values ​​of all pixels in the surface image is used as the semiconductor pixel reference value; the absolute value of the difference between the grayscale value of each pixel in the surface image and the semiconductor pixel reference value is used as the overall grayscale difference value of each pixel in the surface image; Performing a primary comparison and screening on all the pixels in the surface image using the overall grayscale difference value to obtain each singular pixel and each background pixel; Constructing a corresponding neighborhood window with each singular pixel as the center, and calculating a neighborhood grayscale difference value of each singular pixel based on the grayscale difference characteristics of each singular pixel and all pixels in the corresponding neighborhood window; A secondary comparison and screening is performed on all the singular pixel points in the surface image using the neighborhood grayscale difference value to obtain various outlier pixel points.

[0010] As one of the preferred solutions, the fusion update process is designed as follows: The point cluster with the smallest distance from each pixel to be fused is used as the cluster to be merged into each pixel to be fused; The distance between each pixel to be fused and the corresponding outlier pixel in the cluster to be fused is normalized to obtain the contribution weight of each outlier pixel in the cluster to be fused; the grayscale values ​​of all the outlier pixel in the cluster to be fused for each pixel to be fused are weightedly summed with the contribution weight to obtain the fused grayscale weighted value of each pixel to be fused.

[0011] As one of the preferred solutions, the step of acquiring each dust area according to the contour feature of each area to be identified in the surface image includes: Extracting the contour line of each of the areas to be identified, numbering all the pixels on each of the contour lines in a clockwise direction, and calculating the direction change rate of each pixel on each of the contour lines; The standard deviation of the direction change rate of all pixel points on each contour line is used as the contour complexity index of each contour line; all the areas to be identified are screened by the contour complexity index to obtain various dust areas.

[0012] As one of the preferred solutions, the dust removal strategy for each dust area is dynamically determined based on the usage time of the semi-solidified dust removal rod, the size characteristics of each dust area and the corresponding grayscale gradient characteristics, including: Taking the pixel point with the largest difference between the semiconductor pixel reference value and each dust region as the dust accumulation point of each dust region; The ray obtained by connecting the dust accumulation point of each dust area and each pixel point on the corresponding contour line is used as each dust accumulation line of each dust area, and the dust accumulation index of each dust area is calculated based on the gradient degree of the grayscale values ​​of the pixel points on all the dust accumulation lines of each dust area; wherein the dust accumulation index reflects the grayscale gradient characteristics of the corresponding dust area; Based on the linear relationship between the dust accumulation index of each dust area and the total number of pixels in each dust area, the dust cleaning urgency index of each dust area is calculated; wherein the total number of pixels in each dust area reflects the size characteristics of the corresponding dust area; Inputting the use time of the semi-solidified dust removal rod and the dust cleaning urgency index of each dust area into the mode switching control expression, and calculating the non-contact mode switching factor of each dust area; The dust removal strategy of each of the dust areas is adjusted according to the non-contact mode switching factor.

[0013] As one of the preferred options, The step of adjusting the dust removal strategy of each dust area according to the non-contact mode switching factor includes: When the non-contact mode switching factor is less than a preset threshold, the dust removal mode of the corresponding dust area is adjusted to the contact dust removal mode, and the dust removal intensity of the corresponding dust area is adjusted to level one; When the non-contact mode switching factor is greater than or equal to a preset threshold, the dust removal mode of the dust area is adjusted to the non-contact dust removal mode, and the dust removal intensity of the corresponding dust area is adjusted to level two.

[0014] As one of the preferred solutions, the dust accumulation index of each dust area is calculated based on the gradient of the grayscale values ​​of the pixels on all the dust accumulation lines of each dust area, including: The difference between the grayscale value of each pixel point on each dust accumulation line and the grayscale value of the corresponding next pixel point is used as the grayscale gradient value of each pixel point on each dust accumulation line; The sum of the grayscale gradient values ​​of all pixels on each dust accumulation line is used as the linear gradient intensity of each dust accumulation line, and the average value of the linear gradient intensities of all the dust accumulation lines in each dust area is used as the dust accumulation index of each dust area.

[0015] As one of the preferred solutions, the mode switching control expression is: ; in, is the non-contact dust removal mode switching factor for dust area j, is the normalization function, The use time of the semi-solid dust removal rod. is an exponential function with a natural constant as base, is the dust cleaning urgency index of dust zone j.

[0016] Yet another embodiment of the present invention provides a semiconductor product dust removal system based on semi-solidified solids, the system comprising a control console, a manipulator and a semi-solidified solid dust removal rod, the manipulator being fixedly arranged in a control area on the control console, the dust removal rod being placed in a storage area on the control console, a computer device being arranged inside a counter of the control console, the computer device comprising a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, the processor being communicatively connected to the manipulator, and controlling the manipulator to grab the semi-solidified solid dust removal rod when executing the computer program, so as to implement the semiconductor product dust removal method based on semi-solidified solids as described above.

[0017] Compared with the prior art, the embodiments of the present invention have the following advantages: During the dust removal process, the surface image of the semiconductor product to be dusted is collected in real time. Considering the granular characteristics of dust, it may present a discrete distribution in the form of points. In order to avoid segmenting large dust areas with point-like but clustered distribution into discrete small dust areas, which will lead to size recognition errors and then adopt wrong dust removal strategies, the surface image is first subjected to difference regionalization processing to obtain a dust condensation image; the difference regionalization processing is designed to diffuse the differences of outlier pixels in the surface image to merge discretely distributed point-like dust into regions, thereby improving the accuracy of subsequent semiconductor device segmentation; the dust condensation image is segmented to obtain various areas to be identified; considering that the distribution of dust is often irregular and the contour is relatively complex, while the contours of the components on the surface of the semiconductor device are relatively simple, each dust area is obtained according to the contour features of each area to be identified in the surface image; based on the usage time of the semi-solidified dust removal rod and the size of each dust area The dust removal strategy of each dust area is dynamically determined by the features and the corresponding grayscale gradient features; wherein the dust removal strategy includes a dust removal mode; the dust removal mode includes a contact dust removal mode and a non-contact dust removal mode, the contact dust removal mode is designed so that the dust removal rod contacts the surface dust of the semiconductor product and the dust removal rod does not contact the semiconductor product, and the non-contact dust removal mode is designed so that the dust removal rod does not contact the semiconductor product and the surface dust of the semiconductor product does not contact each other; when the non-contact mode switching factor is less than a preset threshold, the dust removal mode of the corresponding dust area is adjusted to the contact dust removal mode, and the dust removal intensity of the corresponding dust area is adjusted to level one; when the non-contact mode switching factor is greater than or equal to the preset threshold, the dust removal mode of the dust area is adjusted to the non-contact dust removal mode, and the dust removal intensity of the corresponding dust area is adjusted to level two, and the dust removal strategy is adjusted in real time according to the dust distribution on the semiconductor surface, thereby improving the dust removal effect on semiconductor devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1It is a schematic flow chart of a dust removal method for semiconductor products based on a semi-solidified body in one embodiment of the present invention; Figure 2 A schematic diagram showing a comparison of image segmentation results of a dust condensation image and a surface image in one embodiment of the present invention; Figure 3 A schematic diagram of a semi-solidified dust removal rod in one embodiment of the present invention; Figure 4 A schematic diagram of dust removal of a semiconductor device in one embodiment of the present invention; Figure 5 A schematic diagram of an implementation of a dust removal method for a semiconductor product based on a semi-solidified body in one embodiment of the present invention; Figure 6 is a schematic diagram of a dust accumulation line in one embodiment of the present invention; Figure 7 A schematic diagram of obtaining a dust removal strategy in one embodiment of the present invention; Figure 8 It is a schematic diagram of a control console in a dust removal system based on a semi-solidified solid in one embodiment of the present invention; Fig. 9 It is a schematic diagram of the hardware structure of a dust removal system for semiconductor products based on semi-solidified solid in one embodiment of the present invention; Fig.10 It is a schematic diagram of the structure of a computer device in one embodiment of the present invention.

[0019] Description of labels: 101. Computer equipment; 102. Robot arm; 103. Visual inspection module; 104. First mobile module; 105. Second mobile module. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0022] In the description of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two elements. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are only for illustrative purposes, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0023] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0024] One embodiment of the present invention provides a semiconductor product dust removal method based on a semi-solidified solid. For details, see Figure 1 , Figure 1 Shown is a schematic flow chart of a dust removal method for semiconductor products based on a semi-solidified body in one embodiment of the present invention.

[0025] Step S201 , in the dust removal process, a surface image of a semiconductor product to be dusted is collected in real time, and a difference regionalization process is performed on the surface image to obtain a dust condensation image.

[0026] During the dust removal process, the surface image of the semiconductor product to be dusted is collected in real time. Considering the granular characteristics of dust, it may present a discrete distribution in the form of points. In order to avoid the point-like but clustered dust area, which leads to size recognition errors and then the wrong dust removal strategy, the surface image is first processed by difference regionalization, and the grayscale values ​​of the pixels belonging to the background in the dust area are replaced with grayscale values ​​similar to the surrounding pixels belonging to the dust.

[0027] It should be noted that the difference regionalization process is designed to diffuse the differences of outlier pixels in the surface image so as to merge discretely distributed point-shaped dust into regions.

[0028] Step S202, performing image segmentation on the dust condensation image to obtain various areas to be identified.

[0029] It should be noted that, through the difference regionalization processing, the point-like but clustered dust in the surface image appears as a patch-distributed dust area in the dust condensation image. At this time, image segmentation of the dust condensation image can completely divide the point-like but clustered dust area and avoid incorrect recognition of the size of the dust area.

[0030] Specifically, an image segmentation algorithm is used to segment the dust condensation image to obtain various areas to be identified. Figure 2 , Figure 2 A schematic diagram showing comparison of image segmentation results of a dust condensation image and a surface image in one embodiment of the present invention is shown. Figure 2 The left side of the figure shows the image segmentation result of the surface image, which divides each point-like but clustered dust into separate to-be-identified area 1 and to-be-identified area 2. Figure 2 The right side of the middle is the image segmentation result of the dust condensation image, which divides the point-like but clustered dust into an overall area to be identified 3.

[0031] It should be noted that since the size of the area to be identified 3 is relatively large, the non-contact dust removal mode should be used to avoid secondary pollution; while the sizes of the areas to be identified 1 and 2 are relatively small, and the contact dust removal mode is likely to be used to increase the dust removal speed. If there is no differential regionalization processing, the area to be identified 3 may be mistakenly identified as dust in small areas such as the areas to be identified 1 and 2, and the contact dust removal mode is used for the dust in small areas such as the areas to be identified 1 and 2. As a result, when the contact dust removal is performed on the areas to be identified 1 or 2, the dust in other small areas around it is contaminated on the dust removal rod, resulting in secondary pollution.

[0032] It should be noted that the image segmentation algorithm is a well-known technology and will not be described in detail in this embodiment.

[0033] Step S203, obtaining each dust area according to the contour features of each area to be identified in the surface image, and dynamically determining the dust removal strategy for each dust area based on the usage time of the semi-solidified dust removal rod, the size features of each dust area and the corresponding grayscale gradient features.

[0034] It should be noted that the distribution of dust is often irregular and has a relatively complex outline, while the outlines of components on the surface of semiconductor devices are relatively simple. Therefore, the dust area can be accurately identified by the complexity of the outline.

[0035] Specifically, each dust region is obtained according to the contour feature of each to-be-identified region in the surface image.

[0036] It should be noted that the dust removal strategy includes dust removal modes, and the dust removal modes include contact dust removal mode and non-contact dust removal mode. The contact dust removal mode is designed so that the dust removal rod contacts the dust on the surface of the semiconductor product, and the dust removal rod does not contact the semiconductor product; the non-contact dust removal mode is designed so that the dust removal rod does not contact the semiconductor product, and the dust on the surface of the semiconductor product does not contact either.

[0037] For details, see Figure 3 , Figure 3 FIG. 1 is a schematic diagram of a semi-solidified dust removal rod in one embodiment of the present invention. Figure 4 , Figure 4 A schematic diagram of dust removal of a semiconductor device in one embodiment of the present invention is shown.

[0038] It should be further explained that the contact dust removal mode may cause dust or pollutants to adhere to the surface of the dust removal rod. If it is not cleaned in time, it may cause secondary pollution in the subsequent dust removal process. By reducing the frequency of use of contact dust removal, the cleaning cycle of the dust removal rod can be extended and the risk of secondary pollution can be reduced. The non-contact dust removal mode causes less pollution to the dust removal rod and can reduce the frequency of cleaning the dust removal rod, thereby improving cleaning efficiency.

[0039] Specifically, the dust removal strategy for each dust area is dynamically determined based on the usage time of the semi-solidified dust removal rod, the size characteristics of each dust area, and the corresponding grayscale gradient characteristics.

[0040] It should be noted that the semi-solid solid is a material state between solid and liquid, with a certain viscosity and fluidity. In the air, the surface of the semi-solid solid will gradually dry, and its internal molecular movement will intensify during the drying process, forming a strong adsorption capacity. The working principle of the non-contact dust removal mode is: the robotic arm controls the dust removal rod to reach the target position to be dusted, the semi-solid solid at the bottom of the dust removal rod is separated from the dust by a preset distance, the semi-solid solid will continue to dry in the air, and energy will be released during the drying process. The adsorption force generated by the internal molecular movement can adsorb the dust on the surface of the semiconductor product to the rod head of the semi-solid dust removal rod, thereby achieving dust removal on the surface of the semiconductor product. The working principle of the contact dust removal mode is: the robotic arm controls the dust removal rod to reach the target position to be dusted, the semi-solid solid at the bottom of the dust removal rod contacts the dust, and the energy will be released during the drying process of the semi-solid solid. The adsorption force generated by the internal molecular movement can adsorb the dust on the surface of the semiconductor product to the rod head of the semi-solid dust removal rod, thereby achieving dust removal on the surface of the semiconductor product.

[0041] When the semi-solid dust removal rod is used for a longer time and the size characteristics and corresponding grayscale gradient characteristics of the dust area are more obvious, the non-contact dust removal mode is adopted to reduce the risk of secondary pollution; when the semi-solid dust removal rod is used for a shorter time and the size characteristics and corresponding grayscale gradient characteristics of the dust area are less obvious, the contact dust removal mode is adopted to improve cleaning efficiency.

[0042] Step S204, controlling the functional component matched with the dust removal rod to execute the corresponding dust removal strategy for each dust area.

[0043] Specifically, the functional components matched with the dust removal rod are controlled to execute the corresponding dust removal strategy for each dust area. As an embodiment of the present application, the functional components matched with the dust removal rod are controlled to execute the corresponding dust removal strategy for each dust area, and a simulation model is used to perform a simulation experiment on the dust removal strategy for each dust removal area.

[0044] It should be noted that by experimenting with dust removal strategies through simulation models, the feasibility and effectiveness of the strategies can be verified before actual application, thus reducing the risks in actual applications.

[0045] The semiconductor product dust removal method of this embodiment performs difference regionalization processing on the real-time collected surface image of the semiconductor product, merges the discretely distributed point-like dust into regions, and uses an image segmentation algorithm to completely divide the dust region in the dust condensation image, so as to ensure the accurate identification of the dust region and avoid the size recognition error caused by the distribution of dust clusters; according to the size characteristics, grayscale gradient characteristics and the use time of the dust removal rod of the dust region, the contact or non-contact dust removal mode is dynamically selected; for larger dust regions, the dust removal rod is prevented from contacting the surface to reduce the risk of secondary pollution; according to the actual situation of the dust region, the dust removal intensity is adjusted to ensure the cleaning effect, so as to realize the intelligence and refinement of the dust removal strategy, significantly reduce the risk of secondary pollution, and improve the dust removal efficiency and effect; before actual application, the feasibility and effectiveness of the dust removal strategy are verified through a simulation model to reduce the risk in actual application.

[0046] For details, see Figure 5 , Figure 5 It is a schematic diagram showing an execution method of dust removal of semiconductor products based on semi-solidified bodies in one embodiment of the present invention. Figure 5 Where a is the AI ​​system, executing steps S202 and S203, b is the camera, executing step S201, c is the execution system, executing step S204, and d is the chip.

[0047] In one embodiment, Figure 1 In step S201, the surface image is subjected to difference regionalization processing to obtain a dust condensation image, including steps S301 to S303: Step S301, performing a two-stage screening process on all pixels in the surface image based on the grayscale difference between each pixel in the surface image and the overall image, and the grayscale difference between each pixel in the surface image and the neighborhood range, to obtain each outlier pixel and each background pixel.

[0048] It should be noted that the background pixel points represent the pixel points belonging to the background and located in the dust area in the surface image, and the outlier pixel points represent the pixel points belonging to the dust and located in the dust area in the surface image.

[0049] In this step, a two-level screening process is designed as follows: The average value of the grayscale values ​​of all pixels in the surface image is used as the semiconductor pixel reference value; the absolute value of the difference between the grayscale value of each pixel in the surface image and the semiconductor pixel reference value is used as the overall grayscale difference value of each pixel in the surface image; all pixels in the surface image are subjected to primary comparison and screening using the overall grayscale difference value to obtain each singular pixel and each background pixel; a corresponding neighborhood window is constructed with each singular pixel as the center, and the neighborhood grayscale difference value of each singular pixel is calculated based on the grayscale difference characteristics of each singular pixel and all pixels in the corresponding neighborhood window; all singular pixels in the surface image are subjected to secondary comparison and screening using the neighborhood grayscale difference value to obtain each outlier pixel.

[0050] It should be noted that the difference between each pixel and the overall grayscale reference value is calculated through the first-level comparison screening, and all pixels that may belong to the dust area are initially screened out as singular pixels, that is, singular pixels, to reduce the possibility of missed detection. This step uses the overall grayscale average value of the image as a reference value, which can adapt to different lighting conditions and image background changes, avoid errors caused by fixed thresholds, and quickly separate pixels that may belong to the dust area; then, through the second-level comparison screening, a neighborhood window is constructed and the neighborhood grayscale difference value is calculated, which can combine local grayscale features to more accurately determine whether the pixel belongs to the dust area, effectively avoid misjudging background pixels as dust pixels, and reduce the false detection rate. The first-level screening quickly narrows the detection range, and the second-level screening is only performed on singular pixels, avoiding the complex calculation of pixels in the entire image and improving the efficiency of the algorithm.

[0051] As an embodiment of the present application, the average value of the overall grayscale difference values ​​of all pixels in the surface image is used as the overall grayscale difference reference value, and all pixels whose overall grayscale difference values ​​are greater than or equal to the overall grayscale difference reference value are used as singular pixels. All singular pixels are classified into two categories based on all neighborhood grayscale difference values, and the average value of the neighborhood grayscale difference values ​​of all singular pixels in each category is used as the neighborhood grayscale average difference, all singular pixels in the category with the largest neighborhood grayscale average difference are used as outlier pixels, and all singular pixels in the category with the smallest neighborhood grayscale average difference are used as background pixels.

[0052] Step S302, based on the position proximity and corresponding grayscale similarity between the two outlier pixels, the same-region belonging similarity between the two outlier pixels is calculated; the same-region belonging similarity is used as the similarity measurement criterion of the clustering algorithm, and the clustering algorithm is used to cluster all the outlier pixels to obtain various point clusters.

[0053] It should be noted that by considering the spatial distance between outlier pixels through position proximity, dust pixels with similar positions can be grouped into the same cluster, avoiding the incorrect merging of scattered dust areas; by considering the grayscale similarity between outlier pixels, dust pixels with similar grayscale features can be grouped into the same cluster, avoiding the incorrect merging of dust areas with large grayscale differences. Combining position proximity and grayscale similarity can more accurately reflect the correlation between dust pixels, thereby achieving accurate clustering of dust areas, merging scattered pixels that belong to the same dust area into one cluster, avoiding the over-segmentation problem caused by scattered pixels, and improving the subsequent accurate judgment of dust size.

[0054] As an embodiment of the present application, the absolute value of the grayscale difference between two outlier pixels is used as the point-to-point grayscale difference value of the two pixels, the sum of the position distance between the two outlier pixels and the corresponding point-to-point grayscale difference value is used as the corresponding same-region belonging difference index, and the calculation result of an exponential function with a natural constant as the base and the opposite of the same-region belonging difference index as the exponent is used as the same-region belonging similarity between the two outlier pixels.

[0055] Step S303, based on the distance between each background pixel and each point cluster, background pixels are screened to obtain each pixel to be fused; the grayscale value of each pixel to be fused is fused and updated according to the grayscale values ​​of all outlier pixels to obtain a fused grayscale weighted value of each pixel to be fused; the grayscale value of each pixel to be fused in the surface image is replaced with the fused grayscale weighted value to obtain a dust condensation image.

[0056] It should be noted that by calculating the distance between background pixels and point clusters, background pixels close to the dust area, i.e., pixels to be fused, can be screened out. This screening mechanism ensures that only background pixels related to the dust area are included in the fusion process to avoid interference with irrelevant background areas. The dust condensation image can better adapt to the dispersion, aggregation or irregular distribution of dust areas, clearly display the distribution and morphology of dust areas, and provide more intuitive support for the evaluation of dust size, thereby improving the accuracy of subsequent dust removal strategy determination.

[0057] In this step, the fusion update process is designed as follows: The point cluster with the smallest distance from each pixel to be fused is used as the cluster to be integrated for each pixel to be fused; the distance between each pixel to be fused and all outlier pixels in the corresponding cluster to be integrated is normalized to obtain the contribution weight of each outlier pixel in the cluster to be integrated; the grayscale values ​​of all outlier pixels in the cluster to be integrated for each pixel to be fused are weightedly summed with the contribution weight to obtain the fusion grayscale weighted value of each pixel to be fused.

[0058] It should be noted that, by taking the point clustering cluster closest to the pixel to be fused as the clustering cluster to be merged, it can ensure that the pixel to be fused is merged with the most relevant dust area, avoiding erroneous association of the pixel to be fused with an irrelevant dust area, and normalizing the distance between the pixel to be fused and the outlier pixel in the clustering cluster to be merged, so that the distance information can be converted into a contribution weight. The closer the outlier pixel is, the greater its contribution to the fusion grayscale value, which can dynamically reflect the influence of the outlier pixel on the pixel to be fused, avoiding obvious edges or discontinuous areas in the fused image. The fused dust condensation image can more clearly show the characteristics of the dust area, which is convenient for counting the number, area, distribution and other information of the dust area.

[0059] The semiconductor product dust removal method of this embodiment calculates the difference between each pixel and the overall grayscale reference value to preliminarily screen out singular pixels to reduce the possibility of missed detection, and further screens out outlier pixels to reduce the false detection rate by constructing a neighborhood window and calculating the neighborhood grayscale difference value. The two-stage screening process significantly improves the accuracy of dust detection and reduces missed detection and false detection. Combining position proximity and grayscale similarity, the similarity between outlier pixels is calculated as the similarity measurement criterion of the clustering algorithm, which can more accurately reflect the correlation between dust pixels and avoid the over-segmentation problem caused by pixel dispersion. The clustering algorithm is further used to cluster all outlier pixels into point clusters to achieve accurate division of dust areas, and through fusion update processing, it is ensured that only background pixels related to the dust area are included in the fusion processing to avoid interference with irrelevant background areas. The grayscale value of the pixel to be fused is replaced by the fusion grayscale weighted value to generate a dust condensation image, which clearly shows the distribution and morphology of the dust area, provides more intuitive support for the evaluation of dust size, and thus improves the reliability of dust removal mode determination.

[0060] In one embodiment, Figure 1 In step S203, each dust area is obtained according to the contour feature of each area to be identified in the surface image, including: The contour lines of each area to be identified in the surface image are extracted, all the pixels on each contour line are numbered in a clockwise direction, and the directional change rate of each pixel on each contour line is calculated; the standard deviation of the directional change rate of all the pixels on each contour line is used as the contour complexity index of each contour line; all the areas to be identified are screened with the contour complexity index to obtain various dust areas.

[0061] It should be noted that by mapping the area to be identified segmented from the dust condensation image to the corresponding area in the surface image, the point-like but clustered dust area can be analyzed as a whole while retaining the original surface image features. Dust areas often have irregular shapes and complex contours. Further, by calculating the directional change rate of each pixel on the contour line, the local geometric features of the contour line can be quantified, reflecting the degree of curvature and change trend of the contour line. The standard deviation of the directional change rate is used as the contour complexity index, which can comprehensively reflect the complexity of the entire contour line. Dust areas usually have irregular contours and have a higher contour complexity index, while the contour complexity index of components on the semiconductor surface is lower.

[0062] As an embodiment of the present application, the average value of contour complexity indexes of all to-be-identified areas is used as a contour complexity reference threshold, and all to-be-identified areas corresponding to contour complexity indexes greater than or equal to the contour complexity reference threshold are used as dust areas.

[0063] In one embodiment, Figure 1 In step S203, the dust removal strategy for each dust area is dynamically determined based on the use time of the semi-solidified dust removal rod, the size characteristics of each dust area and the corresponding grayscale gradient characteristics, including steps S401 to S403: Step S401, taking the pixel point in each dust area with the largest difference from the semiconductor pixel reference value as the dust accumulation point of each dust area; taking the rays connecting the dust accumulation point of each dust area and each pixel point on the corresponding contour line as the dust accumulation lines of each dust area, and calculating the dust accumulation index of each dust area based on the gradient of the grayscale values ​​of the pixel points on all dust accumulation lines of each dust area; wherein the dust accumulation index reflects the grayscale gradient characteristics of the corresponding dust area.

[0064] It should be noted that by selecting the pixel point with the largest difference from the semiconductor pixel reference value as the dust accumulation point, the core position of the dust area, that is, the area with the most serious dust accumulation, can be accurately reflected. Based on the change in the grayscale value of the pixel point on the dust accumulation line, the grayscale gradient characteristics of the dust area can be quantified. The dust area usually has a grayscale gradient from the center to the edge, while the grayscale changes in the background area or regular area are more uniform.

[0065] The dust accumulation index of each dust area is calculated based on the gradient of the grayscale values ​​of the pixels on all the dust accumulation lines of each dust area, including: The difference between the grayscale value of each pixel point on each dust accumulation line and the grayscale value of the corresponding next pixel point is taken as the grayscale gradient value of each pixel point on each dust accumulation line; the sum of the grayscale gradient values ​​of all pixels on each dust accumulation line is taken as the linear gradient intensity of each dust accumulation line; and the average value of the linear gradient intensities of all dust accumulation lines in each dust area is taken as the dust accumulation index of each dust area.

[0066] It should be noted that the difference between the grayscale value of each pixel on the dust accumulation line and the grayscale value of the corresponding next pixel reflects the radial grayscale change of the dust area. The dust accumulation line can reflect the distribution of dust from the center to the edge. The radially distributed dust accumulation line can more comprehensively capture the grayscale gradient characteristics of the dust area, indirectly reflect the dust accumulation thickness, and improve the accuracy of subsequent assessment of the severity of dust accumulation.

[0067] For details, see Figure 6 , Figure 6 A schematic diagram of a dust accumulation line in one embodiment of the present invention is shown.

[0068] Step S402, based on the linear relationship between the dust accumulation index of each dust area and the total number of pixels in each dust area, the dust cleaning urgency index of each dust area is calculated; wherein the total number of pixels in each dust area reflects the size characteristics of the corresponding dust area.

[0069] It should be noted that the dust accumulation index reflects the grayscale gradient characteristics of the dust area, and the total number of pixels in the dust area reflects the size characteristics of the dust area. By combining the dust accumulation index and the dust area size, the severity of dust accumulation in the dust area can be comprehensively evaluated, thereby improving the accuracy of the subsequent dust removal strategy determination.

[0070] The product of the dust accumulation index of each dust region and the total number of pixels in each dust region may be used as the dust cleaning urgency index of each dust region.

[0071] It should be noted that the dust cleaning urgency index is used to evaluate the severity of dust accumulation. It comprehensively considers the thickness and area size of dust accumulation, improves the accuracy of the assessment of the severity of dust accumulation, and thus improves the reliability of the subsequent dust removal strategy determination.

[0072] Step S403, input the usage time of the semi-solidified dust removal rod and the dust cleaning urgency index of each dust area into the mode switching control expression, calculate the non-contact mode switching factor of each dust area; adjust the dust removal strategy of each dust area with the non-contact mode switching factor.

[0073] It should be noted that each semi-solidified dust removal rod is provided with a timer, which can record the usage time of the semi-solidified dust removal rod during its use. The surface of the semiconductor device may include multiple dust areas. When using the dust removal rod to remove dust from the surface of the semiconductor device, for each dust area, it is necessary to comprehensively consider the characteristics of the dust area and the usage of the dust removal rod. Through the mode switching control expression, the non-contact mode switching factor is automatically calculated and the dust removal strategy is adjusted to reduce manual intervention and reduce the risk of secondary pollution caused by improper operation.

[0074] In this step, the mode switching control expression is: ; in, is the non-contact dust removal mode switching factor for dust area j, is the normalization function, The use time of the semi-solid dust removal rod. is an exponential function with a natural constant as base, is the dust cleaning urgency index of dust zone j.

[0075] It should be noted that the dust cleaning urgency index The value range of is from 0 to positive infinity. The value range of is standardized to 0 to 1. The value of the non-contact dust removal mode switching factor is limited to between 0 and 1 through the normalization function, which effectively prevents the occurrence of gradient explosion and protects the stability and reliability of the model. When the semi-solidified dust removal rod is used for a shorter time and the severity of dust accumulation is lower, it means that the surface of the semi-solidified dust removal rod adsorbs less dust. At this time, the contact dust removal is more obvious than the non-contact dust removal for improving the cleaning efficiency. The contact dust removal mode should be used more to increase the dust removal rate, and the non-contact dust removal mode switching factor value is smaller; when the semi-solidified dust removal rod is used for a longer time and the severity of dust accumulation is higher, it means that the surface of the semi-solidified dust removal rod is more likely to have adsorbed more dust, resulting in weakened adsorption capacity. At this time, the contact dust removal is not obvious compared with the non-contact dust removal for improving the cleaning efficiency. The non-contact dust removal mode should be used more to reduce the risk of secondary pollution, and the non-contact dust removal mode switching factor value is larger.

[0076] The semiconductor product dust removal method of this embodiment selects the pixel point with the largest difference from the semiconductor pixel reference value as the dust accumulation point, which can accurately reflect the core position of the dust area and realize the precise positioning of the dust accumulation point. The dust accumulation is quantitatively evaluated by calculating the dust accumulation index, which reflects the grayscale gradient characteristics of the dust area and improves the accuracy of the assessment of the severity of dust accumulation. In combination with the total number of pixels in the dust area, the thickness and area size of the dust accumulation are comprehensively considered, so that the severity of dust accumulation can be more comprehensively evaluated, providing a reliable basis for determining the dust removal strategy. Based on the use time of the semi-solid dust removal rod and the dust cleaning urgency index, the non-contact mode switching factor is calculated to realize the automatic adjustment of the dust removal strategy, which significantly reduces the risk of secondary pollution while improving the dust removal efficiency, shortening the cleaning cycle, and reducing the cleaning cost.

[0077] In one embodiment, the dust removal strategy of each dust area is adjusted using the non-contact mode switching factor in step S403, including: When the non-contact mode switching factor is less than a preset threshold, the dust removal mode of the corresponding dust area is adjusted to the contact dust removal mode, and the dust removal intensity of the corresponding dust area is adjusted to level one; When the non-contact mode switching factor is greater than or equal to a preset threshold, the dust removal mode of the dust area is adjusted to the non-contact dust removal mode, and the dust removal intensity of the corresponding dust area is adjusted to level two.

[0078] It should be noted that by combining the non-contact dust removal mode with the secondary dust removal intensity, the dust removal rate can be increased as much as possible while reducing the risk of secondary pollution.

[0079] It should be noted that the preset threshold is a manually preset value. In this embodiment, the preset threshold is 0.5. For example, when the use time of the semi-solidified dust removal rod is 0, the non-contact mode switching factor is 0, the contact dust removal mode is adopted, and the dust removal intensity is adjusted to level one; when the dust cleaning urgency index is 0.5 and the use time of the semi-solidified dust removal rod is 15, the non-contact mode switching factor is 0.39, the contact dust removal mode is adopted, and the dust removal intensity is adjusted to level one; when the dust cleaning urgency index is 5 and the use time of the semi-solidified dust removal rod is 15, the non-contact mode switching factor is 0.99, the non-contact dust removal mode is adopted, and the dust removal intensity is adjusted to level two.

[0080] For details, see Figure 7 , Figure 7 It is a schematic diagram showing the acquisition of dust removal strategy in one embodiment of the present invention.

[0081] Compared with the prior art, the embodiments of the present invention have the following advantages: During the dust removal process, the surface image of the semiconductor product to be dusted is collected in real time. Considering the granular characteristics of dust, it may present a discrete distribution in the form of points. In order to avoid the point-like but clustered dust area, which leads to size recognition errors and then adopts the wrong dust removal strategy, the surface image is first subjected to difference regionalization processing to obtain a dust condensation image; the difference regionalization processing is designed to diffuse the differences of outlier pixels in the surface image to merge the discretely distributed point-like dust into regions, thereby improving the accuracy of subsequent semiconductor device segmentation; the dust condensation image is segmented to obtain various areas to be identified; considering that the distribution of dust is often irregular and the contour is relatively complex, while the contours of the components on the surface of the semiconductor device are relatively simple, each dust area is obtained according to the contour features of each area to be identified in the surface image; based on the use time of the semi-solidified dust removal rod, the size characteristics of each dust area and the corresponding The grayscale gradient feature dynamically determines the dust removal strategy for each dust area; wherein the dust removal strategy includes a dust removal mode; the dust removal mode includes a contact dust removal mode and a non-contact dust removal mode, the contact dust removal mode is designed so that the dust removal rod contacts the dust on the surface of the semiconductor product and the dust removal rod does not contact the semiconductor product, and the non-contact dust removal mode is designed so that the dust removal rod does not contact the semiconductor product or the dust on the surface of the semiconductor product; when the non-contact mode switching factor is less than a preset threshold, the dust removal mode of the corresponding dust area is adjusted to the contact dust removal mode, and the dust removal intensity of the corresponding dust area is adjusted to level one; when the non-contact mode switching factor is greater than or equal to the preset threshold, the dust removal mode of the dust area is adjusted to the non-contact dust removal mode, and the dust removal intensity of the corresponding dust area is adjusted to level two, and the dust removal strategy is adjusted in real time according to the dust distribution on the semiconductor surface, thereby improving the dust removal effect on semiconductor devices.

[0082] One embodiment of the present invention provides a semiconductor product dust removal system based on a semi-solidified body. For details, see Figure 8 , Figure 8 A schematic diagram of a control console in a semi-solidified solid dust removal system in one embodiment of the present invention is shown, wherein the dust removal system includes a control console, a manipulator and a semi-solidified solid dust removal rod, wherein the manipulator is fixedly arranged in a control area on the control console, and the dust removal rod is placed in a storage area on the control console, and a computer device is arranged inside the counter of the console, wherein the computer device includes a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, wherein the processor is communicatively connected with the manipulator, and when the processor executes the computer program, the manipulator is controlled to grab the semi-solidified solid dust removal rod, so as to realize the semiconductor product dust removal method based on the semi-solidified solid as described above.

[0083] As an embodiment of the present application, the system further includes a visual detection module, a first mobile module, and a second mobile module. Fig. 9 , Fig. 9 Shown is a schematic diagram of the hardware structure of a dust removal system for semiconductor products based on a semi-solidified body in one embodiment of the present invention. Fig. 9 The computer device 101 is arranged inside the counter of the console, the first output end of the computer device 101 is connected to the manipulator 102 through the first mobile module 104, and the second output end of the computer device 101 is connected to the visual detection module 103 through the second mobile module 105. When the system is working: the computer device 101 controls the second mobile module 105 to move the visual detection module 103 to the target position, obtains the image of the target to be measured, detects the dust position, dust area and dust removal strategy (contact or non-contact) in the image, determines the size of the dust removal rod to be used, and then controls the first mobile module 104 to move the manipulator 102 to the top of the dust removal rod in the corresponding storage area, grabs the dust removal rod of matching size, and then controls the manipulator 102 to move to the dust area to perform dust removal operation.

[0084] For details, see Fig.10 , Fig.10 FIG. 1 is a schematic diagram showing the structure of a computer device in one embodiment of the present invention. Fig.10 As shown, the computer device of this embodiment includes: at least one processor ( Fig.10 Only one is shown in the figure), a memory, and a computer program stored in the memory and executable on at least one processor, wherein when the processor executes the computer program, the steps in any of the above-mentioned abnormal display detection method embodiments are implemented.

[0085] The computer device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that Fig.10 This is merely an example of a computer device and does not constitute a limitation on the computer device. The computer device may include more or fewer components than those shown in the figure, or a combination of certain components, or different components. For example, it may also include a network interface, a display screen, and an input device.

[0086] The processor may be a CPU, or other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0087] The memory includes a readable storage medium, an internal memory, etc., wherein the internal memory may be the memory of a computer device, and the internal memory provides an environment for the operation of the operating system and computer-readable instructions in the readable storage medium. The readable storage medium may be a hard disk of a computer device, and in other embodiments, it may also be an external storage device of the computer device, for example, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device. Further, the memory may also include both an internal storage unit of the computer device and an external storage device. The memory is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of a computer program, etc. The memory may also be used to temporarily store data that has been output or is to be output.

[0088] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A dust removal method for semiconductor products based on semi-solidified solids, characterized in that: The method comprises: In the dust removal process, a surface image of the semiconductor product to be dusted is collected in real time, and a difference regionalization process is performed on the surface image to obtain a dust condensation image; the difference regionalization process is designed to diffuse the difference of outlier pixel points in the surface image to merge discretely distributed point-shaped dust into regions; Performing image segmentation on the dust condensation image to obtain various areas to be identified; Acquire each dust area according to the contour features of each area to be identified in the surface image; dynamically determine the dust removal strategy for each dust area based on the use time of the semi-solidified dust removal rod, the size features of each dust area and the corresponding grayscale gradient features; Wherein, the dust removal strategy includes a dust removal mode; The dust removal mode includes a contact dust removal mode and a non-contact dust removal mode. The contact dust removal mode is designed so that the dust removal rod contacts the dust on the surface of the semiconductor product, and the dust removal rod does not contact the semiconductor product; the non-contact dust removal mode is designed so that the dust removal rod does not contact the semiconductor product, and the dust on the surface of the semiconductor product does not contact each other. The functional components matched with the dust removal rod are controlled to execute the corresponding dust removal strategy for each dust area.

2. The dust removal method for semiconductor products based on semi-solidified solid according to claim 1, characterized in that: The performing difference regionalization processing on the surface image to obtain the dust condensation image includes: Based on the grayscale difference between each pixel in the surface image and the overall image, and the grayscale difference between each pixel in the surface image and the neighborhood range, a two-stage screening process is performed on all the pixels in the surface image to obtain each outlier pixel and each background pixel; Calculating the similarity of belonging to the same region between the two outlier pixels based on the position proximity and the corresponding grayscale similarity between the two outlier pixels; The same region belonging similarity is used as a similarity measurement criterion of a clustering algorithm, and the clustering algorithm is used to cluster all the outlier pixels to obtain various point clusters; The background pixels are screened based on the distance between each background pixel and each point-shaped cluster to obtain each pixel to be fused; Performing a fusion update process on the grayscale value of each pixel to be fused according to the grayscale values ​​of all the outlier pixels, so as to obtain a fusion grayscale weighted value of each pixel to be fused; The grayscale value of each pixel to be fused in the surface image is replaced with the fused grayscale weighted value to obtain a dust condensation image.

3. The dust removal method for semiconductor products based on semi-solidified solid according to claim 2, characterized in that: The two-stage screening process is designed to: Taking the average grayscale value of all pixels in the surface image as the semiconductor pixel reference value; Taking the absolute value of the difference between the grayscale value of each pixel in the surface image and the semiconductor pixel reference value as the overall grayscale difference value of each pixel in the surface image; Performing a primary comparison and screening on all the pixels in the surface image using the overall grayscale difference value to obtain each singular pixel and each background pixel; Constructing a corresponding neighborhood window with each singular pixel as the center, and calculating a neighborhood grayscale difference value of each singular pixel based on the grayscale difference characteristics of each singular pixel and all pixels in the corresponding neighborhood window; A secondary comparison and screening is performed on all the singular pixel points in the surface image using the neighborhood grayscale difference value to obtain various outlier pixel points.

4. The semiconductor product dust removal method based on semi-solidified body according to claim 2, characterized in that: The fusion update process is designed to: The point cluster with the smallest distance from each pixel to be fused is used as the cluster to be merged into each pixel to be fused; Normalizing the distance between each pixel to be fused and all the outlier pixels in the corresponding cluster to be fused, to obtain a contribution weight of each outlier pixel in the cluster to be fused; The grayscale values ​​of all the outlier pixels in the cluster to be integrated into each pixel to be integrated are weightedly summed according to the contribution weight to obtain a fused grayscale weighted value of each pixel to be integrated.

5. The semiconductor product dust removal method based on semi-solidified body according to claim 3, characterized in that: The step of acquiring each dust area according to the contour feature of each area to be identified in the surface image comprises: Extracting the contour line of each of the areas to be identified, numbering all the pixels on each of the contour lines in a clockwise direction, and calculating the direction change rate of each pixel on each of the contour lines; The standard deviation of the direction change rate of all pixel points on each contour line is used as the contour complexity index of each contour line; all the areas to be identified are screened by the contour complexity index to obtain various dust areas.

6. The semiconductor product dust removal method based on semi-solidified body according to claim 5, characterized in that: The method of dynamically determining the dust removal strategy for each dust area based on the use time of the semi-solidified dust removal rod, the size characteristics of each dust area, and the corresponding grayscale gradient characteristics includes: Taking the pixel point with the largest difference between the semiconductor pixel reference value and each dust region as the dust accumulation point of each dust region; The ray obtained by connecting the dust accumulation point of each dust area and each pixel point on the corresponding contour line is used as each dust accumulation line of each dust area, and the dust accumulation index of each dust area is calculated based on the gradient degree of the grayscale values ​​of the pixel points on all the dust accumulation lines of each dust area; wherein the dust accumulation index reflects the grayscale gradient characteristics of the corresponding dust area; Based on the linear relationship between the dust accumulation index of each dust area and the total number of pixels in each dust area, the dust cleaning urgency index of each dust area is calculated; wherein the total number of pixels in each dust area reflects the size characteristics of the corresponding dust area; Inputting the use time of the semi-solidified dust removal rod and the dust cleaning urgency index of each dust area into the mode switching control expression, and calculating the non-contact mode switching factor of each dust area; The dust removal strategy of each of the dust areas is adjusted according to the non-contact mode switching factor.

7. The dust removal method for semiconductor products based on semi-solidified solid according to claim 6, characterized in that: The step of adjusting the dust removal strategy of each dust area according to the non-contact mode switching factor includes: When the non-contact mode switching factor is less than a preset threshold, the dust removal mode of the corresponding dust area is adjusted to the contact dust removal mode, and the dust removal intensity of the corresponding dust area is adjusted to level one; When the non-contact mode switching factor is greater than or equal to a preset threshold, the dust removal mode of the dust area is adjusted to the non-contact dust removal mode, and the dust removal intensity of the corresponding dust area is adjusted to level two.

8. The dust removal method for semiconductor products based on semi-solidified solid according to claim 6, characterized in that: The calculating the dust accumulation index of each dust region based on the gradient of the grayscale values ​​of the pixels on all the dust accumulation lines of each dust region comprises: The difference between the grayscale value of each pixel point on each dust accumulation line and the grayscale value of the corresponding next pixel point is used as the grayscale gradient value of each pixel point on each dust accumulation line; The sum of the grayscale gradient values ​​of all pixels on each dust accumulation line is used as the linear gradient intensity of each dust accumulation line, and the average value of the linear gradient intensities of all the dust accumulation lines in each dust area is used as the dust accumulation index of each dust area.

9. The dust removal method for semiconductor products based on semi-solidified solid according to claim 6, characterized in that: The mode switching control expression is: ； in, is the non-contact dust removal mode switching factor for dust area j, is the normalization function, The use time of the semi-solid dust removal rod. is an exponential function with a natural constant as base, is the dust cleaning urgency index of dust zone j.

10. A semiconductor product dust removal system based on semi-solidified solid, characterized in that: The system includes a control console, a manipulator and a semi-solidified dust removal rod. The manipulator is fixedly arranged in a control area on the control console, and the dust removal rod is placed in a storage area on the control console. A computer device is arranged inside the counter of the control console. The computer device includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. The processor is communicatively connected with the manipulator. When the processor executes the computer program, the manipulator controls the manipulator to grab the semi-solidified dust removal rod, so as to realize the dust removal method for semiconductor products based on semi-solidified solids as described in any one of claims 1 to 9.

Images