Quantity detection equipment and formula setting method and application method based on real-object-free scene

Abstract

The invention provides quantity detection equipment and a formula setting method and application method based on a real-object-free scene. A detection image of a to-be-detected object is generated through analogue simulation by obtaining specification information of the to-be-detected object and pre-configuring an attitude adjustment initial parameter of the to-be-detected object, an attitude adjustment parameter adaptive algorithm, a detection signal acquisition initial parameter and a detection signal acquisition parameter adaptive algorithm, and the to-be-detected object does not need to reach current equipment; a formula creation and configuration method can be carried out in advance; and the quantity detection equipment completes formula editing of quantity detection in advance, so that the production efficiency is improved. According to the invention, image simulation is carried out according to the specification information of the to-be-detected object without depending on actual image acquisition, so that the utilization rate of equipment is improved. The configuration parameters in the formula are divided into the pre-configuration attitude adjustment initial parameters, the attitude adjustment parameter adaptive algorithm, the detection signal acquisition initial parameters and the detection signal acquisition parameter adaptive algorithm, so that the reusability of the formula of the quantity detection equipment is improved.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor technology, and in particular to a quantity detection device and a recipe setting method and an application method based on a non-physical scene. Background Art

[0002] The recipe of mass testing equipment defines the specific operation process that the equipment needs to perform on the object to be tested, the control strategy and static data in each operation process during the mass production batch process, and finally saves it in the form of a recipe file so that it can be loaded during batch processing.

[0003] In the traditional quantity detection recipe configuration process, it is necessary to first upload the detection object to the detection object carrier, and move the detection area of ​​the detection object to the effective detection range of the signal detector. Then, professional process personnel will repeatedly design strategies and configure parameters for each operation process based on the description information of the detection object and the real-time detection results of the detector, and decide whether to save the quantity detection recipe based on the final detection results.

[0004] It can be seen that in the traditional quantity detection recipe configuration process, the quantity detection equipment requires the actual detection object during the recipe configuration process to complete the configuration and storage of the recipe parameters. In addition, the quantity detection equipment needs to capture images of the detection object when editing the quantity detection area, and perform area editing based on the captured image; and the parameters in each operation link in the recipe are fixed, which lacks adaptability to process changes and other system fluctuations. Summary of the invention

[0005] The purpose of the present invention is to provide a quantity detection device and a recipe setting method and application method based on a non-physical scene, so as to solve at least one of the problems that the traditional quantity detection recipe configuration process must rely on the actual detection object, perform actual image acquisition on the detection object, and lack adaptability when the process changes.

[0006] In order to solve the above technical problems, the present invention provides a recipe setting method based on a non-physical scene, comprising:

[0007] Acquire specification information of the object to be detected, and complete configuration of basic information of the object to be detected according to the specification information;

[0008] Pre-configuring initial parameters for posture adjustment of the object to be detected, an adaptive algorithm for posture adjustment parameters, initial parameters for detection signal acquisition, an adaptive algorithm for detection signal acquisition parameters, and parameters for quantity detection algorithm;

[0009] Based on the specification information and the preconfigured detection signal acquisition initial parameters and detection signal acquisition parameter adaptive algorithm, a detection image of the object to be detected is generated by simulation;

[0010] A recipe is saved, wherein the recipe includes at least one of basic information of the object to be detected, initial parameters for posture adjustment, an adaptive algorithm for posture adjustment parameters, initial parameters for detection signal acquisition, an adaptive algorithm for detection signal acquisition parameters, and parameters of a quantity detection algorithm.

[0011] Optionally, the posture adjustment parameters of the object to be detected include horizontal posture adjustment parameters and vertical posture adjustment parameters.

[0012] Optionally, in the horizontal posture adjustment parameter configuration process, the horizontal posture adjustment parameter is configured as an initial optical parameter for image acquisition in the alignment phase.

[0013] Optionally, in the horizontal posture adjustment parameter configuration process, an optical parameter adaptive algorithm is used to adjust the horizontal posture adjustment parameters.

[0014] Optionally, the adjustment method of the optical parameter adaptive algorithm includes:

[0015] Capturing images of the object under test;

[0016] Perform grayscale histogram analysis on the collected image to obtain characteristic information of grayscale distribution of the object to be detected after imaging;

[0017] Adopting an automatic parameter adjustment algorithm to adjust optical parameters and re-execute the image acquisition step according to the characteristic information of the grayscale distribution and a preset recommended interval, until the overall distribution of the grayscale distribution approaches the recommended interval;

[0018] After the grayscale distribution reaches the standard of the recommended interval, the smoothness of the distribution curve of the grayscale histogram is determined and the optical parameters are adjusted to meet the measurement requirements.

[0019] Optionally, the initial parameters of the horizontal posture adjustment are obtained by matching feature image templates to obtain a mapping relationship between nominal positions and actual positions of multiple alignment marks, and a horizontal coordinate system transformation model is solved based on the mapping relationship of multiple groups of alignment marks.

[0020] Optionally, in the mass production stage, the method for adjusting the initial parameters according to the preconfigured horizontal posture to complete the feature template extraction of the pattern includes:

[0021] The image acquisition is performed on the object to be detected, and the image acquisition covers a range greater than 1.5 times the size of the pattern periodically distributed on the surface of the object to be detected;

[0022] Generate a binary image of the outer contour of the periodic pattern according to the mapping relationship between the physical size of the information of the object to be detected and the pixels;

[0023] Using the generated binary image of the outer contour, step-slide matching is performed along the X direction and the Y direction with the origin of the acquired image as the starting position;

[0024] In the sliding matching process, a series of matching results are formed, data analysis is performed on the matching results, and the distribution frequency of the highest matching score intervals in the X direction and the Y direction is obtained. The position of the periodic pattern in the matched image is identified through the distribution frequency;

[0025] The feature entropy of a single periodic pattern image is calculated, and the area with the largest feature entropy is automatically intercepted as the feature matching template in the horizontal alignment process.

[0026] Optionally, the judgment criterion for data analysis of the matching results is: performing image similarity comparison and scoring between the binarized image of the outer contour and the image obtained during sliding, setting a similarity threshold in the adaptive parameters, and when the similarity exceeds the threshold and the results appear periodically, determining the periodic position as the location of the pattern.

[0027] Optionally, edge position recognition of the detection object is performed by performing grayscale projection on the image of the periodic pattern.

[0028] Optionally, the initial parameters for adjusting the vertical posture include vertical control parameters of the object to be detected during the signal detection process.

[0029] Optionally, the method for adjusting the vertical posture includes:

[0030] Using a measuring sensor to measure the vertical distance between the surface of the object to be detected and the imaging lens, and adjusting the vertical position of the object to be detected or the imaging lens according to a theoretical optimal focal plane height;

[0031] Collecting the surface image of the object to be detected and calculating the image clarity;

[0032] Adjusting the vertical position of the object to be detected or the imaging lens at a set step distance, so that the surface of the object to be detected is away from the imaging lens, and then re-capturing the image and re-calculating the image clarity;

[0033] After the object to be detected is away from the actual optimal focal plane, completing adjustment detection in a single direction;

[0034] Restoring the vertical position of the object to be detected or the imaging lens to the theoretical focal plane height, and gradually adjusting the vertical position of the object to be detected or the imaging lens in the opposite direction and collecting images to calculate the clarity;

[0035] Automatically identify and select the best focal plane posture based on the measured statistical clarity changes.

[0036] Optionally, in the vertical attitude adjustment initial parameter configuration process, an optical parameter adaptive algorithm is used to adjust the vertical attitude adjustment parameters.

[0037] Optionally, after the simulation generates the detection image, the following is further included:

[0038] Editing amount detection effective area, wherein different areas are distinguished according to the characteristics of the object to be detected;

[0039] Pre-configure detection parameters and parameter adaptation algorithms for different areas.

[0040] Optionally, based on the overall specification information of the object to be detected, configure the scope that needs to be involved in the detection during the actual detection process.

[0041] Optionally, the valid specification information of the object to be detected includes: at least one of the size of the object to be detected, whether there is a valid pattern to be detected on the object to be detected, and a distribution pattern of the pattern to be detected.

[0042] Optionally, the valid specification information of the object to be inspected includes an exposure process design drawing of the object to be inspected.

[0043] Based on the same inventive concept, an application method of a recipe based on a non-physical scene adopts a recipe setting method based on a non-physical scene as described in any of the above items to create a quantity detection recipe, adopts the quantity detection recipe to perform quantity detection, and adjusts the adaptive algorithm parameters in the quantity detection recipe according to the detection results in the mass production process.

[0044] Based on the same inventive concept, the present invention also provides a quantity detection device, which uses the recipe setting method based on the non-physical scene as described in any of the above items to create a recipe for quantity detection.

[0045] The recipe setting method based on the non-physical scene provided by the present invention generates a detection image of the object to be detected by simulating and simulating by acquiring the specification information of the object to be detected and pre-configuring the initial parameters for posture adjustment of the object to be detected, the adaptive algorithm for posture adjustment parameters, the initial parameters for detection signal acquisition, and the adaptive algorithm for detection signal acquisition parameters. The method can create and configure the recipe in advance without the object to be detected arriving at the current device; in a semiconductor manufacturing plant, the quantity detection equipment completes the recipe editing of quantity detection in advance, which helps to improve production efficiency. The present invention can perform image simulation according to the specification information of the object to be detected, and can be independent of actual image acquisition, thereby improving the utilization rate of the equipment. The present invention divides the configuration parameters in the recipe into the initial parameters for posture adjustment of the object to be detected, the adaptive algorithm for posture adjustment parameters, the initial parameters for detection signal acquisition, and the adaptive algorithm for detection signal acquisition parameters, which is conducive to improving the reusability of the recipe of the quantity detection equipment, and can quickly create a new recipe by reusing the old recipe. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Those skilled in the art will appreciate that the accompanying drawings are provided for a better understanding of the present invention and do not constitute any limitation on the scope of the present invention.

[0047] Figure 1 It is a flow chart of a method for setting a recipe without physical objects according to an embodiment of the present invention.

[0048] Figure 2 Schematic diagram of a spatial coordinate system according to an embodiment of the present invention.

[0049] Figure 3 It is a flow chart of posture control configuration according to an embodiment of the present invention.

[0050] Figure 4-Figure 6 It is a schematic diagram of adjusting optical parameters based on a grayscale histogram according to an embodiment of the present invention.

[0051] Figure 7 1 is a top view of an object to be detected according to an embodiment of the present invention.

[0052] Figure 8 Yes Yes Figure 7 An imaging diagram of the local area of ​​the object to be detected.

[0053] Fig. 9 It is a schematic diagram of a matching template graphic and a geometric outline according to an embodiment of the present invention.

[0054] Fig.10 4 is a diagram of a periodic pattern position recognition process according to an embodiment of the present invention.

[0055] Fig.11 4 is an image simulation flow chart of an embodiment of the present invention.

[0056] In the attached figure:

[0057] 10- object to be detected; 11- local imaging image;

[0058] 20-matching template image; 21-geometric contour. DETAILED DESCRIPTION

[0059] In order to make the purpose, advantages and features of the present invention clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the drawings are all in a very simplified form and are not drawn to scale, and are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention. In addition, the structure shown in the drawings is often a part of the actual structure. In particular, the emphasis of each drawing is different, and sometimes different scales are used.

[0060] As used in the present invention, the singular forms "one", "an" and "the" include plural objects, the term "or" is usually used to include the meaning of "and / or", the term "several" is usually used to include the meaning of "at least one", and the term "at least two" is usually used to include the meaning of "two or more". In addition, the terms "first", "second" and "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined as "first", "second" and "third" can explicitly or implicitly include one or at least two of the features. In addition, as used in the present invention, an element is arranged on another element, which usually only indicates that there is a connection, coupling, matching or transmission relationship between the two elements, and the connection, coupling, matching or transmission between the two elements can be direct or indirect through an intermediate element, and cannot be understood as indicating or implying the spatial position relationship between the two elements, that is, an element can be in any orientation such as inside, outside, above, below or on one side of another element, unless the content clearly indicates otherwise. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0061] Figure 1 1 is a flow chart of a method for setting a recipe without physical objects according to an embodiment of the present invention. Figure 1 As shown, this embodiment provides a recipe setting method based on a non-physical scene, including:

[0062] Step S10, obtaining specification information of the object to be detected, and completing configuration of basic information of the object to be detected according to the specification information;

[0063] Step S20, pre-configuring the initial parameters for posture adjustment of the object to be detected, the adaptive algorithm for posture adjustment parameters, the initial parameters for detection signal acquisition, the adaptive algorithm for detection signal acquisition parameters and the quantity detection algorithm parameters;

[0064] Step S30, based on the specification information and the preconfigured detection signal acquisition initial parameters and detection signal acquisition parameter adaptive algorithm, simulate and generate a detection image of the object to be detected;

[0065] Step S40, saving a recipe, wherein the recipe includes at least one of basic information of the object to be detected, initial parameters for posture adjustment, an adaptive algorithm for posture adjustment parameters, initial parameters for detection signal acquisition, an adaptive algorithm for detection signal acquisition parameters, and quantity detection algorithm parameters.

[0066] In step S10, during the recipe configuration process, the user first needs to obtain valid specification information of the object to be detected. In this embodiment, the valid specification information of the object to be detected is obtained to form a quantity detection recipe, and the specification information includes, for example, process parameters of the previous exposure of the detection object, the shape of the mark, the layout of the mark, etc. That is, the method of creating and configuring a recipe in advance does not require the object to be detected to arrive at the current device; in a semiconductor manufacturing plant, the quantity detection equipment completes the recipe editing of the quantity detection in advance, which helps to improve production efficiency.

[0067] In one embodiment, the effective specification information of the object to be detected includes the size, process information and layout information of the object to be detected, whether there is an effective pattern to be detected on the object to be detected and the distribution law of the pattern to be detected, etc. The object to be detected is, for example, a whole silicon wafer or a chip on a silicon wafer.

[0068] In another embodiment, the effective specification information of the object to be detected includes not only the size, process information and layout information of the object to be detected, whether there is an effective detected pattern on the object to be detected and the distribution pattern of the detected pattern, but also the exposure process design drawings of the object to be detected. In this embodiment, the object to be detected is, for example, a chip. The recipe creation stage will perform imaging simulation based on chip design drawings, process information, optical parameters, imaging field information, etc., to generate and correct simulated images. The final generated simulated image will be adjusted according to the changes in the input parameters, and there will be a certain mapping relationship.

[0069] In step S10, after obtaining the valid specification information of the object to be detected, the user needs to add the specification information of the object to be detected to the configuration interface according to the requirements of the recipe setting software, or automatically load the configuration information by importing the configuration file. For example, according to the valid specification information of the object to be detected, complete the configuration of the basic information of the object to be detected, including process and size, etc. After completing the parameter configuration, the configuration software automatically generates a layout image of the object to be detected based on the input information.

[0070] In step S20, the initial parameters for posture adjustment of the object to be detected in the mass production stage, the adaptive algorithm for posture adjustment parameters, the initial parameters for detection signal acquisition, the adaptive algorithm for detection signal acquisition parameters, and the parameters for the detection algorithm are preconfigured. Specifically, in the recipe configuration process, the user can configure the posture adjustment strategy and parameters of the object to be detected in the mass production process based on the acquired valid specification information of the object to be detected. Figure 2 Schematic diagram of the spatial coordinate system of an embodiment of the present invention. Figure 2As shown, during the actual detection process, the object to be detected 10 is in a three-dimensional orthogonal spatial coordinate system. In order to make the object to be detected 10 in a suitable static or dynamic posture during the detection process, the overall posture of the object to be detected needs to be adjusted. During the recipe setting process, relevant parameters need to be configured.

[0071] Figure 3 : is a flow chart of the posture control configuration of an embodiment of the present invention. Figure 3 As shown, the posture adjustment parameters of the object to be detected include horizontal posture adjustment parameters and vertical posture adjustment parameters. In the horizontal posture adjustment parameter configuration process, the horizontal posture adjustment parameters are configured as the initial optical parameters for image acquisition in the alignment stage. Optical parameters include, for example: light source mode, objective lens magnification, digital and analog gain of the camera, white balance coefficient, signal strength of different acquisition channels, etc. In order to ensure that the set formula can complete the optical parameter adaptation in the mass production stage, it is necessary to select the optical parameter adaptation algorithm and the corresponding standard in the formula setting stage. The parameter standard is mainly derived from the feature information of the image. The parameter standard includes, for example: grayscale histogram, contrast, protection, brightness, sharpness, etc. One or more feature combinations can be selected as the standard for whether the automatically adjusted optical parameters meet the standards.

[0072] Figure 4-Figure 6 It is a schematic diagram of adjusting optical parameters based on a grayscale histogram according to an embodiment of the present invention. Figure 4-Figure 6 The horizontal axis is the grayscale value, which ranges from 0 to 255, and the vertical axis is the number of pixels, which ranges from 0 to 14*10 4 Specifically, a feasible adjustment method of the optical parameter adaptive algorithm includes:

[0073] Step S21, when the recipe file is used to perform quantity detection on the actual detection object, the optical system uses the initial optical parameters used in the image acquisition in the alignment stage to perform image acquisition.

[0074] Step S22, after completing the image acquisition of the detection object, grayscale histogram analysis is performed on the image to obtain characteristic information of the grayscale distribution of the detection object after imaging, such as Figure 4 As shown; the characteristic information of the grayscale distribution is, for example, the distribution law of the grayscale distribution.

[0075] Step S23, using the automatic parameter adjustment algorithm to adjust the optical parameters and re-execute the image acquisition step according to the distribution law of the grayscale distribution and the preset recommended interval, repeating step S23, and continuously iterating this process until the overall distribution of the grayscale distribution approaches the preset recommended interval; in this embodiment, the recommended interval preset by the automatic parameter adjustment algorithm is, for example, to set the pixel points with grayscale values ​​between 100 and 200 to exceed 90%. When the statistical distribution of image pixel points is mainly concentrated in the area with lower grayscale (for example, the grayscale values ​​between 0 and 100 are 30%, then it is obvious that the pixel points with grayscale values ​​between 100 and 200 cannot exceed 90%, and there are too many pixels with lower grayscale values, at this time, the automatic parameter adjustment algorithm actively improves the light source illumination brightness according to the distribution law of the grayscale distribution and the parameters of the preset recommended interval, and re-executes the image acquisition step, repeating this step, and continuously iterating this process until the overall distribution of the grayscale distribution approaches the above-mentioned recommended interval, such as Figure 5 shown.

[0076] Step S24, after the grayscale distribution reaches the standard of the recommended interval, the smoothness of the distribution curve of the grayscale histogram is determined and the optical parameters are adjusted to meet the measurement requirements. The determination method is, for example, to obtain the distribution function by linear fitting and perform high-order parameter analysis, or to adjust the analog gain or the light source band according to the variance based on the median, so that the overall distribution is smoother and the characteristic signal with discrete distribution is enhanced, such as Figure 6 As shown in the figure, the median is the median value of the number of pixels at each gray level after the gray values ​​of all pixels in the image are counted.

[0077] During the optical parameter adjustment process, all optical parameters can be adjusted positively or negatively based on the current values, and steps S23 and S24 can be repeatedly executed in a loop to meet the quantity detection requirements.

[0078] The purpose of horizontal posture adjustment is to establish a coordinate system conversion relationship between the support platform for the object to be detected, the effective chip or graphic on the object to be detected and the imaging system. The mapping relationship between the nominal position and the actual position of multiple alignment points can be obtained by matching the feature image template, and the horizontal coordinate system conversion model is solved based on the mapping relationship of multiple groups of marks using methods including but not limited to polynomial fitting. The recipe configuration process provided in this embodiment is based on a scenario without a physical object to be detected. The configuration process does not extract a standard matching template. When the recipe configured in this embodiment is used for actual mass production, it is necessary to complete the automatic chip (Die) or graphic position correction and feature template extraction according to the pre-configured parameters.

[0079] Figure 7 1 is a top view of an object to be detected according to an embodiment of the present invention. Figure 8Yes Yes Figure 7 An imaging diagram of the local area of ​​the object to be detected. Fig. 9 It is a schematic diagram of a matching template graphic and a geometric outline according to an embodiment of the present invention.

[0080] Figure 7 In the example, the object to be inspected 10 is a wafer, and a local area on the wafer is inspected and photographed to form a local image. Fig.11 ; Figure 8 Local imaging in Fig.11 for Figure 7 Local imaging in Fig.11 The enlarged image shows that the local imaging Fig.11 The plurality of periodic patterns are included, and the plurality of periodic patterns are, for example, chips. Fig. 9 , which is a matching template image 20 and a geometric contour 21 located around the matching template image 20. Figure 7-Figure 9 As shown, in the mass production stage, the feature template extraction of the pattern is completed according to the pre-configured specification information of the object to be tested. The specific method includes:

[0081] The image acquisition is performed on the object to be detected 10, and the effective range covered by the image acquisition is at least 1.5 times larger than the size of the periodically distributed pattern on the surface of the object to be detected, so as to match a complete single periodically distributed pattern; Figure 8 As shown, local imaging Fig.11 The image includes multiple periodically distributed patterns. When the imaging field of view cannot cover the image acquisition range, a low-magnification objective lens or multiple image stitching can be used to generate an effective image.

[0082] According to the mapping relationship between the physical size of the object to be detected and the pixels, a binary image of the outer contour of the periodic pattern is generated; specifically, according to the size information of the object to be detected configured in step S10, the imaging system magnification used for image acquisition, and the mapping relationship between the physical size and the pixels under the imaging system magnification, a binary image of the outer contour of the periodic pattern is generated. For example, when the imaging system magnification A is used for image acquisition, the conversion relationship between the single pixel after imaging and the actual physical size of the object to be detected is f (x，y) According to the size information (m*n) of the periodic pattern, we can get the size of f in the image. (m,n) Geometry matching template.

[0083] Fig.10 : is a diagram of the periodic pattern position recognition process of an embodiment of the present invention. Using the generated binary image of the outer contour of the periodic pattern, the origin of the captured image is used as the starting position, and the origin is located at the upper left corner of the image, and sliding matching is performed along the X and Y directions with a certain sliding step. In the sliding matching process, a series of matching results M are formed. 1 、M 2…M x …M n ,like Fig.10 Perform data analysis on the matching results to obtain the distribution frequency H of the highest matching score intervals in the X direction and the Y direction x , H y , the position of the periodic pattern in the matched image is automatically identified through the distribution frequency, and the positioning of the periodic pattern is completed. Among them, the judgment criteria for data analysis of the matching results are, for example: the image similarity is compared and scored by the peripheral contour binary image and the image obtained in the sliding, and the similarity threshold is set in the adaptive parameters. When the similarity exceeds the threshold and the result appears periodically, the periodic position can be determined as the location of the pattern. And the edge position of the detection object is identified by grayscale projection of the image of the periodic pattern.

[0084] After completing the positioning correction, the characteristic entropy calculation is performed on the image of a single periodic pattern, and the area with the largest characteristic entropy is automatically intercepted as the characteristic matching template in the horizontal alignment process. In this embodiment, the periodic pattern can also be a chip (die).

[0085] Please continue to refer to Figure 3 , the vertical attitude adjustment parameters include the control parameters of the vertical direction of the object to be detected during the signal detection process. Specifically, in the vertical attitude adjustment parameter configuration process, the strategy of vertical control of the object to be detected during the signal detection process can be configured, and the strategy of vertical control includes configuring the control parameters of the selected control method. The control parameters include but are not limited to image contrast, clarity, adjustment step, vertical measurement angle, etc.

[0086] The feasible adjustment methods of the vertical attitude parameters include:

[0087] Step S201, using a measuring sensor to measure the vertical distance between the surface of the object to be detected and the imaging lens, and adjusting the vertical position of the object to be detected or the imaging lens according to the theoretical optimal focal plane height; specifically, the vertical position of the object to be detected can be adjusted by adjusting the vertical position of the object support platform.

[0088] Step S202, after completing the vertical position adjustment, the surface image of the object to be detected is collected and the image clarity is calculated.

[0089] Step S203, adjusting the vertical position of the object to be detected or the imaging lens with a set step distance, so that the surface of the object to be detected is away from the imaging lens, re-capturing the image of the object to be detected and re-calculating the image clarity.

[0090] Repeat step S202 and step S203. After the object to be detected is away from the actual optimal focal plane, the image clarity will be significantly reduced. At this time, the adjustment detection in a single direction is completed.

[0091] Step S204, restoring the vertical position of the object to be detected or the imaging lens to the theoretical focal plane height, and gradually adjusting the vertical position of the object to be detected or the imaging lens in the opposite direction and collecting images to calculate the clarity.

[0092] Step S205: automatically identifying and selecting the best focal plane posture according to the measured and statistically analyzed clarity changes.

[0093] In the vertical attitude adjustment parameter configuration process, the vertical attitude adjustment parameter selects the optical parameter adaptive algorithm and the corresponding standard. Specifically, the user needs to configure the parameters of signal acquisition during the detection process of the object to be detected in the recipe configuration process. In the image-based detection scenario, the configuration parameters include: light source mode, objective lens magnification used, digital and analog gain of the camera, white balance coefficient, signal strength of different acquisition channels, etc. In order to ensure that the set recipe can complete the acquisition parameter adaptation in the mass production stage, it is necessary to select the optical parameter adaptive algorithm and the corresponding standard in the recipe setting stage.

[0094] In step S30, based on the specification information and the preconfigured detection signal acquisition initial parameters and detection signal acquisition parameter adaptive algorithm, a detection image of the object to be detected is generated by simulation. Specifically, after completing the preconfiguration of the detector signal acquisition related parameters, that is, after saving the posture control parameters, the device software combines the preconfigured detection signal acquisition initial parameters and detection signal acquisition parameter adaptive algorithm and specification information to perform image simulation of the object to be detected. Fig.11 FIG. 1 is an image simulation flow chart of an embodiment of the present invention. Fig.11 As shown, the data information involved in image simulation includes: the size of the object to be detected, the process information of the object to be detected, the lighting information, the camera gain and white balance coefficient, the camera imaging correction parameters, the effective area of ​​the imaging field of view, the imaging magnification, the signal strength of each acquisition channel, etc. The lighting information includes: lighting type, light source intensity, wavelength, each channel coefficient, etc. The specific process of imaging simulation is as follows:

[0095] Step S31, based on the size of the object to be detected, imaging magnification, effective area of ​​imaging field of view and camera imaging correction parameters, calculate the image pixel size information corresponding to the image of the area of ​​the object to be detected. According to the image pixel size information, further calculate the number of times of local imaging of the object to be detected to cover the complete area to be detected, and these local imaging images are defined as P 1 , P 2 …P n(where n≥1).

[0096] Step S32, based on the imaging magnification, lighting information, camera parameters, imaging correction parameters, process information of the object to be inspected, signal strength of each acquisition channel and other information, optical imaging simulation is performed on the object to be inspected. After the imaging simulation, P 1 To P n Simulated image with current detection parameters.

[0097] Step S33: correcting the internal lines and contours of all generated simulated images based on the camera imaging correction parameters.

[0098] After the above steps, a simulated image of the actual image of the object to be detected in the batch processing process can be obtained.

[0099] After the simulation generates the detection image, in some embodiments, the method further includes:

[0100] Edit the effective area for quantity detection, and distinguish different areas in the effective area according to the characteristics of the object to be detected. Specifically, after the equipment software completes the image simulation, the user can edit the effective area for quantity detection according to the acquired simulation image. The editing of the effective area can divide the detection area according to user needs or process characteristics. The shape of the area division can be rectangular, circular, elliptical, irregular shape, etc. At the same time, the user can configure whether it is valid for different areas. These divided sub-areas are defined as R 1 , R 2 ,…R m (where m≥1).

[0101] After completing the area division or editing, the user preconfigures the detection parameters and parameter adaptive algorithms for different areas. 1 To R m The corresponding algorithm type and detection parameters can be pre-configured for each region. The algorithm type is, for example, image comparison based on statistics or image comparison based on thresholds. In order to ensure that the set formula can complete the detection parameter adaptation in the mass production stage, it is necessary to select the detection parameter adaptation algorithm and parameter quality evaluation method in the formula setting stage. 1 To R m The configuration of the regions does not affect each other and can use the same or different parameters.

[0102] The user can select the range of repeated units involved in the detection in the object to be detected during the recipe setting stage. If the above process is not executed, the device software defaults to the entire object to be detected as the operating range. This step can be performed at any stage after the layout image of the object to be detected is generated in step S10. The present invention can perform image simulation based on the specification information or design drawings of the object to be detected, and can generate data for regional editing without relying on actual image acquisition; it can significantly improve the convenience and editing efficiency of the recipe editing of the quantity detection equipment, and improve the utilization rate of the equipment.

[0103] In step S40, after completing the above process operation, the user can complete the saving of the recipe file through the recipe saving function, including the configuration parameters of each stage and some intermediate data generated in the process. Specifically, it includes at least one of the basic information of the object to be detected, the initial parameters of posture adjustment, the adaptive algorithm of posture adjustment parameters, the initial parameters of detection signal acquisition, the adaptive algorithm of detection signal acquisition parameters and the parameters of the quantity detection algorithm. Furthermore, it also includes the intermediate data generated in the above parameter adaptation process.

[0104] The present embodiment provides a recipe setting method based on a non-physical scene. First, the specification information of the object to be detected can be obtained in the recipe creation stage, and it does not rely on the physical entity to be detected. Secondly, in the alignment and detection parameter configuration interface during the recipe creation process, in addition to hardware parameters such as light source, camera, objective lens, and aperture, it is also necessary to select the algorithm type for automatically adjusting parameters and the multi-dimensional merit interval for optical parameter configuration, wherein the multi-dimensional merit interval includes but is not limited to: target grayscale distribution interval, target contrast interval, target clarity interval, etc. In addition, in the recipe creation process, in the scene without the object to be detected, the configuration of the detection area needs to rely on the generation of detection simulation images based on the theoretical size and detection magnification. The virtual image generated by the interface will be adjusted according to the chip size and magnification switching, and there is a certain mapping relationship. In addition, in the actual mass production process based on the recipe or the automatic adjustment stage based on the physical scene, there is an obvious chip position recognition or optical parameter adjustment process. In addition, the relevant parameters defined in the recipe cannot usually be directly used for mass production, and the parameters used in the mass production stage are different from the recipe definition results. After actual mass production based on the recipe or automatic adjustment based on a physical scene, the adaptive parameters will be saved, that is, the original recipe file will be updated or saved as a new recipe, including but not limited to silicon wafer layout parameters, optical hardware parameters used in each image acquisition scene, vertical focal plane compensation, detection algorithm parameters, etc.

[0105] This embodiment also provides an application method of a recipe based on a non-physical scenario, which uses a recipe setting method based on a non-physical scenario as described in any of the above items to create a quantity detection recipe, uses the quantity detection recipe to perform quantity detection, and adjusts the adaptive algorithm parameters in the quantity detection recipe according to the detection results in the mass production process.

[0106] This embodiment also provides a quantity detection device, which uses any of the above-mentioned recipe setting methods based on non-physical scenarios to create a recipe for quantity detection.

[0107] In summary, it can be seen that in a quantity detection device and a recipe setting method and application method based on a non-physical scene provided by an embodiment of the present invention, by obtaining the specification information of the object to be detected and pre-configuring the initial parameters of the posture adjustment of the object to be detected, the posture adjustment parameter adaptive algorithm, the initial parameters of the detection signal acquisition and the detection signal acquisition parameter adaptive algorithm to simulate and generate the detection image of the object to be detected, the method of formula creation and configuration can be performed in advance without the object to be detected arriving at the current device; in a semiconductor manufacturing plant, the quantity detection device completes the recipe editing of quantity detection in advance, which helps to improve production efficiency. The present invention can perform image simulation according to the specification information or design drawings of the object to be detected, and can generate data for regional editing without relying on actual image acquisition; it can significantly improve the convenience and editing efficiency of the recipe editing of the quantity detection device, and improve the utilization rate of the device. The present invention divides the configuration parameters in the recipe into the initial parameters of the posture adjustment of the object to be detected, the adaptive algorithm of the posture adjustment parameter, the initial parameters of the detection signal acquisition and the adaptive algorithm of the detection signal acquisition parameter; it can improve the reusability of the recipe of the quantity detection device, and can quickly create a new recipe by reusing the old recipe.

[0108] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other. In addition, the different parts between the various embodiments can also be used in combination with each other, and the present invention is not limited to this.

[0109] In addition, it should be recognized that although the present invention has been disclosed as a preferred embodiment, the above embodiment is not intended to limit the present invention. For any technician familiar with the art, without departing from the scope of the technical solution of the present invention, the technical content disclosed above can be used to make many possible changes and modifications to the technical solution of the present invention, or modified into equivalent embodiments of equivalent changes. Therefore, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still belongs to the scope of protection of the technical solution of the present invention.

Claims

1. A recipe setting method based on non-physical scene, It is characterized in that include: Obtaining specification information of the object to be detected, and completing configuration of basic information of the object to be detected according to the specification information; Pre-configuring initial parameters for posture adjustment of the object to be detected, an adaptive algorithm for posture adjustment parameters, initial parameters for detection signal acquisition, an adaptive algorithm for detection signal acquisition parameters, and parameters for quantity detection algorithm; Based on the specification information and the preconfigured detection signal acquisition initial parameters and detection signal acquisition parameter adaptive algorithm, a detection image of the object to be detected is generated by simulation; A recipe is saved, wherein the recipe includes at least one of basic information of the object to be detected, initial parameters for posture adjustment, an adaptive algorithm for posture adjustment parameters, initial parameters for detection signal acquisition, an adaptive algorithm for detection signal acquisition parameters, and parameters of a quantity detection algorithm.

2. The recipe setting method based on non-physical scene according to claim 1, It is characterized in that The posture adjustment parameters of the object to be detected include horizontal posture adjustment parameters and vertical posture adjustment parameters.

3. The recipe setting method based on non-physical scene according to claim 2, It is characterized in that In the horizontal posture adjustment parameter configuration process, the horizontal posture adjustment parameters are configured as initial optical parameters for image acquisition in the alignment phase.

4. The recipe setting method based on non-physical scene according to claim 2, It is characterized in that In the horizontal posture adjustment parameter configuration process, an optical parameter adaptive algorithm is used to adjust the horizontal posture adjustment parameters.

5. The recipe setting method based on non-physical scene according to claim 4, It is characterized in that The adjustment method of the optical parameter adaptive algorithm includes: Capturing images of the object under test; Perform grayscale histogram analysis on the collected image to obtain characteristic information of grayscale distribution of the object to be detected after imaging; Adopting an automatic parameter adjustment algorithm to adjust optical parameters and re-execute the image acquisition step according to the characteristic information of the grayscale distribution and a preset recommended interval, until the overall distribution of the grayscale distribution approaches the recommended interval; After the grayscale distribution reaches the standard of the recommended interval, the smoothness of the distribution curve of the grayscale histogram is determined and the optical parameters are adjusted to meet the measurement requirements.

6. The recipe setting method based on non-physical scene according to claim 2, It is characterized in that The horizontal posture adjustment parameters are obtained by matching the feature image template to obtain the mapping relationship between the nominal positions and actual positions of multiple alignment marks, and the horizontal coordinate system transformation model is solved based on the mapping relationship of multiple groups of alignment marks.

7. The recipe setting method based on non-physical scene according to claim 6, It is characterized in that In the mass production stage, the method of extracting the feature template of the pattern according to the pre-configured horizontal posture adjustment parameters includes: The image acquisition is performed on the object to be detected, and the image acquisition covers a range greater than 1.5 times the size of the pattern periodically distributed on the surface of the object to be detected; Generate a binary image of the outer contour of the periodic pattern according to the mapping relationship between the physical size of the information of the object to be detected and the pixels; Using the generated binary image of the outer contour, step-slide matching is performed along the X direction and the Y direction with the origin of the acquired image as the starting position; In the sliding matching process, a series of matching results are formed, data analysis is performed on the matching results, and the distribution frequency of the highest matching score intervals in the X direction and the Y direction is obtained. The position of the periodic pattern in the matched image is identified through the distribution frequency; The feature entropy of a single periodic pattern image is calculated, and the area with the largest feature entropy is automatically intercepted as the feature matching template in the horizontal alignment process.

8. The recipe setting method based on non-physical scene according to claim 7, It is characterized in that The judgment criteria for data analysis of the matching results are: image similarity comparison and scoring are performed between the peripheral contour binary image and the image obtained during sliding, a similarity threshold is set in the adaptive parameters, and when the similarity exceeds the threshold and the results appear periodically, the periodic position is determined to be the location of the pattern.

9. The recipe setting method based on non-physical scene according to claim 7, It is characterized in that The edge position of the detection object is identified by grayscale projection of the image of the periodic pattern.

10. The recipe setting method based on non-physical scene according to claim 2, It is characterized in that The vertical attitude adjustment parameters include vertical control parameters of the object to be detected during the signal detection process.

11. The recipe setting method based on non-physical scene according to claim 10, It is characterized in that The method for adjusting the vertical posture includes: Using a measuring sensor to measure the vertical distance between the surface of the object to be detected and the imaging lens, and adjusting the vertical position of the object to be detected or the imaging lens according to a theoretical optimal focal plane height; Collecting the surface image of the object to be detected and calculating the image clarity; Adjusting the vertical position of the object to be detected or the imaging lens at a set step distance, so that the surface of the object to be detected is away from the imaging lens, and then re-capturing the image and re-calculating the image clarity; After the object to be detected is away from the actual optimal focal plane, completing adjustment detection in a single direction; Restoring the vertical position of the object to be detected or the imaging lens to the theoretical focal plane height, and gradually adjusting the vertical position of the object to be detected or the imaging lens in the opposite direction and collecting images to calculate the clarity; Automatically identify and select the best focal plane posture based on the measured statistical clarity changes.

12. The recipe setting method based on non-physical scene according to claim 2, It is characterized in that In the vertical attitude adjustment parameter configuration process, an optical parameter adaptive algorithm is used to adjust the vertical attitude adjustment parameters.

13. The recipe setting method based on non-physical scene according to claim 1, It is characterized in that After the simulation generates the detection image, it also includes: The editing amount detects an effective area, wherein different areas are distinguished according to the characteristics of the object to be detected; Pre-configure detection parameters and parameter adaptation algorithms for different areas.

14. The recipe setting method based on non-physical scene according to claim 1, It is characterized in that Based on the overall specification information of the object to be detected, configure the scope that needs to be involved in the actual detection process.

15. The recipe setting method based on non-physical scene according to claim 1, It is characterized in that The effective specification information of the object to be detected includes at least one of the size of the object to be detected, whether there is an effective detected pattern on the object to be detected, and the distribution regularity of the detected pattern.

16. The recipe setting method based on non-physical scene according to claim 1, It is characterized in that The valid specification information of the object to be inspected includes an exposure process design drawing of the object to be inspected.

17. An application method based on a recipe without physical scene, It is characterized in that A quantity detection recipe is created by using the recipe setting method based on a non-physical scenario as described in any one of claims 1 to 16, quantity detection is performed using the quantity detection recipe, and the adaptive algorithm parameters in the quantity detection recipe are adjusted according to the detection results in the mass production process.

18. A quantity detection device, It is characterized in that A recipe for quantity detection is created by using the recipe setting method based on a non-physical scenario as described in any one of claims 1 to 16.

Images