Focal plane calibration method and device, computer equipment and storage medium

Abstract

The invention provides a focal plane calibration method and device, computer equipment and a storage medium, and belongs to the technical field of digital exposure, and the focal plane calibration method comprises the steps: carrying out the image collection of a preset sample plate, and obtaining a target image; the target image comprises a plurality of first object groups arranged side by side along a first direction and a plurality of second object groups arranged side by side along a second direction; the preset sample plate comprises a plurality of test areas, and the target object in one test area is obtained by one-time exposure of exposure equipment; determining pixel position information corresponding to each target object according to the target image; based on the pixel position information of each target object, judging whether the exposure state of the target object is normal or defocused, and determining predicted focal plane heights respectively corresponding to different test areas according to a judgment result and pre-stored exposure heights corresponding to each first object group; based on each predicted focal plane height, calibration information of an optimal focal plane of the exposure apparatus is determined.

Description

Technical Field

[0001] The present disclosure belongs to the field of exposure technology, and in particular relates to a focal plane calibration method, device, computer equipment and storage medium. Background Art

[0002] Exposure is an important process in the photolithography process. The exposure machine is set with an optimal focal plane, also known as the focus plane. However, usually, as the exposure machine ages and is affected by external factors such as the working environment, the optimal focal plane of the exposure machine will shift, causing the size of the product prepared after the exposure process to change, affecting the product quality, and in severe cases, affecting the formation of the entire product.

[0003] At present, the focal plane is often calibrated manually. However, there are differences in the calibration results of different engineers, so the accuracy of manual calibration is low. In addition, the efficiency of manual calibration is low, which seriously affects the production line process and increases the product manufacturing cost. Summary of the invention

[0004] The present disclosure aims to solve at least one of the technical problems existing in the prior art and provides a focal plane calibration method, device, computer equipment and storage medium.

[0005] In a first aspect, the technical solution adopted to solve the technical problem of the present disclosure is a focal plane calibration method, comprising:

[0006] Capturing an image of a preset template to obtain a target image; the target image includes a plurality of first object groups arranged side by side along a first direction, and a plurality of second object groups arranged side by side along a second direction; the first object group includes a plurality of target objects arranged side by side along the second direction; the second object group includes a plurality of target objects arranged side by side along the first direction; the preset template includes a plurality of test areas, and a target object in one of the test areas is obtained by one exposure of an exposure device, and target objects in different test areas are obtained by different exposures of the exposure device;

[0007] Determining pixel position information corresponding to each target object according to the target image;

[0008] Based on the pixel position information of each of the target objects, determining whether the exposure state of the target object is normal or out of focus, and determining the predicted focal plane heights corresponding to the different test areas according to the determination result and the pre-stored exposure heights corresponding to each of the first object groups;

[0009] Based on each of the predicted focal plane heights, calibration information of an optimal focal plane of the exposure device is determined.

[0010] In some embodiments, determining pixel position information corresponding to each target object according to the target image includes:

[0011] Extracting contour features in the target image to determine the center position of the target object;

[0012] Based on the center position of the target object and a preset size range of the target object, pixel position information of the target object is determined.

[0013] In some embodiments, determining pixel position information corresponding to each target object according to the target image includes:

[0014] Extracting contour features from the target image, performing binarization processing on the target image to obtain a first binary image, and determining the center position of the target object according to the extracted contour features;

[0015] Based on the center position of the target object and a preset size range of the target object, pixel position information of the target object is determined.

[0016] In some embodiments, determining pixel position information corresponding to each target object according to the target image includes:

[0017] Extracting contour features from the target image, and performing binarization processing on the target image to obtain a first binary image;

[0018] Performing dilation processing on the first binary image to obtain a first intermediate image;

[0019] performing corrosion processing on the first intermediate image to obtain a second intermediate image;

[0020] Extracting contour features in the second intermediate image to determine the center position of the target object;

[0021] Based on the center position of the target object and a preset size range of the target object, pixel position information of the target object is determined.

[0022] In some embodiments, binarizing the target image to obtain a first binary image includes:

[0023] Extracting contour features from the target image, and performing binarization processing on the target image to obtain an initial binary image;

[0024] Based on the pixel value and position information of each pixel point in the initial binary image, determining the area to be cropped; the pixel points in the area to be cropped are an area in the second direction that contains multiple segments of pixel values ​​that are continuously 0 and is surrounded by multiple segments of pixel values ​​that are continuously 255;

[0025] The portion of the initial binary image located in the to-be-cropped area is cut off to obtain the first binary image.

[0026] In some embodiments, the size range of the target object includes a width range and a height range;

[0027] The determining the pixel position information of the target object based on the center position of the target object and a preset size range of the target object includes:

[0028] Taking the center position of the target object as the center, based on a random width within the width range and a random height within the height range, the target object is determined, and pixel position information of the target object is determined.

[0029] In some embodiments, judging whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object includes:

[0030] Determining the area within the outline of each target object based on the pixel position information of each target object;

[0031] Based on the comparison result between the areas within the contours of the target objects, a reference object is selected from the target objects, and pixel position information of the reference object is determined;

[0032] Based on the pixel position information of each of the target objects and the pixel position information of the reference object, it is determined whether the exposure state of the target object is normal or out of focus.

[0033] In some embodiments, judging whether the exposure state of the target object is normal or out of focus based on the pixel position information of each of the target objects and the pixel position information of the reference object includes:

[0034] Based on the pixel position information of each of the target objects in the target image and the pixel position information of the reference object, using a preset similarity algorithm, determining the similarity between each of the target objects and the reference object;

[0035] Based on the comparison result between the similarity between each of the target objects and the reference object and a first preset threshold, it is determined whether the exposure state of the target object is normal or out of focus.

[0036] In some embodiments, judging whether the exposure state of the target object is normal or out of focus based on the pixel position information of each of the target objects and the pixel position information of the reference object includes:

[0037] Based on the pixel position information of each of the target objects in the first binary image and the pixel position information of the reference object, using a preset similarity algorithm, determining the similarity between each of the target objects and the reference object;

[0038] Based on the comparison result between the similarity between each of the target objects and the reference object and a first preset threshold, it is determined whether the exposure state of the target object is normal or out of focus.

[0039] In some embodiments, judging whether the exposure state of the target object is normal or out of focus based on the pixel position information of each of the target objects and the pixel position information of the reference object includes:

[0040] Based on the pixel position information of each of the target objects in the second intermediate image and the pixel position information of the reference object, using a preset similarity algorithm, determining the similarity between each of the target objects and the reference object;

[0041] Based on the comparison result between the similarity between each of the target objects and the reference object and a first preset threshold, it is determined whether the exposure state of the target object is normal or out of focus.

[0042] In some embodiments, judging whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object includes:

[0043] According to the pixel position information of the target object, intercepting the target sub-image corresponding to the target object from the target image, and determining the area within the outline of the target object in each of the target sub-images;

[0044] Based on the comparison results between the areas within the contours of the target objects, a reference object is selected from the target objects, and a reference subgraph of the reference object is determined;

[0045] Binarization is performed on each of the target sub-images and the reference sub-images to obtain each target binary sub-image and reference binary sub-image;

[0046] Based on each of the target binary subgraphs and the reference binary subgraphs, using a preset similarity algorithm, determining the similarity between each of the target objects and the reference object;

[0047] Based on the comparison result between the similarity between each of the target objects and the reference object and a first preset threshold, it is determined whether the exposure state of the target object is normal or out of focus.

[0048] In some embodiments, judging whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object includes:

[0049] According to the pixel position information of the target object, intercepting the target sub-image corresponding to the target object from the target image, and determining the area within the outline of the target object in each of the target sub-images;

[0050] Based on the comparison results between the areas within the contours of the target objects, a reference object is selected from the target objects, and a reference subgraph of the reference object is determined;

[0051] The target sub-image and the reference sub-image are reduced according to a preset scaling factor to obtain the reduced target sub-image and the reduced reference sub-image;

[0052] Binarization is performed on each of the reduced target sub-images and the reduced reference sub-images to obtain each target binary sub-image and reference binary sub-image;

[0053] Based on each of the target binary subgraphs and the reference binary subgraph, determining a Hamming distance of each of the target binary subgraphs;

[0054] Based on the comparison result of the Hamming distance of each of the target binary sub-images and the second preset threshold, it is determined whether the exposure state of the target object is normal or out of focus.

[0055] In some embodiments, judging whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object includes:

[0056] According to the pixel position information of the target object, intercepting a target sub-image corresponding to the target object from the target image;

[0057] Binarization is performed on each of the target subgraphs to obtain each target binary subgraph;

[0058] Determining the Hamming distance of each of the target binary sub-images based on each of the target binary sub-images and a preset pure color image; the size of the pure color image is the same as the preset size;

[0059] Based on the comparison result of the Hamming distance of each of the target binary sub-images and the third preset threshold, it is determined whether the exposure state of the target object is normal or out of focus.

[0060] In some embodiments, determining the predicted focal plane heights corresponding to different test areas respectively according to the judgment result and the pre-stored exposure heights of each column of the target object includes:

[0061] For the target objects in each of the second object groups arranged side by side along the second direction, based on the exposure status of each of the target objects and the pixel position information of each of the target objects, determining the pixel position information of the defocus boundary corresponding to each of the second object groups;

[0062] Determining regional defocus boundaries corresponding to each of the test areas based on pixel position information of the defocus boundaries corresponding to each of the second object groups and position information of the test area;

[0063] For the regional defocus boundaries respectively corresponding to the test areas, based on the pixel position information of the regional defocus boundaries and the pre-stored exposure heights corresponding to the first object groups, determining the exposure heights of the regional defocus boundaries;

[0064] For the regional defocus boundaries respectively corresponding to the test areas, the predicted focal plane height of the test area is determined based on the exposure height of the regional defocus boundaries.

[0065] In some embodiments, the regional out-of-focus boundary includes a first boundary and a second boundary;

[0066] The step of determining the predicted focal plane height of the test area based on the exposure height of the defocus boundary of the area includes:

[0067] The average of the exposure height of the first boundary and the exposure height of the second boundary is used as the predicted focal plane height of the test area.

[0068] In some embodiments, the calibration information of the best focal plane of the exposure device includes the height and tilt of the best focal plane; the test area includes two;

[0069] The step of determining calibration information of an optimal focal plane of the exposure device based on each of the predicted focal plane heights comprises:

[0070] Taking the average of the predicted focal plane heights as the height of the optimal focal plane;

[0071] The difference between the two predicted focal plane heights is used as the tilt of the optimal focal plane.

[0072] In a second aspect, the embodiments of the present disclosure further provide a focal plane calibration device, including an image acquisition module, an image processing module, a prediction module, and a calibration information determination module:

[0073] The image acquisition module is configured to acquire an image of a preset template to obtain a target image; the target image includes a plurality of first object groups arranged side by side along a first direction, and a plurality of second object groups arranged side by side along a second direction; the first object group includes a plurality of target objects arranged side by side along the second direction; the second object group includes a plurality of target objects arranged side by side along the first direction; the preset template includes a plurality of test areas, and a target object in one of the test areas is obtained by one exposure, and target objects in different test areas are obtained by different exposures;

[0074] The image processing module is configured to determine pixel position information corresponding to each target object according to the target image;

[0075] The prediction module is configured to judge whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object, and determine the predicted focal plane heights corresponding to different test areas according to the judgment result and the pre-stored exposure heights corresponding to each of the first object groups;

[0076] The calibration information determination module is configured to determine the calibration information of the optimal focal plane of the exposure device based on each of the predicted focal plane heights.

[0077] In a third aspect, an embodiment of the present disclosure further provides a computer device, comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor communicates with the memory through the bus, and when the machine-readable instructions are executed by the processor, the steps of the focal plane calibration method described in any one of the first aspects are performed.

[0078] In a fourth aspect, an embodiment of the present disclosure further provides a computer non-volatile readable storage medium, wherein a computer program is stored on the computer non-volatile readable storage medium, and when the computer program is executed by a processor, the steps of the focal plane calibration method as described in any one of the first aspects are executed. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 A flow chart of a focal plane calibration method provided by an embodiment of the present disclosure;

[0080] Figure 2 A schematic diagram of an exemplary preset template after exposure provided in an embodiment of the present disclosure;

[0081] Figure 3 A specific flow chart of an exemplary focal plane calibration method provided in an embodiment of the present disclosure;

[0082] Figure 4A schematic diagram of Laplacian boundary recognition provided in an embodiment of the present disclosure;

[0083] Figure 5a A schematic diagram of boundary recognition in the X direction based on Sobel provided in an embodiment of the present disclosure;

[0084] Figure 5b A schematic diagram of boundary recognition in the Y direction based on Sobel provided in an embodiment of the present disclosure;

[0085] Figure 6 A schematic diagram based on Candy boundary recognition provided in an embodiment of the present disclosure;

[0086] Figure 7 A schematic diagram of a first binary image provided by an embodiment of the present disclosure;

[0087] Figure 8 for Figure 7 Schematic diagram of the first binary image after removing redundant images;

[0088] Fig. 9 A schematic diagram of a first intermediate image after dilation processing is performed on a first binary image provided by an embodiment of the present disclosure;

[0089] Fig.10 A schematic diagram of a second intermediate image after the first intermediate image is corroded according to an embodiment of the present disclosure;

[0090] Fig.11 for Figure 2 A schematic diagram of a sub-image corresponding to the target object shown;

[0091] Fig.12 for Figure 8 A schematic diagram of a sub-image corresponding to the target object shown;

[0092] Fig.13 for Fig.10 A schematic diagram of a sub-image corresponding to the target object shown;

[0093] Fig.14 A schematic diagram of converting a target sub-image provided by an embodiment of the present disclosure into a reference sub-image;

[0094] Fig.15 A schematic diagram of a defocus boundary corresponding to a second object group provided in an embodiment of the present disclosure;

[0095] Fig.16 A schematic diagram of a defocused boundary of a region on a target image provided by an embodiment of the present disclosure;

[0096] Fig.17A schematic diagram of a focal plane calibration device provided in an embodiment of the present disclosure;

[0097] Fig.18 A schematic diagram of the structure of a computer device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0098] In order to make the purpose, technical scheme and advantages of the embodiments of the present disclosure clearer, the technical scheme in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The components of the embodiments of the present disclosure generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure provided in the drawings is not intended to limit the scope of the present disclosure for protection, but merely represents the selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present disclosure.

[0099] Unless otherwise defined, the technical terms or scientific terms used in the present disclosure should be understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, similar words such as "one", "one" or "the" do not indicate quantity restrictions, but indicate that there is at least one. Similar words such as "include" or "comprise" mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Similar words such as "connect" or "connected" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0100] The "multiple or several" mentioned in this disclosure refers to two or more. "And / or" describes the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0101] Before describing the embodiments of the present disclosure, some special terms provided in the embodiments of the present disclosure are further explained.

[0102] 1. OCD (Optical Critical Dimension) is a commonly used measuring device in electronic manufacturing technology. Its principle is based on the phenomenon of optical reflection. In this device, after a beam of light passes through the surface of the object being measured, it will be affected by the surface morphology and reflected. Finally, the morphological information of the object being measured is measured by detecting the change of the light path. Therefore, the OCD device can accurately measure the geometric morphology of tiny structures in semiconductor chips, such as line width, line spacing, radius, etc., to meet the requirements of advanced manufacturing technology for manufacturing accuracy.

[0103] 2. The Focus Energy Matrix (FEM) is a set of data obtained by changing the focus value in one direction with a fixed step size and changing the exposure energy in another direction with a fixed step size during exposure. However, in the disclosed embodiment, the FEM matrix is ​​visualized as target objects arranged in an array in a preset template, or is understood as a target object in a target image.

[0104] 3. The Laplacian operator is a second-order differential operator in n-dimensional Euclidean space, defined as the divergence div of the gradient grad.

[0105] 4. Sobel operator, or Sobel operator, is an important processing method in the field of computer vision. It is mainly used to obtain the first-order gradient of digital images. Its common application and physical meaning is edge detection.

[0106] 5. The Canny edge detection operator is a multi-level edge detection algorithm.

[0107] 6. The Hamming distance is used in data transmission error control coding. The Hamming distance is a concept that represents the number of different characters in the corresponding positions of two (same length) strings. We use d(x,y) to represent the Hamming distance between two characters x and y. Perform an XOR operation on two strings and count the number of results that are 1, then this number is the Hamming distance.

[0108] In related technologies, the focal plane of an exposure machine is often calibrated manually. However, the calculation results of different personnel for the same exposure sample are often inconsistent, and usually the differences are large, so the accuracy of manual calibration is low. In addition, the subsequent processing and judgment of different calculation results and finding a more accurate optimal focal plane require a long manual calibration cycle and low efficiency, which seriously affects the production line process and increases the product preparation cost.

[0109] In view of this, the embodiments of the present disclosure provide a focal plane calibration method, apparatus, computer equipment and storage medium, which can automatically and accurately calculate the calibration information of the optimal focal plane of the exposure device, thereby improving the calibration efficiency.

[0110] The focal plane calibration method provided by the embodiment of the present disclosure performs image acquisition on a preset template to obtain a target image; the target image includes multiple groups of first object groups arranged side by side along a first direction, and multiple groups of second object groups arranged side by side along a second direction; the first object group includes multiple target objects arranged side by side along the second direction; the second object group includes multiple target objects arranged side by side along the first direction; the preset template includes multiple test areas, and the target object in one test area is obtained by one exposure of the exposure device, and the target objects in different test areas are obtained by different exposures of the exposure device; according to the target image, the pixel position information corresponding to each target object is determined; based on the pixel position information of each target object, the exposure state of the target object is judged to be normal or out of focus, and the predicted focal plane heights corresponding to different test areas are determined according to the judgment result and the exposure heights corresponding to each first object group stored in advance; based on each predicted focal plane height, the calibration information of the optimal focal plane of the exposure device is determined.

[0111] The disclosed embodiment first collects images of a preset template with multiple target objects, applies image processing technology to the identification and calibration of the focal plane, realizes automated detection, and improves calibration efficiency. At the same time, based on the pixel position information of each target object, the exposure state of each target object is refined to prepare for the subsequent improvement of calibration accuracy. Through the judgment results of the exposure state of each target object and the exposure heights corresponding to each target object pre-stored in groups, the predicted focal plane heights corresponding to different test areas can be determined. Since different test areas correspond to different exposures of the same exposure device, the calibration information of the optimal focal plane can be accurately obtained according to different prediction results.

[0112] The specific process of a focal plane calibration method provided by an embodiment of the present disclosure is described in detail below.

[0113] Figure 1 A flow chart of a focal plane calibration method provided by an embodiment of the present disclosure is as follows: Figure 1 As shown, it includes steps S11 to S14, wherein:

[0114] S11, collecting images of a preset template to obtain a target image.

[0115] The target image includes a plurality of first object groups 31 arranged side by side along a first direction X, and a plurality of second object groups 32 arranged side by side along a second direction Y; the first object group 31 includes a plurality of target objects arranged side by side along the second direction Y; the second object group 32 includes a plurality of target objects arranged side by side along the first direction X.

[0116] The preset template includes multiple test areas, and the target object in one test area is obtained by one exposure of the exposure device, and the target objects in different test areas are obtained by different exposures of the exposure device. The target image is an image obtained by image acquisition of the preset template, and the content in the target image is also the information of the preset template. The preset template is an exposed template, and different test areas are exposed different times to form multiple target objects. The specific arrangement of the multiple target objects in the preset template is as follows: multiple groups of first object groups 31 are arranged side by side along a first direction X, and each group of first object group 31 includes multiple target objects arranged side by side along a second direction Y; and multiple groups of second object groups 32 are arranged side by side along the second direction Y, and each group of second object group 32 includes multiple target objects arranged side by side along the first direction X (such as Figure 2 A) shown.

[0117] During exposure, different first object groups 31 correspond to different exposure heights. Figure 2 A schematic diagram of an exemplary preset template after exposure provided by the embodiment of the present disclosure, such as Figure 2 As shown, the preset template is divided into two test areas, which are respectively recorded as the first test area 201 and the second test area 202. Using the same exposure device to be calibrated, the two test areas are respectively exposed to form a plurality of target objects arranged in an array. The first direction X is, for example, the row direction, and the second direction Y is, for example, the column direction. Different first object groups 31 correspond to different exposure heights. Taking a test area as an example, a test area includes 25 groups of first object groups 31. When exposing, the exposure height (which can also be understood as the focal plane height or the focusing height) of the exposure device is adjusted to scan in sequence. For example, along the positive direction of the first direction X, the exposure height corresponding to each group of first object groups 31 decreases in sequence, and the exposure height difference between two adjacent columns of first object groups 31 is the same, for example, p. Exemplarily, the 2n groups of first object groups 31 arranged side by side along the first direction X are respectively recorded as the first group of first object groups 31, the second group of first object groups 31, ..., and the 2nth group of first object groups 31. Taking the best focal plane 0μm currently stored in the exposure device as a reference, set the exposure height of the nth group as the best focal plane currently stored in the exposure device. Specifically, the exposure height corresponding to the first object group 31 of the first group is xμm, the exposure height corresponding to the first object group 31 of the second group is xp, the exposure height corresponding to the first object group 31 of the third group is x-2p, ..., the exposure height corresponding to the first object group 31 of the nth group is 0μm, the exposure height corresponding to the first object group 31 of the n+1th group is -p, ..., and the exposure height corresponding to the first object group 31 of the 2nth group is -x. The exposure heights corresponding to the first object groups 31 are stored for subsequent algorithm calls.

[0118] It should be noted that the exposure process to obtain the preset template is a test process for the exposure device, so for the exposure process of a test area, the exposure height of the exposure device can be adjusted for scanning of different columns to collect multiple groups of first object groups 31 corresponding to different exposure heights. However, in the actual application stage, the exposure height will not be adjusted during the exposure process of the exposure device, that is, in the actual application stage, the exposure device performs exposure at an optimal exposure height (optimal focal plane).

[0119] The image acquisition process is performed on the preset template. Specifically, the OCD device can be used to locate and photograph the FEM preset template to obtain a target image, thereby obtaining a FEM image.

[0120] S12. Determine pixel position information corresponding to each target object according to the target image.

[0121] During specific implementation, a preset algorithm can be used to identify object features of the target image to determine the target object, and further determine the pixel position information corresponding to the target object. The pixel position information can characterize the position of the target object in the target object. Specifically, the pixel position information may include the position of at least part of the pixel points where the target object is located.

[0122] S13, based on the pixel position information of each target object, determine whether the exposure state of the target object is normal or out of focus, and determine the predicted focal plane heights corresponding to different test areas according to the determination result and the pre-stored exposure heights corresponding to each first object group 31.

[0123] Based on the pixel position information of each target object, the shape of each target object can be determined, and the exposure state can be judged based on the shape. For example, the exposure state of a target object with a regular shape can be determined to be normal, and the exposure state of a target object with an irregular shape can be determined to be defocused. Defocus means that the focal plane of the exposure device deviates from the optimal focal plane.

[0124] Afterwards, based on the judgment result of the exposure status of each target object, for each group of second object groups 32, it is possible to determine the target object among the multiple target objects in the normal state that is closest to the defocused state, or to determine the target object among the multiple target objects in the defocused state that is closest to the normal state, thereby determining the defocus boundary corresponding to each group of second object groups 32.

[0125] Afterwards, the predicted focal plane heights corresponding to different test areas are determined based on the defocus boundaries corresponding to each group of second object groups 32. Taking a test area as an example, for each group of second object groups 32 in the test area, the defocus boundaries corresponding to each group of second object groups 32 in the test area are integrated to determine the defocus boundaries corresponding to the test area, and then the predicted focal plane height corresponding to the test area is determined based on the exposure heights corresponding to each pre-stored first object group 31.

[0126] S14. Determine calibration information of an optimal focal plane of an exposure device based on each predicted focal plane height.

[0127] The calibration information of the best focal plane includes, for example, the height and tilt of the best focal plane.

[0128] Specifically, the height of the best focal plane of the exposure device can be determined by using a first preset algorithm in combination with each predicted focal plane height; the tilt of the best focal plane of the exposure device can be determined by using a second preset algorithm in combination with each predicted focal plane height. Finally, based on the height of the best focal plane and the tilt of the best focal plane, the parameters of the exposure device are adjusted so that the exposure device is adjusted from the current focal plane to the best focal plane, thereby realizing automatic calibration of the focal plane of the exposure device.

[0129] For example, Figure 3 A specific flow chart of an exemplary focal plane calibration method provided in an embodiment of the present disclosure, such as Figure 3 As shown, it includes steps S21 to S25, wherein: S21, using the OCD device to collect FEM images; S22, identifying the target object; S23, determining the defocus boundary corresponding to each group of second objects 32; S24, determining the defocus boundary corresponding to each test area; S25, determining the calibration information of the best focal plane of the exposure device.

[0130] In some embodiments, with respect to step S12, determining the pixel position information corresponding to each target object specifically includes the following steps S12-1-1 to S12-1-2, wherein:

[0131] S12-1-1. Extract contour features in the target image and determine the center position of the target object.

[0132] A preset algorithm may be used to identify the target image and extract contour features in the target image, wherein the preset algorithm may include, but is not limited to, edge detection algorithms, such as Laplacian, Sobel (two-dimensional direction search), Canny, and the like.

[0133] Afterwards, the extracted contour features may be determined to be the contour of the target object, and the center position of the contour, such as the geometric center position, may be determined and recorded as the center position of the target object.

[0134] Figure 4 A schematic diagram of Laplacian boundary recognition provided by an embodiment of the present disclosure is provided. Figure 5a A schematic diagram of boundary recognition in the X direction based on Sobel provided in an embodiment of the present disclosure, Figure 5b A schematic diagram of boundary recognition in the Y direction based on Sobel provided in an embodiment of the present disclosure, Figure 6 A schematic diagram based on Candy boundary recognition is provided for an embodiment of the present disclosure.

[0135] like Figure 4 , Figure 5a , Figure 5b and Figure 6 As shown, for contour features within a certain range, it can be determined that it is a contour of a target object A, and based on the position of the pixel point where the contour of the target object A is located, the center position of the contour of the target object A is determined and recorded as the center position of the target object A.

[0136] Exemplarily, a preset algorithm may be used to extract closed contour features in the target image, thereby avoiding the influence of noise such as non-closed contour features on the accuracy of the recognition result.

[0137] S12-1-2. Determine pixel position information of the target object based on the center position of the target object and a preset size range of the target object.

[0138] The preset size range of the target object may be a preset width range and / or width range of the target object, and the size range is sufficient to cover the entire target object and does not include adjacent target objects.

[0139] Exemplarily, with the center position of the target object as the center, based on a random width within the width range and a random height within the height range, the target object is determined, and the pixel position information of the target object is determined. For example, the width range is between m1 pixels and m2 pixels; the height range is between m3 pixels and m4 pixels. Among them, m1, m2, m3 and m4 are positive integers, and m2 is greater than m1, and m4 is greater than m3. The width of the target object is randomly selected as n1 in the preset width range, m1≤n1≤m2; the height of the target object is randomly selected as n2 in the preset height range, m3≤n2≤m4, to determine a n1×n2 target object, and determine the pixel position information of each pixel in the n1×n2 area, that is, the pixel position information of the target object.

[0140] This embodiment uses an image recognition algorithm to accurately determine the pixel position information of each target object, thereby improving the accuracy of the subsequent judgment results of the exposure status of the target object, further improving the accuracy of determining the defocus boundary, and finally determining more accurate calibration information.

[0141] In some embodiments, with respect to step S12, determining the pixel position information corresponding to each target object specifically includes the following steps S12-2-1 to S12-2-2, wherein:

[0142] S12-2-1. Extract contour features from the target image, perform binarization on the target image to obtain a first binary image, and determine the center position of the target object based on the extracted contour features.

[0143] In one possible implementation, Figure 7 A schematic diagram of a first binary image provided by an embodiment of the present disclosure, such as Figure 7 As shown, a preset algorithm can be used to identify edge information in the target image and extract contour features therefrom. The preset algorithm can be, for example, a canny algorithm, and in the process of extracting contours, the target image is binarized, the pixel values ​​of the pixels corresponding to the contour features are set to 255, and the pixel values ​​of other pixels except the contour features are set to 0. The pixel values ​​of 255 are white pixels, and the pixel values ​​of 0 are black pixels, thereby obtaining a first binary image.

[0144] In another possible implementation, Figure 8 for Figure 7 A schematic diagram of the first binary image after removing redundant images, as shown in Figure 8 As shown in FIG. 1 , for the target image, the redundant image can be firstly eliminated, and then the target image after the redundant image is eliminated is converted to obtain the first binary image. The redundant image here can be understood as the relatively obvious interference feature image (such as Figure 7 The redundant images can be eliminated by, for example, using an image recognition algorithm to identify redundant targets for elimination. Specifically, the corresponding area can be directly cropped, or the pixel values ​​of the pixels corresponding to the redundant targets can be changed. Specifically, the data after the pixel value is changed can be set according to the actual situation, which is not limited in the embodiments of the present disclosure.

[0145] Afterwards, the extracted contour features may be determined to be the contour of the target object, and the center position of the contour, such as the geometric center position, may be determined and recorded as the center position of the target object.

[0146] like Figure 7 or Figure 8As shown, for contour features within a certain range, it can be determined that it is a contour of a target object A, and based on the position of the pixel point where the contour of the target object A is located, the center position of the contour of the target object A is determined and recorded as the center position of the target object A.

[0147] Exemplarily, a preset algorithm may be used to extract closed contour features in the first binary image, thereby avoiding the influence of noise such as non-closed contour features on the accuracy of the recognition result.

[0148] S12-2-2. Determine pixel position information of the target object based on the center position of the target object and a preset size range of the target object.

[0149] The preset size range of the target object may be a preset width range and / or width range of the target object.

[0150] Exemplarily, with the center position of the target object as the center, based on a random width within the width range and a random height within the height range, the target object is determined, and the pixel position information of the target object is determined. For example, the width range is between m1 pixels and m2 pixels; the height range is between m3 pixels and m4 pixels. Among them, m1, m2, m3 and m4 are positive integers, and m2 is greater than m1, and m4 is greater than m3. The width of the target object is randomly selected as n1 in the preset width range, m1≤n1≤m2; the height of the target object is randomly selected as n2 in the preset height range, m3≤n2≤m4, to determine a n1×n2 target object, and determine the pixel position information of each pixel in the n1×n2 area, that is, the pixel position information of the target object.

[0151] This embodiment converts the target image into a first binary image before feature extraction, which can improve contour recognition, thereby making it easier to extract the contour features of the target object and improve contour recognition efficiency. In addition, the image recognition algorithm can accurately determine the pixel position information of each target object, thereby improving the accuracy of the subsequent judgment result of the exposure state of the target object, thereby improving the accuracy of determining the defocus boundary, and finally determining more accurate calibration information.

[0152] In some embodiments, with respect to step S12, determining the pixel position information corresponding to each target object specifically includes the following steps S12-3-1 to S12-3-5, wherein:

[0153] S12-3-1. Extract contour features from the target image, perform binarization on the target image, and obtain a first binary image.

[0154] In a possible implementation, Figure 7As shown, a preset algorithm can be used to identify edge information in the target image and extract contour features from it. Among them, the preset algorithm can be, for example, a canny algorithm, and in the process of extracting contours, the target image is binarized, the pixel values ​​of the pixels corresponding to the contour features are set to 255, and the pixel values ​​of other pixels except the contour features are set to 0, so as to obtain a first binary image. The pixel values ​​of 255 are pixels showing white, and the pixel values ​​of 0 are pixels showing black. In another possible implementation, for the target image, redundant images can be first eliminated, and then the target image after the redundant images are eliminated is converted to obtain a first binary image. The redundant images here can be understood as the interference feature images that are more obvious on the preset template except the target object when the preset template is collected. The method of eliminating redundant images can be, for example, using an image recognition algorithm to identify redundant targets for elimination. Specifically, the pixel values ​​of the pixels corresponding to the redundant targets can be changed, and the data after the pixel values ​​are changed can be set according to the actual situation, which is not limited in the embodiments of the present disclosure.

[0155] S12-3-2. Dilate the first binary image to obtain a first intermediate image.

[0156] Fig. 9 A schematic diagram of a first intermediate image after dilation processing is performed on a first binary image provided in an embodiment of the present disclosure, such as Fig. 9 As shown, the first binary image can be expanded by an image expansion algorithm to obtain a first intermediate image. By expanding the first binary image, the contour features of the target object can be increased, thereby enlarging the contour of the target object as a whole, so that multiple tiny closed contours formed by defocused exposure are merged into one closed contour.

[0157] S12-3-3. Perform corrosion processing on the first intermediate image to obtain a second intermediate image.

[0158] Fig.10 A schematic diagram of a second intermediate image after the first intermediate image is corroded according to an embodiment of the present disclosure, as shown in FIG. Fig.10 As shown, the first intermediate image can be eroded by an image erosion algorithm to obtain a second intermediate image. By eroding the first intermediate image, the expanded contour is restored to its original state as much as possible, so as to more accurately determine the center position of the target object.

[0159] In the above steps S12-3-1 to S12-3-3, the edges of connected white pixels in the first binary image are used as contour feedback, and the contour can be maximized using dilation and erosion methods.

[0160] S12-3-4. Extract contour features in the second intermediate image and determine the center position of the target object.

[0161] A preset algorithm may be used to identify the second intermediate image, extract contour features in the second intermediate image, and set a threshold for the area contained in the maximized contour, filter out contours smaller than the set area threshold, and retain contour information closest to the actual contour. The preset algorithm may include, but is not limited to, edge detection algorithms, such as Laplacian, Sobel (two-dimensional direction search), Canny, and other algorithms.

[0162] Afterwards, the extracted contour features may be determined to be the contour of the target object, and the center position of the contour, such as the geometric center position, may be determined and recorded as the center position of the target object.

[0163] like Fig.10 As shown, for contour features within a certain range, it can be determined that it is a contour of a target object A, and based on the position of the pixel points where the contour of the target object A is located, the boundary of the two pixel points farthest from each other in the first direction X along the second direction and the center of the rectangular frame formed by the two pixel points farthest from each other in the second direction Y along the first direction X as the boundary are selected as the center position of the contour of the target object A.

[0164] Exemplarily, a preset algorithm may be used to extract closed contour features in the second intermediate image, thereby avoiding the influence of noise such as non-closed contour features on the accuracy of the recognition result.

[0165] S12-3-5. Determine pixel position information of the target object based on the center position of the target object and a preset size range of the target object.

[0166] The preset size range of the target object may be a preset width range and / or width range of the target object.

[0167] Exemplarily, with the center position of the target object as the center, based on a random width within the width range and a random height within the height range, the target object is determined, and the pixel position information of the target object is determined. For example, the width range is between m1 pixels and m2 pixels; the height range is between m3 pixels and m4 pixels. Among them, m1, m2, m3 and m4 are positive integers, and m2 is greater than m1, and m4 is greater than m3. The width of the target object is randomly selected as n1 in the preset width range, m1≤n1≤m2; the height of the target object is randomly selected as n2 in the preset height range, m3≤n2≤m4, to determine a n1×n2 target object, and determine the pixel position information of each pixel in the n1×n2 area, that is, the pixel position information of the target object.

[0168] Since the exposure heights corresponding to the first object groups 31 of each group are different, the target objects in the different first object groups 31 formed by exposure are different. The contour shape of the target object formed by the exposure height closer to the optimal focal plane is more regular; on the contrary, the contour shape of the target object formed by the exposure height further away from the optimal focal plane is more irregular, and there may be multiple tiny closed contours. If the closed contour features of the first binary image are directly extracted, for some irregular contour shapes, the complete contour features may not be extracted, resulting in inaccurate screening of the target object. Therefore, this embodiment adopts the processing process of the above steps S12-3-5 to S12-3-5, and by adding expansion and corrosion processing, it can eliminate the noise information caused by defocused exposure, improve the accuracy of the edge contour, thereby improving the precise positioning of the target object, and then improve the accuracy of the subsequent judgment results of the exposure state of the target object, improve the accuracy of determining the defocus boundary, and finally determine more accurate calibration information.

[0169] In some embodiments, for the above step S12-2-1 or S12-3-1, determining the first binary image may specifically include steps S121 to S123, wherein:

[0170] S121, extracting contour features in the target image, and performing binarization processing on the target image to obtain an initial binary image.

[0171] like Figure 7 As shown, a preset algorithm can be used to identify edge information in the target image and extract contour features from it. The preset algorithm can be, for example, a canny algorithm, and in the process of extracting contours, the target image is binarized, the pixel values ​​of the pixels corresponding to the contour features are set to 255, and the pixel values ​​of other pixels except the contour features are set to 0. The pixel values ​​of 255 are white pixels, and the pixel values ​​of 0 are black pixels, and an initial binary image is obtained.

[0172] S122: Determine the area to be cropped based on the pixel value and position information of each pixel point in the initial binary image.

[0173] The pixel points in the to-be-cropped area are an area in the second direction Y, which includes multiple segments of pixel values ​​that are continuously 0 and is surrounded by multiple segments of pixel values ​​that are continuously 255.

[0174] Specifically, based on the pixel value and position information of each pixel point in the initial binary image, an array obtained by summing along the second direction Y (for example, the column direction) can be determined, and an area containing multiple segments of pixel values ​​that are continuously 0 and surrounded by multiple segments of pixel points that are continuously 255 can be determined, such as Figure 7 As shown, 71 represents the area to be cropped.

[0175] S123 , cutting off a portion of the initial binary image located in the area to be cut, to obtain a first binary image.

[0176] like Figure 8 As shown, this is the first binary image after eliminating interference features.

[0177] This embodiment further improves the authenticity of the target object's contour and improves the efficiency and accuracy of subsequent contour extraction by eliminating features that have a significant interference with the target object's contour.

[0178] In some embodiments, step S13 determines whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object, which specifically includes the following steps S131 to S133, wherein:

[0179] S131. Determine the area within the contour of each target object based on the pixel position information of each target object.

[0180] Specifically, taking a target object as an example, according to the pixel position information of the target object, the pixel position information here may be the position of the contour of the target object, so that storage space can be saved during data storage. According to the position of the contour of the target object, the area of ​​the target object within the contour is determined.

[0181] The area of ​​other target objects within the contour is calculated in the same way as above, and the repeated parts are not repeated here. Finally, the area within the contour of each target object is obtained.

[0182] S132: based on the comparison result between the areas within the contours of the target objects, select a reference object from the target objects, and determine the pixel position information of the reference object.

[0183] During the exposure process, the closer the exposure height is to the optimal focal plane, the sharper the image is and the smaller the astigmatism is, resulting in a relatively smaller imaging area of ​​the target object. Therefore, the area within the contour of each target object can be compared to screen out the target object with the smallest contour area as the reference object. The pixel position information of the target object with the smallest contour area is used as the pixel position information of the reference object.

[0184] S133 . Based on the pixel position information of each target object and the pixel position information of the reference object, determine whether the exposure state of the target object is normal or out of focus.

[0185] Specifically, based on the pixel position information of each target object and the pixel position information of the reference object, the correlation information or difference information between each target object and the reference object can be determined with the reference object as the standard. Then, the exposure state is determined according to whether there is correlation or difference. For example, if there is a strong correlation between the target object and the reference object, the exposure state of the target object is determined to be normal; conversely, if there is a large difference between the target object and the reference object, the exposure state of the target object is determined to be out of focus.

[0186] The following describes in detail the different methods for determining the exposure status of a target object provided by the embodiments of the present disclosure.

[0187] In a possible implementation, correlation comparison may be performed directly based on the target image to determine the exposure state of the target object, which specifically includes the following steps S133-1-1 to S133-1-2, wherein:

[0188] S133-1-1. Based on the pixel position information of each target object in the target image and the pixel position information of the reference object, a preset similarity algorithm is used to determine the similarity between each target object and the reference object.

[0189] The target image in this step, that is, Figure 2 The target image shown can be understood as the original image before image processing.

[0190] The preset similarity algorithm includes, but is not limited to, a cosine similarity algorithm, an algorithm for calculating similarity using Euclidean distance, an algorithm for calculating similarity using Manhattan distance, and the like.

[0191] Exemplarily, based on the positions of contour pixel points of each target object in the target image and the positions of contour pixel points of the reference object, a preset similarity algorithm is used to determine the similarity between each target object and the reference object.

[0192] For example, Fig.11 for Figure 2 A schematic diagram of the sub-image corresponding to the target object shown in FIG. Fig.11 As shown, based on the pixel position information of each target object in the target image, a first sub-image corresponding to the target object is intercepted from the target image. One sub-image includes only one target object, and different sub-images include different target objects. Based on the pixel position information of the reference object in the target image, a second sub-image (i.e., a standard image) corresponding to the reference object is intercepted from the target image. Based on the position of the contour pixel points of the target object in each first sub-image and the position of the contour pixel points of the reference object in the second sub-image, a preset similarity algorithm is used to determine the similarity between each target object and the reference object. Here, by intercepting and saving the sub-images, storage space can be saved.

[0193] S133-1-2: Based on the comparison result of the similarity between each target object and the reference object and the first preset threshold, determine whether the exposure state of the target object is normal or out of focus.

[0194] Taking a target object as an example, determine the comparison result of the similarity between the target object and the reference object and the first preset threshold; if the similarity is greater than the first preset threshold, it can be determined that the exposure state of the target object is normal; conversely, if the similarity is less than or equal to the first preset threshold, it can be determined that the exposure state of the target object is out of focus.

[0195] It should be noted that the first preset threshold in the embodiment of the present disclosure can be set based on experience and practical applications, and the embodiment of the present disclosure does not specifically limit it.

[0196] In another possible implementation, correlation comparison may be performed based on the first binary image to determine the exposure state of the target object, which specifically includes the following steps S133-2-1 to S133-2-2, wherein:

[0197] S133-2-1. Based on the pixel position information of each target object in the first binary image and the pixel position information of the reference object, a preset similarity algorithm is used to determine the similarity between each target object and the reference object.

[0198] The first binary image in this step can be, for example, Figure 7 The first binary image shown, or, alternatively, may be Figure 8 In order to improve the processing efficiency, the embodiment of the present disclosure selects Figure 8 The first binary image shown is taken as an example for explanation.

[0199] The preset similarity algorithm includes, but is not limited to, a cosine similarity algorithm, an algorithm for calculating similarity using Euclidean distance, an algorithm for calculating similarity using Manhattan distance, and the like.

[0200] Exemplarily, based on the positions of contour pixel points of each target object and the positions of contour pixel points of the reference object in the first binary image, a preset similarity algorithm is used to determine the similarity between each target object and the reference object.

[0201] For example, Fig.12 for Figure 8 A schematic diagram of the sub-image corresponding to the target object shown in FIG. Fig.12As shown, based on the pixel position information of each target object in the first binary image, a third sub-image corresponding to the target object is intercepted from the first binary image. One sub-image includes only one target object, and different sub-images include different target objects. Based on the pixel position information of the reference object in the first binary image, a fourth sub-image (i.e., a standard image) corresponding to the reference object is intercepted from the first binary image. Based on the position of the contour pixel points of the target object in each third sub-image and the position of the contour pixel points of the reference object in the fourth sub-image, a preset similarity algorithm is used to determine the similarity between each target object and the reference object. Here, by intercepting and saving the sub-images, storage space can be saved.

[0202] S133-2-2: Based on the comparison result between the similarity between each target object and the reference object and the first preset threshold, determine whether the exposure state of the target object is normal or out of focus.

[0203] Taking a target object as an example, determine the comparison result of the similarity between the target object and the reference object and the first preset threshold; if the similarity is greater than the first preset threshold, it can be determined that the exposure state of the target object is normal; conversely, if the similarity is less than or equal to the first preset threshold, it can be determined that the exposure state of the target object is out of focus.

[0204] It should be noted that the output results of different correlation algorithms may be smaller, indicating more similarity, or larger, indicating more similarity. For the embodiments of the present disclosure, the similarity here indicates the degree of similarity between the target object and the reference object. Therefore, the larger the similarity, which is greater than the first preset threshold, the closer the target object is to the reference object, and the exposure state of the target object can be determined to be normal; otherwise, the exposure state of the target object is determined to be out of focus.

[0205] The disclosed embodiment can improve the accuracy of the similarity by calculating the similarity between the target object and the reference object on the first binary image, thereby improving the accuracy of the exposure state of the determined target object, further improving the accuracy of determining the defocus boundary, and ultimately determining more accurate calibration information.

[0206] In another possible implementation, correlation comparison may be performed based on the first binary image to determine the exposure state of the target object, which specifically includes the following steps S133-3-1 to S133-3-2, wherein:

[0207] S133-3-1. Based on the pixel position information of each target object in the second intermediate image and the pixel position information of the reference object, a preset similarity algorithm is used to determine the similarity between each target object and the reference object.

[0208] The second intermediate image in this step can be, for example, Fig.10 The second intermediate image is shown.

[0209] The preset similarity algorithm includes, but is not limited to, a cosine similarity algorithm, an algorithm for calculating similarity using Euclidean distance, an algorithm for calculating similarity using Manhattan distance, and the like.

[0210] Exemplarily, based on the positions of contour pixel points of each target object in the second intermediate image and the positions of contour pixel points of the reference object, a preset similarity algorithm is used to determine the similarity between each target object and the reference object.

[0211] For example, Fig.13 for Fig.10 A schematic diagram of the sub-image corresponding to the target object shown in FIG. Fig.13 As shown, based on the pixel position information of each target object in the second intermediate image, a fifth sub-image corresponding to the target object is intercepted from the second intermediate image. One sub-image includes only one target object, and different sub-images include different target objects. Based on the pixel position information of the reference object in the second intermediate image, a sixth sub-image (i.e., a standard image) corresponding to the reference object is intercepted from the second intermediate image. Based on the position of the contour pixel points of the target object in each fifth sub-image and the position of the contour pixel points of the reference object in the sixth sub-image, a preset similarity algorithm is used to determine the similarity between each target object and the reference object. Here, by intercepting and saving the sub-images, storage space can be saved.

[0212] S133-3-2. Based on the comparison result between the similarity between each target object and the reference object and the first preset threshold, determine whether the exposure state of the target object is normal or out of focus.

[0213] Taking a target object as an example, determine the comparison result of the similarity between the target object and the reference object and the first preset threshold; if the similarity is greater than the first preset threshold, it can be determined that the exposure state of the target object is normal; conversely, if the similarity is less than or equal to the first preset threshold, it can be determined that the exposure state of the target object is out of focus.

[0214] The disclosed embodiment can further improve the accuracy of the similarity by calculating the similarity between the target object and the reference object on the image after corrosion and expansion (i.e., the second intermediate image), thereby improving the accuracy of the exposure state of the determined target object, and further improving the accuracy of determining the defocus boundary, and ultimately determining more accurate calibration information.

[0215] In another possible implementation, correlation comparison may be performed based on the target binary sub-image to determine the exposure state of the target object, which specifically includes the following steps S13-1-1 to S13-1-5, wherein:

[0216] S13-1-1. According to the pixel position information of the target object, a target sub-image corresponding to the target object is intercepted from the target image, and the area within the outline of the target object in each target sub-image is determined.

[0217] The pixel position information of the target object may be the position of the contour of the target object. According to the position of the contour of the target object, the target sub-image corresponding to the target object can be intercepted from the target image; at the same time, according to the position of the contour of the target object, the area of ​​the target object within the contour can be determined.

[0218] S13-1-2. Based on the comparison results between the areas within the contours of the target objects, select a reference object from the target objects and determine a reference subgraph of the reference object.

[0219] The areas within the contours of the target objects can be compared to select the target object with the smallest contour area as the reference object. The target sub-graph corresponding to the target object with the smallest contour area is used as the reference sub-graph of the reference object.

[0220] S13-1-3. Binarize each target sub-image and reference sub-image respectively to obtain each target binary sub-image and reference binary sub-image.

[0221] Fig.14 A schematic diagram of converting a target sub-graph provided in an embodiment of the present disclosure into a reference sub-graph, such as Fig.14 As shown, binarization is directly performed according to the pixel values ​​in the target sub-image, for example, the comparison result between the pixel value of each pixel in the target sub-image and the fourth preset threshold is determined; when the pixel value of a pixel is greater than the fourth preset threshold, the pixel value of the pixel is updated to the second set value; when the pixel value of a pixel is less than or equal to the fourth preset threshold, the pixel value of the pixel is updated to the first set value. For example, when converting the target image to a black and white image, the value range of the fourth preset threshold can be between 100 and 200; the first set value can be set to 0 and the second set value can be set to 255. The pixel point with a pixel value of 255 is a pixel point that displays white, and the pixel point with a pixel value of 0 is a pixel point that displays black.

[0222] S13-1-4. Based on each target binary sub-graph and the reference binary sub-graph, a preset similarity algorithm is used to determine the similarity between each target object and the reference object.

[0223] For example, Fig.14As shown, taking the determination of the similarity between a target object and a reference object as an example, based on the pixel values ​​of each pixel point in the target binary sub-image and the pixel values ​​of each pixel point in the reference binary sub-image, the similarity between the target object and the reference object is determined using a preset similarity algorithm. The preset similarity algorithm includes, but is not limited to, a cosine similarity algorithm, an algorithm for calculating similarity using Euclidean distance, an algorithm for calculating similarity using Manhattan distance, and the like.

[0224] S133-4-5. Based on the comparison result between the similarity between each target object and the reference object and the first preset threshold, determine whether the exposure state of the target object is normal or out of focus.

[0225] Taking a target object as an example, determine the comparison result of the similarity between the target object and the reference object and the first preset threshold; if the similarity is greater than the first preset threshold, it can be determined that the exposure state of the target object is normal; conversely, if the similarity is less than or equal to the first preset threshold, it can be determined that the exposure state of the target object is out of focus.

[0226] The disclosed embodiment can improve the accuracy of the similarity by calculating the similarity between the target object and the reference object in the target binary sub-image, thereby improving the accuracy of the exposure state of the determined target object, and further improving the accuracy of determining the defocus boundary, and ultimately determining more accurate calibration information.

[0227] In another possible implementation, correlation comparison can be performed based on the target binary sub-image to determine the exposure state of the target object, which specifically includes the following steps S13-2-1 to S13-2-6, wherein:

[0228] S13-2-1. According to the pixel position information of the target object, a target sub-image corresponding to the target object is intercepted from the target image, and the area within the outline of the target object in each target sub-image is determined.

[0229] The implementation principle of this step is the same as that of step S13-1-1, and the repeated parts will not be repeated.

[0230] S13-2-2. Based on the comparison results between the areas within the contours of the target objects, select a reference object from the target objects and determine a reference subgraph of the reference object.

[0231] The implementation principle of this step is the same as that of step S13-1-2, and the repeated parts will not be repeated.

[0232] S13-2-3. Reduce the target sub-image and the reference sub-image according to a preset scale factor to obtain reduced target sub-images and reduced reference sub-images.

[0233] The preset scale factor can be understood as a reduction scale factor pre-set for the target object, which can be set based on experience or actual scenarios, and the embodiments of the present disclosure do not limit specific data. Here, reducing the target sub-image and the reference sub-image can reduce the amount of data processing and improve efficiency.

[0234] S13-2-4. Binarization is performed on each of the reduced target sub-images and the reduced reference sub-images to obtain each target binary sub-image and reference binary sub-image.

[0235] The principle of binarization in this step is the same as the principle of binarization in step S13-1-3, and the repeated parts will not be repeated.

[0236] S13-2-5. Based on each target binary subgraph and the reference binary subgraph, determine the Hamming distance of each target binary subgraph.

[0237] Taking a target binary sub-image as an example, the Hamming distance algorithm is used to determine the number of operations for converting the target binary sub-image into the reference binary sub-image, which is the Hamming distance. The Hamming distance can reflect the degree of difference between the reduced target object in the target binary sub-image and the reduced reference object in the reference binary sub-image. For example, the larger the Hamming distance, the greater the difference.

[0238] S13-2-6. Based on the comparison result of the Hamming distance of each target binary sub-image and the second preset threshold, determine whether the exposure state of the target object is normal or out of focus.

[0239] Taking a target binary sub-image as an example, determine the comparison result of the Hamming distance of the target binary sub-image and the second preset threshold; if the Hamming distance is less than the second preset threshold, it can be determined that the exposure state of the target object is normal; conversely, if the Hamming distance is greater than or equal to the second preset threshold, it can be determined that the exposure state of the target object is out of focus.

[0240] In another possible implementation, correlation comparison can be performed based on the target binary sub-image to determine the exposure state of the target object, which specifically includes the following steps S13-3-1 to S13-3-4, wherein:

[0241] S13-3-1. According to the pixel position information of the target object, a target sub-image corresponding to the target object is intercepted from the target image.

[0242] The implementation principle of this step is the same as that of step S13-1-1, and the repeated parts will not be repeated.

[0243] S13-3-2, binarizing each target sub-graph to obtain each target binary sub-graph;

[0244] The principle of binarization in this step is the same as the principle of binarization in step S13-1-3, and the repeated parts will not be repeated.

[0245] S13-3-3. Based on each target binary sub-image and a preset pure color image, determine the Hamming distance of each target binary sub-image.

[0246] The size of the solid color image is the same as the preset size.

[0247] The size of the pure color image is the same as the preset size. The pixel value of each pixel in the pure color image is the same as the pixel value of the non-target object in the target binary sub-image, that is, the pixel value of the target object is opposite. The pixel value of the target object is 255, and the pixel value of each pixel in the pure color image is 0, that is, the pure color image is a pure black image.

[0248] Specifically, taking a target binary sub-image as an example, the Hamming distance algorithm is used to determine the Hamming distance of the target binary sub-image converted to a pure color image. The calculated Hamming distance here can be understood as the number or area of ​​white pixels in the target binary sub-image. The more white pixels or the larger the area of ​​white pixels, the larger the Hamming distance. The larger the Hamming distance, the more abnormal the exposure state, that is, the greater the possibility of defocus.

[0249] S13-3-4. Based on the comparison result of the Hamming distance of each target binary sub-image and the third preset threshold, determine whether the exposure state of the target object is normal or out of focus.

[0250] Taking a target binary sub-image as an example, determine the comparison result between the difference between the Hamming distance of the target binary sub-image and the Hamming distance of the reference binary sub-image converted into a pure color image and the third preset threshold; if the difference is less than the third preset threshold, it can be determined that the exposure state of the target object is normal; conversely, if the difference is greater than or equal to the third preset threshold, it can be determined that the exposure state of the target object is out of focus.

[0251] In some embodiments, with respect to step S12, determining the pixel position information corresponding to each target object specifically includes the following steps S12-4-1 to S12-4-3, wherein:

[0252] S12-4-1. Binarize the target image to obtain a second binary image.

[0253] In a possible implementation, Fig.14As shown, taking a target object as an example, the target image can be converted to directly obtain a second binary image. Exemplarily, the comparison result between the pixel value of each pixel in the target image and the fourth preset threshold is determined; when the pixel value of a pixel is greater than the fourth preset threshold, the pixel value of the pixel is updated to the second set value; when the pixel value of a pixel is less than or equal to the fourth preset threshold, the pixel value of the pixel is updated to the first set value. For example, when converting the target image to a black and white image, the value range of the fourth preset threshold can be between 100 and 200; the first set value can be set to 0 and the second set value can be set to 255. A pixel with a pixel value of 255 is a pixel that displays white, and a pixel with a pixel value of 0 is a pixel that displays black.

[0254] In another possible implementation, for the target image, the redundant image can be first eliminated, and then the target image after the redundant image is eliminated is converted to obtain a second binary image. The redundant image here can be understood as an image of interference features that is more obvious on the preset template except for the target object when the preset template is acquired. The method of eliminating redundant images, for example, can be to use an image recognition algorithm to identify redundant targets for elimination. Specifically, the pixel value of the pixel point corresponding to the redundant target can be changed, and the data after the pixel value is changed can be set according to the actual situation, which is not limited in the embodiments of the present disclosure.

[0255] S12-4-2. Extract contour features in the second binary image and determine the center position of the target object.

[0256] A preset algorithm may be used to identify the second binary image and extract contour features in the second binary image. The preset algorithm may be, for example, a canny algorithm, which uses the edges of connected white pixels in the second binary image as contour feedback, maximizes the contour by using dilation and erosion methods, and sets a threshold on the area contained in the contour, filters out contours that are smaller than the set area threshold, and retains contour information that is closest to the actual contour.

[0257] Afterwards, the extracted contour features may be determined to be the contour of the target object, and the center position of the contour, such as the geometric center position, may be determined and recorded as the center position of the target object.

[0258] For contour features within a certain range, it can be determined that it is the contour of a target object A, and based on the position of the pixel points where the contour of the target object A is located, the center of the rectangular frame formed by the boundary of the two pixel points farthest in the first direction X along the second direction and the two pixel points farthest in the second direction Y along the first direction X as the boundary is selected as the center position of the contour of the target object A, which is recorded as the center position of the target object A.

[0259] Exemplarily, a preset algorithm may be used to extract closed contour features in the first binary image, thereby avoiding the influence of noise such as non-closed contour features on the accuracy of the recognition result.

[0260] S12-4-3. Determine pixel position information of the target object based on the center position of the target object and a preset size range of the target object.

[0261] The preset size range of the target object may be a preset width range and / or width range of the target object.

[0262] Exemplarily, with the center position of the target object as the center, based on a random width within the width range and a random height within the height range, the target object is determined, and the pixel position information of the target object is determined. For example, the width range is between m1 pixels and m2 pixels; the height range is between m3 pixels and m4 pixels. Among them, m1, m2, m3 and m4 are positive integers, and m2 is greater than m1, and m4 is greater than m3. The width of the target object is randomly selected as n1 in the preset width range, m1≤n1≤m2; the height of the target object is randomly selected as n2 in the preset height range, m3≤n2≤m4, to determine a n1×n2 target object, and determine the pixel position information of each pixel in the n1×n2 area, that is, the pixel position information of the target object.

[0263] This embodiment converts the target image into a second binary image before feature extraction, which can improve contour recognition, thereby making it easier to extract the contour features of the target object and improve contour recognition efficiency. In addition, the image recognition algorithm can accurately determine the pixel position information of each target object, thereby improving the accuracy of the subsequent judgment result of the exposure state of the target object, thereby improving the accuracy of determining the defocus boundary, and finally determining more accurate calibration information.

[0264] In some embodiments, correlation comparison can be performed based on the second binary image to determine the exposure state of the target object, which specifically includes the following steps S13-4-1 to S13-4-4, wherein:

[0265] S13-4-1. Determine the area within the outline of each target object based on the pixel position information of each target object in the second binary image.

[0266] The implementation principle of this step is the same as that of step S131, and the repeated parts will not be repeated.

[0267] S13-4-2. Based on the comparison results between the areas within the contours of the target objects, select a reference object from the target objects and determine the pixel position information of the reference object.

[0268] The implementation principle of this step is the same as that of step S132, and the repeated parts will not be repeated.

[0269] S13-4-3. Based on the pixel position information of each target object in the second binary image and the pixel position information of the reference object, a preset similarity algorithm is used to determine the similarity between each target object and the reference object.

[0270] The preset similarity algorithm includes, but is not limited to, a cosine similarity algorithm, an algorithm for calculating similarity using Euclidean distance, an algorithm for calculating similarity using Manhattan distance, and the like.

[0271] For example, Fig.14 As shown, based on the positions of the contour pixel points of each target object and the positions of the contour pixel points of the reference object in the second binary image, a preset similarity algorithm is used to determine the similarity between each target object and the reference object.

[0272] S13-4-4. Based on the comparison result between the similarity between each target object and the reference object and the first preset threshold, determine whether the exposure state of the target object is normal or out of focus.

[0273] Taking a target object as an example, determine the comparison result of the similarity between the target object and the reference object and the first preset threshold; if the similarity is greater than the first preset threshold, it can be determined that the exposure state of the target object is normal; conversely, if the similarity is less than or equal to the first preset threshold, it can be determined that the exposure state of the target object is out of focus.

[0274] The disclosed embodiment can improve the accuracy of the similarity by calculating the similarity between the target object and the reference object on the second binary image, thereby improving the accuracy of the exposure state of the determined target object, further improving the accuracy of determining the defocus boundary, and ultimately determining more accurate calibration information.

[0275] In some embodiments, correlation comparison can be performed based on the second binary image to determine the exposure state of the target object, which specifically includes the following steps S13-5-1 to S13-5-6, wherein:

[0276] S13-5-1. Determine the area within the contour of each target object based on the pixel position information of each target object in the second binary image.

[0277] The implementation principle of this step is the same as that of step S131, and the repeated parts will not be repeated.

[0278] S13-5-2. Based on the comparison results between the areas within the contours of the target objects, select a reference object from the target objects and determine the pixel position information of the reference object.

[0279] The implementation principle of this step is the same as that of step S132, and the repeated parts will not be repeated.

[0280] S13-5-3. Based on the pixel position information of each target object and the pixel position information of the reference object in the second binary image, the target object and the reference object are reduced according to a preset scale factor to obtain reduced target objects and reduced reference objects.

[0281] The second binary image in this step can be found in Fig.14 As shown, it is a sub-image corresponding to the target object in the second binary image.

[0282] The preset scale factor can be understood as a reduction scale factor pre-set for the target object, which can be set based on experience or actual scenarios, and the embodiments of the present disclosure do not limit specific data. Exemplarily, based on the positions of the contour pixel points of each target object and the positions of the contour pixel points of the reference object in the second binary image, the target object and the reference object are reduced according to the preset scale factor to obtain each reduced target object and the reduced reference object. Here, reducing the target object and the reference object can reduce the amount of data processing and improve efficiency.

[0283] S13-5-4. Based on the pixel position information of each reduced target object and the pixel position information of the reduced reference object, a first sub-image corresponding to the reduced target object is cut out from the second binary image according to a preset size, and a second sub-image corresponding to the reduced reference object is cut out from the second binary image.

[0284] The preset size may be pre-set based on experience or actual application scenarios, and the specific data is not specifically limited in the embodiments of the present disclosure.

[0285] A first sub-graph includes only one target object, and different first sub-graphs include different target objects.

[0286] S13-5-5. Based on each first subgraph and the second subgraph, determine the Hamming distance of each first subgraph.

[0287] Taking a first sub-image as an example, the Hamming distance algorithm is used to determine the number of operations for converting the first sub-image into the second sub-image, which is the Hamming distance. The Hamming distance can reflect the degree of difference between the reduced target object in the first sub-image and the reduced reference object in the second sub-image. For example, the larger the Hamming distance, the greater the difference.

[0288] S13-5-6. Based on the comparison result of the Hamming distance of each first sub-image and the second preset threshold, determine whether the exposure state of the target object is normal or out of focus.

[0289] Taking a first sub-image as an example, determine the comparison result of the Hamming distance of the first sub-image and the second preset threshold; if the Hamming distance is less than the second preset threshold, it can be determined that the exposure state of the target object is normal; conversely, if the Hamming distance is greater than or equal to the second preset threshold, it can be determined that the exposure state of the target object is out of focus.

[0290] In some embodiments, correlation comparison can be performed based on the second binary image to determine the exposure state of the target object, which specifically includes the following steps S13-6-1 to S13-6-3, wherein:

[0291] S13-6-1. Based on the pixel position information of each target object in the second binary image, third sub-images corresponding to each target object are cut out from the second binary image according to a preset size.

[0292] The second binary image in this step is Fig.14 shown.

[0293] The preset size may be pre-set based on experience or actual application scenarios, and the specific data is not specifically limited in the embodiments of the present disclosure.

[0294] A third sub-graph includes only one target object, and different third sub-graphs include different target objects.

[0295] S13-6-2. Based on each third sub-image and a preset pure color image, determine the Hamming distance of each third sub-image.

[0296] The size of the pure color image is the same as the preset size. The pixel value of each pixel in the pure color image is the same as the pixel value of the non-target object in the third sub-image, that is, the pixel value of the target object is opposite. The pixel value of the target object is 255, and the pixel value of each pixel in the pure color image is 0, that is, the pure color image is a pure black image.

[0297] Specifically, taking a third sub-image as an example, the Hamming distance algorithm is used to determine the Hamming distance of the third sub-image converted into a pure color image. The calculated Hamming distance here can be understood as the number or area of ​​white pixels in the third sub-image. The more white pixels or the larger the area of ​​white pixels, the larger the Hamming distance. The larger the Hamming distance, the more abnormal the exposure state, that is, the greater the possibility of defocus.

[0298] S13-6-3. Based on the comparison result of the Hamming distance of each third sub-image and the third preset threshold, determine whether the exposure state of the target object is normal or out of focus.

[0299] Taking a target binary sub-image as an example, determine the comparison result between the difference between the Hamming distance of the target binary sub-image and the Hamming distance of the reference binary sub-image converted into a pure color image and the third preset threshold; if the difference is less than the third preset threshold, it can be determined that the exposure state of the target object is normal; conversely, if the difference is greater than or equal to the third preset threshold, it can be determined that the exposure state of the target object is out of focus.

[0300] In some embodiments, according to the judgment result and the pre-stored exposure heights corresponding to each first object group 31, the predicted focal plane heights corresponding to different test areas are determined in step S13, which specifically includes the following steps S134 to S137, wherein:

[0301] S134. For target objects in each second object group 32 arranged side by side along the second direction Y, based on the exposure status of each target object and the pixel position information of each target object, determine the pixel position information of the defocus boundary corresponding to each second object group 32.

[0302] In one possible implementation, Fig.15 A schematic diagram of a defocus boundary corresponding to a second object group provided in an embodiment of the present disclosure, such as Fig.15 As shown, for a group of second object groups 32, based on the exposure status and pixel position information of each target object in the second object group 32, a target object 322 (for ease of understanding, the target object is recorded as the first target object 322) adjacent to the target object 321 with a defocused exposure status and a normal exposure status is determined. The position of the first target object 322 is determined to be the defocus boundary corresponding to the second object group 32. For example, based on the pixel position information of the first target object, the position of the pixel point closest to the defocused target object in the first target object is used as the position of the defocus boundary; or, the position of the geometric center point of the first target object is used as the position of the defocus boundary.

[0303] Since the exposure state of the first target object is close to being out of focus, this embodiment uses the position of the first target object closest to the defocus state as the defocus boundary, which can improve the accuracy of subsequent prediction of the optimal focal plane exposure height.

[0304] It should be noted that if Fig.15 As shown, since the preset template sets the exposure height of the 13th group of first object groups 31 as the optimal focal plane (the optimal focal plane may deviate and is the focal plane to be calibrated) during the exposure stage, the exposure heights of the other first object groups 31 except the 13th group of first object groups 31 deviate and are symmetrical on the left and right, so a group of second object groups 32 includes two defocus boundaries 323.

[0305] Similarly, the method for determining the defocus boundary corresponding to each second object group 32 is the same as above, and the repeated parts are not repeated here.

[0306] In another possible implementation, for a group of second object groups 32, based on the exposure status and pixel position information of each target object in the second object group 32, a target object adjacent to the target object with a defocused exposure status and a normal exposure status (for ease of understanding, the target object is recorded as the first target object 322) is determined, and a target object adjacent to the first target object and with a defocused exposure status (for ease of understanding, the target object is recorded as the second target object 321) is determined. The center position between the first target object 322 and the second target object 321 is determined as the defocus boundary corresponding to the second object group 32, and the pixel position information of the defocus boundary can be determined according to the pixel position information of the first target object and the pixel position information of the second target object. For example, the midpoint of the line connecting the position of the geometric center point of the first target object and the position of the geometric center point of the second target object is recorded as the position of the defocus boundary.

[0307] S135 , determining the regional defocus boundaries corresponding to each test area based on the pixel position information of the defocus boundaries corresponding to each second object group 32 and the position information of the test area.

[0308] Specifically, it is known that the preset template includes a plurality of test areas, and the positions of the test areas are fixed, so each second object group 32 in each test area can be determined.

[0309] Exemplarily, since each second object group 32 includes two oppositely disposed boundaries, for the convenience of distinction, they are respectively recorded as a first sub-boundary and a second sub-boundary. The regional defocus boundary includes two oppositely disposed boundaries, for the convenience of distinction, they are respectively recorded as a first boundary and a second boundary. Fig.16 A schematic diagram of a defocused boundary of a region on a target image provided by an embodiment of the present disclosure, such as Fig.16 As shown, 331 represents the first boundary, and 332 represents the second boundary. Fig.16 Two test areas are shown (referred to as the first test area 201 and the second test area 202 for convenience of description). Taking the determination of the first boundary 331 in one test area as an example, based on the positions of the first sub-boundaries corresponding to the second object groups 32 in the test area, the mean of the first sub-boundaries in the first direction X is determined, which is the position of the first boundary in the first direction X. Similarly, based on the positions of the second sub-boundaries corresponding to the second object groups 32 in the test area, the mean of the second sub-boundaries in the first direction X is determined, which is the position of the second boundary in the first direction X. The first boundary 331 and the second boundary 332 constitute the regional defocus boundary of the test area.

[0310] The method for determining the regional defocus boundaries corresponding to other test areas is similar, and the repeated parts will not be repeated.

[0311] S136 , for the regional defocus boundaries corresponding to the test areas, determine the exposure heights of the regional defocus boundaries based on the pixel position information of the regional defocus boundaries and the pre-stored exposure heights corresponding to the first object groups 31 .

[0312] The regional defocus boundary includes a first boundary 331 and a second boundary 332. If the first boundary 331 is located in the first object group 31A (not shown in the figure), the exposure height of the first boundary 331 is determined to be the exposure height corresponding to the first object group 31A; if the second boundary 332 is located in the first object group 31B (not shown in the figure), the exposure height of the second boundary 332 is determined to be the exposure height corresponding to the first object group 31B. If the first boundary 331 is not located in the first object group 31, the first object group 31C (not shown in the figure) closest to the first boundary 331 is determined, and the exposure height of the first boundary 331 is determined to be the exposure height corresponding to the first object group 31C; if the second boundary 332 is not located in the first object group 31, the first object group 31D (not shown in the figure) closest to the second boundary 332 is determined, and the exposure height of the second boundary 332 is determined to be the exposure height corresponding to the first object group 31D.

[0313] S137 . For the regional defocus boundaries corresponding to the respective test regions, determine the predicted focal plane height of the test region based on the exposure height of the regional defocus boundaries.

[0314] Specifically, the average of the exposure height of the first boundary 331 and the exposure height of the second boundary 332 may be used as the predicted focal plane height of the test area.

[0315] In some embodiments, the calibration information of the best focal plane of the exposure device includes the height and tilt of the best focal plane; the test area includes two (recorded as the first test area 201 and the second test area 202 for the convenience of description). Specifically, the first preset algorithm is an algorithm for calculating the average of the predicted focal plane heights; the second preset algorithm is an algorithm for calculating the difference between the two predicted focal plane heights. The average of the predicted focal plane heights can be used as the height of the best focal plane. The difference between the two predicted focal plane heights can be used as the tilt of the best focal plane.

[0316] Exemplarily, the predicted focal plane height UpBest of the first test area 201 is determined by the following formula:

[0317] UpBest=(UL+UR) / 2, wherein UL represents the exposure height of the first border of the first test area 201 , and UR represents the exposure height of the second border of the first test area 201 .

[0318] Exemplarily, the predicted focal plane height DownBest of the second test area 202 is determined by the following formula:

[0319] DownBest=(DL+DR) / 2, wherein DL represents the exposure height of the first border of the second test area 202 , and DR represents the exposure height of the second border of the second test area 202 .

[0320] Exemplarily, the height of the best focal plane Best is determined by the following formula:

[0321] Best=(UpBest+DownBest) / 2

[0322] Exemplarily, the tilt amount Tilt of the optimal focal plane is determined by the following specific formula:

[0323] Tilt=UpBest–DownBest

[0324] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.

[0325] In addition, a focal plane calibration device corresponding to the focal plane calibration method is also provided in an embodiment of the present disclosure. Since the principle of solving the problem by the device in the embodiment of the present disclosure is similar to the above-mentioned focal plane calibration method in the embodiment of the present disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0326] Fig.17 A schematic diagram of a focal plane calibration device provided in an embodiment of the present disclosure is shown in FIG. Fig.17 As shown, it includes an image acquisition module 21, an image processing module 22, a prediction module 23 and a calibration information determination module 24.

[0327] in:

[0328] The image acquisition module 21 is configured to acquire an image of a preset template to obtain a target image; the target image includes a plurality of first object groups arranged side by side along a first direction, and a plurality of second object groups arranged side by side along a second direction; the first object group includes a plurality of target objects arranged side by side along the second direction; the second object group includes a plurality of target objects arranged side by side along the first direction; the preset template includes a plurality of test areas, and the target object in one test area is obtained by one exposure, and the target objects in different test areas are obtained by different exposures.

[0329] It should be noted that the image acquisition module 21 in the embodiment of the present disclosure is configured to execute step S11 in the above-mentioned focal plane calibration method.

[0330] The image processing module 22 is configured to determine pixel position information corresponding to each target object according to the target image.

[0331] It should be noted that the image processing module 22 in the embodiment of the present disclosure is configured to execute step S12 in the above-mentioned focal plane calibration method.

[0332] The prediction module 23 is configured to judge whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object, and determine the predicted focal plane heights corresponding to different test areas according to the judgment result and the exposure heights corresponding to each pre-stored first object group.

[0333] It should be noted that the prediction module 23 in the embodiment of the present disclosure is configured to execute step S13 in the above-mentioned focal plane calibration method.

[0334] The calibration information determination module 24 is configured to determine calibration information of an optimal focal plane of the exposure device based on each predicted focal plane height.

[0335] It should be noted that the calibration information determination module 24 in the embodiment of the present disclosure is configured to execute step S14 in the above-mentioned focal plane calibration method.

[0336] In some embodiments, the image processing module 22 is specifically configured to extract contour features in the target image and determine the center position of the target object; based on the center position of the target object and a preset size range of the target object, determine the pixel position information of the target object.

[0337] It should be noted that the image processing module 22 in the embodiment of the present disclosure is specifically configured to execute steps S12 - 1 - 1 to S12 - 1 - 2 in the above-mentioned focal plane calibration method.

[0338] In some embodiments, the image processing module 22 is specifically configured to extract contour features in the target image, and binarize the target image to obtain a first binary image, and determine the center position of the target object based on the extracted contour features; based on the center position of the target object and a pre-set size range of the target object, determine the pixel position information of the target object.

[0339] It should be noted that the image processing module 22 in the embodiment of the present disclosure is specifically configured to execute steps S12-2-1 to S12-2-2 in the above-mentioned focal plane calibration method.

[0340] In some embodiments, the image processing module 22 is specifically configured to extract contour features in the target image, and binarize the target image to obtain a first binary image; dilate the first binary image to obtain a first intermediate image; erode the first intermediate image to obtain a second intermediate image; extract contour features in the second intermediate image to determine the center position of the target object; determine the pixel position information of the target object based on the center position of the target object and a pre-set size range of the target object.

[0341] It should be noted that the image processing module 22 in the embodiment of the present disclosure is specifically configured to execute steps S12-3-1 to S12-3-2 in the above-mentioned focal plane calibration method.

[0342] In some embodiments, the image processing module 22 performs binarization processing on the target image to obtain a first binary image, including: extracting contour features in the target image, and performing binarization processing on the target image to obtain an initial binary image; determining the area to be cropped based on the pixel value and position information of each pixel point in the initial binary image; the pixel points in the area to be cropped are an area in the second direction that contains multiple segments of pixel values ​​that are continuously 0 and is surrounded by multiple segments of pixel values ​​that are continuously 255; and cutting off the portion of the image in the initial binary image that is located in the area to be cropped to obtain a first binary image.

[0343] It should be noted that the image processing module 22 in the embodiment of the present disclosure is specifically configured to execute the specific implementation process of step S12-2-1 or S12-3-1 in the above-mentioned focal plane calibration method.

[0344] In some embodiments, the size range of the target object includes a width range and a height range;

[0345] The image processing module 22 determines the pixel position information of the target object based on the center position of the target object and the preset size range of the target object, including:

[0346] Taking the center position of the target object as the center, based on a random width within a width range and a random height within a height range, the target object is determined, and pixel position information of the target object is determined.

[0347] It should be noted that the image processing module 22 in the embodiment of the present disclosure is specifically configured to execute the specific implementation process of step S12-2-3 or S12-3-5 in the above-mentioned focal plane calibration method.

[0348] In some embodiments, the prediction module 23 includes a judgment unit 231 and a prediction unit 232; wherein the judgment unit 231 is configured to determine the area within the outline of each target object based on the pixel position information of each target object; based on the comparison results between the areas within the outlines of each target object, filter out a reference object from each target object and determine the pixel position information of the reference object; based on the pixel position information of each target object and the pixel position information of the reference object, judge whether the exposure state of the target object is normal or out of focus.

[0349] It should be noted that the determination unit 231 in the embodiment of the present disclosure is configured to execute steps S131 to S133 in the above-mentioned focal plane calibration method.

[0350] In some embodiments, the judgment unit 231 is specifically configured to determine the similarity between each target object and the reference object based on the pixel position information of each target object in the target image and the pixel position information of the reference object, using a preset similarity algorithm; and judge whether the exposure state of the target object is normal or out of focus based on the comparison result of the similarity between each target object and the reference object and a first preset threshold.

[0351] It should be noted that the determination unit 231 in the embodiment of the present disclosure is specifically configured to execute steps S133 - 1 - 1 to S133 - 1 - 2 in the above-mentioned focal plane calibration method.

[0352] In some embodiments, the judgment unit 231 is specifically configured to determine the similarity between each target object and the reference object based on the pixel position information of each target object in the first binary image and the pixel position information of the reference object, using a preset similarity algorithm; and judge whether the exposure state of the target object is normal or out of focus based on the comparison result of the similarity between each target object and the reference object and the first preset threshold.

[0353] It should be noted that the determination unit 231 in the embodiment of the present disclosure is specifically configured to execute steps S133-2-1 to S133-2-2 in the above-mentioned focal plane calibration method.

[0354] In some embodiments, the judgment unit 231 is specifically configured to determine the similarity between each target object and the reference object based on the pixel position information of each target object in the second intermediate image and the pixel position information of the reference object, using a preset similarity algorithm; and judge whether the exposure state of the target object is normal or out of focus based on the comparison result of the similarity between each target object and the reference object and the first preset threshold.

[0355] It should be noted that the judgment unit 231 in the embodiment of the present disclosure is specifically configured to execute steps S133-3-1 to S133-3-2 in the above-mentioned focal plane calibration method.

[0356] In some embodiments, the judgment unit 231 is specifically configured to intercept a target sub-image corresponding to the target object from the target image according to the pixel position information of the target object, and determine the area within the outline of the target object in each target sub-image; based on the comparison result between the areas within the outlines of each target object, screen out a reference object from each target object, and determine the reference sub-image of the reference object; binarize each target sub-image and the reference sub-image respectively to obtain each target binary sub-image and a reference binary sub-image; based on each target binary sub-image and the reference binary sub-image, using a preset similarity algorithm, determine the similarity between each target object and the reference object; based on the comparison result of the similarity between each target object and the reference object and the first preset threshold, judge whether the exposure state of the target object is normal or out of focus.

[0357] It should be noted that the judgment unit 231 in the embodiment of the present disclosure is specifically configured to execute steps S13-1-1 to S13-1-6 in the above-mentioned focal plane calibration method.

[0358] In some embodiments, the judgment unit 231 is specifically configured to intercept a target sub-image corresponding to the target object from the target image according to the pixel position information of the target object, and determine the area within the outline of the target object in each target sub-image; based on the comparison result between the areas within the outlines of each target object, screen out a reference object from each target object, and determine the reference sub-image of the reference object; according to a preset scale factor, reduce the target sub-image and the reference sub-image to obtain each reduced target sub-image and the reduced reference sub-image; binarize each reduced target sub-image and the reduced reference sub-image respectively to obtain each target binary sub-image and the reference binary sub-image; determine the Hamming distance of each target binary sub-image based on each target binary sub-image and the reference binary sub-image; based on the comparison result of the Hamming distance of each target binary sub-image and the second preset threshold, judge whether the exposure state of the target object is normal or out of focus.

[0359] It should be noted that the judgment unit 231 in the embodiment of the present disclosure is specifically configured to execute steps S13-2-1 to S13-2-6 in the above-mentioned focal plane calibration method.

[0360] In some embodiments, the judgment unit 231 is specifically configured to intercept a target sub-image corresponding to the target object from the target image according to the pixel position information of the target object; perform binarization processing on each target sub-image respectively to obtain each target binary sub-image; determine the Hamming distance of each target binary sub-image based on each target binary sub-image and a preset solid color image; the size of the solid color image is the same as the preset size; based on the comparison result of the Hamming distance of each target binary sub-image and the third preset threshold value, judge whether the exposure state of the target object is normal or out of focus.

[0361] It should be noted that the judgment unit 231 in the embodiment of the present disclosure is specifically configured to execute steps S13-3-1 to S13-3-6 in the above-mentioned focal plane calibration method.

[0362] In some embodiments, the image processing module 22 is specifically configured to perform binarization processing on the target image to obtain a second binary image; extract contour features in the second binary image to determine the center position of the target object; and determine the pixel position information of the target object based on the center position of the target object and a pre-set size range of the target object.

[0363] It should be noted that the image processing module 22 in the embodiment of the present disclosure is specifically configured to execute steps S12-4-1 to S12-4-3 in the above-mentioned focal plane calibration method.

[0364] In some embodiments, the judgment unit 231 is specifically configured to determine the area within the outline of each target object based on the pixel position information of each target object in the second binary image; based on the comparison result between the areas within the outlines of each target object, screen out a reference object from each target object and determine the pixel position information of the reference object; based on the pixel position information of each target object in the second binary image and the pixel position information of the reference object, using a preset similarity algorithm, determine the similarity between each target object and the reference object; based on the comparison result of the similarity between each target object and the reference object and the first preset threshold, judge whether the exposure state of the target object is normal or out of focus.

[0365] It should be noted that the judgment unit 231 in the embodiment of the present disclosure is specifically configured to execute steps S13-1-1 to S13-1-4 in the above-mentioned focal plane calibration method.

[0366] In some embodiments, the judgment unit 231 is specifically configured to determine the area within the outline of each target object based on the pixel position information of each target object in the second binary image; based on the comparison result between the areas within the outlines of each target object, screen out a reference object from each target object and determine the pixel position information of the reference object; based on the pixel position information of each target object in the second binary image and the pixel position information of the reference object, reduce the target object and the reference object according to a preset scale factor to obtain each reduced target object and the reduced reference object; based on the pixel position information of each reduced target object and the pixel position information of the reduced reference object, according to a preset size, intercept a first sub-image corresponding to the reduced target object from the second binary image and intercept a second sub-image corresponding to the reduced target object from the second binary image; based on each first sub-image and the second sub-image, determine the Hamming distance of each first sub-image; based on the comparison result of the Hamming distance of each first sub-image and the second preset threshold, judge whether the exposure state of the target object is normal or out of focus.

[0367] It should be noted that the judgment unit 231 in the embodiment of the present disclosure is specifically configured to execute steps S13-2-1 to S13-2-6 in the above-mentioned focal plane calibration method.

[0368] In some embodiments, the judgment unit 231 is specifically configured to intercept, based on the pixel position information of each target object in the second binary image and in accordance with a preset size, a third sub-image corresponding to each target object; determine the Hamming distance of each third sub-image based on each third sub-image and a preset solid color image; the size of the solid color image is the same as the preset size; and judge whether the exposure state of the target object is normal or out of focus based on the comparison result of the Hamming distance of each third sub-image and a third preset threshold.

[0369] It should be noted that the judgment unit 231 in the embodiment of the present disclosure is specifically configured to execute steps S13-3-1 to S13-3-3 in the above-mentioned focal plane calibration method.

[0370] In some embodiments, the prediction unit 232 is configured to determine, for target objects in each second object group arranged side by side along the second direction, pixel position information of the defocus boundary corresponding to each second object group based on the exposure status of each target object and the pixel position information of each target object; determine the regional defocus boundary corresponding to each test area based on the pixel position information of the defocus boundary corresponding to each second object group and the position information of the test area; for the regional defocus boundary corresponding to each test area, determine the exposure height of the regional defocus boundary based on the pixel position information of the regional defocus boundary and the exposure heights corresponding to each first object group stored in advance; for the regional defocus boundary corresponding to each test area, determine the predicted focal plane height of the test area based on the exposure height of the regional defocus boundary.

[0371] It should be noted that the prediction unit 232 in the embodiment of the present disclosure is specifically configured to execute steps S134 to S137 in the above-mentioned focal plane calibration method.

[0372] In some embodiments, the regional defocus boundary includes a first boundary and a second boundary; the prediction unit 232 determines the predicted focal plane height of the test area based on the exposure height of the regional defocus boundary, including: taking the average between the exposure height of the first boundary and the exposure height of the second boundary as the predicted focal plane height of the test area.

[0373] It should be noted that the prediction unit 232 in the embodiment of the present disclosure is specifically configured to perform the specific implementation process of step S137 in the above-mentioned focal plane calibration method.

[0374] In some embodiments, the calibration information of the best focal plane of the exposure device includes the height and tilt of the best focal plane; the test area includes two;

[0375] The calibration information determination module 24 is specifically configured to use the average of the predicted focal plane heights as the height of the best focal plane; and use the difference between the two predicted focal plane heights as the tilt of the best focal plane.

[0376] It should be noted that the calibration information determination module 24 in the embodiment of the present disclosure is configured to execute the specific implementation process in step S14 in the above-mentioned focal plane calibration method.

[0377] Fig.18 The following is a schematic diagram of the structure of a computer device provided in an embodiment of the present disclosure. Fig.18As shown, an embodiment of the present disclosure provides a computer device including: one or more processors 301, a memory 302, and one or more I / O interfaces 303. The memory 302 stores one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement any focal plane calibration method in the above embodiments; the one or more I / O interfaces 303 are connected between the processor and the memory, and are configured to implement information interaction between the processor and the memory.

[0378] Among them, the processor 301 is a device with data processing capabilities, including but not limited to a central processing unit (CPU), etc.; the memory 302 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH); the I / O interface (read-write interface) 303 is connected between the processor 301 and the memory 302, and can realize information interaction between the processor 301 and the memory 302, including but not limited to a data bus (Bus), etc.

[0379] In some embodiments, the processor 301 , the memory 302 , and the I / O interface 303 are connected to each other via a bus 304 , and further connected to other components of the computing device.

[0380] According to an embodiment of the present disclosure, a non-transitory computer readable storage medium is further provided. The non-transitory computer readable storage medium stores a computer program, wherein when the program is executed by a processor, the steps in any of the focal plane calibration methods in the above embodiments are implemented.

[0381] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a machine-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), the above-mentioned functions defined in the system of the present disclosure are executed.

[0382] It should be noted that the computer non-transient readable medium shown in the present disclosure may be a computer readable signal medium or a computer readable storage medium or any combination of the above two. The computer readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable non-transient storage medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable non-transient storage medium may be transmitted using any suitable medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0383] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the devices, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the aforementioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two connected boxes can actually represent that they are executed in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0384] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present disclosure, but the present disclosure is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and substance of the present disclosure, and these modifications and improvements are also considered to be within the scope of protection of the present disclosure.

Claims

1. A focal plane calibration method, wherein: include: Capturing an image of a preset template to obtain a target image; the target image includes a plurality of first object groups arranged side by side along a first direction, and a plurality of second object groups arranged side by side along a second direction; the first object group includes a plurality of target objects arranged side by side along the second direction; the second object group includes a plurality of target objects arranged side by side along the first direction; the preset template includes a plurality of test areas, and a target object in one of the test areas is obtained by one exposure of an exposure device, and target objects in different test areas are obtained by different exposures of the exposure device; Determining pixel position information corresponding to each target object according to the target image; Based on the pixel position information of each of the target objects, determining whether the exposure state of the target object is normal or out of focus, and determining the predicted focal plane heights corresponding to the different test areas according to the determination result and the pre-stored exposure heights corresponding to each of the first object groups; Based on each of the predicted focal plane heights, calibration information of an optimal focal plane of the exposure device is determined.

2. The focal plane calibration method according to claim 1, wherein: Determining pixel position information corresponding to each target object according to the target image includes: Extracting contour features in the target image to determine the center position of the target object; Based on the center position of the target object and a preset size range of the target object, pixel position information of the target object is determined.

3. The focal plane calibration method according to claim 1, wherein: Determining pixel position information corresponding to each target object according to the target image includes: Extracting contour features from the target image, performing binarization processing on the target image to obtain a first binary image, and determining the center position of the target object according to the extracted contour features; Based on the center position of the target object and a preset size range of the target object, pixel position information of the target object is determined.

4. The focal plane calibration method according to claim 1, wherein: Determining pixel position information corresponding to each target object according to the target image includes: Extracting contour features from the target image, and performing binarization processing on the target image to obtain a first binary image; Performing dilation processing on the first binary image to obtain a first intermediate image; performing corrosion processing on the first intermediate image to obtain a second intermediate image; Extracting contour features in the second intermediate image to determine the center position of the target object; Based on the center position of the target object and a preset size range of the target object, pixel position information of the target object is determined.

5. The focal plane calibration method according to claim 3 or 4, wherein: The step of performing binarization processing on the target image to obtain a first binary image includes: Extracting contour features from the target image, and performing binarization processing on the target image to obtain an initial binary image; Based on the pixel value and position information of each pixel point in the initial binary image, determining the area to be cropped; the pixel points in the area to be cropped are an area in the second direction that contains multiple segments of pixel values ​​that are continuously 0 and is surrounded by multiple segments of pixel values ​​that are continuously 255; The portion of the initial binary image located in the area to be cropped is cut off to obtain the first binary image.

6. The focal plane calibration method according to any one of claims 2 to 4, wherein: The size range of the target object includes a width range and a height range; The determining the pixel position information of the target object based on the center position of the target object and a preset size range of the target object includes: Taking the center position of the target object as the center, based on a random width within the width range and a random height within the height range, the target object is determined, and pixel position information of the target object is determined.

7. The focal plane calibration method according to claim 4, wherein: The determining, based on the pixel position information of each target object, whether the exposure state of the target object is normal or out of focus includes: Determining the area within the outline of each target object based on the pixel position information of each target object; Based on the comparison result between the areas within the contours of the target objects, a reference object is selected from the target objects, and pixel position information of the reference object is determined; Based on the pixel position information of each of the target objects and the pixel position information of the reference object, it is determined whether the exposure state of the target object is normal or out of focus.

8. The focal plane calibration method according to claim 7, wherein: The step of judging whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object and the pixel position information of the reference object includes: Based on the pixel position information of each of the target objects in the target image and the pixel position information of the reference object, using a preset similarity algorithm, determining the similarity between each of the target objects and the reference object; Based on the comparison result between the similarity between each of the target objects and the reference object and a first preset threshold, it is determined whether the exposure state of the target object is normal or out of focus.

9. The focal plane calibration method according to claim 7, wherein: The step of judging whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object and the pixel position information of the reference object includes: Based on the pixel position information of each of the target objects in the first binary image and the pixel position information of the reference object, using a preset similarity algorithm, determining the similarity between each of the target objects and the reference object; Based on the comparison result between the similarity between each of the target objects and the reference object and a first preset threshold, it is determined whether the exposure state of the target object is normal or out of focus.

10. The focal plane calibration method according to claim 7, wherein: The step of judging whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object and the pixel position information of the reference object includes: Based on the pixel position information of each of the target objects in the second intermediate image and the pixel position information of the reference object, using a preset similarity algorithm, determining the similarity between each of the target objects and the reference object; Based on the comparison result between the similarity between each of the target objects and the reference object and a first preset threshold, it is determined whether the exposure state of the target object is normal or out of focus.

11. The focal plane calibration method according to claim 4, wherein: The step of judging whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object includes: According to the pixel position information of the target object, intercepting the target sub-image corresponding to the target object from the target image, and determining the area within the outline of the target object in each of the target sub-images; Based on the comparison results between the areas within the contours of the target objects, a reference object is selected from the target objects, and a reference subgraph of the reference object is determined; Binarization is performed on each of the target sub-images and the reference sub-images to obtain each target binary sub-image and reference binary sub-image; Based on each of the target binary subgraphs and the reference binary subgraphs, using a preset similarity algorithm, determining the similarity between each of the target objects and the reference object; Based on the comparison result between the similarity between each of the target objects and the reference object and a first preset threshold, it is determined whether the exposure state of the target object is normal or out of focus.

12. The focal plane calibration method according to claim 4, wherein: The step of judging whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object includes: According to the pixel position information of the target object, intercepting the target sub-image corresponding to the target object from the target image, and determining the area within the outline of the target object in each of the target sub-images; Based on the comparison results between the areas within the contours of the target objects, a reference object is selected from the target objects, and a reference subgraph of the reference object is determined; According to a preset scaling factor, the target sub-image and the reference sub-image are reduced to obtain the reduced target sub-image and the reduced reference sub-image; Binarization is performed on each of the reduced target sub-images and the reduced reference sub-images to obtain each target binary sub-image and reference binary sub-image; Based on each of the target binary subgraphs and the reference binary subgraph, determining a Hamming distance of each of the target binary subgraphs; Based on the comparison result of the Hamming distance of each of the target binary sub-images and the second preset threshold, it is determined whether the exposure state of the target object is normal or out of focus.

13. The focal plane calibration method according to claim 4, wherein: The step of judging whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object includes: According to the pixel position information of the target object, intercepting a target sub-image corresponding to the target object from the target image; Binarization is performed on each of the target subgraphs to obtain each target binary subgraph; Determining the Hamming distance of each of the target binary sub-images based on each of the target binary sub-images and a preset pure color image; the size of the pure color image is the same as the preset size; Based on the comparison result of the Hamming distance of each of the target binary sub-images and the third preset threshold, it is determined whether the exposure state of the target object is normal or out of focus.

14. The focal plane calibration method according to any one of claims 7 to 13, wherein: Determining predicted focal plane heights corresponding to different test areas respectively according to the judgment result and the pre-stored exposure heights of the target objects in each column includes: For the target objects in each of the second object groups arranged side by side along the second direction, based on the exposure status of each of the target objects and the pixel position information of each of the target objects, determining the pixel position information of the defocus boundary corresponding to each of the second object groups; Determining regional defocus boundaries corresponding to each of the test areas based on pixel position information of the defocus boundaries corresponding to each of the second object groups and position information of the test area; For the regional defocus boundaries respectively corresponding to the test areas, based on the pixel position information of the regional defocus boundaries and the pre-stored exposure heights corresponding to the first object groups, determining the exposure heights of the regional defocus boundaries; For the regional defocus boundaries respectively corresponding to the test areas, the predicted focal plane height of the test area is determined based on the exposure height of the regional defocus boundaries.

15. The focal plane calibration method according to claim 14, wherein: The regional defocus boundary includes a first boundary and a second boundary; The step of determining the predicted focal plane height of the test area based on the exposure height of the defocus boundary of the area includes: The average of the exposure height of the first boundary and the exposure height of the second boundary is used as the predicted focal plane height of the test area.

16. The focal plane calibration method according to claim 1, wherein: The calibration information of the best focal plane of the exposure device includes the height and the tilt of the best focal plane; the test area includes two; The step of determining calibration information of an optimal focal plane of the exposure device based on each of the predicted focal plane heights comprises: Taking the average of the predicted focal plane heights as the height of the optimal focal plane; The difference between the two predicted focal plane heights is used as the tilt of the optimal focal plane.

17. A focal plane calibration device, wherein: It includes image acquisition module, image processing module, prediction module and calibration information determination module: The image acquisition module is configured to acquire an image of a preset template to obtain a target image; the target image includes a plurality of first object groups arranged side by side along a first direction, and a plurality of second object groups arranged side by side along a second direction; the first object group includes a plurality of target objects arranged side by side along the second direction; the second object group includes a plurality of target objects arranged side by side along the first direction; the preset template includes a plurality of test areas, and a target object in one of the test areas is obtained by one exposure, and target objects in different test areas are obtained by different exposures; The image processing module is configured to determine pixel position information corresponding to each target object according to the target image; The prediction module is configured to judge whether the exposure state of the target object is normal or out of focus based on the pixel position information of each target object, and determine the predicted focal plane heights corresponding to different test areas according to the judgment result and the pre-stored exposure heights corresponding to each of the first object groups; The calibration information determination module is configured to determine the calibration information of the optimal focal plane of the exposure device based on each of the predicted focal plane heights.

18. A computer device, wherein: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the focal plane calibration method according to any one of claims 1 to 16 are performed.

19. A computer non-transitory readable storage medium, wherein: The computer non-transitory readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the focal plane calibration method according to any one of claims 1 to 16 are executed.

Images