An image detection method

Abstract

The present application provides an image detection method, which includes: obtaining an alignment mark image of a semiconductor mask workpiece; extracting at least one local image from the alignment mark image; for each local image, determining a local image detection value of the local image; determining a global image detection value of the alignment mark image; and determining the image quality of the alignment mark image according to at least one local image detection value and the global image detection value. The present application evaluates the quality of the entire alignment mark image and also evaluates the image quality of the local alignment marks in the alignment mark image, achieving the effect of accurately judging the quality of the alignment mark image.

Description

Technical Field

[0001] This application relates to the field of image recognition technology, and more particularly, to an image detection method. Background Art

[0002] In the manufacturing process of semiconductor process equipment, it is necessary to etch the image on the mask workpiece onto the wafer workpiece. The off-axis alignment system can be used to determine whether the mask workpiece and the wafer workpiece are aligned by the alignment marks on the mask workpiece and the wafer workpiece. When collecting the alignment mark image, the alignment mark image is affected by many factors such as the illumination range, illumination wavelength, illumination intensity, and illumination angle of the light source, which may cause the loss, blurring, noise, or distortion of the mark texture information. At the same time, the alignment mark image itself is accompanied by a reduction in image quality and distortion during the processes of signal acquisition, compression, transmission, processing, and reconstruction. The above adverse factors have a serious impact on the detection, recognition, and positioning of the alignment marks in the alignment mark image by the off-axis alignment system.

[0003] Currently, image quality evaluation mainly includes subjective evaluation and objective evaluation. Subjective evaluation requires human participation, scoring and statistics for all collected images, which is time-consuming and laborious. At the same time, due to the sensory differences of each person in the manual operation, the stability of subjective evaluation is insufficient.

[0004] Objective evaluation is divided into two methods: with reference and without reference. Since most images do not have a fixed reference scene, the method without reference is generally used. Traditional image quality evaluation mainly focuses on the evaluation of the overall image quality of the image, and does not judge the quality of the local area of interest in the image. It often occurs that although the overall image quality of the alignment mark image meets the standard, due to the insufficient quality of the image of the alignment mark part, the off-axis alignment system still cannot determine whether the semiconductor mask workpiece is aligned. Summary of the Invention

[0005] In view of this, the purpose of this application is to provide an image detection method, which can evaluate the quality of the entire alignment mark image and the local image quality of the alignment marks in the alignment mark image, solve the problem in the prior art that the alignment mark image meets the detection standard but the image quality of the alignment mark part is not high, and achieve the effect of accurately judging the quality of the alignment mark image.

[0006] First aspect, an embodiment of the present application provides an image detection method, and the method includes: obtaining an alignment mark image of a semiconductor mask workpiece, where the alignment mark image includes at least one alignment mark, and the at least one alignment mark is a mark preset on the semiconductor mask workpiece for aligning the semiconductor mask workpiece; extracting at least one local image from the alignment mark image, where each local image contains a corresponding alignment mark; for each local image, determining a local image detection value of the local image; determining a global image detection value of the alignment mark image; and determining the image quality of the alignment mark image according to at least one local image detection value and the global image detection value.

[0007] Optionally, the step of extracting at least one local image from the alignment mark image includes: determining the image position of each alignment mark in the alignment mark image; for each alignment mark, generating a detection box corresponding to the alignment mark, and the detection box contains the alignment mark; and for each alignment mark, determining the image in the detection box corresponding to the alignment mark as a local image.

[0008] Optionally, the local image detection value of each local image is determined in the following manner: determining the edge lines and edge points of the alignment mark in the local image; determining an alignment mark region corresponding to the alignment mark in the local image according to the edge lines and the edge points; determining the region outside the alignment mark region in the local image as a blank region; and determining the local image detection value of the local image according to the gray value of the alignment mark region and the gray value of the blank region.

[0009] Optionally, each local image detection value includes a local image contrast value, and the local image contrast value of each local image is determined through the following steps: determining a first average gray value of the alignment mark region; determining a second average gray value of the blank region; and determining the ratio of the first average gray value of the alignment mark region in the local image to the second average gray value of the blank region in the local image as the local image contrast value of the local image.

[0010] Optionally, each local image detection value further includes a local image sharpness value, and the local image sharpness value of each local image is determined through the following steps: determining the local image sharpness value of the local image according to the gray value of each pixel point in the blank region of the local image and the gray value of a reference pixel point adjacent to each pixel point.

[0011] Optionally, the first average gray value of the alignment mark region is determined by the following formula:

[0012]

[0013] Among them, GrayMeanValue is the first average gray value of the image of the alignment mark area of the local image, f(x, y) is the gray value of each pixel point of the image of the alignment mark area of the local image, and Area mean is the area of the alignment mark area of the local image.

[0014] Optionally, the second average gray value of the blank area is determined by the following formula:

[0015]

[0016] Among them, GrayValue outer is the second average gray value of the blank area of the local image, f(x, y) outer is the gray value of each pixel point of the blank area of the local image, and Area outer is the area of the blank area of the local image.

[0017] Optionally, the reference pixel points adjacent to each pixel point in the blank area include a first reference pixel point and a second reference pixel point. The first reference pixel point is the pixel point with the same abscissa as the corresponding pixel point and adjacent ordinate; the second reference pixel point is the pixel point with the same ordinate as the corresponding pixel point and adjacent abscissa;

[0018] Among them, the local image sharpness value of each local image is determined by the following formula:

[0019]

[0020] Among them, DR is the local image sharpness value of the local image, f(x i , y i ) outer is the gray value of each target pixel point, f(x i + 1, y i ) outer is the gray value of the first reference pixel point adjacent to each target pixel point, f(x i , y i + 1) outer is the gray value of the second reference pixel point adjacent to each target pixel point, and m is the number of pixel points in the blank area of the local image.

[0021] Optionally, the step of determining the global image detection value of the alignment mark image includes: determining the area of the effective gray image in the alignment mark image that is greater than the preset gray level; determining the reference gray value of the alignment mark image according to the ratio of the area of the effective gray image to the area of the global image; determining the global image detection value according to the area of the effective gray image and the reference gray value.

[0022] Optionally, the global image detection value includes a first global image ratio. Wherein, the first global image ratio of the alignment mark image is determined through the following steps: Select a preset number of target rectangular regions with a preset size at different positions in the alignment mark image, and the alignment mark region is not included in the target rectangular region; For each target rectangular region, determine the deviation from the reference gray value of the target rectangular region according to the gray value of each pixel point in the target rectangular region and the reference gray value; For each target rectangular region, determine the average deviation from the reference gray value of the target rectangular region according to the ratio of the deviation from the reference gray value to the area of the target rectangular region; Determine the average deviation of the alignment mark image according to the reference gray value, the deviation from the reference gray value, the area of the pixels at each gray level, and the area of the global image; Determine the first global image ratio of the global image according to the ratio of the average deviation from the reference gray value to the average deviation.

[0023] Optionally, the global image detection value further includes a second global image ratio. Wherein, the second global image ratio of the alignment mark image is determined through the following steps: Determine the second global image ratio of the alignment mark image according to the ratio of the area of the effective gray image in the alignment mark image to the area of the alignment mark image.

[0024] Optionally, the area of the effective gray image in the alignment mark image is determined by the following formula:

[0025]

[0026] Wherein, SumArea2 is the area of the effective gray image, Area i is the area of the i-th pixel point, and n is the number of pixel points in the global image that are greater than the preset gray level.

[0027] Optionally, the deviation from the reference gray value of the target rectangular region is determined by the following formula:

[0028]

[0029] Wherein, SumGrayOffset is the deviation from the reference gray value, y iGrayValue_i is the gray value of the i-th pixel point, GrayBaseValue is the reference gray value, w is the number of target rectangular regions, a is the first side length of each target rectangular region, and b is the second side length of each target rectangular region.

[0030] Optionally, the average deviation of the alignment mark image is determined by the following formula:

[0031]

[0032] Where Sig is the average deviation, GrayBaseValue is the reference gray value, L is the deviation from the average reference gray value, Hist(j) is the area of all pixel points with gray level j in the global image, w is the number of target rectangular regions, a is the first side length of each target rectangular region, and b is the second side length of each target rectangular region.

[0033] Optionally, the first global image ratio of the global image is determined by the following formula:

[0034]

[0035] Where LR is the first global image ratio, L is the deviation from the average reference gray value, and sig is the average deviation.

[0036] Optionally, the second global image ratio of the alignment mark image is determined by the following formula:

[0037]

[0038] Where AR is the second image ratio, SumArea2 is the area of the effective gray image, and SumArea1 is the area of the alignment mark image.

[0039] Optionally, the local image detection values include a contrast value and a sharpness value, and the global image detection values include a first global image ratio and a second global image ratio. Among them, the steps of determining the image quality of the global image according to the local image detection values and the global image detection values include: calculating the average value of multiple local image detection values to obtain a local image detection average value, where the local image detection average value includes: a contrast average value and a sharpness average value; determining whether the contrast average value is greater than a standard contrast value, whether the sharpness average value is greater than a standard sharpness value, whether the first global image ratio is greater than a first global image standard value, and whether the second global image ratio is greater than a second global image standard value; if the contrast average value is greater than the standard contrast value, the sharpness average value is greater than the standard sharpness value, the first global image ratio is greater than the first global image standard value, and the second global image ratio is greater than the second global image standard value, then determine that the global image is a high-quality image; if the contrast average value is not greater than the standard contrast value and / or the sharpness average value is not greater than the standard sharpness value and / or the first global image ratio is not greater than the first global image standard value and / or the second global image ratio is not greater than the second global image standard value, then determine that the global image is a low-quality image.

[0040] In a second aspect, an embodiment of the present application further provides an image detection device, and the device includes:

[0041] An image acquisition module, configured to acquire an alignment mark image of a semiconductor mask workpiece, where the alignment mark image includes at least one alignment mark, and the at least one alignment mark is a mark pre-set on the semiconductor mask workpiece for aligning the semiconductor mask workpiece;

[0042] A local image extraction module, configured to extract at least one local image from the alignment mark image, where each local image contains a corresponding alignment mark;

[0043] A local image detection value calculation module, configured to determine a local image detection value of each local image;

[0044] A global image detection value calculation module, configured to determine a global image detection value of the alignment mark image;

[0045] An image quality determination module, configured to determine the image quality of the alignment mark image according to at least one local image detection value and the global image detection value.

[0046] In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the memory through the bus. When the machine-readable instructions are executed by the processor, the steps of the image detection method as described above are executed.

[0047] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is run by a processor, the steps of the image detection method as described above are executed.

[0048] The image detection method provided by the embodiment of the present application can solve the problem in the prior art that although the alignment mark image meets the detection standard, the image quality of the alignment mark part is not high by a method that performs quality evaluation on the global part of the alignment mark image and also performs image quality evaluation on the local part of the alignment mark in the alignment mark image, achieving the effect of accurately judging the image quality of the alignment mark image.

[0049] To make the above objects, features, and advantages of the present application more obvious and understandable, the following specifically enumerates preferred embodiments and, in conjunction with the accompanying drawings, makes a detailed description as follows. Description of the Drawings

[0050] To more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application and should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other relevant drawings can be obtained based on these drawings.

[0051] Figure 1 It is a flowchart of an image detection method provided by an embodiment of the present application;

[0052] Figure 2 It is a schematic structural diagram of an off-axis alignment system provided by an embodiment of the present application;

[0053] Figure 3 It is a schematic diagram of a semiconductor mask workpiece provided by an embodiment of the present application;

[0054] Figure 4 It is a schematic diagram of a local image provided by an embodiment of the present application. Detailed Embodiments

[0055] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the accompanying drawings in the embodiments of this application. Apparently, the described embodiments are only some, but not all, of the embodiments of this application. Components of the embodiments of this application described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. Based on the embodiments of this application, every other embodiment obtained by those skilled in the art without creative efforts belongs to the scope of protection of this application.

[0056] First, an application scenario applicable to this application is introduced. This application can be applied to image detection.

[0057] After research, it is found that traditional image quality evaluation mainly focuses on the evaluation of the overall image area quality and does not judge the quality of the local area of interest in the image. It often occurs that although the overall image quality of the alignment mark image meets the standard, due to the insufficient quality of the image of the alignment mark part, the off-axis alignment system still cannot determine whether the semiconductor mask workpiece is aligned.

[0058] Based on this, the embodiments of this application provide an image detection method to detect the quality of the alignment mark image.

[0059] Please refer to Figure 1 , Figure 1 which is a flowchart of an image detection method provided by the embodiments of this application. As shown in Figure 1 , the image detection method provided by the embodiments of this application includes:

[0060] S101. Obtain an alignment mark image of a semiconductor mask workpiece.

[0061] Among them, the alignment mark image includes at least one alignment mark, and the at least one alignment mark is a mark preset on the semiconductor mask workpiece for aligning the semiconductor mask workpiece.

[0062] It should be noted that the alignment mark image can be captured by an off-axis alignment system. As shown in Figure 2 , the off-axis alignment system 200 includes: a semiconductor mask workpiece 201 to be detected, a first imaging lens 202, a reflection prism 203, a second imaging lens 204, a first beam splitter prism 205, a light source 206, a third imaging lens 207, a second beam splitter prism 208, a fourth imaging lens 209, and a camera 210.

[0063] Here, the light source 206 illuminates the semiconductor mask workpiece. The light rays of the light source 206 are reflected by the first beam splitter prism 205, the second imaging lens 204, and the reflecting prism 203, and then provide illumination for the semiconductor mask workpiece through the first imaging lens 202. The camera 210 acquires the alignment mark image of the semiconductor mask workpiece by detecting the first imaging lens 202, the reflecting prism 203, the second imaging lens 204, the first beam splitter prism 205, the third imaging lens 207, the second beam splitter prism 208, and the fourth imaging lens 209.

[0064] Among them, the camera can use a CCD (Charge-coupled Device) camera or a CMOS (Complementary Metal Oxide Semiconductor) camera.

[0065] Optionally, after obtaining the alignment mark image, downsampling and filtering processing can be performed on the alignment mark image. The downsampling process is to scale the alignment mark pattern to the size of the standard alignment mark image. The filtering process can optimize the brightness value of the alignment mark image, reducing the brightness of particularly bright parts and increasing the brightness of particularly dark parts of the alignment mark image. In this way, an alignment mark image processed by image processing can be obtained.

[0066] Exemplarily, as Figure 3 shown, at least one alignment mark 303 is provided on the semiconductor mask workpiece for aligning the semiconductor mask workpiece.

[0067] As Figure 3 shown, in the alignment mark image, the areas 301 and 302 outside the reticle may also be photographed. The reticle area needs to be deleted from the alignment marks to avoid the influence of the area outside the reticle on the calculation of the alignment mark image.

[0068] S102. Extract at least one local image from the alignment mark image.

[0069] Among them, each local image contains a corresponding alignment mark.

[0070] Among them, the step of extracting at least one local image from the alignment mark image includes: determining the image position of each alignment mark in the alignment mark image; for each alignment mark, generating a detection box corresponding to the alignment mark, and including the alignment mark in the detection box; for each alignment mark, determining the image in the detection box corresponding to the alignment mark as a local image.

[0071] Here, local images can be extracted from the alignment mark images through image segmentation technology. Before extracting the local images, morphological dilation needs to be performed on the detection frames to ensure that the local images contain complete alignment marks.

[0072] S103. For each local image, determine the local image detection value of this local image.

[0073] Here, the local image detection value of each local image can be determined in the following way: Determine the edge lines and edge points of the alignment marks in this local image; According to the edge lines and the edge points, determine the alignment mark region corresponding to the alignment marks in this local image; Determine the region outside the alignment mark region in this local image as the blank region; According to the gray value of the alignment mark region and the gray value of the blank region, determine the local image detection value of this local image.

[0074] Specifically, as Figure 4 shown, the alignment mark region 303 in the detection frame 401 is the alignment mark region, and the blank region between the detection frame 401 and the alignment mark 303 is the above-mentioned blank region.

[0075] Among them, the local image detection value includes: the local image contrast value and the local image sharpness value.

[0076] Among them, the local image contrast value of each local image is determined through the following steps: Determine the first average gray value of the alignment mark region; Determine the second average gray value of the blank region; Determine the ratio of the first average gray value of the alignment mark region in this local image to the second average gray value of the blank region in this local image as the local image contrast value of this local image.

[0077] The local image sharpness value of each local image can be determined through the following steps: According to the gray value of each pixel point in the blank region of this local image and the gray value of the reference pixel points adjacent to each pixel point, determine the local image sharpness value of this local image.

[0078] Specifically, the first average gray value of the alignment mark region is determined through the following formula:

[0079]

[0080] Among them, GrayMeanValue is the first average gray value of the image of the alignment mark region of this local image, f(x, y) is the gray value of each pixel point of the image of the alignment mark region of this local image, and Area mean is the area of the alignment mark region of this local image.

[0081] Specifically, the second average gray value of the blank area is determined by the following formula:

[0082]

[0083] where GrayValue outer is the second average gray value of the blank area of the local image, and f(x, y) outer is the gray value of each pixel point in the blank area of the local image, and Area outer is the area of the blank area of the local image.

[0084] Among them, the reference pixel points adjacent to each pixel point in the blank area include a first reference pixel point and a second reference pixel point. The first reference pixel point is the pixel point with the same abscissa and adjacent ordinate as the corresponding pixel point; the second reference pixel point is the pixel point with the same ordinate and adjacent abscissa as the corresponding pixel point.

[0085] The local image sharpness value of each local image can be determined by the following formula:

[0086]

[0087] where DR is the local image sharpness value of the local image, and f(x i , y i ) outer is the gray value of each target pixel point, and f(x i + 1, y i ) outer is the gray value of the first reference pixel point adjacent to each target pixel point, and f(x i , y i + 1) outer is the gray value of the second reference pixel point adjacent to each target pixel point, and m is the number of pixel points in the blank area of the local image.

[0088] S104. Determine the global image detection value of the alignment mark image.

[0089] Specifically, the steps for determining the global image detection value of the alignment mark image include: determining the area of the effective gray image in the alignment mark image that is greater than the preset gray level; determining the reference gray value of the alignment mark image according to the ratio of the area of the effective gray image to the area of the global image; and determining the global image detection value according to the area of the effective gray image and the reference gray value.

[0090] Among them, the global image detection value includes a first global image ratio and a second global image ratio.

[0091] Here, the gray level of each pixel is divided into 255 gray levels from 1 to 255. Exemplarily, the preset gray level can be from 200 to 255 gray levels. In this way, all pixel points with gray levels from 200 to 255 in the alignment mark image can be determined, and the image composed of the pixel points with gray levels from 200 to 255 is determined as the effective gray image.

[0092] Among them, the first global image ratio of the alignment mark image is determined through the following steps: Select a preset number of target rectangular regions with a preset size at different positions in the alignment mark image, and the alignment mark region is not included in the target rectangular region; For each target rectangular region, determine the deviation from the reference gray value of the target rectangular region according to the gray value of each pixel point in the target rectangular region and the reference gray value; For each target rectangular region, determine the average deviation from the reference gray value of the target rectangular region according to the ratio of the deviation from the reference gray value to the area of the target rectangular region; Determine the average deviation of the alignment mark image according to the reference gray value, the deviation from the reference gray value, the area of the pixels of each gray level, and the area of the global image; Determine the first image ratio of the global image according to the ratio of the average deviation from the reference gray value to the average deviation.

[0093] Exemplarily, the alignment mark image can be divided into 9 target rectangular acquisition regions, and the target rectangular regions are acquired in the 9 target rectangular acquisition regions respectively to obtain 9 target rectangular regions.

[0094] In this way, the target rectangular regions can be acquired in the alignment mark image on average, making the data of the first image ratio more accurate and reliable.

[0095] Among them, the area of the effective gray image in the alignment mark image can be determined through the following formula:

[0096]

[0097] Among them, SumArea2 is the area of the effective gray image, Area i is the area of the i-th pixel point, and n is the number of pixel points in the global image that are greater than the preset gray level.

[0098] The deviation from the reference gray value of the target rectangular region can be determined through the following formula:

[0099]

[0100] Among them, SumGrayOffset is the deviation from the reference gray value, y iGrayValue_i is the grayscale value of the i-th pixel, GrayBaseValue is the reference grayscale value, w is the number of target rectangular regions, a is the first side length of each target rectangular region, and b is the second side length of each target rectangular region.

[0101] The average deviation of the alignment mark image can be determined by the following formula:

[0102]

[0103] Where Sig is the average deviation, GrayBaseValue is the reference grayscale value, L is the deviation from the average reference grayscale, Hist(j) is the area of all pixels with grayscale level j in the global image, w is the number of target rectangular regions, a is the first side length of each target rectangular region, and b is the second side length of each target rectangular region.

[0104] The first image ratio of the global image can be determined by the following formula:

[0105]

[0106] Where LR is the first image ratio, L is the deviation from the average reference grayscale, and sig is the average deviation.

[0107] Among them, the second global image ratio of the alignment mark image can be determined by the following steps: Determine the second global image ratio of the alignment mark image according to the ratio of the area of the effective grayscale image in the alignment mark image to the area of the alignment mark image.

[0108] Among them, the second global image ratio of the alignment mark image is determined by the following formula:

[0109]

[0110] Where AR is the second global image ratio, SumArea2 is the area of the effective grayscale image, and SumArea1 is the area of the alignment mark image.

[0111] S105. Determine the image quality of the alignment mark image according to at least one local image detection value and the global image detection value.

[0112] Among them, the steps of determining the image quality of the global image according to the local image detection value and the global image detection value include:

[0113] Calculate the average value of multiple local image detection values to obtain the local image detection average value, where the local image detection average value includes: the average contrast value and the average sharpness value;

[0114] Determine whether the average contrast value is greater than the standard contrast value, whether the average sharpness value is greater than the standard sharpness value, whether the first global image ratio is greater than the first global image standard value, and whether the second global image ratio is greater than the second global image standard value;

[0115] If the average contrast value is greater than the standard contrast value, the average sharpness value is greater than the standard sharpness value, the first global image ratio is greater than the first global image standard value, and the second global image ratio is greater than the second global image standard value, determine that the global image is a high-quality image;

[0116] If the average contrast value is not greater than the standard contrast value and / or the average sharpness value is not greater than the standard sharpness value and / or the first global image ratio is not greater than the first global image standard value and / or the second global image ratio is not greater than the second global image standard value, determine that the global image is a low-quality image.

[0117] Optionally, if the alignment mark image is a low-quality image, the low-quality image can be deleted or the alignment mark image can be reacquired, and the detection is performed again.

[0118] The image detection method provided by the embodiments of the present application can solve the problem in the prior art that the alignment mark image meets the detection standard, but the image quality of the alignment mark part is not high by a method of performing quality evaluation on the whole of the alignment mark image and also performing image quality evaluation on the local alignment marks of the alignment mark image, achieving the effect of accurately judging the quality of the alignment mark image.

[0119] Based on the same inventive concept, an image detection device corresponding to the image detection method is further provided in the embodiments of the present application. Since the principle of solving problems by the device in the embodiments of the present application is similar to the above image detection method in the embodiments of the present application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0120] Specifically, the image detection device includes: an image acquisition module, configured to acquire an alignment mark image of a semiconductor mask workpiece, where the alignment mark image includes at least one alignment mark, and the at least one alignment mark is a mark preset on the semiconductor mask workpiece for aligning the semiconductor mask workpiece;

[0121] A local image extraction module, configured to extract at least one local image from the alignment mark image, where each local image contains a corresponding alignment mark;

[0122] A local image detection value calculation module, configured to determine a local image detection value of each local image for each local image;

[0123] A global image detection value calculation module for determining the global image detection value of the alignment mark image;

[0124] An image quality determination module for determining the image quality of the alignment mark image according to at least one local image detection value and the global image detection value.

[0125] The image detection device provided by the embodiments of the present application can solve the problem in the prior art that although the alignment mark image meets the detection standard, the image quality of the alignment mark part is not high, and achieve the effect of accurately judging the image quality of the alignment mark image by evaluating the quality of the global alignment mark image and the local image quality of the alignment mark of the alignment mark image.

[0126] An electronic device provided by the embodiments of the present application. The electronic device includes a processor, a memory, and a bus.

[0127] The memory stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the memory through the bus. When the machine-readable instructions are executed by the processor, the steps of the image detection method in the method embodiment as described above can be executed. The specific implementation manner can be referred to the method embodiment and will not be elaborated here. Figure 1 As shown in the method embodiment, the steps of the image detection method can be executed. The specific implementation manner can be referred to the method embodiment and will not be elaborated here.

[0128] The embodiments of the present application also provide a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is run by a processor, the steps of the image detection method in the method embodiment as described above can be executed. The specific implementation manner can be referred to the method embodiment and will not be elaborated here. Figure 1 As shown in the method embodiment, the steps of the image detection method can be executed. The specific implementation manner can be referred to the method embodiment and will not be elaborated here.

[0129] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system, device, and unit can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated here.

[0130] In several embodiments provided by the present application, it should be understood that the disclosed system, device, and method can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For another example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the displayed or discussed mutual coupling or direct coupling or communication connection can be through some communication interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical, mechanical, or other form.

[0131] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0132] In addition, in each embodiment of the present application, each functional unit may be integrated in a processing unit, may exist physically alone for each unit, or two or more units may be integrated in one unit.

[0133] If the described function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0134] Finally, it should be noted that: the above-described embodiments are only specific implementation manners of the present application, used to illustrate the technical solutions of the present application, rather than limiting it. The protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: any person skilled in the art within the technical scope disclosed by the present application can still modify the technical solutions recorded in the foregoing embodiments or can easily think of changes, or perform equivalent replacements on some of the technical features; and these modifications, changes, or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An image detection method, characterized in that, The method includes: Obtaining an alignment mark image of a semiconductor mask workpiece, where the alignment mark image includes at least one alignment mark, and the at least one alignment mark is a mark preset on the semiconductor mask workpiece for aligning the semiconductor mask workpiece; Extracting at least one local image from the alignment mark image, where each local image contains a corresponding one of the alignment marks; For each local image, determining a local image detection value of the local image; Determining a global image detection value of the alignment mark image; Determining the image quality of the alignment mark image according to at least one local image detection value and the global image detection value; Determining the local image detection value of each local image in the following manner: determining the edge lines and edge points of the alignment mark in the local image; determining the alignment mark region corresponding to the alignment mark in the local image according to the edge lines and the edge points; determining the region outside the alignment mark region in the local image as the blank region; determining the local image detection value of the local image according to the gray value of the alignment mark region and the gray value of the blank region; Each local image detection value includes a local image contrast value, where the local image contrast value of each local image is determined by the following steps: determining a first average gray value of the alignment mark region; determining a second average gray value of the blank region; determining the ratio of the first average gray value of the alignment mark region in the local image to the second average gray value of the blank region in the local image as the local image contrast value of the local image; Determining the first average gray value of the alignment mark region by the following formula: Among them, GrayMeanValue is the first average gray value of the image of the alignment mark area of the local image, f(x, y) is the gray value of each pixel point of the image of the alignment mark area of the local image, and Area mean is the area of the alignment mark area of the local image; Determining the second average gray value of the blank region by the following formula: Among them, GrayValue outer is the second average gray value of the blank area of the local image, and f(x, y) outer is the gray value of each pixel point in the blank area of the local image, and Area outer is the area of the blank area of the local image; The reference pixel points adjacent to each pixel point in the blank region include a first reference pixel point and a second reference pixel point. The first reference pixel point is the pixel point with the same abscissa and adjacent ordinate as the corresponding pixel point; the second reference pixel point is the pixel point with the same ordinate and adjacent abscissa as the corresponding pixel point. Wherein, the local image sharpness value of each local image is determined by the following formula: where DR is the local image sharpness value of the local image, f(x i ,y i ) outer is the grayscale value of each target pixel, f(x i +1,y i ) outer is the grayscale value of the first reference pixel adjacent to each target pixel, f(x i ,y i +1) outer is the grayscale value of the second reference pixel adjacent to each target pixel, and m is the number of pixels in the blank area of the local image.

2. The method according to claim 1, characterized in that, The step of extracting at least one local image from the alignment mark image includes: Determining the image position of each alignment mark in the alignment mark image; For each alignment mark, generating a detection frame corresponding to the alignment mark, and the detection frame contains the alignment mark; For each alignment mark, determining the image in the detection frame corresponding to the alignment mark as a local image.

3. The method according to claim 1, wherein Each local image detection value further includes a local image sharpness value, where the local image sharpness value of each local image is determined by the following steps: Determining the local image sharpness value of the local image according to the gray value of each pixel point in the blank region of the local image and the gray value of the reference pixel points adjacent to each pixel point.

4. The method according to claim 1, characterized in that, The step of determining the global image detection value of the alignment mark image includes: Determining the area of the effective gray image with a gray level greater than a preset gray level in the alignment mark image; Determine the reference gray value of the alignment mark image according to the ratio of the area of the effective gray image to the area of the global image; Determine the global image detection value according to the area of the effective gray image and the reference gray value.

5. The method according to claim 4, wherein The global image detection value includes a first global image ratio, wherein, the first global image ratio of the alignment mark image is determined by the following steps: Select a preset number of target rectangular regions with a preset size at different positions in the alignment mark image, and the alignment mark region is not included in the target rectangular region; For each target rectangular region, determine the deviation reference gray value of the target rectangular region according to the gray value of each pixel point in the target rectangular region and the reference gray value; For each target rectangular region, determine the average deviation reference gray value of the target rectangular region according to the ratio of the deviation reference gray value to the area of the target rectangular region; Determine the average deviation of the alignment mark image according to the reference gray value, the deviation reference gray value, the area of the pixels of each gray level, and the area of the global image; Determine the first global image ratio of the global image according to the ratio of the average deviation reference gray value to the average deviation.

6. The method according to claim 4, wherein The global image detection value further includes a second global image ratio, wherein, the second global image ratio of the alignment mark image is determined by the following steps: Determine the second global image ratio of the alignment mark image according to the ratio of the area of the effective gray image in the alignment mark image to the area of the alignment mark image.

7. The method according to claim 4, wherein Determine the area of the effective gray image in the alignment mark image through the following formula: Among them, SumArea2 is the area of the effective grayscale image, Area i is the area of the i-th pixel point, and n is the number of pixel points greater than the preset grayscale level in the global image.

8. The method according to claim 5, wherein Determine the deviation reference gray value of the target rectangular region through the following formula: Among them, SumGrayOffset is the offset from the reference gray value, y i is the gray value of the i-th pixel, GrayBaseValue is the reference gray value, w is the number of target rectangular regions, a is the first side length of each target rectangular region, and b is the second side length of each target rectangular region.

9. The method according to claim 5, wherein Determine the average deviation of the alignment mark image through the following formula: wherein, Sig is the average deviation, GrayBaseValue is the reference gray value, L is the average deviation reference gray value, Hist(j) is the area of all pixel points with gray level j in the global image, w is the number of target rectangular regions, a is the first side length of each target rectangular region, and b is the second side length of each target rectangular region.

10. The method according to claim 5, characterized in that Determine the first global image ratio of the global image through the following formula: wherein, LR is the first global image ratio, L is the average deviation reference gray value, and sig is the average deviation.

11. The method according to claim 6, wherein Determine the second global image ratio of the alignment mark image through the following formula: wherein, AR is the second global image ratio, SumArea2 is the area of the effective gray image, and SumArea1 is the area of the alignment mark image.

12. The method according to claim 1, characterized in that, The local image detection value includes a contrast value and a sharpness value, and the global image detection value includes a first global image ratio and a second global image ratio. wherein, the steps of determining the image quality of the global image according to the local image detection value and the global image detection value include: Calculate the average value of multiple local image detection values to obtain the local image detection average value, wherein the local image detection average value includes: the contrast average value and the sharpness average value; Determine whether the average contrast value is greater than the standard contrast value, whether the average sharpness value is greater than the standard sharpness value, whether the first global image ratio is greater than the first global image standard value, and whether the second global image ratio is greater than the second global image standard value; If the average contrast value is greater than the standard contrast value, the average sharpness value is greater than the standard sharpness value, the first global image ratio is greater than the first global image standard value, and the second global image ratio is greater than the second global image standard value, then determine that the global image is a high-quality image; If the average contrast value is not greater than the standard contrast value and / or the average sharpness value is not greater than the standard sharpness value and / or the first global image ratio is not greater than the first global image standard value and / or the second global image ratio is not greater than the second global image standard value, then determine that the global image is a low-quality image.

Images