Detection method for shielded mask mark and correction method for mask position

Abstract

The present disclosure provides a detection method for a shielded mask mark, comprising: S1, converting an acquired image of a mask mark into a gray-scale image, wherein the acquired image is at least partially shielded; S2, performing image segmentation on the gray-scale image, and filling the non-mark part with white; S3, obtaining the gray-scale median of the gray-scale image, taking the gray-scale median as an initial binarization threshold to perform binarization processing on the gray-scale image, performing contour detection, and obtaining multiple minimum bounding rectangles of an image contour; S4, determining a fuzzy inference rule on the basis of length-width characteristics of the minimum bounding rectangles of the mask mark, and obtaining the probability that each minimum bounding rectangle is an actual bounding rectangle of the shielded mark; S5, judging whether there is a probability among the probabilities that is greater than a preset threshold, if not, adaptively adjusting the binarization threshold and repeating S3-S5, and if yes, determining the minimum bounding rectangle with the maximum probability as a first detected object; and S6, determining first center coordinates on the basis of the first detected object to complete positioning detection of the mask mark.

Description

Method for detecting mask marks with occlusion and method for correcting mask position

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This disclosure claims priority to Chinese patent application number 202311455987.1 filed with the State Intellectual Property Office of China on November 2, 2023, entitled “Method for detecting mask marks with occlusion and method for correcting mask position,” the entire contents of which are incorporated by reference into this disclosure. Technical Field

[0003] The present disclosure relates to the field of optical detection technology, and in particular to a method for detecting mask marks with occlusion and a method for correcting the position of a mask. Background Art

[0004] During mask transport, position correction is required at the end. A CCD camera uses a transparent mask suction cup to capture alignment marks on the mask for positional deviation detection. The higher the accuracy of the detection results, the better the correction effect. However, due to the suction cup's air passages or other structures obscuring the marks, the CCD camera may capture incomplete marks, partially obscured marks, and uneven grayscale distribution in the captured image, all of which affect the positioning accuracy of the detected marks.

[0005] The binarization method using an appropriate threshold can reduce the interference information of the image. That is, under the premise of retaining the integrity of the mark of the acquired image, the binarization processing of the acquired image can improve the positioning accuracy of the detected mark. However, the appropriate binarization threshold is difficult to determine. In the same image, after binarization processing with different thresholds, the mark detection results will show obvious differences, which will affect the mark positioning accuracy.

[0006] Summary of the Invention

[0007] (1) Technical issues to be resolved

[0008] In response to the above problems, the present disclosure provides a method for detecting mask marks with occlusion and a method for correcting the mask position, which are used to solve technical problems such as the difficulty in obtaining an appropriate binarization threshold in traditional methods and improving the positioning accuracy of detection marks.

[0009] (2) Technical solution

[0010] On the one hand, the present disclosure provides a method for detecting mask marks with occlusion, including: S1, using a first camera to capture an image of the mask mark and converting the captured image into a grayscale image; wherein the captured image is at least partially occluded; S2, performing image segmentation on the grayscale image and filling the non-marked part with white; S3, obtaining the grayscale median of the grayscale image obtained in S2, and binarizing the grayscale image using the grayscale median as an initial binarization threshold; performing contour detection and obtaining multiple minimum bounding rectangles of the image contour; S4, determining a fuzzy inference rule based on the length and width features of the minimum bounding rectangle of the mask mark, and obtaining the probability that each minimum bounding rectangle obtained in S3 is the actual bounding rectangle of the occluded mark; S5, judging whether there is a probability greater than a preset threshold among the probabilities obtained in S4; if not, adaptively adjusting the binarization threshold and repeating S3 to S5; if so, determining the minimum bounding rectangle with the largest probability as the first detection target; S6, determining the first center coordinates based on the first detection target to complete the positioning detection of the mask mark.

[0011] According to an embodiment of the present disclosure, S1 further includes: filtering the grayscale image to suppress noise.

[0012] According to an embodiment of the present disclosure, image segmentation in S2 includes: selecting the overlapping point of the mask mark and the occluded part as the segmentation point based on the graphic features of the mask mark; and performing image segmentation on the grayscale image using the segmentation point as an end point or edge point of the segmentation rectangle.

[0013] According to an embodiment of the present disclosure, the method for performing contour detection in S3 includes any one of a connectivity-based contour detection algorithm, an edge detection-based contour detection algorithm, and a segmentation-based contour detection algorithm; the method for obtaining multiple minimum enclosing rectangles of the image contour includes using the BoundingRect function to calculate multiple minimum enclosing rectangles.

[0014] According to an embodiment of the present disclosure, S4 includes: S41, estimating the length and width of the bounding rectangle in the captured image; S42, dividing the pixel interval of the length into multiple segments according to the length, and determining the length membership function; dividing the pixel interval of the width into multiple segments according to the width, and determining the width membership function; S43, obtaining a fuzzy reasoning result based on the length membership function and the width membership function; S44, defuzzifying the fuzzy reasoning result using the maximum membership method to obtain the probability that each minimum bounding rectangle is the actual bounding rectangle of the occluded mark.

[0015] According to an embodiment of the present disclosure, in S41, the length and width of the circumscribed rectangle are estimated according to the following formula: 图像 =l 实际 ×n÷Pix 像元

[0016] Among them, l图像 represents the side length in the image, l 实际 Indicates the actual side length, n indicates the magnification of the camera lens, Pix 像元 Indicates the pixel size.

[0017] According to an embodiment of the present disclosure, the length membership function and the width membership function in S42 are respectively one of a triangle membership function, a trapezoidal membership function, and a piecewise linear membership function.

[0018] According to an embodiment of the present disclosure, S5 includes: adaptively adjusting the binarization threshold according to the following formula: V self-adaption =V Threshold ±f%×V Threshold ×m

[0019] Among them, V Threshold is the current binarization threshold, V self-adaption is the adaptively adjusted binarization threshold, m is the number of adaptive adjustments, and f% is the adjustment amplitude. The smaller f is, the more accurate the adaptively adjusted binarization threshold is, and the longer the algorithm takes.

[0020] On the other hand, the present disclosure provides a method for correcting the position of a mask, comprising: obtaining a first center coordinate of a first detection target according to the above-mentioned mask mark detection method with occlusion; S7, repeating S1 to S6 using a second camera to obtain a second center coordinate of a second detection target; S8, obtaining an X, Y deviation and a deflection angle of an upper mask piece according to the first center coordinate, the second center coordinate and the third center coordinate of a calibration figure; S9, correcting the position of the mask according to the X, Y deviation and the deflection angle.

[0021] According to an embodiment of the present disclosure, S8 includes: calculating the deflection angle θ of the mask sheet according to the following formula:

[0022] Where R is the distance between the two mask marks, and L is the difference between the distances between the centers of the two detection targets and the center of the calibration pattern. (3) Beneficial effects

[0023] The disclosed method for detecting obscured mask marks and correcting mask position utilizes a single camera to detect obscured marks on the mask. Fuzzy inference rules are determined based on the length and width characteristics detected after binarization of the mark image. The inference results serve as the termination condition for the threshold optimization process of adaptive threshold image binarization. This result-driven approach determines a suitable binarization threshold, reducing image interference while improving the positioning accuracy of the detected marks. Furthermore, two alignment marks are collected to calculate the distance between the center coordinates of the detected mark pattern and the center coordinates of the calibration pattern. The X and Y deviations and the deflection angle θ of the mask can be calculated, yielding accurate data for the required mask position correction. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] FIG1 schematically shows a flow chart of a method for detecting mask marks with occlusion according to an embodiment of the present disclosure;

[0025] FIG2 schematically shows a schematic diagram of a grayscale image before and after image segmentation according to an embodiment of the present disclosure;

[0026] FIG3 schematically shows a result diagram of binarization processing performed on a grayscale image according to an embodiment of the present disclosure;

[0027] FIG4 schematically shows a result diagram of contour detection according to an embodiment of the present disclosure;

[0028] FIG5 schematically shows a result diagram of a minimum circumscribed rectangle group of an image outline obtained according to an embodiment of the present disclosure;

[0029] FIG6 schematically shows a schematic diagram of a detection mark input / output membership function according to an embodiment of the present disclosure;

[0030] FIG7 schematically shows a result diagram of a defuzzified plane of marking probabilities obtained according to an embodiment of the present disclosure;

[0031] FIG8 schematically shows a detection result diagram after fuzzy reasoning according to an embodiment of the present disclosure;

[0032] FIG9 schematically shows a schematic diagram of each stage of target detection according to an embodiment of the present disclosure;

[0033] FIG10 schematically shows an image captured by a CCD camera and a detection target image using two markers according to an embodiment of the present disclosure;

[0034] FIG11 schematically shows a partial flow chart of a method for correcting a mask position according to an embodiment of the present disclosure;

[0035] FIG12 schematically shows a schematic diagram of a detection device according to an embodiment of the present disclosure;

[0036] FIG13 schematically shows a schematic diagram of the principle of calculating the deflection angle according to an embodiment of the present disclosure;

[0037] FIG14 schematically shows an example of calculating deflection angle data according to an embodiment of the present disclosure;

[0038] FIG15 schematically shows a schematic diagram of real-time marker detection results according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0039] In order to make the objectives, technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0040] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise," "include," etc. used herein indicate the presence of the features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0041] It should be noted that if directional indications are involved in the embodiments of the present disclosure, the directional indications are only used to explain the relative positional relationship, movement status, etc. between the components under a certain specific posture. If the specific posture changes, the directional indications will also change accordingly.

[0042] The use of ordinal numbers such as "first," "second," and "third" in the specification and claims to modify corresponding elements does not in itself imply or represent any ordinal number of the elements, nor does it represent the order of one element relative to another or the order in the manufacturing method. The use of such ordinal numbers is only used to clearly distinguish one element with a certain name from another element with the same name.

[0043] In the process of determining the marker's position information, the CCD camera is first used to detect the obscured marker on the mask. For the obscured portion of the captured image, the CCD camera's captured marker image can be segmented, and then the non-marked portion can be filled with white to ensure the success rate of marker position detection. However, CCD camera-captured images often have unevenly distributed shadow areas and small residual non-marked areas that are easily produced after image segmentation. The use of binarization with an appropriate threshold can reduce image interference information and improve the positioning accuracy of the detected marker. However, the appropriate binarization threshold is difficult to determine, and the results of adaptive threshold image binarization show significant differences in the marker image depending on the termination conditions. This makes it impossible to locate the marker using the binarized image.

[0044] Based on this, the present disclosure provides a method for detecting mask marks with occlusion, see Figure 1, including: S1, using a first camera to capture an image of the mask mark and converting the captured image into a grayscale image; wherein the captured image is at least partially occluded; S2, performing image segmentation on the grayscale image and filling the non-marked part with white; S3, obtaining the grayscale median of the grayscale image obtained in S2, and binarizing the grayscale image using the grayscale median as an initial binarization threshold; performing contour detection and obtaining multiple minimum bounding rectangles of the image contour; S4, determining a fuzzy inference rule based on the length and width features of the minimum bounding rectangle of the mask mark, and obtaining the probability that each minimum bounding rectangle obtained in S3 is the actual bounding rectangle of the occluded mark; S5, judging whether there is a probability greater than a preset threshold among the probabilities obtained in S4; if not, adaptively adjusting the binarization threshold and repeating S3 to S5; if so, determining the minimum bounding rectangle with the largest probability as the first detection target; S6, determining the first center coordinates based on the first detection target to complete the positioning detection of the mask mark.

[0045] This method uses a single camera to detect the obscured mark on the mask. According to the length and width features detected after the mark image is binarized, the fuzzy inference rules are determined, and the inference results are used as the termination condition of the threshold optimization process of the adaptive threshold image binarization. The method is guided by the results to obtain a suitable binarization threshold, which improves the positioning accuracy of the detected mark while reducing image interference information.

[0046] Based on the above embodiment, S1 further includes: filtering the grayscale image to suppress noise.

[0047] The image captured by the CCD camera is converted into a grayscale image and filtered to suppress image noise while preserving image details as much as possible.

[0048] Based on the above embodiment, image segmentation in S2 includes: selecting the overlapping point of the mask mark and the occluded part as the segmentation point according to the graphic features of the mask mark; and performing image segmentation on the grayscale image using the segmentation point as an end point or edge point of the segmentation rectangle.

[0049] Image segmentation is performed based on the captured image of the occluded marker, and the non-marked portion is filled with white. Based on the marker's graphical characteristics, the image segmentation points should be selected to preserve the marker's features as much as possible to ensure successful marker detection. As shown in Figure 2, selecting the point where the marker and the occluded portion overlap as an endpoint or edge of the segmentation rectangle maximizes the preservation of the marker's features.

[0050] Based on the above embodiment, the method for contour detection in S3 includes any one of a connectivity-based contour detection algorithm, an edge detection-based contour detection algorithm, and a segmentation-based contour detection algorithm; the method for obtaining multiple minimum enclosing rectangles of the image contour includes using the BoundingRect function to calculate multiple minimum enclosing rectangles.

[0051] The grayscale image is binarized by traversing its grayscale values ​​to obtain its median value. This median value serves as the initial value for the binarization threshold. The labeled image after binarization is shown in Figure 3. Contour detection is then performed, as shown in Figure 4. This method uses the OpenCV algorithm. After contour detection, the BoundingRect method in OpenCV is used to obtain the minimum bounding rectangle (Rectangle) of the image contour. These rectangles are the detected targets, as shown in Figure 5. As can be seen in Figure 5, due to the uneven position and grayscale distribution of the acquired image, as well as the remaining unmarked portions after image segmentation, multiple minimum bounding rectangles for the unmarked portions appear.

[0052] Based on the above embodiment, S4 includes: S41, estimating the length and width of the bounding rectangle in the captured image; S42, dividing the pixel interval of the length into multiple segments according to the length, and determining the length membership function; dividing the pixel interval of the width into multiple segments according to the width, and determining the width membership function; S43, obtaining the fuzzy reasoning result according to the length membership function and the width membership function; S44, using the maximum membership method to defuzzify the fuzzy reasoning result to obtain the probability that each minimum bounding rectangle is the actual bounding rectangle of the occluded mark.

[0053] Fuzzy inference rules are designed based on the length and width features of the minimum bounding rectangle marked in the CCD image. Classical logic (Boolean logic) insists that everything can be represented by binary terms (0 or 1, black or white, yes or no), while the logic of fuzzy inference uses membership instead of Boolean values, that is, values ​​between 0 and 1 can be used to represent the relationship value between members.

[0054] The fuzzy inference system has two input fuzzy sets: {length, width}, where the former represents the length of the minimum bounding rectangle and the latter represents the width. It has one output, {Rectangle}, which represents the fuzzy inference output of the minimum bounding rectangle for the marker. Its physical meaning indicates the degree of similarity between the minimum bounding rectangle detected in the first two steps and the actual bounding rectangle of the occluded marker. The fuzzy subset of length is {short, normal, long}; the fuzzy subset of width is {narrow, general, wide}. Both of the above fuzzy subsets correspond to a membership function. The membership function is the membership degree of each member in the fuzzy subset, such as short, normal1, and long within its value range. The membership degree is between [0, 1], and the membership function can take any form. Common ones include triangle, trapezoid, and segmented line. It can be selected based on experience. For example, a trapezoidal membership function is used, as shown in Figure 6; the fuzzy subset of Rectangle is {small, medium, large}, which means that when the corresponding input is {length, width}, the similarity between the minimum bounding rectangle detected in the first two steps of fuzzy reasoning and the actual marked bounding rectangle is low, medium, and high.

[0055] For example, the input and output membership functions of the selected detection mark are shown in Figure 6, and the fuzzy inference rules are shown in Table 1. The membership functions of the fuzzy subsets of length and width are determined using the same rules. Here, only the determination method of the membership function of length is introduced.

[0056] Table 1 Fuzzy reasoning rules table

[0057] S41, estimate the side length (unit: pixel) of the circumscribed rectangle marked in the image imaged by the CCD after image segmentation. Estimate the length and width of the circumscribed rectangle according to the following formula: 图像 =l 实际 ×n÷Pix 像元 (1)

[0058] Among them, l 图像 represents the side length in the image, l 实际 Indicates the actual side length, n indicates the magnification of the camera lens, Pix 像元 Indicates the pixel size.

[0059] S42, since the length of the bounding rectangle of the marker in the CCD image after image segmentation is approximately 1000 pixels, the pixel interval of length is divided into five segments (0,600], [600,900], [900,1100], [1100,1300], and [1300,1900). The vertical axis membership is used to characterize the degree to which an object belongs to a certain definition. As shown in Figure 6 (a), when the length of length is less than 600 pixels, its membership to short is 1, indicating that if the length of the minimum bounding rectangle in the CCD image is less than 600 pixels, the length belongs to short. The minimum bounding rectangle is too short and is not the minimum bounding rectangle corresponding to the marker; when the length of length is between 900 and 1100, its membership to normal is 1; when the length of length is greater than 1300 pixels, its membership to long is 1. In the pixel intervals [600, 900] and [1100, 1300], since there are two memberships, the maximum membership function is selected here, that is, the maximum membership value is selected as the membership corresponding to the pixel.

[0060] S43~S44, the output of the fuzzy system is the union of the inference results of each rule. The maximum membership method is used for defuzzification, and the results of fuzzy inference (small, medium, large) are converted into accurate data. The defuzzified plane is shown in Figure 7. The probability P of a minimum bounding rectangle in the minimum bounding rectangle group of the image contour is the actual bounding rectangle of the occluded mark is obtained. Rectangle .

[0061] Set the probability P Rectangle The preset threshold is generally set according to the accuracy of the image information to be retained. If only image information that completely matches the inference result needs to be retained, the preset threshold can be set larger. If more image information related to the inference result needs to be retained, the preset threshold can be set smaller. For example, the selected probability P Rectangle The threshold value is 0.75. The retention probability P Rectangle The minimum enclosing rectangle of the image contour that is greater than the threshold and has the highest probability is obtained, and the loop ends at the same time.

[0062] Based on the above embodiment, S5 includes: adaptively adjusting the binarization threshold according to the following formula: V self-adaptiom =V Threshold ±f%×V Threshold ×m (2)

[0063] Among them, V Threshold is the current binarization threshold, V self-adaptionis the adaptively adjusted binarization threshold, m is the number of adaptive adjustments, and f% is the adjustment amplitude. The smaller f is, the more accurate the adaptively adjusted binarization threshold is, and the longer the algorithm takes, for example, f% is 2%.

[0064] If the maximum P of the minimum bounding rectangle of the image contour Rectangle Does not meet the probability P Rectangle As shown in Figure 8, for example, the probability of the detected target being marked is 0.13, which is less than the set probability P Rectangle The detected marker image is incomplete. This is because the selected binary value is too large or too small, resulting in too many or too few retained image features. Threshold In this case, the binarization threshold is adaptively adjusted according to the formula, and steps S3 to S5 are repeated until a probability P is satisfied. Rectangle For example, find the minimum bounding rectangle of the image contour that satisfies the probability P Rectangle The image of the minimum circumscribed rectangle of the image contour during the process is shown in Figure 9.

[0065] The formula selection of "+" or "-" in a certain cycle is based on the fuzzy reasoning result after the binary threshold is selected. For example, if "+" is selected first in the first optimization, if the maximum P Rectangle Increases, it means that increasing the binarization threshold can increase the maximum P Rectangle , in the next cycle, you can still use "+"; at the maximum P Rectangle Greater than probability P Rectangle Before the threshold, if in a certain cycle, after using "+" to update the binary threshold, the maximum P in the next cycle Rectangle decreases, it means that increasing the binarization threshold will lead to the maximum P Rectangle Decrease, then you need to use "-" to update the binarization threshold.

[0066] For example, after multiple cycles, when the fuzzy reasoning obtains the maximum P Rectangle is 0.856, which is greater than the probability P Rectangle The preset threshold, the binary image of the occluded mark, the contour detection image, the image contour minimum bounding rectangle group and the occluded mark image contour minimum bounding rectangle (detection target) obtained by fuzzy inference are shown in Figure 9, and the contour detection results of the two groups of marks are shown in Figure 10.

[0067] The center coordinates of the minimum circumscribed rectangle (detection target) of the occluded marker image outline are calculated according to the following formula.

[0068] P x 、P yThey represent the x-coordinate and y-coordinate of the center of the minimum circumscribed rectangle (detection target) of the occluded mark image collected by the CCD camera, respectively. xRectangle-TopLeft and yRectangle-TopLeft represent the x-coordinate and y-coordinate of the upper left corner of the detection target obtained by the fuzzy inference process, respectively. W Rectangle , L Rectangle Indicates the x- and y-lengths of the detection target, thereby achieving accurate positioning of the mask mark.

[0069] On the basis of the above, the present disclosure also provides a method for correcting the position of a mask, see Figure 11, including: obtaining the first center coordinates of the first detection target according to the above-mentioned obstructed mask mark detection method; S7, using the second camera to repeat S1 to S6 to obtain the second center coordinates of the second detection target; S8, obtaining the X, Y deviation and deflection angle of the mask sheet according to the first center coordinate, the second center coordinate and the third center coordinate of the calibration figure; S9, correcting the position of the mask according to the X, Y deviation and deflection angle.

[0070] This method uses two CCD devices to capture images for mark detection. The detection device includes a lens assembly, a transfer lens, a light source, and a CCD camera. A schematic diagram of the detection device is shown in Figure 12. These two CCD devices are used to detect and correct position deviations on the mask. When performing image processing and positioning of obscured mask marks, a single detection device can complete the image processing and positioning functions of the obscured mask marks.

[0071] Specifically, two sets of alignment mark positions are designed based on the mask correction accuracy requirements. According to the inverse tangent function (i.e., the following formula (4)), the greater the distance between the two mask marks (the larger R), the smaller the calculated angle (the smaller θ), the higher the detection accuracy, and the better the effect of subsequent mask position correction using the marks.

[0072] Where R is the distance between the two mask marks, L is the difference between the centers of the two detection targets and the center of the calibration pattern, the y-coordinate difference between the center of one detection target and the center of the calibration pattern is L1, and the y-coordinate difference between the center of the other detection target and the center of the calibration pattern is L2, where L is the difference between L2 and L1, as shown in Figure 14. θ is the deflection angle of the detection image from the calibration image.

[0073] The schematic diagram of calculating the rotation angle is shown in Figure 13. The mask is installed according to the parallelism of the X / Y direction, and the CCD camera collects images for calibration. OP1 and OP2 are the centers of the two sets of mark calibration graphics. This center is the detection contour center of the calibration image. The center can be obtained by using formula (3) and other methods. P1 and P2 are the detection target center coordinates of the mark image collected by the CCD camera after the robot transfers the mask to the workstation (P x , P y )

[0074] L1 and L2 are calculated according to formula (5), where L1 is the y-coordinate difference between points P1 and OP1, and L2 is the y-coordinate difference between points P2 and OP2.

[0075] Using the detection contour centerline OP1-OP2 of the calibration image as a reference, first perform position correction on the detection target centerline P1-P2 in the x-direction. For example, in the figure above, point OP1 is the correction origin, and the detection target centerline P1-P2 is shifted -ΔX in the x-direction. The calculation formula is as follows:

[0076] After the movement, the x coordinates of point P1 and point OP1 are the same, and the graph after the movement is shown in Figure 14. At this time, the deflection angle θ between the OP1-OP2 line and the P1-P2 line is calculated:

[0077] The calculation of θ can be summarized into the following six cases. An example of angle calculation data is shown in Figure 14.

[0078] After calculating the deflection angle θ and correcting the deflection angle at point P1, the OP1-OP2 line and the P1-P2 line are parallel to each other. At this time, the center line of the detection target P1-P2 is moved upward by -ΔY in the y direction to complete the correction of the mask position.

[0079] Please note that this is for the convenience of understanding. Therefore, with OP1 as the correction origin, the P1-P2 line is first moved upward in x, then the deflection angle is corrected, and finally the y movement is performed. Those skilled in the art can also obtain the detection target center coordinates of the calibration image and the marker image captured by the CCD camera (P x , P y ), take OP2 as the correction origin, or adjust the correction order. The specific calculation formula can be derived according to the above process and will not be repeated here.

[0080] The present invention first uses a single camera to detect the obscured mark on the mask through a transparent lens to determine its position coordinates; then two cameras are used to detect the mask position deviation using the detected position coordinates of the mark on the mask. The deflection angle of the mask sheet can be obtained through inverse trigonometric functions, thereby improving the accuracy of mask position correction.

[0081] The present disclosure is further described below through specific implementations. The following examples specifically illustrate the above-mentioned mask mark detection method and mask position correction method. However, the following examples are merely illustrative of the present disclosure and are not intended to limit the scope of the present disclosure.

[0082] The mask mark detection method and mask position correction method disclosed in the present invention, as shown in Figures 2 to 15, include performing the following steps in sequence:

[0083] Step 1: Convert the image captured by a single camera into a grayscale image and filter it to suppress image noise while preserving image details as much as possible. Segment the image based on the captured image of the occluded mark and fill the unmarked area with white. This is equivalent to steps S1 to S2 above, as shown in Figure 2.

[0084] Step 2: Traverse the grayscale image's grayscale values ​​and obtain the grayscale median for binarization. This grayscale median serves as the initial value for the binarization threshold. The labeled image after binarization is shown in Figure 3. Contour detection is then performed, with the results shown in Figure 4. The BoundingRect method in OpenCV is used to obtain the minimum bounding rectangle (Rectangle) of the image's contour. These rectangles represent the detected targets, as shown in Figure 5. This is equivalent to step S3 above.

[0085] Step 3: Detection Mark Input\Output Membership Function is shown in Figure 6, and the fuzzy inference rule table is shown in Table 1. In this embodiment, the membership function of the fuzzy subsets of length and width is determined using the same rule. The output of the fuzzy system is the union of the inference results of each rule. The maximum membership method is used for defuzzification, and the results of fuzzy inference (small, medium, large) are converted into accurate data. The defuzzified plane is shown in Figure 7, and the probability P that a minimum bounding rectangle in the minimum bounding rectangle group of the image contour is the actual bounding rectangle of the occluded mark is obtained. Rectangle . Equivalent to the above step S4.

[0086] Step 4: Determine whether there is a probability greater than a preset threshold among the probabilities obtained in S4; if not, adaptively adjust the binarization threshold and repeat S3 to S5; if so, determine the minimum bounding rectangle with the highest probability as the first detection target. In this embodiment, the maximum P of the initial minimum bounding rectangle is Rectangle Does not meet the probability PRectangle The preset threshold is 0.75. As shown in Figure 8, the probability of the detected target being marked is 0.13, which is less than the set probability P Rectangle After multiple cycles, fuzzy reasoning obtains the maximum P Rectangle is 0.856, which is greater than the probability P Rectangle The preset threshold is 0.75, the binary image of the occluded mark, the contour detection image, the image contour minimum bounding rectangle group and the occluded mark image contour minimum bounding rectangle (detection target) obtained by fuzzy reasoning are shown in Figure 9. This is equivalent to the above step S5.

[0087] Step 5: Determine the first center coordinates based on the first detection target to complete the positioning detection of the mask mark, which is equivalent to the above step S6.

[0088] Step 6: Based on the mask correction accuracy requirements, design the positions of two sets of alignment marks. The minimum bounding rectangle detection results of the two sets of marks are shown in Figure 10. Obtain the center coordinates of the two sets of detection targets. The detection results are shown in Figure 15. The dashed rectangle represents the detection outline of the calibration pattern, and the solid rectangle represents the detection outline of the real-time image captured by the CCD camera. The two marks are set 120 mm apart, and each pixel is approximately 2.2 μm. The pre-alignment algorithm detection accuracy can reach 0.001°. This is equivalent to step S7 above.

[0089] Step 7: Calculate the X and Y deviations and deflection angles of the mask sheet based on the center coordinates of the two sets of detection targets and the center coordinates of the calibration pattern. This is equivalent to the above step S8.

[0090] Step 8: Move the mask according to the X and Y deviations and the deflection angle to complete the correction of the mask position. This is equivalent to the above step S9.

[0091] The present invention uses a single camera to determine the accurate position information of the mark, thereby achieving high-precision detection of the mask mark; further, two cameras are used for detection to calculate the X and Y deviations and deflection angle θ of the mask sheet, and obtain accurate data for the position correction required for the mask.

[0092] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present disclosure. It should be understood that the above are only specific embodiments of the present disclosure and are not intended to limit the present disclosure. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present disclosure should be included in the scope of protection of the present disclosure. Industrial Applicability

[0093] The present disclosure provides a method for detecting mask marks with occlusion, comprising: S1, converting a collected mask mark image into a grayscale image; wherein the collected image is at least partially occluded; S2, performing image segmentation on the grayscale image and filling the non-marked part with white; S3, obtaining the grayscale median of the grayscale image and binarizing the grayscale image as an initial binarization threshold; performing contour detection and obtaining multiple minimum bounding rectangles of the image contour; S4, determining a fuzzy inference rule based on the length and width features of the minimum bounding rectangle of the mask mark, and obtaining the probability that each minimum bounding rectangle is the actual bounding rectangle of the occluded mark; S5, judging whether there is a probability greater than a preset threshold in the probability; if not, adaptively adjusting the binarization threshold and repeating S3 to S5; if so, determining the minimum bounding rectangle with the largest probability as the first detection target; S6, determining the first center coordinates based on the first detection target to complete the positioning detection of the mask mark.

[0094] Furthermore, it is understood that the disclosed methods for detecting obstructed mask marks and correcting mask positions are reproducible and can be used in a variety of applications. For example, the disclosed methods for detecting obstructed mask marks and correcting mask positions can be used in the field of optical inspection technology.

Claims

1. A method for detecting mask marks with occlusion, characterized in that: include: S1, using a first camera to capture an image of a mask mark, and converting the captured image into a grayscale image; wherein the captured image is at least partially blocked; S2, performing image segmentation on the grayscale image and filling the non-marked parts with white; S3, obtaining the grayscale median value of the grayscale image obtained in S2, and binarizing the grayscale image using the grayscale median value as an initial binarization threshold; performing contour detection, and obtaining multiple minimum circumscribed rectangles of the image contour; S4, determining a fuzzy inference rule according to the length and width features of the minimum bounding rectangle of the mask mark, and obtaining the probability that each minimum bounding rectangle obtained in S3 is the actual bounding rectangle of the masked mark; S5, determining whether there is a probability greater than a preset threshold among the probabilities obtained in S4; if not, adaptively adjusting the binarization threshold, and repeating S3 to S5; if so, determining the minimum bounding rectangle with the highest probability as the first detection target; S6, determining the first center coordinates according to the first detection target, and completing the positioning detection of the mask mark.

2. The method for detecting mask marks with occlusion according to claim 1, characterized in that: The S1 further comprises: The grayscale image is filtered to suppress noise.

3. The method for detecting mask marks with occlusion according to claim 1, characterized in that: The image segmentation in S2 includes: According to the graphic features of the mask mark, the overlapping point of the mask mark and the blocked part is selected as the segmentation point; The grayscale image is segmented using the segmentation point as an end point or an edge point of a segmentation rectangle.

4. The method for detecting mask marks with occlusion according to claim 1, characterized in that: The method for performing contour detection in S3 includes any one of a contour detection algorithm based on connectivity, a contour detection algorithm based on edge detection, and a contour detection algorithm based on segmentation; The method for obtaining multiple minimum bounding rectangles of the image contour includes using a BoundingRect function to calculate and obtain multiple minimum bounding rectangles.

5. The method for detecting mask marks with occlusion according to claim 1, characterized in that: The S4 includes: S41, estimating the length and width of a circumscribed rectangle in the acquired image; S42, dividing the pixel interval of the length into multiple segments according to the length, and determining the length membership function; dividing the pixel interval of the width into multiple segments according to the width, and determining the width membership function; S43, obtaining a fuzzy reasoning result according to the length membership function and the width membership function; S44, using the maximum membership method to defuzzify the fuzzy reasoning result, and obtain the probability that each minimum bounding rectangle is the actual bounding rectangle of the obscured mark.

6. The method for detecting mask marks with occlusion according to claim 5, characterized in that: In S41, the length and width of the circumscribed rectangle are estimated according to the following formula: 图像 = l 实际 ×n÷Pix 像元 Among them, l 图像 represents the edge length in the image, l 实际 represents the actual side length, n represents the magnification of the camera lens, Pix 像元 Represents the pixel size.

7. The method for detecting mask marks with occlusion according to claim 5, characterized in that: The length membership function and the width membership function in S42 are respectively one of a triangle membership function, a trapezoidal membership function, and a piecewise linear membership function.

8. The method for detecting mask marks with occlusion according to claim 1, characterized in that: The S5 includes: adaptively adjusting the binarization threshold according to the following formula: V self-adaption =V Threshold ±f%×V Threshold ×m Among them, V Threshold is the current binarization threshold, V self-adaption is the adaptively adjusted binarization threshold, m is the number of adaptive adjustments, and f% is the adjustment amplitude.

9. A method for correcting a mask position, characterized in that: include: Obtaining the first center coordinates of the first detection target according to the mask mark detection method with occlusion according to any one of claims 1 to 7; S7, using a second camera to repeat S1 to S6 to obtain a second center coordinate of a second detection target; S8, obtaining the X and Y deviations and the deflection angle of the mask sheet according to the first center coordinate, the second center coordinate and the third center coordinate of the calibration pattern; S9, correcting the position of the mask according to the X and Y deviations and the deflection angle.

10. The method for correcting the mask position according to claim 9, characterized in that: The step S8 includes: calculating the deflection angle θ of the mask sheet according to the following formula: Among them, R is the distance between the two mask marks, and L is the difference between the distances between the centers of the two detection targets and the center of the calibration pattern.

Images