Method for detecting mask surface particle defects
By acquiring transmitted and reflected light images, utilizing grayscale difference and polarity judgment, and combining the defect result matrix, the problem of particle defect detection on the mask surface was solved, achieving efficient detection of particles and soft defects on the mask surface and improving the yield of photolithography process.
Patent Information
- Application Number
- CN202510082672.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-20
- Publication Date
- 2025-11-21
- Estimated Expiration
- 2045-01-20
AI Technical Summary
Existing technologies cannot effectively detect particle defects on the mask surface, leading to a decrease in the yield of photolithography processes. In particular, defects on the glass can cause fatal defects in photolithography.
By acquiring transmitted and reflected light images, using grayscale difference and polarity judgment, and combining the defect result matrix, particle defects on the mask surface are detected. The method does not require mask design documents.
It improves the detection effect of particle defects on the mask surface, and can detect defects in the glass area, chromium area and edge area with high sensitivity, and can also detect other soft defects, such as fingerprints and chemical stains.
Smart Images
Figure CN120009306B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to a mask surface particle defect detection method, and belongs to the technical field of semiconductor mask defect detection. BACKGROUND
[0002] At present, once a mask plate has a defect, the defect will be repeatedly transmitted to each chip or wafer, directly affecting the yield of the photolithography process. In photolithography, compared with dust and particles on the chromium layer, defects on the glass are more serious and can cause fatal defects in photolithography. When the particles are located at the position of the fine line width glass pattern, they can cause insufficient exposure energy of the photoresist at this position, resulting in abnormal integrated circuits after photolithography, thereby affecting the logic operation and function of the chip. Therefore, how to effectively detect these non-logic defects is particularly important. SUMMARY
[0003] The technical problem to be solved by the present application is to overcome the defects of the prior art and provide a mask surface particle defect detection method, which can be used without providing a mask design file and improves the detection effect.
[0004] In order to solve the above technical problems, the technical scheme of the present application is as follows: a mask surface particle defect detection method, the steps of the method comprising: S1: collecting a transmission light image under transmission light as a detection image, and collecting a reflection light image under reflection light as a shielding signal to detect the defects of the mask under transmission light; wherein step S1 is specifically as follows:
[0005] S11: initializing a defect result matrix C;
[0006] S12: shifting the transmission light image of the mask collected under transmission light by N pixels to obtain a shifted transmission light image, and calculating the gray difference between the shifted transmission light image and the original transmission light image to obtain a transmission difference value image;
[0007] S13: inverting the color of the reflection light image of the mask collected under reflection light to obtain an inverted color reflection light image;
[0008] S14: shifting the inverted color reflection light image by N pixels to obtain a shifted reflection light image, and calculating the gray difference between the shifted reflection light image and the inverted color reflection light image to obtain a reflection difference value image;
[0009] S15: setting a transmission gray scale threshold value for the transmission difference value map, traversing each pixel point in the transmission difference value map, when the absolute value of the gray scale difference of the corresponding pixel point is less than the transmission gray scale threshold value, considering that the corresponding pixel point has no difference, and setting the transmission gray scale difference polarity value at this position to 0; when the absolute value of the gray scale difference of the corresponding pixel point is greater than or equal to the transmission gray scale threshold value, considering that the corresponding pixel point has difference, and then judging the polarity of the gray scale difference of the corresponding pixel point: when the gray scale difference is greater than 0, setting the transmission gray scale difference polarity value at this position to 1, and when the gray scale difference is less than 0, setting the transmission gray scale difference polarity value at this position to -1;
[0010] S16: setting a reflection gray scale threshold value for the reflection difference value map, traversing each pixel point in the reflection difference value map, when the absolute value of the gray scale difference of the corresponding pixel point is less than the reflection gray scale threshold value, considering that the corresponding pixel point has no difference, and setting the reflection gray scale difference polarity value at this position to 0; when the absolute value of the gray scale difference of the corresponding pixel point is greater than or equal to the reflection gray scale threshold value, considering that the corresponding pixel point has difference, and then judging the polarity of the gray scale difference of the corresponding pixel point: when the gray scale difference is greater than 0, setting the reflection gray scale difference polarity value at this position to 1, and when the gray scale difference is less than 0, setting the reflection gray scale difference polarity value at this position to -1;
[0011] S17: when the absolute value of the gray scale difference of the corresponding pixel point in the transmission difference value map is less than the transmission gray scale threshold value, setting the defect result value of the corresponding pixel point in the defect result matrix C to the first mark value representing no defect; when the absolute value of the gray scale difference of the corresponding pixel point in the transmission difference value map is greater than or equal to the transmission gray scale threshold value, and the absolute value of the gray scale difference of the corresponding pixel point in the reflection difference value map is greater than or equal to the reflection gray scale threshold value, if the transmission gray scale difference polarity value of the corresponding pixel point in the transmission difference value map is the same as the reflection gray scale difference polarity value of the neighborhood pixel point in the reflection difference value map, setting the defect result value of the corresponding pixel point in the defect result matrix C to the no defect mark value representing no defect, and if the transmission gray scale difference polarity value of the corresponding pixel point in the transmission difference value map is not the same as the reflection gray scale difference polarity value of the neighborhood pixel point in the reflection difference value map, setting the defect result value of the corresponding pixel point in the defect result matrix C to the defect mark value representing defect.
[0012] Further, in step S17, a neighborhood distance threshold value range is set, the coordinates of the corresponding pixel point are [x1, y1], the coordinates of the neighborhood pixel point are [x2, y2], and the neighborhood distance threshold value range is , . .
[0013] Further, the steps of the method further include: S2: taking the image acquired under the reflection light as a detection image, and taking the image acquired under the transmission light as a shielding signal to detect the defects of the mask under the reflection light; wherein step S2 is specifically:
[0014] Steps S21-S26 correspond to steps S11-S16 respectively;
[0015] Step S27: When the absolute value of the gray scale difference of the pixel point at the corresponding position in the reflection difference value map is < the reflection gray scale threshold value, then the defect result value of the pixel point at the corresponding position in the defect result matrix C is set as the first mark value representing no defect; when the absolute value of the gray scale difference of the pixel point at the corresponding position in the reflection difference value map is ≥ the reflection gray scale threshold value, and the absolute value of the gray scale difference of the pixel point at the corresponding position in the transmission difference value map is ≥ the transmission gray scale threshold value, if the reflection gray scale difference polarity value of the pixel point at the corresponding position in the reflection difference value map is the same as the transmission gray scale difference polarity value of the pixel point at the neighborhood position in the transmission difference value map, then the defect result value of the pixel point at the corresponding position in the defect result matrix C is set as the no defect mark value representing no defect; if the reflection gray scale difference polarity value of the pixel point at the corresponding position in the reflection difference value map is not the same as the transmission gray scale difference polarity value of the pixel point at the neighborhood position in the transmission difference value map, then the defect result value of the pixel point at the corresponding position in the defect result matrix C is set as the defect mark value representing defect.
[0016] Further, the reflection gray scale threshold value in step S2 is greater than the reflection gray scale threshold value in step S1, and the transmission gray scale threshold value in step S2 is smaller than the transmission gray scale threshold value in step S1.
[0017] Further, the displacement in steps S12 and S14 is at least one of horizontal movement, vertical movement, +45° direction movement, and -45° direction movement, and an or operation is performed on the corresponding defect result matrix C obtained by the corresponding direction movement.
[0018] Further, the method further comprises the following steps:
[0019] A closing operation is performed on the defect result matrix C, then the defect positions are counted, and defect filtering is performed according to the length and / or width parameters of the defects, to obtain final defect data.
[0020] After the above technical scheme is adopted, the normal pattern on the mask plate is divided into three parts, i.e., a glass area, a chromium area, and an edge area. The particle capture sensitivity of the glass area is very high by the method, and the edge area and the chromium area defects can be captured at the same time. In addition to particle defects, the invention is also effective for various soft defects such as fingerprints, water marks, or chemical contamination marks. In addition to bipolar mask plates (BIM), the invention can also detect phase shift mask plates (PSM). BRIEF DESCRIPTION OF DRAWINGS
[0021] Fig. 1(a) is a transmission light collection image of the present application;
[0022] Fig. 1(b) is a local gray scale difference value map of the particle position of the transmission light of the present application;
[0023] Fig. 2(a) is a reflected light collection image of the present application;
[0024] Fig. 2(b) is a reflected light collection image after color inversion of the present application;
[0025] Fig. 2(c) is a reflected light particle position local gray scale difference value image of the present application;
[0026] Fig. 3(a) is a local image after high threshold of a transmission light gray scale difference value image;
[0027] Fig. 3(b) is a local image after low threshold of a reflected light gray scale difference value image;
[0028] Fig. 3(c) is a defect result calculated under transmission light;
[0029] Fig. 4(a) is a local image after high threshold of a reflected light gray scale difference value image;
[0030] Fig. 4(b) is a local image after low threshold of a transmission light gray scale difference value image;
[0031] Figure 5 is a defect position image;
[0032] Figure 6 is a mask area transmission and reflection relationship image of the present application. DETAILED DESCRIPTION
[0033] In order to make the content of the present application more easily and clearly understood, the present application is further described in detail below according to specific embodiments and in conjunction with the accompanying drawings.
[0034] As shown in Figs. 1-6, the mask surface particle defect detection method includes the following steps: S1: collecting a transmission light image of the mask under transmission light as a detection image, and collecting a reflected light image of the mask under reflected light as a shielding signal to detect defects of the mask under transmission light; wherein step S1 is specifically:
[0035] S11: initializing a defect result matrix C as a full 0 matrix;
[0036] S12: horizontally shifting the transmission light image of the mask collected under transmission light by N pixels to obtain a shifted transmission light image, and calculating a gray scale difference between the shifted transmission light image and the original transmission light image to obtain a transmission difference value image ; the original transmission light image T is shown in Fig. 1(a), and the difference value at the particle is shown in Fig. 1(b); the calculation formula of the gray scale difference is as follows: ;
[0037] S13: inverting the color of the reflected light image of the mask collected under reflected light to obtain a reflected light image after color inversion ; the original reflected light image is shown in Fig. 2(a); the reflected light image after color inversion As shown in Fig. 2(b);
[0038] S14: obtaining a reflected light image after color inversion displacing the reflected light image horizontally by N pixels, obtaining a displaced reflected light image, and obtaining a reflected difference value image by calculating the gray scale difference between the displaced reflected light image and the reflected light image after color inversion The difference value at the particle is shown in Fig. 2(c). The formula for calculating the gray scale difference is as follows: ;
[0039] S15: processing the transmission difference value image Setting a transmission gray scale threshold Thrd, and traversing each pixel point at position [x, y] in the transmission difference value image When the absolute value of the gray scale difference of the pixel point at the corresponding position is less than the transmission gray scale threshold Thrd, it is considered that the pixel point at the corresponding position has no difference, i.e., the pixel position that is not red in Fig. 3(a), and the transmission gray scale difference polarity value at this position is set to 0. When the absolute value of the gray scale difference of the pixel point at the corresponding position is greater than or equal to the transmission gray scale threshold Thrd, it is considered that the pixel point at the corresponding position has a difference, i.e., the pixel position marked in red in Fig. 3(a), and then the polarity of the gray scale difference of the pixel point at the corresponding position is judged: when the gray scale difference is greater than 0, the transmission gray scale difference polarity value at this position is set to 1, and when the gray scale difference is less than 0, the transmission gray scale difference polarity value at this position is set to -1. This step is summarized as follows:
[0040] ;
[0041] S16: processing the reflected difference value image Setting a reflection gray scale threshold Thrd_r, and traversing each pixel point at position [x, y] in the reflected difference value image When the absolute value of the gray scale difference of the pixel point at the corresponding position is less than the reflection gray scale threshold Thrd_r, it is considered that the pixel point at the corresponding position has no difference, i.e., the pixel position that is not red in Fig. 3(b), and the reflection gray scale difference polarity value at this position is set to 0. When the absolute value of the gray scale difference of the pixel point at the corresponding position is greater than or equal to the reflection gray scale threshold Thrd_r, it is considered that the pixel point at the corresponding position has a difference, i.e., the pixel position marked in red in Fig. 3(b), and then the polarity of the gray scale difference of the pixel point at the corresponding position is judged: when the gray scale difference is greater than 0, the reflection gray scale difference polarity value at this position is set to 1, and when the gray scale difference is less than 0, the reflection gray scale difference polarity value at this position is set to -1. This step is summarized as follows:
[0042] ;
[0043] S17: when the transmission difference value image If the absolute value of the gray scale difference of the pixel point at [x1, y1] is less than the transmission gray scale threshold Thrd, then the defect result value of the pixel point at [x1, y1] in the defect result matrix C is set as the first mark value representing no defect, and the transmission difference graph If the absolute value of the gray scale difference of the pixel point at [x1, y1] is greater than or equal to the transmission gray scale threshold Thrd, and the reflection difference graph If the absolute value of the gray scale difference of the pixel point at [x1, y1] is greater than or equal to the reflection gray scale threshold Thrd_r, then the defect result value of the pixel point at [x1, y1] in the defect result matrix C is set as the first mark value representing no defect, and the transmission difference graph If the absolute value of the gray scale difference of the pixel point at [x1, y1] is greater than or equal to the transmission gray scale threshold Thrd, and the reflection difference graph If the absolute value of the gray scale difference of the pixel point at [x1, y1] is greater than or equal to the reflection gray scale threshold Thrd_r, then the defect result value of the pixel point at [x1, y1] in the defect result matrix C is set as the first mark value representing no defect, and the transmission difference graph If the absolute value of the gray scale difference of the pixel point at [x1, y1] is greater than or equal to the transmission gray scale threshold Thrd, and the reflection difference graph If the absolute value of the gray scale difference of the pixel point at [x1, y1] is greater than or equal to the reflection gray scale threshold Thrd_r, then the defect result value of the pixel point at [x1, y1] in the defect result matrix C is set as the first mark value representing no defect, and the transmission difference graph
[0044] ;
[0045] Specifically, in step S17, a neighborhood distance threshold range is set, and the corresponding position pixel point coordinates are [x1, y1], the neighborhood position pixel point coordinates are [x1+range, y1+range], [x1-range, y1-range], [x1+range, y1-range], [x1-range, y1+range], [x1, y1+range] and [x1, y1-range]. , ] wherein, .
[0046] Specifically, the steps of the method further comprise: S2: taking the reflection light image of the mask under the reflection light as a detection image, and taking the transmission light image of the mask under the transmission light as a shielding signal to detect the defects of the mask under the reflection light; wherein, step S2 specifically comprises:
[0047] Steps S21-S26 correspond to steps S11-S16 respectively;
[0048] Step S27: When the absolute value of the gray scale difference of the pixel point at the corresponding position in the reflection difference value map is < the reflection gray scale threshold value, then the defect result value of the pixel point at the corresponding position in the defect result matrix C is set as the first mark value representing no defect. When the absolute value of the gray scale difference of the pixel point at the corresponding position in the reflection difference value map is ≥ the reflection gray scale threshold value, and the absolute value of the gray scale difference of the pixel point at the corresponding position in the transmission difference value map is ≥ the transmission gray scale threshold value, if the reflection gray scale difference polarity value of the pixel point at the corresponding position in the reflection difference value map is the same as the transmission gray scale difference polarity value of the pixel point at the neighborhood position in the transmission difference value map, then the defect result value of the pixel point at the corresponding position in the defect result matrix C is set as the no defect mark value representing no defect. If the reflection gray scale difference polarity value of the pixel point at the corresponding position in the reflection difference value map is not the same as the transmission gray scale difference polarity value of the pixel point at the neighborhood position in the transmission difference value map, then the defect result value of the pixel point at the corresponding position in the defect result matrix C is set as the defect mark value representing defect. Details are shown in Figure 4(a) 、 4(b)
[0049] Specifically, the reflection gray scale threshold value in step S2 is greater than the reflection gray scale threshold value in step S1, and the transmission gray scale threshold value in step S2 is less than the transmission gray scale threshold value in step S1, which is used to reduce false defects.
[0050] Specifically, the thin lines in the example diagram are horizontally distributed, and the gray scale difference of the glass and the particles is detected by horizontal movement to detect the particle defects. The actual mask pattern may be of any angle, and therefore vertical movement, +45° direction movement and -45° direction movement are added to detect thin line width patterns and other patterns of various possible direction distributions. The transmission and reflection gray scale threshold values of each direction can be independently set and do not affect each other. That is, the displacement in step S12 and step S14 can be horizontal movement, vertical movement, +45° direction movement and -45° direction movement, and the defect results of the four direction transmission and reflection mutual detection are counted, and the corresponding defect result matrix C obtained by the corresponding direction movement is calculated. As long as a defect is detected in a certain direction, it is considered that the detected pattern has a defect.
[0051] Wherein, the no defect mark value can be 0, and the defect mark value can be 255.
[0052] Specifically, the steps of the method further include:
[0053] The closed operation is performed in the defect result matrix C, and then the defect positions are counted, the defect filtering is performed according to the length (preset parameter) and / or width (preset parameter) of the defect, and the final defect data desired by the user is obtained. Then, the non-0 positions of the matrix C are mapped to the original diagram for marking, and finally presented to the user as shown in Figure 5
[0054] In the embodiment, the reflected light image contrast is less than the transmitted light image. The horizontal direction moving distance N can be specifically 5; the transmitted gray threshold Thrd can be specifically 25; the reflected gray threshold Thrd_r can be specifically 9; and the neighborhood range range can be specifically 3. In step S2, the reflected gray threshold can be specifically 15; and the transmitted gray threshold can be specifically 12.
[0055] In the embodiment, when detecting, only the imaging data of the transmitted light and the reflected light is needed to be provided, without providing a mask design file. The normal pattern on the mask is divided into three parts, i.e., a glass area, a chrome area, and an edge area, which is a transition area of the glass area and the chrome area. The method can detect the particle defects in the glass area, the chrome area, and the edge area, and has good detection effect on the pattern with small line width. The relationship between the theoretical transmission and the reflection of the mask is shown by the dashed line in Figure 6 .
[0056] Specifically, when the particle is located in the glass area or the edge area, the particle defect changes the transmission and the reflection of the light. The dust particle is darker (easier to absorb) than the glass in the reflection, and darker than the glass in the transmission, as shown by the area 1 in Figure 6 . Therefore, at the particle position, the transmission and reflection relationship does not satisfy the negative correlation relationship shown by the dashed line in the above figure, and the method detects the particle defect based on the principle. Specifically, the transmitted light image is moved by N pixel units in any direction (horizontal, vertical, +45°, -45°), the gray difference between the moved transmitted light image and the original transmitted light image is calculated, so that the gray difference is generated between the glass area, the edge area, the chrome area, and the particle area, to obtain a polarity image (the gray difference greater than 0 is positive, and the gray difference less than 0 is negative). Similarly, the reflected light image is moved, the gray difference between the moved reflected light image and the reflected light image after color inversion is calculated, to also obtain a polarity image. In the normal pattern position of the mask, the glass area gray value > the edge gray value > the chrome area gray value in the transmission, and the glass area gray value > the edge gray value > the chrome area gray value in the reflection (after color inversion). The transmission gray value and the reflection gray value are positively correlated, and the corresponding gray difference under the transmission and the corresponding gray difference under the reflection at the same [x, y] position have the same polarity. At the particle position, the glass gray value under the transmission is greater than the particle gray value, and the polarity is positive; the glass gray value under the reflection (after color inversion) is less than the particle gray value, and the polarity is negative. When the polarity of the transmission and reflection gray difference is opposite, it is considered that there is a defect. The region near the origin is noise interference, and no polarity judgment is made. When the absolute value of the corresponding gray difference under the transmission and the absolute value of the corresponding gray difference under the reflection are small, it is considered that it is noise interference, and no polarity judgment is made or the polarity value is 0.
[0057] When the particle is located in the chrome area, the particle is not imaged under the transmitted light, and is imaged only under the reflected light, as shown in region 2 in the above figure. At the particle, the transmitted light gray scale difference is 0, and the reflected light (after the reverse color) has a gray scale difference between the chrome area and the particle, and at this time, it is considered that there is a defect at this position.
[0058] In summary, the particle capture sensitivity of the method of the present embodiment to the glass area is very high, and the edge area and chrome area defects can be captured at the same time. In addition to the particle defects, the method is also effective for various soft defects such as fingerprints, water marks, or chemical contamination marks. In addition to the bipolar mask (BIM), the present application can also detect the phase shift mask (PSM).
[0059] The specific embodiments described above further illustrate the technical problems solved by the present application, the technical solutions and the beneficial effects. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
Claims
1. A method of detecting a particle defect on a mask surface, characterized by, The steps of the method comprise: S1: collecting a transmission light image under transmission light as a detection image and collecting a reflection light image under reflection light as a shielding signal to detect defects of a mask under transmission light; wherein the step S1 is specifically: S11: initializing a defect result matrix C; S12: shifting the transmission light image of the mask collected under transmission light by N pixels to obtain a shifted transmission light image, and calculating a gray difference between the shifted transmission light image and the original transmission light image to obtain a transmission difference value image; S13: inverting the reflection light image of the mask collected under reflection light to obtain an inverted reflection light image; S14: shifting the inverted reflection light image by N pixels to obtain a shifted inverted reflection light image, and calculating a gray difference between the shifted inverted reflection light image and the inverted reflection light image to obtain a reflection difference value image; S15: setting a transmission gray threshold for the transmission difference value image, and traversing each pixel point in the transmission difference value image; when the absolute value of the gray difference of the corresponding pixel point is less than the transmission gray threshold, it is considered that the corresponding pixel point has no difference, and the transmission gray difference polarity value at this position is set to 0; when the absolute value of the gray difference of the corresponding pixel point is greater than or equal to the transmission gray threshold, it is considered that the corresponding pixel point has a difference, and then the polarity of the gray difference of the corresponding pixel point is judged: when the gray difference is greater than 0, the transmission gray difference polarity value at this position is set to 1; when the gray difference is less than 0, the transmission gray difference polarity value at this position is set to -1; S16: setting a reflection gray threshold for the reflection difference value image, and traversing each pixel point in the reflection difference value image; when the absolute value of the gray difference of the corresponding pixel point is less than the reflection gray threshold, it is considered that the corresponding pixel point has no difference, and the reflection gray difference polarity value at this position is set to 0; when the absolute value of the gray difference of the corresponding pixel point is greater than or equal to the reflection gray threshold, it is considered that the corresponding pixel point has a difference, and then the polarity of the gray difference of the corresponding pixel point is judged: when the gray difference is greater than 0, the reflection gray difference polarity value at this position is set to 1; when the gray difference is less than 0, the reflection gray difference polarity value at this position is set to -1; S17: when the absolute value of the gray difference of the corresponding pixel point in the transmission difference value image is less than the transmission gray threshold, the defect result value of the corresponding pixel point in the defect result matrix C is set to a first mark value representing no defect; when the absolute value of the gray difference of the corresponding pixel point in the transmission difference value image is greater than or equal to the transmission gray threshold and the absolute value of the gray difference of the corresponding pixel point in the reflection difference value image is greater than or equal to the reflection gray threshold, if the transmission gray difference polarity value of the corresponding pixel point in the transmission difference value image is the same as the reflection gray difference polarity value of the pixel point in the neighborhood position in the reflection difference value image, the defect result value of the corresponding pixel point in the defect result matrix C is set to a no defect mark value representing no defect; if the transmission gray difference polarity value of the corresponding pixel point in the transmission difference value image is not the same as the reflection gray difference polarity value of the pixel point in the neighborhood position in the reflection difference value image, the defect result value of the corresponding pixel point in the defect result matrix C is set to a defect mark value representing a defect.
2. The method according to claim 1, characterized in that, In step S17, a neighborhood distance threshold range is set, and the corresponding position pixel point coordinates are [x1, y1], the neighborhood position pixel point coordinates are [x2, y2], and the neighborhood distance threshold range is range. , ] where .
3. The method according to claim 1 or 2, characterized in that, The method further comprises the following steps in the method: The step S2 comprises: The steps S21-S26 correspond to the steps S11-S16 respectively. The step S27 comprises: when the absolute value of the gray difference of the pixel point at the corresponding position in the reflection difference value image is less than the reflection gray threshold, the defect result value of the pixel point at the corresponding position in the defect result matrix C is set as the first mark value representing no defect; when the absolute value of the gray difference of the pixel point at the corresponding position in the reflection difference value image is greater than or equal to the reflection gray threshold and the absolute value of the gray difference of the pixel point at the corresponding position in the transmission difference value image is greater than or equal to the transmission gray threshold, if the reflection gray difference polarity value of the pixel point at the corresponding position in the reflection difference value image is the same as the transmission gray difference polarity value of the pixel point at the neighborhood position in the transmission difference value image, the defect result value of the pixel point at the corresponding position in the defect result matrix C is set as the no-defect mark value representing no defect; if the reflection gray difference polarity value of the pixel point at the corresponding position in the reflection difference value image is different from the transmission gray difference polarity value of the pixel point at the neighborhood position in the transmission difference value image, the defect result value of the pixel point at the corresponding position in the defect result matrix C is set as the defect mark value representing defect.
4. The method of claim 3, wherein the reflection gray threshold in the step S2 is greater than the reflection gray threshold in the step S1, and the transmission gray threshold in the step S2 is less than the transmission gray threshold in the step S1.
5. The method of claim 1, wherein the displacement in the step S12 and the step S14 is at least one of horizontal movement, vertical movement, +45° direction movement and -45° direction movement, and an OR operation is performed on the corresponding defect result matrix C obtained by the corresponding direction movement. The method further comprises the following steps in the method:
6. The method of claim 1, wherein, The closed operation is performed on the defect result matrix C, then the defect positions are counted, the defect filtering is performed according to the length and / or width parameters of the defects, and finally the final defect data is obtained.
Citation Information
Patent Citations
Mask substrate surface micro-nano defect detection device and detection method
CN114965369A
Pattern inspection method
JP2006030518A