A method for detecting mold deformation during nanoimprinting

Abstract

The present invention discloses a method for detecting mold deformation during nanoimprinting, which belongs to the field of deformation monitoring technology and includes the following steps: S1, obtaining a processing image of the mold at the current moment, and determining the first spatial unit, the second spatial unit, the third spatial unit and the fourth spatial unit of the processing image; S2, extracting deformation monitoring points of the processing image; S3, determining whether the mold is deformed according to the positional relationship between the deformation monitoring points and the first spatial unit, the second spatial unit, the third spatial unit and the fourth spatial unit. In the deformation monitoring process, the pixel point values ​​of the units segmented by the processing image are considered, and the pixel points of the processing image that do not participate in the deformation monitoring points are eliminated, so as to ensure that the mold deformation monitoring of nanoimprinting is accurate, the error rate is low, and the work efficiency of the imprinting is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of deformation monitoring, and in particular relates to a method for detecting mold deformation during a nanoimprinting process. Background Art

[0002] Nanoimprint technology is a high-precision and high-efficiency nanomanufacturing technology that is widely used in semiconductors, microelectronics, optical devices and other fields. However, during the nanoimprint process, the mold is prone to deformation under conditions such as high pressure and high temperature, resulting in a decrease in the accuracy of the imprinted pattern and affecting product quality. Therefore, how to accurately detect and control mold deformation has become a key issue in nanoimprint technology. However, existing solutions for deformation monitoring usually rely on manual visual monitoring, which has low detection accuracy and is time-consuming and labor-intensive. Summary of the invention

[0003] In order to solve the above problems, the present invention proposes a method for detecting mold deformation during nanoimprinting process.

[0004] The technical solution of the present invention is: a method for detecting mold deformation during nanoimprinting process comprises the following steps:

[0005] S1, obtaining a processing image of the mold at the current moment, and determining a first spatial unit, a second spatial unit, a third spatial unit, and a fourth spatial unit of the processing image;

[0006] S2, extracting deformation monitoring points of the processed image;

[0007] S3. Determine whether the mold is deformed according to the positional relationship between the deformation monitoring point and the first space unit, the second space unit, the third space unit, and the fourth space unit.

[0008] Furthermore, S1 includes the following sub-steps:

[0009] S11, using the row and column where the center point of the processed image is located as segmentation lines, splitting the processed image into four segmentation units;

[0010] S12, calculating the image change amount of each segmentation unit;

[0011] S13, arranging the image change amounts of the segmentation units in order from small to large, as the first spatial unit, the second spatial unit, the third spatial unit and the fourth spatial unit respectively.

[0012] The beneficial effect of the above further scheme is that in the present invention, the processed image is segmented using the rows and columns of the center of the processed image as the segmentation line, so as to facilitate the subsequent steps to determine the weight of each spatial unit to determine the monitoring area related to the deformation monitoring point.

[0013] Furthermore, in S12, the calculation formula of the image change amount γ of the segmentation unit is: ; In the formula, Represents the pixel value of the kth pixel in the segmentation unit, represents the pixel value of the k+1th pixel in the segmentation unit, c represents a constant, and K represents the number of pixels in the segmentation unit.

[0014] Furthermore, S2 includes the following sub-steps:

[0015] S21, removing local feature points of the processed image, and extracting LoG spots in the remaining area of ​​the processed image;

[0016] S22, calculating the spot coverage value of each pixel in the LoG spot;

[0017] S23, taking the pixel point corresponding to the maximum spot coverage value as the deformation monitoring point.

[0018] The beneficial effect of the above further scheme is: in the present invention, the local feature points of an image refer to the local structures in the image that are unique, stable and repeatable, usually composed of some key points in the image and the feature descriptors of a small area around them, while the deformation monitoring points are usually abnormal pixels, so the local feature points do not participate in the extraction of deformation monitoring points. The LoG spots refer to those areas in the image that have significant color or grayscale differences from the surrounding areas. The colors or grayscales inside these areas are basically the same, but there are obvious boundaries with the surrounding areas, so they participate in the extraction of deformation monitoring points.

[0019] Furthermore, in S22, the calculation formula of the spot coverage value F of the pixel point in the LoG spot is: ; where σ represents the standard deviation of the Gaussian kernel, Represents the pixel value of the pixel in the LoG spot, Indicates a round-up operation.

[0020] Furthermore, S3 includes the following sub-steps:

[0021] S31, setting the weights of the first spatial unit, the second spatial unit, the third spatial unit and the fourth spatial unit to 1, 2, 3 and 4 respectively;

[0022] S32, determining the weight of the spatial unit where the deformation monitoring point is located;

[0023] S33, determining a deformation area to be monitored according to the weight of the spatial unit where the deformation monitoring point is located and the weights of the remaining spatial units;

[0024] S34. Determine whether the mold is deformed based on the deformation area to be monitored.

[0025] The beneficial effect of the above further scheme is: in the present invention, the order of spatial units is determined according to the size of the image change, and their weights are determined according to the order of spatial units. According to the size of the weight of the spatial unit, the distance threshold is determined, and then combined with the deformation monitoring points, the area used for deformation monitoring is determined. This can reduce the workload of deformation monitoring and improve monitoring accuracy.

[0026] Furthermore, in S33, twice the weight of the spatial unit where the deformation monitoring point is located is summed with the weights of the remaining spatial units, and the ratio between the processed image and the summation result is used as the distance threshold, and the pixel points whose straight-line distance to the deformation monitoring point is less than the distance threshold are used as the deformation area to be monitored.

[0027] The beneficial effect of the above further solution is that: in the present invention, when generating the deformation area to be monitored, it is necessary to use the weight of each spatial unit, and it is necessary to increase the size of the weight of the spatial unit where the deformation monitoring point is located.

[0028] Furthermore, in S34, an adversarial neural network is used to process the deformation area to be monitored to determine whether the mold is deformed.

[0029] The generative adversarial network consists of a generator and a discriminator, which can evaluate the degree of deformation of the image.

[0030] The beneficial effects of the present invention are as follows: the present invention discloses a method for detecting mold deformation during nanoimprinting, which performs spatial unit segmentation on a processing image of a nanoimprinting mold, and extracts deformation monitoring points of the processing image; determines the area to be monitored by the distance relationship between the deformation monitoring points and each spatial unit, and finally uses an adversarial network to determine whether the mold is deformed; in the deformation monitoring process, the pixel point values ​​of the units of the processing image segmentation are considered, and the pixel points of the processing image that do not participate in the deformation monitoring points are eliminated, thereby ensuring that the mold deformation monitoring of nanoimprinting is accurate, the error rate is low, and the work efficiency of imprinting is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 Flow chart of the mold deformation detection method during nanoimprinting process. DETAILED DESCRIPTION

[0032] The embodiments of the present invention will be further described below in conjunction with the accompanying drawings.

[0033] like Figure 1 As shown, the present invention provides a method for detecting mold deformation during nanoimprinting, comprising the following steps:

[0034] S1, obtaining a processing image of the mold at the current moment, and determining a first spatial unit, a second spatial unit, a third spatial unit, and a fourth spatial unit of the processing image;

[0035] S2, extracting deformation monitoring points of the processed image;

[0036] S3. Determine whether the mold is deformed according to the positional relationship between the deformation monitoring point and the first space unit, the second space unit, the third space unit, and the fourth space unit.

[0037] In this embodiment of the present invention, S1 includes the following sub-steps:

[0038] S11, using the row and column where the center point of the processed image is located as segmentation lines, splitting the processed image into four segmentation units;

[0039] S12, calculating the image change amount of each segmentation unit;

[0040] S13, arranging the image change amounts of the segmentation units in order from small to large, as the first spatial unit, the second spatial unit, the third spatial unit and the fourth spatial unit respectively.

[0041] In the present invention, the processed image is segmented using the central row and column of the processed image as a segmentation line, so as to facilitate the subsequent steps to determine the weight of each spatial unit to determine the monitoring area related to the deformation monitoring point.

[0042] In the embodiment of the present invention, in S12, the calculation formula of the image change amount γ of the segmentation unit is: ; In the formula, Represents the pixel value of the kth pixel in the segmentation unit, represents the pixel value of the k+1th pixel in the segmentation unit, c represents a constant, and K represents the number of pixels in the segmentation unit.

[0043] In this embodiment of the present invention, S2 includes the following sub-steps:

[0044] S21, removing local feature points of the processed image, and extracting LoG spots in the remaining area of ​​the processed image;

[0045] S22, calculating the spot coverage value of each pixel in the LoG spot;

[0046] S23, taking the pixel point corresponding to the maximum spot coverage value as the deformation monitoring point.

[0047] In the present invention, the local feature points of an image refer to the local structures in the image that are unique, stable and repeatable, usually composed of some key points in the image and the feature descriptors of a small area around them, while the deformation monitoring points are usually abnormal pixels, so the local feature points do not participate in the extraction of deformation monitoring points. The LoG spots refer to those areas in the image that have significant color or grayscale differences from the surrounding areas. The colors or grayscales inside these areas are basically the same, but there are obvious boundaries with the surrounding areas, so they participate in the extraction of deformation monitoring points.

[0048] In the embodiment of the present invention, in S22, the calculation formula of the spot coverage value F of the pixel point in the LoG spot is: ; where σ represents the standard deviation of the Gaussian kernel, Represents the pixel value of the pixel in the LoG spot, Indicates a round-up operation.

[0049] In this embodiment of the present invention, S3 includes the following sub-steps:

[0050] S31, setting the weights of the first spatial unit, the second spatial unit, the third spatial unit and the fourth spatial unit to 1, 2, 3 and 4 respectively;

[0051] S32, determining the weight of the spatial unit where the deformation monitoring point is located;

[0052] S33, determining a deformation area to be monitored according to the weight of the spatial unit where the deformation monitoring point is located and the weights of the remaining spatial units;

[0053] S34. Determine whether the mold is deformed based on the deformation area to be monitored.

[0054] In the present invention, the order of spatial units is determined according to the size of image change, and their weights are determined according to the order of spatial units. According to the size of the weight of the spatial units, the distance threshold is determined, and then combined with the deformation monitoring points, the area used for deformation monitoring is determined. This can reduce the workload of deformation monitoring and improve monitoring accuracy.

[0055] In an embodiment of the present invention, in S33, twice the weight of the spatial unit where the deformation monitoring point is located is summed with the weights of the remaining spatial units, and the ratio between the processed image and the summation result is used as the distance threshold, and the pixel points whose straight-line distance to the deformation monitoring point is less than the distance threshold are used as the deformation area to be monitored.

[0056] In the present invention, when generating the deformation area to be monitored, it is necessary to use the weight of each spatial unit, and it is necessary to increase the weight of the spatial unit where the deformation monitoring point is located.

[0057] In the embodiment of the present invention, in S34, an adversarial neural network is used to process the deformation area to be monitored to determine whether the mold is deformed.

[0058] The generative adversarial network consists of a generator and a discriminator, which can evaluate the degree of deformation of the image.

[0059] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A method for detecting mold deformation during nanoimprinting, characterized in that: The following steps are involved: S1, obtaining a processing image of the mold at the current moment, and determining a first spatial unit, a second spatial unit, a third spatial unit, and a fourth spatial unit of the processing image; S2, extracting deformation monitoring points of the processed image; S3, determining whether the mold is deformed according to the positional relationship between the deformation monitoring point and the first space unit, the second space unit, the third space unit, and the fourth space unit; The S1 comprises the following sub-steps: S11, using the row and column where the center point of the processed image is located as segmentation lines, splitting the processed image into four segmentation units; S12, calculating the image change amount of each segmentation unit; S13, arranging the image changes of the segmentation units in order from small to large, as the first spatial unit, the second spatial unit, the third spatial unit and the fourth spatial unit respectively; In S12, the calculation formula of the image change amount γ of the segmentation unit is: ; In the formula, Represents the pixel value of the kth pixel in the segmentation unit, represents the pixel value of the k+1th pixel in the segmentation unit, c represents a constant, and K represents the number of pixels in the segmentation unit; The S3 comprises the following sub-steps: S31, setting the weights of the first spatial unit, the second spatial unit, the third spatial unit and the fourth spatial unit to 1, 2, 3 and 4 respectively; S32, determining the weight of the spatial unit where the deformation monitoring point is located; S33, determining a deformation area to be monitored according to the weight of the spatial unit where the deformation monitoring point is located and the weights of the remaining spatial units; S34, determining whether the mold is deformed based on the deformation area to be monitored; In S33, twice the weight of the spatial unit where the deformation monitoring point is located is summed with the weights of the remaining spatial units, and the ratio between the processed image and the summation result is used as the distance threshold, and the pixel points whose straight-line distance to the deformation monitoring point is less than the distance threshold are used as the deformation area to be monitored.

2. The method for detecting mold deformation during nanoimprinting according to claim 1, characterized in that: The S2 comprises the following sub-steps: S21, removing local feature points of the processed image, and extracting LoG spots in the remaining area of ​​the processed image; S22, calculating the spot coverage value of each pixel in the LoG spot; S23, taking the pixel point corresponding to the maximum spot coverage value as the deformation monitoring point.

3. The method for detecting mold deformation during nanoimprinting according to claim 2, characterized in that: In S22, the calculation formula of the spot coverage value F of the pixel point in the LoG spot is: ; where σ represents the standard deviation of the Gaussian kernel, Represents the pixel value of the pixel in the LoG spot, Indicates a round-up operation.

4. The method for detecting mold deformation during nanoimprinting according to claim 1, characterized in that: In S34, the deformation area to be monitored is processed using an adversarial neural network to determine whether the mold is deformed.

Images