Pattern inspection apparatus, corrected image generation method, and computer-readable recording medium storing program

1. A pattern inspection apparatus having a microprocessor comprising:

an optical image acquisition unit configured to acquire an optical image of a target workpiece to be inspected on which a pattern is formed;

a reference image generation unit configured to generate a reference image to be compared with the optical image;

a least-squares method calculation unit configured, by using a mathematical model in which a parallel shift amount displaced from a position where the optical image and the reference image are aligned by a pixel unit, an expansion and contraction error coefficient, a rotation error coefficient, a gray-level offset and an image transmission loss ratio are parameters, to calculate each of the parameters by a least-squares method;

a corrected image generation unit configured to generate a corrected image by shifting a position of the reference image by a displacement amount displaced from the position aligned by the pixel unit, based on the each of the parameters; and

a comparison unit implemented by the microprocessor and configured to compare the corrected image with the optical image,

wherein the least-squares method calculation unit calculates a gray-level offset, an image transmission loss ratio, a parallel shift amount in X direction, a parallel shift amount in Y direction, an expansion and contraction error coefficient in the X direction, an expansion and contraction error coefficient in the Y direction, a rotation error coefficient in the X direction, and a rotation error coefficient of the Y direction so that a first difference obtained by subtracting a second sum that is a sum of a product of the expansion and contraction error coefficient in the X direction, a value obtained by partially differentiating a pixel value of the reference image with respect to x, and a variable x, a product of the expansion and contraction error coefficient in the Y direction, a value obtained by partially differentiating the pixel value of the reference image with respect to y, and a variable y, and a product of the rotation error coefficient in the Y direction, the value obtained by partially differentiating the pixel value of the reference image with respect to y, and the variable x from a first sum that is a sum of the gray-level offset, a product of the image transmission loss ratio and the pixel value of the reference image, a product of the parallel shift amount in the X direction and the value obtained by partially differentiating the pixel value of the reference image with respect to x, a product of the parallel shift amount in the Y direction and the value obtained by partially differentiating the pixel value of the reference image with respect to y, and aproduct of the rotation error coefficient in the X direction, the value obtained by partially differentiating the pixel value of the reference image with respect to x, and the variable y may be equal to a second difference obtained by subtracting a pixel value of the optical image from the pixel value of the reference image.

2. The apparatus according to claim 1 , wherein the corrected image generation unit generates the corrected image by performing a linear interpolation in a 4-point weighted model.

3. The apparatus according to claim 1 , wherein the corrected image generation unit generates the corrected image by performing a linear interpolation in a model which simply divides an image of a predetermined size into N×N.

4. The apparatus according to claim 1 , wherein the least-squares method calculation unit is implemented by the microprocessor.

5. A pattern inspection apparatus having a microprocessor comprising:

an optical image acquisition unit configured to acquire an optical image of a target workpiece to be inspected on which a pattern is formed;

a reference image generation unit configured to generate a reference image to be compared with the optical image;

a least-squares method calculation unit configured, by using a mathematical model in which a parallel shift amount displaced from a position where the optical image and the reference image are aligned by a pixel unit, an expansion and contraction error coefficient, a rotation error coefficient, a gray-level offset and an image transmission loss ratio are parameters, to calculate each of the parameters by a least-squares method;

a judgment unit configured to judge whether a calculated parallel shift amount, a calculated expansion and contraction error coefficient, a calculated rotation error coefficient, a calculated_gray-level offset and a calculated image transmission loss ratio are suitable or unsuitable, by using each threshold value set up for each of the parameters;

a second least-squares method calculation unit configured to delete at least one parameter judged to be unsuitable or set up a value equivalent to deletion for the parameter judged to be unsuitable, and recalculate other parameters in the mathematical model by the least-squares method;

a corrected image generation unit configured to generate a corrected image by shifting a position of the reference image by a displacement amount displaced from the position aligned by the pixel unit, based on recalculated parameters; and

a comparison unit implemented by the microprocessor and configured to compare the corrected image with the optical image,

wherein the least-squares method calculation unit calculates a gray-level offset, an image transmission loss ratio, a parallel shift amount in X direction, a parallel shift amount in Y direction, an expansion and contraction error coefficient in the X direction, an expansion and contraction error coefficient in the Y direction, a rotation error coefficient in the X direction, and a rotation error coefficient of the Y direction so that a first difference obtained by subtracting a second sum that is a sum of a product of the expansion and contraction error coefficient in the X direction, a value obtained by partially differentiating a pixel value of the reference image with respect to x, and a variable x, a product of the expansion and contraction error coefficient in the Y direction, a value obtained by partially differentiating the pixel value of the reference image with respect to y, and a variable y, and a product of the rotation error coefficient in the Y direction, the value obtained by partially differentiating the pixel value of the reference image with respect to y, and the variable x from a first sum that is a sum of the gray-level offset, a product of the image transmission loss ratio and the pixel value of the reference image, a product of the parallel shift amount in the X direction and the value obtained by partially differentiating the pixel value of the reference image with respect to x, a product of the parallel shift amount in the Y direction and the value obtained by partially differentiating the pixel value of the reference image with respect to y, and a product of the rotation error coefficient in the X direction, the value obtained by partially differentiating the pixel value of the reference image with respect to x, and the variable y may be equal to a second difference obtained by subtracting a pixel value of the optical image from the pixel value of the reference image.

6. The apparatus according to claim 5 , wherein the corrected image generation unit generates the corrected image by performing a linear interpolation in a 4-point weighted model.

7. The apparatus according to claim 5 , wherein the corrected image generation unit generates the corrected image by performing a linear interpolation in a model which simply divides an image of a predetermined size into N×N.

8. The apparatus according to claim 5 , wherein the least-squares method calculation unit is implemented by the microprocessor.

9. A corrected image generation method implemented by a pattern inspection apparatus having a microprocessor comprising:

inputting an optical image and a reference image of a target workpiece to be inspected;

calculating, by using a mathematical model in which a parallel shift amount displaced from a position where the optical image and the reference image are aligned by a pixel unit, an expansion and contraction error coefficient, a rotation error coefficient, a gray-level offset and an image transmission loss ratio are parameters, the parameters by a least-squares method; and

generating, using the microprocessor, a corrected image by shifting a position of the reference image by a displacement amount displaced from the position aligned by the pixel unit, based on calculated parameters, and outputting the corrected image,

wherein a gray-level offset, an image transmission loss ratio, a parallel shift amount in X direction, a parallel shift amount in Y direction, an expansion and contraction error coefficient in the X direction, an expansion and contraction error coefficient in the Y direction, a rotation error coefficient in the X direction, and a rotation error coefficient of the Y direction are calculated, so that a first difference obtained by subtracting a second sum that is a sum of a product of the expansion and contraction error coefficient in the X direction, a value obtained by partially differentiating a pixel value of the reference image with respect to x, and a variable x, a product of the expansion and contraction error coefficient in the Y direction, a value obtained by partially differentiating the pixel value of the reference image with respect to y, and a variable y, and a product of the rotation error coefficient in the Y direction, the value obtained by partially differentiating the pixel value of the reference image with respect to y, and the variable x from a first sum that is a sum of the gray-level offset, a product of the image transmission loss ratio and the pixel value of the reference image, a product of the parallel shift amount in the X direction and the value obtained by partially differentiating the pixel value of the reference image with respect to x, a product of the parallel shift amount in the Y direction and the value obtained by partially differentiating the pixel value of the reference image with respect to y, and a product of the rotation error coefficient in the X direction, the value obtained by partially differentiating the pixel value of the reference image with respect to x, and the variable y may be equal to a second difference obtained by subtracting a pixel value of the optical image from the pixel value of the reference image.

10. The method according to claim 9 , wherein the calculating is performed using the microprocessor.

11. A corrected image generation method implemented by a pattern inspection apparatus having a microprocessor comprising:

inputting an optical image and a reference image of a target workpiece to be inspected;

calculating, by using a mathematical model in which a parallel shift amount displaced from a position where the optical image and the reference image are sparsely aligned by a pixel unit, an expansion and contraction error coefficient, a rotation error coefficient, a gray-level offset and an image transmission loss ratio are parameters, the parameters by a least-squares method;

judging whether a calculated parallel shift amount, a calculated expansion and contraction error coefficient, a calculated rotation error coefficient, a calculated gray-level offset and a calculated image transmission loss ratio are suitable or unsuitable by using each threshold value set up for each of the parameters;

deleting at least one parameter judged to be unsuitable or setting up a value equivalent to deletion for the parameter judged to be unsuitable, and recalculating other parameters in the mathematical model by a least-squares method; and

generating, using the microprocessor, a corrected image by shifting a position of the reference image by a displacement amount displaced from the position aligned by the pixel unit, based on recalculated parameters, and outputting the corrected image,

wherein a gray-level offset, an image transmission loss ratio, a parallel shift amount in X direction, a parallel shift amount in Y direction, an expansion and contraction error coefficient in the X direction, an expansion and contraction error coefficient in the Y direction, a rotation error coefficient in the X direction, and a rotation error coefficient of the Y direction are calculated, so that a first difference obtained by subtracting a second sum that is a sum of a product of the expansion and contraction error coefficient in the X direction, a value obtained by partially differentiating a pixel value of the reference image with respect to x, and a variable x, a product of the expansion and contraction error coefficient in the Y direction, a value obtained by partially differentiating the pixel value of the reference image with respect to y, and a variable y, and a product of the rotation error coefficient in the Y direction, the value obtained by partially differentiating the pixel value of the reference image with respect to y, and the variable x from a first sum that is a sum of the gray-level offset, a product of the image transmission loss ratio and the pixel value of the reference image, a product of the parallel shift amount in the X direction and the value obtained by partially differentiating the pixel value of the reference image with respect to x, a product of the parallel shift amount in the Y direction and the value obtained by partially differentiating the pixel value of the reference image with respect to y, and a product of the rotation error coefficient in the X direction, the value obtained by partially differentiating the pixel value of the reference image with respect to x, and the variable y may be equal to a second difference obtained by subtracting a pixel value of the optical image from the pixel value of the reference image.

12. The method according to claim 11 , wherein the calculating is performed using the microprocessor.

13. A non-transitory computer-readable recording medium storing a program that when executed by a computer causes the computer to implement a method comprising:

storing an optical image and a reference image to be used for a comparative inspection of a target workpiece to be inspected on which a pattern is formed, in a storage device;

reading the optical image and the reference image from the storage device, and by using a mathematical model in which a parallel shift amount displaced from a position where the optical image and the reference image are aligned by a pixel unit, an expansion and contraction error coefficient, a rotation error coefficient, a gray-level offset and an image transmission loss ratio are parameters, calculating the parameters by a least-squares method; and

generating a corrected image by shifting a position of the reference image by a displacement amount displaced from the position aligned by the pixel unit, based on the parameters, and outputting the corrected image,

wherein a gray-level offset, an image transmission loss ratio, a parallel shift amount in X direction, a parallel shift amount in Y direction, an expansion and contraction error coefficient in the X direction, an expansion and contraction error coefficient in the Y direction, a rotation error coefficient in the X direction, and a rotation error coefficient of the Y direction are calculated, so that a first difference obtained by subtracting a second sum that is a sum of a product of the expansion and contraction error coefficient in the X direction, a value obtained by partially differentiating a pixel value of the reference image with respect to x, and a variable x, a product of the expansion and contraction error coefficient in the Y direction, a value obtained by partially differentiating the pixel value of the reference image with respect to y, and a variable y, and a product of the rotation error coefficient in the Y direction, the value obtained by partially differentiating the pixel value of the reference image with respect to y, and the variable x from a first sum that is a sum of the gray-level offset, a product of the image transmission loss ratio and the pixel value of the reference image, a product of the parallel shift amount in the X direction and the value obtained by partially differentiating the pixel value of the reference image with respect to x, a product of the parallel shift amount in the Y direction and the value obtained by partially differentiating the pixel value of the reference image with respect to y, and a product of the rotation error coefficient in the X direction, the value obtained by partially differentiating the pixel value of the reference image with respect to x, and the variable y may be equal to a second difference obtained by subtracting a pixel value of the optical image from the pixel value of the reference image.