JP2009097928A - Defect inspecting device and defect inspection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To suppress effectively generation of pseudo-defects, in defect detection for detecting pattern defects on a sample surface, by inspecting a photographed image on the sample surface. <P>SOLUTION: This defect inspection device 1 for detecting a pattern defect formed on the sample 2 surface is equipped with a pseudo-defect determining part 31 for determining that a defect candidate is a pseudo-defect, when the size of the defect candidate detected on the imaged image photographed by an imaging part 13 is smaller than a resolution limit of an imaging optical system of the imaging part 13, and when a spatial change of the gray level value from a pixel around a defect candidate pixel, showing the defect candidate to the defect candidate, pixel exceeds a resolving power of the imaging optical system. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

The present invention relates to a defect inspection apparatus for detecting defects in a pattern on a surface of a sample by inspecting a captured image obtained by imaging a pattern formed on the surface of the sample with a predetermined imaging device according to a predetermined inspection method. And a defect inspection method.
More specifically, the present invention relates to a technique for preventing detection of a pseudo defect caused by noise or the like included in a captured image being inspected in such a defect inspection apparatus and defect inspection method.

The manufacture of semiconductor devices such as semiconductor wafers, photomask substrates, and liquid crystal display panels consists of a large number of man-hours, and the occurrence of defects in final and intermediate processes is inspected and fed back to the manufacturing process. Is important from the viewpoint of improving yield. In order to detect defects in the middle of the manufacturing process, images obtained by imaging the pattern formed on the surface of a sample such as a semiconductor wafer, a photomask substrate, a liquid crystal display panel substrate, or a liquid crystal device substrate are obtained. A pattern defect inspection for detecting defects existing on the surface of a sample by inspecting the pattern is widely performed.
In the following description, a semiconductor wafer defect inspection apparatus that inspects a defect of a pattern formed on a semiconductor wafer will be described as an example. However, the present invention is not limited to this, and can be widely applied to defect inspection apparatuses for inspecting semiconductor devices such as a semiconductor memory photomask substrate, a liquid crystal device substrate, and a liquid crystal display panel substrate. .

FIG. 1 shows a block diagram of a defect inspection apparatus similar to that proposed by the applicant of the present application in Japanese Patent Application No. 2003-188209 (the following Patent Document 1). The defect inspection apparatus 1 is provided with a stage 11 movable in a three-dimensional direction, and a sample stage (chuck stage) 12 is provided on the upper surface of the stage 11. A semiconductor wafer (hereinafter, simply referred to as “wafer”) 2 as a sample to be inspected is placed and fixed on the sample table 12.
An imaging unit 13 for capturing an optical image of the surface of the wafer 2 is provided above the sample table 12. An image sensor (imaging device) such as a one-dimensional or two-dimensional CCD sensor (preferably a one-dimensional TDI sensor) is used for the imaging unit 13, and an optical image of the surface of the wafer 2 imaged on the light receiving surface is used. Convert to electrical signal. Here, the configuration example of FIG. 1 will be described assuming that the imaging unit 13 uses a one-dimensional TDI sensor as an imaging device.

  By moving the imaging unit 13 and the wafer 2 relatively by moving the stage 11, the imaging unit 13 is scanned in the X direction or the Y direction with respect to the wafer 2 to obtain a two-dimensional image of the surface of the wafer 2. As the illumination optical system for illuminating the wafer 2, a bright field illumination optical system or a dark field illumination optical system is used.

The image signal output from the imaging unit 13 is stored in the image storage unit 14 after being converted into a multi-value digital signal (gray level value).
On the wafer 2, as shown in FIG. 2, a plurality of dies (chips) 3 are repeatedly arranged in a matrix in the X and Y directions. As shown in FIG. 3, the imaging unit 13 including the one-dimensional TDI sensor scans the surface of the wafer 2 along the “scanning direction” indicated by the arrow 90, thereby including the dotted lines including the wafers 3 a to 3 c. Assume that the image 100 of the area surrounded by is acquired.

  The same pattern is formed on each die 3a, 3b, 3c. For this reason, when the wafer 2 is imaged, a pattern generated in the area where each die is imaged is used as a unit pattern, and a repetitive pattern in which these unit patterns repeatedly appear in the captured image of the wafer 2. Accordingly, the gray level values of the corresponding portions of the captured images of the respective dies, that is, the images of the portions captured of the same portion of each die, are essentially the same value. In the example of FIG. 3, the gray level values of the pixels 101a and 101b corresponding to each other in the captured image obtained by imaging the die 3a and the die 3b, that is, the pixels 101a and 101b on which the images of the same portion of the die 3a and the die 3b are originally formed Value.

Therefore, when a difference in gray level values (gray level difference) between corresponding locations (pixels 101a and 101b) that should be the same in the captured images of the two dies is detected, compared to a case where both dies do not have a defect. If one die is defective, the gray level difference becomes large. By detecting such a large gray level difference, defects existing on the die can be detected (die-to-die comparison).
Further, when a repeated pattern such as a memory cell is formed in one die, a defect is detected even if a gray level difference is detected between images obtained by capturing corresponding portions that should be the same in the repeated pattern. Can be detected (cell-to-cell comparison).

  In die-to-die comparison, it is common to compare images obtained by imaging two adjacent dies (single detection). This does not tell which die is defective. Therefore, a comparison is further made with dies adjacent to different sides, and when the gray level difference of the same portion becomes larger than the threshold value again, it is determined that the die is defective (double detection).

  That is, considering the case where it is detected whether or not a defect exists in a part of an image obtained by imaging the die 3a in the image 100, first, a captured image obtained by imaging the die 3a and an image of one die 3b adjacent thereto. A gray level difference between pixels (pixels 101a and 101b) corresponding to each other is detected. Next, a gray level difference between pixels (pixels 101a and 101c) corresponding to each other in a captured image obtained by capturing the die 3a and the image of the other die 3c adjacent thereto is detected. If both the gray level difference between the pixels 101a and 101b and the gray level difference between the pixels 101a and 101c are larger than the detection threshold, a defect occurs in the pixel 101a portion in the image of the die 3a. It is determined that there is (double detection). The same applies to the cell-to-cell comparison.

In the following description, an image to be detected in which detection of defects is to be detected by double detection, such as an image of a portion obtained by imaging the above-described exemplary die 3a, will be referred to as “inspected image”, and Pixels constituting the inspection image may be referred to as “inspected pixels”. In addition, an image that is compared with the image to be inspected, such as an image of a portion obtained by imaging the above-described dies 3b and 3c, is referred to as a “reference image”, and a pixel that is contrasted with the pixel to be inspected is referred to as a “reference pixel”. Sometimes.
In addition, one of images compared with each other in single detection may be referred to as “inspected image” and the other as “reference pixel”. A pixel constituting the inspected image may be referred to as “inspected pixel”, and a pixel contrasted with the inspected pixel may be referred to as “reference pixel”.

  Returning to FIG. 1, the defect inspection apparatus 1 includes a defect detection unit 20. The defect detection unit 20 includes a difference detection unit 21, a detection threshold value determination unit 22, a threshold value comparison unit 23, and a defect information generation unit 24 described below. Hereinafter, the case of performing defect detection by die-to-die comparison in the defect inspection apparatus 1 will be described. The main difference in processing between die-to-die comparison and cell-to-cell comparison is that the interval between the pixel to be inspected and the reference pixel to be compared is different. There is no big difference in the concept of the detection method itself depending on whether it is an integral multiple of the die pitch or an integral multiple of the cell pitch.

  When an output signal of the imaging unit 13 that is a one-dimensional TDI camera is captured while the imaging unit 13 is scanned relative to the wafer 2, a two-dimensional image of the wafer 2 is accumulated in the image storage unit 14. At this time, the difference detection unit 21 is input with the gray level value of the pixel imaged in the same location of the two adjacent dies accumulated in the image storage unit 14. A difference signal (gray level difference) is calculated. One of these two pixels is a pixel to be inspected and the other is a reference pixel. The difference detection unit 21 calculates a gray level difference between the pixel to be inspected and the reference pixel. The calculated gray level difference is input to the detection threshold value determination unit 22 and the threshold value comparison unit 23.

  The detection threshold value determination unit 22 determines a detection threshold value by predetermined statistical processing based on the distribution of gray level differences detected by the difference detection unit 21 for a plurality of pixels to be inspected, and outputs the detection threshold value to the threshold value comparison unit 23. The threshold comparison unit 23 compares the gray level difference input from the difference detection unit 21 with the detection threshold determined by the detection threshold determination unit 22, and detects a defect included in the inspection image by single detection. That is, when the gray level difference exceeds the detection threshold, the threshold comparison unit 23 is one of the inspection candidate pixels and the reference pixels in which such a gray level difference is detected as a defect candidate pixel indicating such a defect. Judge that there is.

The defect information creating unit 24 compares the target die and its adjacent dies at the position where the defect candidate pixel is detected by comparing the current target die with one of the adjacent dies. In the case where a defect candidate pixel is detected between the other die, it is determined that there is a defect candidate pixel at the position of the defect candidate pixel of the target die. That is, in the comparison by the threshold comparison unit 23, two pixels between the pixel to be inspected in the image obtained by imaging a certain target die and each of the reference pixels in the image obtained by imaging two dies adjacent to both sides of the target die. When both of the gray level differences exceed the detection threshold, the defect information creation unit 24 performs double detection for determining that the pixel to be inspected of the target die is a defect candidate pixel.
Then, when pixels determined to be defect candidate pixels are adjacent to each other, image processing such as combining these into one defect is performed, and defect information including position information and size of the detected defect candidate Is generated.
When the defect information creating unit 24 creates defect information, the partial image including the defect candidate pixel and surrounding pixels, the gray level difference between the reference pixel and the defect candidate pixel compared with the defect candidate pixel, A difference image between a partial image including a defective candidate pixel and surrounding pixels and a reference pixel compared with these pixels may be included in the defect information.

JP 2004-177397 A

Due to the influence of noise or the like included in the captured image used for defect detection, the defect inspection apparatus may detect a portion that is not originally a defect as a defective portion. A defect that is not a true defect is called a “pseudo defect”. When the generation amount of pseudo defects is large, a wasteful processing amount increases in automatic defect classification (ADC) performed after defect detection by the defect inspection apparatus. Therefore, it is preferable that pseudo defects are reduced as much as possible.
If the defect detection sensitivity of the defect inspection apparatus is lowered, the generation amount of pseudo defects is reduced. However, decreasing the defect detection sensitivity increases the possibility of missing a minute true defect. Thus, since the suppression of the occurrence of pseudo defects and the detection without leaking true defects are in a trade-off relationship, it is difficult to set the defect detection sensitivity so as to satisfy both.

  In view of the above problems, the present invention detects defects in a pattern on the surface of a sample by inspecting a captured image obtained by imaging a pattern formed on the surface of the sample with a predetermined imaging device. It is an object of the present invention to provide means and method for effectively suppressing the occurrence of pseudo defects.

In the present invention, defect candidates are detected on a captured image in accordance with a predetermined defect inspection method, the size of the defect candidate is smaller than the resolution limit of the imaging optical system of the imaging apparatus, and defect candidate pixels indicating the defect candidate are displayed. When the spatial change of the gray level value from the surrounding pixels to the defect candidate pixel exceeds the resolving power of the imaging optical system, the defect candidate is determined as a pseudo defect.
The image formed on the image sensor provided on the image plane of the imaging apparatus does not include a high frequency component higher than the spatial frequency at the resolution limit of the imaging optical system. Therefore, when a defect candidate with a size smaller than the resolution limit is detected by the specific image processing that performs defect detection processing, the gray level value change near the defect candidate pixel of the captured image exceeds the resolving power of the imaging optical system. Whether the change in the gray level value near the defective candidate pixel is really due to the contrast of the subject, or the gray level value of the defective candidate pixel is caused by other causes such as noise. You can know whether or not

  Therefore, the size of the defect candidate detected on the captured image is smaller than the resolution limit of the imaging optical system of the predetermined imaging device, and the gray level value between the pixel surrounding the defect candidate pixel and the defect candidate pixel When the spatial change exceeds the resolving power of the imaging optical system, this can be determined as a pseudo defect. And the output of the defect information which concerns on a pseudo defect can be suppressed by removing the information determined to be a pseudo defect, for example from the defect information output from a defect inspection apparatus.

In the imaging device, the image formed on the image plane is converted into an electrical signal by an imaging device in which pixels are arranged at a pitch finer than the resolution limit period of the imaging optical system on the image plane of the imaging optical system. An imaging device may be used.
At this time, as a change value indicating the amount of spatial change in the gray level value between the surrounding pixels of the defective candidate pixel and the defective candidate pixel, the gray level difference between the defective candidate pixel and the adjacent pixel May be detected and used. At this time, gray level differences between the defective candidate pixel and a plurality of adjacent pixels may be obtained, and the maximum value may be detected and used as the change value indicating the amount of spatial change.
Further, as a change value indicating the amount of the spatial change, a gray level between a pixel and a defect candidate pixel that is separated from the defect candidate pixel by a predetermined interval that is smaller than a resolution limit period of the imaging optical system. A level difference may be detected and used. At this time, the gray level difference between each of the plurality of pixels separated from the defect candidate pixel by a predetermined interval and the defect candidate pixel is obtained, and the maximum value is detected as the change value indicating the amount of the spatial change. It may be used.

  When detecting a defect candidate pixel, a plurality of reference pixels that are scheduled to be the same as the pixel to be inspected in the captured image are respectively compared with the pixel to be inspected, When any of the gray level differences between them exceeds a predetermined detection threshold, the pixel to be inspected may be detected as a defect candidate pixel. The number of reference pixels to be compared with the pixel to be inspected may be two as in the case of the above-mentioned double detection, but a larger number of reference pixels may be compared with the pixel to be inspected.

  When detecting a pseudo defect, a gray level difference ratio obtained by dividing a change value by a gray level difference between any one of a plurality of reference pixels compared to the defect candidate pixel and the defect candidate pixel is a predetermined upper limit value. May be determined that the spatial change in the gray level value from a pixel around the defect candidate pixel to the defect candidate pixel exceeds the resolving power of the imaging optical system.

  According to the present invention, defect detection in which a captured image is inspected to detect a pattern defect on the surface of the sample is caused by a high-frequency gray level change that cannot be resolved by the imaging optical system of the imaging apparatus. Effectively suppress the generation of pseudo defects.

  The outline of the defect inspection method according to the present invention will be described below with reference to FIGS. 4 and 5A to 5C and FIG. 6A to FIG. FIG. 4 is a schematic diagram showing a picked-up image 100 obtained by picking up the wafer 2 as shown in FIG. 2, and each of the squares arranged in the x-direction and y-direction shown in the drawing indicates one pixel. Consider a case where a defect candidate pixel 101 is detected at the position of coordinates (x0, y0) by die-to-die comparison or cell-to-cell comparison as described with reference to FIG.

  FIG. 5A is a diagram showing the gray level values of each of the pixels around the pixel 101 of the first example detected as a defect candidate pixel. Here, the pixel 101 is centered on the pixel 101 and is parallel to the x-axis. The gray level values of the pixels arranged along the straight line (y = y0) are shown. FIG. 5B is a diagram showing the gray level value of each reference pixel compared with each pixel shown in FIG. 5A by die-to-die comparison or cell-to-cell comparison, and FIG. It is a figure which shows the gray level difference which is each difference with each pixel shown to (A) of FIG. 5, and each pixel shown to (B) of FIG. As shown in FIG. 5C, when the gray level difference between the pixel to be inspected and the reference pixel exceeds the detection threshold TH in the defect candidate pixel 101 at the coordinates (x0, y0), the pixel 101 becomes a defect candidate pixel. Has been detected.

FIG. 6A is a diagram showing the gray level value of each pixel around the pixel 101 of the second example detected as a defect candidate pixel, and the pixel 101 is the same as in FIG. The gray level value of each pixel arranged along a straight line (y = y0) parallel to the x-axis as the center is shown. FIG. 6B is a diagram showing the gray level value of each reference pixel compared with each pixel shown in FIG. 6A, and FIG. 6C is a diagram in FIG. 7 is a diagram showing a gray level difference which is a difference between each pixel shown in FIG. 6 and each pixel shown in FIG. As shown in FIG. 6C, when the gray level difference in the defect candidate pixel 101 exceeds the detection threshold TH, the pixel 101 is detected as a defect candidate pixel.
The spatial change of the gray level value that occurs from the pixels around the defect candidate pixel 101 in FIG. 6A to the defect candidate pixel 101 is steeper than that of the defect candidate pixel 101 in FIG. For example, if the gray level difference between adjacent pixels separated by the pixel pitch Dx is DGL, DGL in FIG. 6A> DGL in FIG.

  When the defect size indicated by the defect candidate pixel 101 detected in FIGS. 5A and 6A is smaller than the resolution limit of the imaging optical system of the imaging apparatus that captures the captured image 100. Originally, an image of such a defect is not formed on the image pickup device of the image pickup apparatus. Therefore, even if a defect detected as a defect having a size smaller than the resolution limit of the imaging optical system is a true defect, it changes relatively slowly as shown in FIG. It can be considered that only the pixel 101 at the vertex of the gray level value is detected as a defect candidate pixel. In this case, the gray level change from the defect candidate pixel 101 to the surrounding pixels is relatively gradual.

That is, when the size of the defect indicated by the defect candidate pixel 101 is smaller than the resolution limit of the imaging optical system of the imaging apparatus, only the gray level value of the defect candidate pixel 101 is the surrounding area as shown in FIG. It is impossible for the gray level values of the pixels to differ greatly. Therefore, it can be seen that such a gray level value of the defect candidate pixel 101 is not reliable.
Therefore, when the gray level change from the defect candidate pixel 101 to the surrounding pixels is steep, it can be determined that the defect candidate pixel 101 is a pseudo defect caused by image noise or the like.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. FIG. 7 is a block diagram showing a schematic configuration of the defect inspection apparatus according to the embodiment of the present invention. The defect inspection apparatus 1 has a configuration similar to the configuration described with reference to FIG. 1, and the same reference numerals are given to the same components and the description of the same functions is omitted.
The defect inspection apparatus 1 shown in FIG. 7 includes a pseudo defect removal unit 30 that inputs defect information generated by the defect information creation unit 24 and removes defect information related to pseudo defects from the input defect information.

When the defect information creating unit 24 creates defect information about the detected defect candidate including certain information such as position and size, at least a partial image including the defect candidate pixel and surrounding pixels, and the defect candidate The gray level difference between the reference pixel compared to the pixel and the defect candidate pixel is included in the defect information. Instead of the gray level difference, a difference image between the partial image including the defect candidate pixel and the surrounding pixels and the reference pixel compared with these pixels may be included in the defect information.
In the description of the defect inspection apparatus 1 shown in FIG. 1, the difference detection unit 21, the threshold comparison unit 23, and the defect information creation unit 24 compare the pixel to be inspected with two reference pixels and detect a gray level difference, respectively. I explained that double-protection will be performed. However, in order to determine that the pixel to be inspected is a defect candidate pixel, the pixel to be inspected may be compared with a plurality of reference pixels, and the number of reference pixels is not necessarily two. Therefore, the scope of the present invention is not limited to a defect inspection apparatus or a defect inspection method for comparing a pixel to be inspected with two reference pixels, and the number of reference pixels may be two or more.

  In addition, as the image pickup element used for the image pickup unit 13, a device in which image pickup pixels are arranged at a pitch finer than a resolution limit cycle on the image plane of the imaging optical system of the image pickup unit 13 is used. That is, the spatial frequency at the resolution limit of the imaging optical system of the imaging unit 13 is f, the period (1 / f) is T, and the magnification between the object and the image projected by this imaging optical system is m. In this case, an image pickup element of the image pickup unit 13 having a pixel pitch p smaller than T / m is used.

  FIG. 8 is a block diagram illustrating a configuration example of the pseudo defect removal unit 30 illustrated in FIG. 6. The pseudo defect removal unit 30 includes a pseudo defect detection unit 31 that determines whether each defect information input from the defect information creation unit 24 is a pseudo defect, and a pseudo defect detection unit based on each input defect information. The data selection part 32 which removes and outputs the defect information determined by 31 as a pseudo defect is provided.

  As shown in the figure, the pseudo defect detection unit 31 outputs the values output from the defect size determination unit 33, the gray level change value detection unit 34, the gray level difference extraction unit 35, and the gray level change value detection unit 34 to the gray level. A divider 36 that outputs a value divided by a value output from the difference extraction unit 35 and a comparator 37 that compares the output value of the divider 36 with a predetermined constant Δx are provided.

The defect size determination unit 33 extracts information on the size of the defect candidate from the input defect information, and determines whether the defect candidate size is equal to or smaller than a predetermined lower limit value THs. It is determined whether the size is smaller than the resolution limit of the imaging optical system of the imaging unit 13.
When the size of the defect candidate is equal to or smaller than a predetermined lower limit value THs, the gray level change value detection unit 34 extracts a partial image including the defect candidate pixel and surrounding pixels from the input defect information. Then, a change value of the gray level value from the pixel adjacent to the defect candidate pixel to the defect candidate pixel is detected.
When the size of the defect candidate is equal to or smaller than a predetermined lower limit value THs, the gray level difference extraction unit 35 determines the gray level difference between the reference pixel compared with the defect candidate pixel and the defect candidate pixel from the input defect information. To extract. When the defect information includes a difference image between the partial image including the defect candidate pixel and surrounding pixels and the reference pixel compared with these pixels, the gray level difference extraction unit 35 determines the reference pixel from the difference image. You may extract the gray level difference between defect candidate pixels.

The divider 36 calculates a gray level difference ratio R obtained by dividing the change value of the gray level value detected by the gray level change value detection unit 34 by the gray level difference extracted by the gray level difference extraction unit 35. The comparator 37 compares the calculated gray level difference ratio R with a predetermined constant Δx.
If the defect size of the defect candidate exceeds the predetermined lower limit value THs as a result of the determination by the defect size determination unit 33, the data selection unit 32 determines that the input defect information is a true defect and outputs it to the subsequent stage. To do.
When the defect size of the defect candidate is equal to or smaller than the predetermined lower limit value THs, the data selection unit 32 refers to the comparison result between the gray level difference ratio R and the predetermined constant Δx by the comparator 37, and determines the gray level difference ratio. If R exceeds a predetermined constant Δx, it is determined that the input defect information is a pseudo defect, and output to the subsequent stage is prohibited. If the gray level difference ratio R is equal to or less than a predetermined constant Δx, the input defect information is determined to be a true defect and output to the subsequent stage.

Hereinafter, the operation of the pseudo defect removal unit 30 will be described in detail with reference to FIG. FIG. 9 is a flowchart illustrating a defect inspection method according to an embodiment of the present invention. In step S <b> 1, the image of the surface pattern of the wafer 2 is captured by the imaging unit 13. The captured image is stored in the image storage unit 14.
In step S2, the difference detection unit 21 extracts the pixel to be inspected and the reference pixel from the captured image stored in the image storage unit 14, and detects a gray level difference between the two.
In step S3, the detection threshold value determination unit 22 determines a detection threshold value by predetermined statistical processing based on the distribution of gray level differences detected by the difference detection unit 21 for a plurality of pixels to be inspected.

  In step S <b> 4, the threshold comparison unit 23 compares the gray level difference input from the difference detection unit 21 with the detection threshold determined by the detection threshold determination unit 22, and detects a defect included in the inspection image by single detection. . The defect information creation unit 24 determines a defect candidate pixel by performing double detection. The defect information creation unit 24 creates defect information including certain information such as position and size regarding the detected defect candidate. At this time, the defect information creation unit 24 includes, in the defect information, a partial image including the defect candidate pixel and surrounding pixels, a gray level difference between the reference pixel and the defect candidate pixel compared with the defect candidate pixel, and a defect candidate. A difference image between a partial image including the pixel and surrounding pixels and a reference pixel compared with these pixels is included.

  In step S5, the defect size determination unit 33 illustrated in FIG. 8 determines whether the defect candidate may be a pseudo defect depending on whether the size of the defect candidate is smaller than the resolution limit of the imaging unit 13. Determine whether. For this reason, the defect size determination unit 33 extracts information on the dimensions of the defect candidates such as the x-direction length and the y-direction length from the defect information created by the defect information creation unit 24. If the size of the defect candidate is larger than the predetermined lower limit value THs, the defect size determination unit 33 determines that the defect related to this defect information is a true defect (step S6). The data selection unit 32 outputs the defect information determined to be a true defect to the subsequent stage.

  The lower limit value THs can be set as appropriate according to the known resolution limit period T of the imaging unit 13. In the present embodiment, since the imaging pixels of the imaging device of the imaging unit 13 are arranged at a pitch finer than the resolution limit cycle on the image plane of the imaging optical system of the imaging unit 13, the imaging unit 13 is arranged. The defect for one pixel of the image generated by is smaller than the resolution limit of the imaging unit 13. In the present embodiment, the defect size determination unit 33 determines that there is a possibility of a pseudo defect when the size of the defect candidate is 1 pixel, and proceeds with the process to step S7. (S6).

  In step S <b> 7, the gray level change value detection unit 34 extracts a partial image including a defect candidate pixel and surrounding pixels from the input defect information. Then, a change value GV of the gray level value from the pixel adjacent to the defect candidate pixel to the defect candidate pixel is detected. An example of a method for calculating the change value GV will be described with reference to FIG. FIG. 10A shows a partial image 100 including defect candidate pixels 101 extracted by the gray level change value detection unit 34 from the input defect information. Each of the squares arranged in the x-direction and the y-direction shown in the drawing represents one pixel.

The gray level value of the defect candidate pixel 101 is Gij, and the pixel values of the adjacent pixels 102a to 102d are G (i-1) j, Gi (j-1), G (i + 1) j, Gi (j + 1), respectively. Then, the gray level differences ΔGL1 to ΔGL4 between the defect candidate pixel 101 and the adjacent pixels 102a to 102d are as follows:
ΔGL1 = | Gij−G (i−1) j |
ΔGL2 = | Gij−Gi (j−1) |
ΔGL3 = | Gij−G (i + 1) j |
ΔGL4 = | Gij−Gi (j + 1) |
Is calculated by The gray level change value detection unit 34 determines the maximum value Max (ΔGL1, ΔGL2, ΔGL3, ΔGL4) among the gray level differences ΔGL1 to ΔGL4 as the change value GV.

The gray level difference extraction unit 35 extracts a gray level difference GLDiff between the reference pixel compared with the defect candidate pixel 101 and the defect candidate pixel 101 from the input defect information. The gray level difference extraction unit 35 calculates a reference pixel, a defect candidate pixel 101, and a defect candidate pixel 101 from a difference image between the defect candidate pixel 101 included in the defect information, a partial image including the surrounding pixels, and a reference pixel compared with these pixels. The gray level difference GLDiff may be extracted.
Then, the divider 36 calculates a gray level difference ratio R obtained by dividing the gray level value change value GV detected by the gray level change value detection unit 34 by the gray level difference GLDiff extracted by the gray level difference extraction unit 35.

  In step S8, the comparator 37 compares the calculated gray level difference ratio R with a predetermined constant Δx, and determines whether the adjacent pixels 102a to 102d depend on whether the gray level difference ratio R exceeds the predetermined constant Δx. To the defect candidate pixel 101, it is determined whether or not the spatial change of the gray level value exceeds the resolving power of the imaging optical system. If the gray level difference ratio R exceeds the predetermined constant Δx, it is determined that the defect candidate related to the input defect information is a pseudo defect (step S9), and the gray level difference ratio R is the predetermined constant Δx. If it is below, the defect candidate is determined to be a true defect (step S6).

  The predetermined constant Δx is, for example, a gray level change ΔGL1 between a defective pixel in which a sample that is known to have a defect in advance is imaged by the imaging unit 13 and a plurality of defective portions are imaged and its adjacent pixels, and the defective pixel It can be determined by determining the gray level difference ΔGL2 between the reference pixel and selecting the lower limit value among these ratios ΔGL1 / ΔGL2.

Further, the predetermined constant Δx is a threshold value for determining whether or not the contrast of the captured image exceeds the resolving power of the imaging optical system of the imaging unit 13, so that the predetermined constant Δx is set to the resolution limit frequency f of the imaging optical system of the imaging unit 13. Dependent. Therefore, the constant Δx may be obtained according to the following equation according to the resolution limit frequency f.
Δx = α × f + β (1)
In the above equation (1), α and β are correction coefficients determined according to the configuration of the imaging optical system of the imaging unit 13. For example, when the imaging unit 13 is a confocal microscope, the resolution of the image is high even at the same resolution limit frequency f, so α needs to be increased. When broadband illumination with a non-single wavelength is used as illumination light, the resolution limit frequency f is not uniquely determined, so adjustment is performed by α and β as correction coefficients.

  When it is determined that the defect information is a pseudo defect (S9), the data selection unit 32 prohibits the defect information from being output to the subsequent stage. If it is determined that the input defect information is a true defect (S6), the defect information is output to the subsequent stage.

  In step S7, the gray level change value detection unit 34 detects the change value of the gray level value from the adjacent pixel of the defect candidate pixel to the defect candidate pixel as the change value GV. You may detect the change of the gray level value between the pixel and the defect candidate pixel which were separated by predetermined distance. An example of a method for calculating such a change value GV will be described with reference to FIG. FIG. 10B shows a partial image 100 including defect candidate pixels 101 extracted by the gray level change value detection unit 34 from the input defect information. Each of the squares arranged in the x-direction and the y-direction shown in the drawing represents one pixel.

The gray level value of the defect candidate pixel 101 is Gij, and the pixel values of the pixels 102a to 102h that are separated from the defect candidate pixel 101 by a predetermined distance R are Ga to Gh, respectively. At this time, the predetermined distance R is set to a value smaller than the resolution limit period on the image plane of the imaging optical system of the imaging unit 13. That is, R <T / m.
Then, the gray level differences ΔGLa to ΔGLh between the defect candidate pixel 101 and the adjacent pixels 102a to 102h are expressed as ΔGLa = | Gij−Ga |, ΔGLb = | Gij−Gb |, ΔGLc = | Gij−Gc |, ΔGLd. = | Gij−Gd |, ΔGLe = | Gij−Ge |, ΔGLf = | Gij−Gf |, ΔGLg = | Gij−Gg |, ΔGLh = | Gij−Gh |. The gray level change value detector 34 determines the maximum value Max (ΔGLa, ΔGLb, ΔGLc, ΔGLd, ΔGLe, ΔGLf, ΔGLg, ΔGLh) among these gray level differences ΔGLa to ΔGLh as the change value GV.

  The present invention relates to a defect inspection apparatus for detecting defects in a pattern on a surface of a sample by inspecting a captured image obtained by imaging a pattern formed on the surface of the sample with a predetermined imaging device according to a predetermined inspection method. And can be used for defect inspection methods.

It is a block diagram which shows schematic structure of the conventional defect inspection apparatus. It is a figure which shows the arrangement | sequence of the die | dye of a semiconductor wafer. It is explanatory drawing of the die-to-die comparison performed with respect to the wafer shown in FIG. It is explanatory drawing (the 1) of the defect inspection method by this invention. (A)-(C) are explanatory drawings (the 2) of the defect inspection method by this invention. (A)-(C) are explanatory drawings (the 3) of the defect inspection method by this invention. It is a block diagram which shows schematic structure of the defect inspection apparatus by the Example of this invention. It is a block diagram which shows the structural example of the pseudo defect removal part shown in FIG. 3 is a flowchart illustrating a defect inspection method according to an embodiment of the present invention. (A) And (B) is a figure which shows the captured image containing a defect candidate pixel.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Defect inspection apparatus 2 Wafer 3, 3a-3c Die 11 Stage 12 Sample stand 13 Imaging part 14 Image storage part 20 Defect detection part 21 Difference detection part 22 Detection threshold value determination part 23 Threshold comparison part 24 Defect information creation part 30 Pseudo defect removal Part 31 Pseudo defect detection part

Claims (14)

A defect inspection apparatus that detects a pattern defect on the surface of a sample by inspecting a captured image obtained by imaging a pattern formed on the surface of the sample with a predetermined imaging device according to a predetermined inspection method. ,
The size of the defect candidate detected on the captured image is smaller than the resolution limit of the imaging optical system of the predetermined imaging device, and the pixels around the defect candidate pixel indicating the defect candidate to the defect candidate pixel A defect inspection apparatus comprising: a pseudo defect determination unit that determines a defect candidate as a pseudo defect when a spatial change in a gray level value exceeds a resolving power of the imaging optical system. The imaging device is configured to provide an electrical signal for an image formed on the image plane by an imaging device in which pixels are arranged at a pitch finer than a resolution limit period of the imaging optical system on the image plane of the imaging optical system. An imaging device for converting to
The pseudo defect detection unit includes a gray level change value detection unit that detects a gray level difference between the defect candidate pixel and a pixel adjacent thereto as a change value indicating the amount of the spatial change.
The defect inspection apparatus according to claim 1. The imaging device is configured to provide an electrical signal for an image formed on the image plane by an imaging device in which pixels are arranged at a pitch finer than a resolution limit period of the imaging optical system on the image plane of the imaging optical system. An imaging device for converting to
The pseudo defect detection unit detects, as a change value indicating the amount of spatial change, a gray level for detecting a maximum value of gray level differences between the defect candidate pixel and a plurality of adjacent pixels. A change value detector;
The defect inspection apparatus according to claim 1. The imaging device is configured to provide an electrical signal for an image formed on the image plane by an imaging element in which pixels are arranged at a pitch finer than a resolution limit period of the imaging optical system on the image plane of the imaging optical system. An imaging device for converting to
The pseudo defect detection unit uses the gray level difference between the defect candidate pixel and a pixel separated from the defect candidate pixel by a predetermined interval set smaller than the period as a change value indicating the amount of the spatial change. A gray level change value detection unit for detecting
The defect inspection apparatus according to claim 1. The imaging device is configured to provide an electrical signal for an image formed on the image plane by an imaging device in which pixels are arranged at a pitch finer than a resolution limit period of the imaging optical system on the image plane of the imaging optical system. An imaging device for converting to
The pseudo defect detection unit, as a change value indicating the amount of the spatial change, between a plurality of pixels separated from the defect candidate pixels by a predetermined interval smaller than the cycle and the defect candidate pixels, respectively A gray level change value detection unit for detecting a maximum value of the gray level differences of
The defect inspection apparatus according to claim 1.   In the captured image, a plurality of reference pixels that are expected to be the same as the pixel to be inspected are respectively compared with the pixel to be inspected, and each gray between the plurality of reference pixels and the pixel to be inspected is compared. 6. The defect according to claim 2, further comprising a defect detection unit that detects the pixel to be inspected as the defect candidate pixel when any of the level differences exceeds a predetermined detection threshold. Inspection device.   The pseudo defect detection unit has a predetermined gray level difference ratio obtained by dividing the change value by a gray level difference between any one of the plurality of reference pixels compared to the defect candidate pixel and the defect candidate pixel. When the upper limit value of the defect candidate pixel is exceeded, it is determined that the spatial change in the gray level value from the surrounding pixels of the defect candidate pixel to the defect candidate pixel exceeds the resolving power of the imaging optical system. The defect inspection apparatus according to claim 6. A defect inspection method for detecting defects in a pattern on a surface of a sample by inspecting a captured image obtained by imaging a pattern formed on the surface of the sample with a predetermined imaging device according to a predetermined inspection method. ,
The size of the defect candidate detected on the captured image is smaller than the resolution limit of the imaging optical system of the predetermined imaging device, and the pixels around the defect candidate pixel indicating the defect candidate to the defect candidate pixel A defect inspection method characterized in that, when a spatial change in the gray level value exceeds the resolving power of the imaging optical system, the defect candidate is determined as a pseudo defect. The imaging device is configured to provide an electrical signal for an image formed on the image plane by an imaging device in which pixels are arranged at a pitch finer than a resolution limit period of the imaging optical system on the image plane of the imaging optical system. An imaging device for converting to
When detecting the pseudo defect, a gray level difference between the defect candidate pixel and a pixel adjacent thereto is detected as a change value indicating the amount of the spatial change.
The defect inspection method according to claim 8. The imaging device is configured to provide an electrical signal for an image formed on the image plane by an imaging device in which pixels are arranged at a pitch finer than a resolution limit period of the imaging optical system on the image plane of the imaging optical system. An imaging device for converting to
When detecting the pseudo defect, a maximum value of gray level differences between the defect candidate pixel and a plurality of adjacent pixels is detected as a change value indicating the amount of the spatial change. ,
The defect inspection method according to claim 8. The imaging device is configured to provide an electrical signal for an image formed on the image plane by an imaging device in which pixels are arranged at a pitch finer than a resolution limit period of the imaging optical system on the image plane of the imaging optical system. An imaging device for converting to
When detecting the pseudo-defect, a gray level between a pixel that is separated from the defect candidate pixel by a predetermined interval smaller than the period and a gray level between the defect candidate pixel is used as a change value indicating the amount of the spatial change. Detect level difference,
The defect inspection method according to claim 8. The imaging device is configured to provide an electrical signal for an image formed on the image plane by an imaging device in which pixels are arranged at a pitch finer than a resolution limit period of the imaging optical system on the image plane of the imaging optical system. An imaging device for converting to
When detecting the pseudo defect, the change value indicating the amount of the spatial change is between a plurality of pixels separated from the defect candidate pixel by a predetermined interval set smaller than the period and the defect candidate pixel. Find the maximum of each gray level difference
The defect inspection method according to claim 8.   When detecting the defect candidate pixels, the plurality of reference pixels that are expected to be the same as the pixel to be inspected in the captured image are respectively compared with the pixel to be inspected, and the plurality of reference pixels 13. The pixel to be inspected is detected as the defect candidate pixel when any of the gray level differences between the pixel to be inspected exceeds a predetermined detection threshold value. The defect inspection method according to item.   When detecting the pseudo defect, a gray level difference ratio obtained by dividing the change value by a gray level difference between any one of the plurality of reference pixels compared to the defect candidate pixel and the defect candidate pixel is obtained. When the predetermined upper limit value is exceeded, it is determined that the spatial change in the gray level value from the surrounding pixels of the defect candidate pixel to the defect candidate pixel exceeds the resolving power of the imaging optical system. The defect inspection method according to claim 13.

Images