JP2010043941A - Image inspection apparatus and image inspection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce influence of noise caused by irregularities in brightness appearing in a captured image in an image inspection for detecting defects existing on the surface of an inspection target by inspecting an inspection image generated by a prescribed image generation means for the inspection target. <P>SOLUTION: An image inspection apparatus 1 includes: an inspection pixel 60 to be determined for the presence or absence of defects in a prescribed image 51 relating to the inspection target 2; minimum gray level difference detection sections 21, 22, 25, 26 for detecting the minimum gray level difference, namely the gray level difference between pixels having the closest gray level value to that of the inspection pixel 60 in a plurality of pixels 61-68 around the inspection pixel 60; and a defect detection section 24 for determining that a defect exists at the position of the pixel 60 when the minimum gray level difference exceeds a prescribed detection threshold. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image inspection apparatus and an image inspection method for detecting a defect present on a surface of an inspection object by inspecting an inspection image generated by a predetermined image generation unit for the inspection object. More specifically, the present invention relates to a technique for reducing the influence of noise existing in an inspection image in such an image inspection.

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 present on the surface of a sample by inspecting the surface is widely performed (for example, Patent Document 1 below).
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 is a diagram showing an array of dies (chips) on a semiconductor wafer (hereinafter referred to as “wafer”) to be inspected (sample). A plurality of dies 3 are repeatedly arranged in a matrix on the wafer 2 in the X and Y directions. Since the same pattern is formed on each die, the images obtained by imaging these dies should be essentially the same, and the pixel values of the corresponding portions of the captured images of each die are essentially the same values.

Therefore, the two dies are imaged to obtain the pixel value (gray level value) of each pixel, and the pixel value difference (gray level difference signal) between corresponding portions that should be essentially the same in the captured images of the two dies. Is detected, the gray level difference signal is greater when one die is defective than when both dies are not defective.
Therefore, using one of the two dies as an inspection image and the other as a reference image, the gray level difference between corresponding pixels between these images is detected, and this gray level difference is compared with a predetermined detection threshold. Thus, defects existing on the die can be detected by detecting a gray level difference exceeding the detection threshold (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 between images obtained by imaging corresponding portions that should be the same in the repeated pattern is detected. 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 a die adjacent to a different side, and it is determined that the die is defective when the gray level difference of the same portion becomes larger than a predetermined detection threshold again (double detection). The same applies to the cell-to-cell comparison.

JP 2004-177397 A

When the gray level value irregularity occurs in the optical image of the sample to be inspected, the gray level difference between the corresponding pixels that should be essentially the same in the captured image tends to increase. The number of detected defects (pseudo defects) increases.
In addition, due to the miniaturization of semiconductor circuit patterns, the pattern cannot often be resolved even if the surface of a sample to be inspected is optically imaged. In an image obtained by imaging a pattern that cannot be resolved, random gray level value irregularities appear, so that the number of detected pseudo defects increases.

In view of the above problems, in the present invention, gray images appearing in a captured image in an image inspection for detecting defects present on the surface of the inspection object by inspecting the inspection image generated by a predetermined image generation unit for the inspection object. Reduces the effects of noise caused by uneven level values.
In the present invention, even if the pattern of the surface to be inspected is not resolved in the captured image, a minute defect that appears in the captured image is detected.

  In the present invention, in an inspection image generated by a predetermined image generation unit for an inspection object, an inspection pixel that is a target of defect determination in the inspection image and an inspection among a plurality of pixels around the inspection pixel When a minimum gray level difference, which is a gray level difference between a pixel closest to the pixel and a gray level value, is detected and the minimum gray level difference exceeds a predetermined detection threshold, there is a defect at the position indicated by the inspection pixel. judge.

As the inspection image, a captured image to be inspected can be used. Usually, in an image having no defect, a pixel having a gray level value comparable to the gray level value of the pixel exists near a certain pixel. For example, in an image in which the pattern on the surface of the sample is not resolved, the change in the gray level value caused by the surface pattern is gradual, and this tendency is further increased.
Therefore, an appropriate number of pixels around a certain pixel (inspection pixel) in the image, for example, an appropriate distance from the inspection pixel, are selected, and the gray level value of each selected pixel is selected. Can be considered to have a minute defect at the position of the inspection pixel.

Further, when a known repetitive pattern is formed on the surface of the inspection target, each pixel in the captured image to be inspected as a test image, and a pixel corresponding to each pixel in the repetitive pattern appearing in the captured image, A gray level difference image having a gray level difference between the pixel values as pixel values at each pixel position may be used. Such a gray level difference image may be, for example, a difference image between adjacent dies in a die-to-die comparison, and may be, for example, a difference image between adjacent cells in a cell-to-cell comparison.
When the above gray level difference image is generated based on a captured image with unevenness, an image with uneven gray level values is obtained. On such a gray level difference image, when a pixel exceeding a predetermined threshold is detected as a defect as in the conventional image inspection, pseudo defects increase due to uneven gray level values.

  Since the unevenness of the gray level value of the captured image is a continuous change of the gray level value, the gray level difference which is a difference between them also appears as a continuous change. Therefore, when a gray level difference image is used as an inspection image, there are pixels having the same gray level value around inspection pixels that are not defective in the inspection image. For this reason, when the gray level value of a certain inspection pixel is greatly different from the gray level values of surrounding pixels, it can be considered that there is a minute defect at the position of this inspection pixel.

In the above defect detection, there is a possibility that a defect that is large to some extent cannot be detected. Because, if the defect is large, even if the pixel to be inspected is at the position of this defect, the position of the pixel that is a predetermined distance away from this pixel is also in this defect, and the gray level between these pixels This is because the difference does not increase and does not exceed the detection threshold.
Therefore, in the present invention, together with the first defect detection described above, an average gray level value of a plurality of pixels distributed in a predetermined range in the inspection image is calculated, and a difference between the inspection pixel and the average gray level value is detected, When this difference exceeds a predetermined detection threshold, second defect detection is performed in which it is determined that a defect exists at the position indicated by the inspection pixel.
In this second defect detection, a defect having a certain size can be detected if the defect is sufficiently small compared to a predetermined range in which pixels used for calculating the average gray level value are selected. It is possible to supplement the first defect detection that detects only a simple defect.

According to the present invention, the present invention reduces the influence of noise caused by uneven gray level values appearing in a captured image.
Even if the surface pattern of the inspection object is not resolved in the captured image, it is possible to detect a minute defect that appears in the captured image.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. FIG. 2 is a block diagram of the defect inspection apparatus according to the first embodiment of the present invention. 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 wafer 2 as a sample to be inspected is placed on the sample stage 12 and fixed.

  Above the sample stage 12, an imaging unit 14 for capturing an optical image of the surface of the wafer 2 is provided. As the imaging unit 14, an image sensor such as a one-dimensional or two-dimensional CCD camera (preferably a TDI camera) is used, and an optical image of the surface of the wafer 2 formed on the light receiving surface is converted into an electric signal. In this configuration example, a one-dimensional TDI camera is used as the imaging unit 14. By moving the imaging unit 14 and the wafer 2 relatively by moving the stage 11, the imaging unit 14 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. Note that either the bright-field illumination optical system or the dark-field illumination optical system may be used for capturing an image of the surface of the wafer 2 by the imaging unit 14. The image signal output from the imaging unit 14 is converted into a multi-value digital signal (gray level signal), and then stored in the image memory 15 as an inspection image.

  Further, the defect inspection apparatus 1 performs image processing on the inspection image stored in the image memory 15 using different algorithms, and first defect inspection for detecting defects on the surface of the wafer 2 appearing in the inspection image. Unit 20 and second defect inspection unit 30, and defect information combining unit 40 that combines and outputs defect information of the defect detected by first defect inspection unit 20 and defect information of the defect detected by second defect inspection unit 30. And comprising.

The first defect inspection unit 20 and the second defect inspection unit 30 determine whether or not there is a defect at each pixel position for each pixel constituting the inspection image stored in the image memory 15; When it is determined that a defect exists, defect information indicating the position of the defect is generated. In the following description, one pixel to be determined when the first defect inspection unit 20 and the second defect inspection unit 30 determine whether or not each pixel has a defect is referred to as “inspection pixel”. I write.
The first defect inspection unit 20 includes a gray level difference detection unit 21, a gray level difference selection unit 22, a first detection threshold value determination unit 23, and a first defect detection unit 24.

The gray level difference detection unit 21 detects a gray level difference between the inspection pixel and M pixels separated from the inspection pixel by a predetermined number R of pixels.
The gray level difference selection unit 22 selects the minimum gray level difference detected by the gray level difference detection unit 21 as the minimum gray level difference. Then, by disposing each pixel having the minimum gray level difference detected for each inspection pixel in the inspection image as a gray level value at a position corresponding to the position in the inspection image of each inspection pixel, the first The difference image is generated.

The first detection threshold value determination unit 23 determines a first detection threshold value by performing predetermined statistical processing on each pixel included in the first difference image.
The first defect detection unit 24 compares each pixel included in the first difference image with the first detection threshold value. When a certain pixel in the first difference image exceeds the first detection threshold, it is detected that a defect exists at the position indicated by the inspection pixel in the inspection image corresponding to the position of this pixel in the first difference image. To do. The first defect detection unit 24 that has detected the defect generates defect information indicating the position of the detected defect.

The second defect inspection unit 30 includes an average gray level calculation unit 31, a difference detection unit 32, a second detection threshold value determination unit 33, and a second defect detection unit 34.
The average gray level calculation unit 31 calculates an average gray level value of a plurality of pixels distributed in a predetermined range in the inspection image.
The difference detection unit 32 detects a difference between the gray level value of the inspection pixel and the average gray level value. The difference detection unit 32 arranges each pixel having the difference detected for each inspection pixel in the inspection image as a gray level value at a position corresponding to the position in the inspection image of each inspection pixel. A second difference image is generated. The average gray level calculation unit 31 calculates an average gray level value for detecting a difference from each inspection pixel for each inspection pixel.

The second detection threshold value determination unit 33 determines a second detection threshold value by performing predetermined statistical processing on each pixel included in the second difference image.
The second defect detection unit 34 compares each pixel included in the second difference image with the second detection threshold value. When a certain pixel in the second difference image exceeds the second detection threshold, it is detected that there is a defect at the position indicated by the inspection pixel in the inspection image corresponding to the position of this pixel in the second difference image. To do. The second defect detector 34 generates defect information indicating the position of the detected defect.

The operation of the defect inspection apparatus 1 according to the first embodiment of the present invention will be described in more detail with reference to the flowchart shown in FIG. FIG. 3 is an overall flowchart of processing executed by the defect inspection apparatus 1 according to the first embodiment of the present invention and the second embodiment described later.
In step S1, the imaging unit 14 which is a one-dimensional TDI camera is scanned to obtain a two-dimensional image of the surface of the wafer 2. FIG. 4 is a diagram illustrating a TDI sensor 14 that images the surface of the wafer 2 and a captured image 51 captured by the TDI sensor 14.

  The one-dimensional TDI camera 14 is realized by scanning a subject to be photographed in the same direction as one of the two-dimensional array directions of the image pickup pixels by a camera having a two-dimensional CCD as an image sensor. The electric signals obtained by the plurality of imaging pixels arranged in the same direction as the scanning direction by transferring the electric charge obtained by photoelectric conversion in each imaging element in the same direction as the scanning direction in synchronization with the scanning speed To generate one-dimensional image information in which pixels are arranged in a line direction orthogonal to the scanning direction. In the following explanation and usage of terms in the claims, “scanning direction” used for the one-dimensional TDI camera 14 means a scanning direction when the subject is scanned by the one-dimensional TDI camera 14. The “line direction” means a direction orthogonal to the scanning direction.

4 shows a state in which the surface of the wafer 2 on which the dies 3a to 3c,... Are formed is scanned by the one-dimensional TDI camera 14 in the scanning direction shown in the figure, and scanning is performed by scanning the one-dimensional TDI camera 14. A picked-up image of a portion of the region 50 on the wafer 2 is indicated by reference numeral 51. Note that the captured image 51 illustrated in FIG. 4 is a schematic diagram, and in the case of an actual captured image, the pattern formed on the wafer 2 is not necessarily resolved as illustrated in FIG. 3.
In the usage of the terms in the following description and claims, the “column direction” used for the two-dimensional captured image 51 captured by the one-dimensional TDI camera 14 refers to imaging arranged in the line direction of the one-dimensional TDI camera 14. It means the direction in which the pixels respectively captured by the pixels are arranged, and the “row direction” used for the two-dimensional captured image 51 means a direction orthogonal to the “column direction”.

Returning to FIG. 3, in step S <b> 2, the first defect inspection unit 20 has a defect at the position of each inspection pixel for each inspection pixel constituting the captured image 51 (inspection image) stored in the image memory 15. If it is determined whether a defect exists, defect information indicating the position of the defect is generated.
FIG. 5 is a flowchart showing a first example of processing performed in the first defect detection processing step S2 shown in FIG. 3, and FIG. 6 is an explanatory diagram of the processing shown in FIG.
In the repetition loop including Steps S11 to S14, the first difference image is generated by repeating Steps S12 and S13 for each of the pixels of the inspection image, with each pixel being the inspection pixel.

  In step S12, the gray level difference detection unit 21 detects a gray level difference between a certain inspection pixel in the inspection image and M pixels separated from the inspection pixel by a predetermined distance R pixels. For example, in the example of the inspection image 51 illustrated in FIG. 6, the eight pixels 61 to 68 that are separated from the inspection pixel 60 by R pixels in the diagonal directions at angles of 45, 135, 225, and 315 degrees. Each gray level difference between is detected. The value of the integer R may be set to 3 to 5 pixels, for example.

In step S13, the gray level difference selection unit 22 selects the smallest one as the minimum gray level difference from the M gray level differences detected by the gray level difference detection unit 21 in step S12. Then, the gray level difference selection unit 22 sets the pixel value of the pixel at the position in the first difference image corresponding to the position of the inspection pixel 60 in the inspection image 51 to the value of the minimum gray level difference.
The first difference image is generated by repeating the above process for each of all the pixels of the inspection image by the repetition loop including Step S11 to Step S14 shown in FIG.

  In step S15, the first detection threshold value determination unit 23 determines a first detection threshold value by performing predetermined statistical processing on the first difference image. FIG. 7 is a flowchart for explaining the processing content of the detection threshold value determination step S15 of FIG. 5, and FIG. 8 is an explanatory diagram of the detection threshold value determination process shown in FIG. Note that the detection threshold value determination process described below with reference to FIGS. 7 and 8 is merely an example, and the detection threshold value determination process used in the present invention is not limited to this.

In step S21, the first detection threshold value determination unit 23 inputs the first difference image, and in step S22, creates a histogram of gray level values of pixels included in the image as shown in FIG. 8A. . In step S23, the first detection threshold value determination unit 23 calculates the cumulative frequency of gray level values from this histogram.
In step S24, the first detection threshold value determining unit 23 assumes that the gray level value follows a certain distribution, and the cumulative frequency calculated in step S23 is linearly related to the gray level value. Convert. At this time, the cumulative frequency is converted on the assumption that the gray level value follows a certain distribution such as a normal distribution, a Poisson distribution, or a chi-square distribution. This conversion cumulative frequency is shown in FIG.

In step S25, an approximate straight line (y = ax + b) indicating the relationship between the gray level value and the conversion cumulative frequency is derived according to the conversion cumulative frequency converted in step S24 (see (C) in FIG. 8).
In step S26, the detection threshold Th is determined from the parameters a and b of the approximate line and the sensitivity setting parameter (fixed value).

Here, VOP and HO are set as fixed sensitivity setting parameters in the approximate straight line of the gray level value and the conversion cumulative frequency, and the cumulative frequency P1 corresponding to the cumulative probability (p) (p is multiplied by the number of samples). )) Is obtained, and the gray level value obtained by moving VOP in the vertical axis direction and HO in the horizontal axis direction from that point is set as the detection threshold Th. Therefore, the detection threshold Th is a predetermined calculation formula Th = (P1−b + VOP) / a + HO (1)
Is calculated by

  Returning to FIG. 5, in step S <b> 16, the first defect detection unit 24 compares each pixel included in the first difference image with the first detection threshold. When a certain pixel in the first difference image exceeds the first detection threshold, a defect is present at the position indicated by the pixel at the position in the inspection image 51 corresponding to the position of this pixel in the first difference image. Detect if it exists. The first defect detection unit 24 generates defect information indicating the position of the detected defect.

  Since each pixel included in the first difference image is a minimum value of a gray level difference between a certain inspection pixel and pixels around the inspection image by R pixels (R is about 3 to 5), When a pixel having a large gray level value in the first difference image is detected in step S16, a pixel having a gray level value different from any of surrounding pixels separated by R pixels can be detected. Therefore, a minute defect having a pixel size smaller than 2R pixels can be detected by this method.

In step S <b> 3 shown in FIG. 3, the second defect inspection unit 30 determines whether or not there is a defect at the position of each inspection pixel for each inspection pixel constituting the inspection image stored in the image memory 15. If it is determined that a defect exists, defect information indicating the position of the defect is generated.
FIG. 9 is a flowchart showing the processing performed in the second defect detection processing step S3 shown in FIG. 3, and FIG. 10 is an explanatory diagram of the processing shown in FIG.
In the repetition loop composed of steps S31 to S34, the second difference image is generated by repeating steps S32 and S33 for each of the pixels of the inspection image, with each pixel being the inspection pixel.

In step S <b> 32, the average gray level calculation unit 31 calculates the average gray level value of the gray level values of the N pixels 69 arranged in the same column as the inspection pixel 60.
In step S33, the difference detection unit 32 calculates the difference between the average gray level value calculated by the average gray level calculation unit 31 in step S32 and the inspection pixel. Then, the difference detection unit 32 sets the pixel value of the pixel at the position in the second difference image corresponding to the position of the inspection pixel 60 in the inspection image 51 to the calculated difference value.
The second difference image is generated by repeating the above process for each of all the pixels of the inspection image by the repetition loop including steps S31 to S34 shown in FIG.

In step S35, the second detection threshold value determination unit 33 determines the second detection threshold value by performing predetermined statistical processing on the second difference image. The determination of the detection threshold by the second detection threshold determination unit 33 is the same as the detection threshold determination processing illustrated with reference to FIGS. 7 and 8 except that the second difference image is used instead of the first difference image. Since it can be performed, description is omitted to avoid duplication.
Returning to FIG. 9, in step S <b> 36, the second defect detection unit 34 compares each pixel included in the second difference image with the second detection threshold. When a certain pixel in the second difference image exceeds the second detection threshold, a defect is present at the position indicated by the pixel in the inspection image 51 corresponding to the position of this pixel in the second difference image. Detect if it exists. The second defect detector 34 generates defect information indicating the position of the detected defect.

  Each pixel included in the second difference image is a gray level difference between the average value of the gray level value of the pixel row of N pixels that is somewhat long and the inspection pixel. Therefore, if the size of the integer N is set to a sufficiently large value (for example, N = 1000 or more), the first defect is detected by detecting a pixel having a large gray level value in the second difference image. A defect larger than 2R pixels that cannot be detected by the inspection unit 20 can be detected.

In step S <b> 32, the average gray level calculation unit 31 calculates the average gray level value of the gray level values of the pixels 69 arranged in the same column as the inspection pixel 60. The reason is as follows.
The gray level signal of another pixel 69 arranged in the same column as the inspection pixel 60 is generated by the same imaging pixel as the imaging pixel of the one-dimensional TDI camera 14 that generates the gray level signal of the inspection pixel 60. Therefore, the signal characteristics such as the S / N ratio of the gray level signal of these pixels 69 are the same as those of the inspection pixel 60.
For this reason, when the difference detection unit 32 detects the difference signal between the average gray level value of the pixel 69 and the inspection pixel 60 in step S33, the difference signal is affected by the difference between the imaging pixels that generated these pixel signals. It is avoided that it occurs.

  In step S4 shown in FIG. 3, the defect information combining unit 40 combines defect defect information detected by the first defect inspection unit 20 and defect information detected by the second defect inspection unit 30, and these defect information. Are both output.

A difference between the first difference image and the second difference image generated by the present invention and a difference image obtained by conventional cell-to-cell comparison will be described with reference to FIGS.
11A is the original captured image used for generating the difference image, and FIG. 11B is the difference obtained by performing the conventional cell-to-cell comparison using the captured image of FIG. FIG. 11C is a first difference image according to the present invention, and FIG. 11D is a second difference image according to the present invention.
Comparing the S / N ratios of the images in (B) to (D) of FIG. 11, if the magnitude of the signal value is the maximum gray level and the magnitude of the noise is defined as 6σ, The S / N ratio of the difference image by the conventional cell-to-cell comparison shown in FIG. 11B is 0.83, and the S / N ratio of the first difference image of FIG. ) Of the second difference image was 2.07.
From the results of this experiment, it can be seen that noise generated in the difference image used for defect detection according to the present invention is effectively suppressed, and defect detection with higher accuracy than before can be performed.

FIG. 12 is a block diagram of a defect inspection apparatus according to the second embodiment of the present invention. The difference between the present embodiment and the first embodiment is the difference in the configuration of the first defect inspection unit 20, and there is a difference in the method of generating the first difference image.
The first defect inspection unit 20 according to the present embodiment includes a reference pixel selection unit 25 and a gray level difference detection unit 26 instead of the gray level difference detection unit 21 and the gray level difference selection unit 22 of the first example.

The reference pixel selection unit 25 selects, as a reference pixel, a pixel having the closest gray level value to the inspection pixel 60 among the plurality of pixels 61 to 68 at a predetermined distance R from the inspection pixel 60 in the inspection image 51 illustrated in FIG. To do.
The gray level difference detection unit 26 detects the gray level difference between the inspection pixel 60 and the reference pixel as the minimum gray level difference, and corresponds to the position of the inspection pixel 60 in the inspection image 51. Is set to the value of the minimum gray level difference.
FIG. 13 is a flowchart showing a second example of processing performed in the first defect detection processing step shown in FIG. The difference from the first example shown in FIG. 5 is the first difference image generation process in steps S41 to S44.

In the repetition loop including Steps S41 to S44, the first difference image is generated by repeating Steps S42 and S43 for each of the pixels of the inspection image, using each pixel as the inspection pixel.
In step S <b> 42, the reference pixel selection unit 25 sets the reference pixel having the closest gray level value to the inspection pixel 60 among the plurality of pixels 61 to 68 at a predetermined distance R from the inspection pixel 60 in the inspection image 51 illustrated in FIG. 6. Select as a pixel.

In step S43, the gray level difference detection unit 26 detects the gray level difference between the inspection pixel 60 and the reference pixel as the minimum gray level difference, and corresponds to the position of the inspection pixel 60 in the inspection image 51. The pixel value of the pixel at the position in the difference image is set to the value of the minimum gray level difference.
The same difference image as the first difference image generated in the first embodiment is generated by repeating the above processing for each of all the pixels of the inspection image by the repetition loop composed of steps S41 to S44 shown in FIG. Is done.

FIG. 14 is a block diagram of a defect inspection apparatus according to a third embodiment of the present invention. This embodiment is a modification of the first embodiment described above, and includes a gray level difference image generation unit 50 and a gray level difference image holding unit 51 in addition to the configuration of the first embodiment.
The gray level difference image generation unit 50 compares the images of adjacent dies in the captured image of the surface of the wafer 2 stored in the image memory 15, and compares the gray level difference between the corresponding pixels of the adjacent die images. Is generated as a pixel value. Alternatively, the gray level difference image generation unit 50 compares the images of the adjacent cells in the captured image of the surface of the wafer 2 stored in the image memory 15 and compares the gray level between the corresponding pixels of the adjacent cell images. A gray level difference image having a level difference as a pixel value is generated.
The gray level difference image holding unit 51 holds the gray level difference image generated by the gray level difference image generation unit 50.

FIG. 15 is an explanatory diagram of processing for generating a gray level difference image by the gray level difference image generation unit 50. In this example, a case where a gray level difference image is generated by die-to-die comparison will be described. However, even when a gray level difference image is generated by cell-to-cell comparison, the difference is the same except that the comparison target is a cell. .
The gray level difference image generation unit 50 corresponds to each other among the pixels of the partial images It and Ir in the range of the adjacent dies 3a and 3b in the captured image 51 of the surface of the wafer 2 stored in the image memory 15. That is, the difference GL0 (P01−P02) between the gray level values (P01 and P02) of the pixels P1 and P2 having the same in-die coordinates is calculated.
The gray level difference image generation unit 50 calculates the gray level difference GL0 for each pixel in the partial image It, and the gray level value of the pixel at the same position as the position (Xt, Yt) of each pixel P1 in the original image is calculated. An image having the gray level difference GL0 calculated for each pixel P1 is generated as a gray level difference image.

  The first defect inspection unit 20 and the second defect inspection unit 30 perform image processing on the inspection image using the gray level difference image held in the gray level difference image holding unit 51 as an inspection image, and the wafer 2 appearing in the inspection image Each surface defect is detected. The processes performed by the first defect inspection unit 20 and the second defect inspection unit 30 are the same as the respective processes in the first embodiment.

FIG. 16 is an overall flowchart of processing executed by the defect inspection apparatus according to the third embodiment of the present invention and the fourth embodiment described later.
In step S1, a two-dimensional image of the surface of the wafer 2 is obtained as in step S1 of the flowchart shown in FIG.
In step S <b> 50, the gray level difference image generation unit 50 compares the images of adjacent dies or cells in the captured image of the surface of the wafer 2 stored in the image memory 15 to generate a gray level difference image. The gray level difference image is held in the gray level difference image holding unit 51.
The subsequent steps S2 to S4 are performed except that the inspection images processed by the first defect inspection unit 20 and the second defect inspection unit 30 are gray level difference images held in the gray level difference image holding unit 51. Thus, the processing is the same as the processing in steps S2 to S4 shown in FIG.

  FIG. 17 is a block diagram of a defect inspection apparatus according to the fourth embodiment of the present invention. This embodiment is a modification of the above-described second embodiment, and includes the above-described gray level difference image generation unit 50 and gray level difference image holding unit 51 in addition to the configuration of the second embodiment. The present embodiment is the second embodiment described above except that the inspection images processed by the first defect inspection unit 20 and the second defect inspection unit 30 are gray level difference images held in the gray level difference image holding unit 51. Similar to the example.

  The present invention can be used for an image inspection apparatus and an image inspection method for detecting a defect present on the surface of an inspection object by inspecting an inspection image generated by a predetermined image generation unit for the inspection object.

It is a figure which shows the arrangement | sequence of the die | dye on a semiconductor wafer. 1 is a block diagram of a defect inspection apparatus according to a first embodiment of the present invention. It is a whole flowchart of the process performed by the defect inspection apparatus by the 1st and 2nd Example of this invention. It is a figure explaining the relationship between the line direction and scanning direction of a TDI sensor which images a wafer surface, and the direction of a captured image. It is a flowchart which shows the 1st example of the process performed at the 1st defect detection process step shown in FIG. It is explanatory drawing of the process shown in FIG. It is a flowchart explaining a detection threshold value determination process. It is explanatory drawing of the detection threshold value determination process shown in FIG. It is a flowchart which shows the process performed in the 2nd defect detection process step shown in FIG. It is explanatory drawing of the process shown in FIG. (A) is a captured image, (B) is a difference image obtained by performing a cell-to-cell comparison using the captured image of (A), (C) is a first difference image according to the present invention, (D ) Is a second difference image according to the present invention. It is a block diagram of the defect inspection apparatus by 2nd Example of this invention. It is a flowchart which shows the 2nd example of the process performed in the 1st defect detection process step shown in FIG. It is a block diagram of the defect inspection apparatus by 3rd Example of this invention. It is explanatory drawing of the process which produces | generates a gray level difference image. It is a whole flowchart of the process performed by the defect inspection apparatus by 3rd and 4th Example of this invention. It is a block diagram of the defect inspection apparatus by 4th Example of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Defect inspection apparatus 2 Wafer 20 1st defect inspection part 21 Gray level difference detection part 22 Gray level difference selection part 23 1st detection threshold value determination part 24 1st defect detection part 30 2nd defect inspection part 31 Average gray level calculation part 32 Difference detection unit 33 Second detection threshold value determination unit 34 Second defect detection unit

Claims (16)

An image inspection apparatus that detects defects present on the surface of an inspection object by inspecting an inspection image generated by a predetermined image generation unit for the inspection object,
A gray level difference between an inspection pixel that is a pixel for which a defect is determined in the inspection image and a pixel that is closest to the inspection pixel among a plurality of pixels around the inspection pixel. A minimum gray level difference detection unit for detecting a certain minimum gray level difference;
A defect detection unit that determines that a defect exists at a position indicated by the inspection pixel when the minimum gray level difference exceeds a predetermined detection threshold;
An image inspection apparatus comprising: The minimum gray level difference detection unit includes:
A gray level difference detection unit for respectively detecting a gray level difference between the inspection pixel and the plurality of pixels around the inspection pixel;
A gray level difference selection unit that selects the minimum gray level difference detected by the gray level difference detection unit as the minimum gray level difference; and
The image inspection apparatus according to claim 1, further comprising: The minimum gray level difference detection unit includes:
A reference pixel selection unit that selects, as a reference pixel, a pixel having a gray level value closest to the inspection pixel among the plurality of pixels around the inspection pixel;
A gray level difference detection unit for detecting a gray level difference between the inspection pixel and the reference pixel as the minimum gray level difference;
The image inspection apparatus according to claim 1, further comprising:   The image inspection apparatus according to claim 1, wherein the inspection image is a captured image of the inspection target. The inspection object has a known repeating pattern on its surface,
The inspection image has a gray level difference between each pixel in the captured image to be inspected and a pixel corresponding to each pixel in a repetitive pattern appearing in the captured image at a position of each pixel. The image inspection apparatus according to claim 1, wherein the image inspection apparatus is a gray level difference image having a pixel value. The inspection pixel, the detection threshold, and the defect detection unit are a first inspection pixel, a first detection threshold, and a first defect detection unit, respectively.
An average gray level calculation unit for calculating an average gray level value of a plurality of pixels distributed in a predetermined range in the inspection image;
A difference detection unit that detects a difference between a gray level value of a second inspection pixel that is a pixel to be subjected to defect determination in the inspection image and the average gray level value;
When the difference exceeds a predetermined second detection threshold, a second defect detection unit that determines that a defect is present at a position indicated by the second inspection pixel;
The image inspection apparatus according to claim 1, further comprising: The predetermined image generation means includes a one-dimensional line sensor for imaging the inspection object, and the inspection object is scanned by scanning the inspection object in a direction orthogonal to the line direction by the one-dimensional line sensor. Image
Pixels in the same row aligned in the column direction in the inspection image are pixels captured by the respective imaging pixels of the one-dimensional line sensor aligned in the line direction, and pixels in the same column aligned in the row direction in the inspection image are the 1 When the same image pickup pixel of the two-dimensional line sensor is a pixel picked up by scanning in a direction orthogonal to the line direction, the average gray level calculation unit includes the second of the pixels of the inspection image. An average of gray level values of a plurality of pixels belonging to the same column as the inspection pixel is calculated as the average gray level value.
The image inspection apparatus according to claim 6. The predetermined image generation means includes a one-dimensional line sensor for imaging the inspection object, and the inspection object is scanned in a direction orthogonal to the line direction by the one-dimensional line sensor. Generate a captured image,
A known repeating pattern is formed on the surface to be inspected,
The inspection image has a gray level difference between each pixel in the captured image and a pixel corresponding to each pixel in the repetitive pattern appearing in the captured image as a pixel value at the position of each pixel. A gray level difference image having
Pixels in the same row aligned in the column direction in the captured image are pixels captured by the respective imaging pixels of the one-dimensional line sensor aligned in the line direction, and pixels in the same column aligned in the row direction in the captured image are the 1 When the same image pickup pixel of the two-dimensional line sensor is a pixel picked up by scanning in a direction orthogonal to the line direction, the average gray level calculation unit includes the second of the pixels of the inspection image. An average of gray level values of a plurality of pixels belonging to the same column as the inspection pixel is calculated as the average gray level value.
The image inspection apparatus according to claim 6. An image inspection method for detecting defects present on the surface of an inspection object by inspecting an inspection image generated by a predetermined image generation means for the inspection object,
A gray level difference between an inspection pixel that is a pixel for which a defect is determined in the inspection image and a pixel that is closest to the inspection pixel among a plurality of pixels around the inspection pixel. Detect a certain minimum gray level difference,
When the minimum gray level difference exceeds a predetermined detection threshold, it is determined that a defect exists at the position indicated by the inspection pixel.
An image inspection method characterized by the above. Detecting a gray level difference between the inspection pixel and the plurality of pixels around the inspection pixel,
Selecting the smallest of these detected gray level differences as the minimum gray level difference;
The image inspection method according to claim 9. Selecting a reference pixel that has a gray level value closest to the inspection pixel among the plurality of pixels around the inspection pixel;
Detecting a gray level difference between the inspection pixel and the reference pixel as the minimum gray level difference;
The image inspection method according to claim 9, further comprising:   The image inspection method according to claim 9, wherein the inspection image is a captured image of the inspection target. The inspection object has a known repeating pattern on its surface,
The inspection image has a gray level difference between each pixel in the captured image to be inspected and a pixel corresponding to each pixel in a repetitive pattern appearing in the captured image at a position of each pixel. The image inspection method according to claim 9, wherein the image inspection method is a gray level difference image having a pixel value. The inspection pixel and the detection threshold are a first inspection pixel and a first detection threshold, respectively.
Calculating an average gray level value of a plurality of pixels distributed in a predetermined range in the inspection image;
Detecting a difference between a gray level value of a second inspection pixel which is a pixel subjected to defect determination in the inspection image and the average gray level value;
When the difference exceeds a predetermined second detection threshold, it is determined that a defect is also present at the position indicated by the second inspection pixel.
The image inspection method according to any one of claims 9 to 11, wherein: The predetermined image generating means includes a one-dimensional line sensor for imaging the inspection object,
The inspection object is imaged by scanning the inspection object in a direction orthogonal to the line direction with the one-dimensional line sensor,
Pixels in the same row aligned in the column direction in the inspection image are pixels captured by the respective imaging pixels of the one-dimensional line sensor aligned in the line direction, and pixels in the same column aligned in the row direction in the inspection image are the 1 A plurality of pixels in the inspection image that belong to the same column as the second inspection pixel, when the same imaging pixel of the dimensional line sensor is taken as a pixel imaged by scanning in a direction orthogonal to the line direction. Calculating the average of the gray level values of the pixels as the average gray level value;
The image inspection method according to claim 14. The predetermined image generating means includes a one-dimensional line sensor for imaging the inspection object,
The captured image of the inspection object is generated by scanning the inspection object in a direction orthogonal to the line direction with the one-dimensional line sensor,
A known repeating pattern is formed on the surface to be inspected,
The inspection image has a gray level difference between each pixel in the captured image and a pixel corresponding to each pixel in the repetitive pattern appearing in the captured image as a pixel value at the position of each pixel. A gray level difference image having
Pixels in the same row aligned in the column direction in the captured image are pixels captured by the respective imaging pixels of the one-dimensional line sensor aligned in the line direction, and pixels in the same column aligned in the row direction in the captured image are the 1 A plurality of pixels in the inspection image that belong to the same column as the second inspection pixel, when the same imaging pixel of the dimensional line sensor is taken as a pixel imaged by scanning in a direction orthogonal to the line direction. Calculating the average of the gray level values of the pixels as the average gray level value;
The image inspection method according to claim 14.