JP2002014056A - Checking apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To more accurately check the defect of an object to be checked of a semiconductor wafer or the like. SOLUTION: A checking apparatus calculates the degree of a blur of image data of a device pattern of the semiconductor wafer with a CPU 31 in an image processor 3, and corrects the degree of the blur so as to properly detect the defect.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an inspection apparatus for inspecting, for example, a defect occurring in a semiconductor wafer.

[0002]

2. Description of the Related Art A semiconductor device is manufactured by forming a fine device pattern on a semiconductor wafer. When such a device pattern is formed, dust or the like may adhere to the semiconductor wafer or may be damaged, thereby causing a defect. A semiconductor device having such a defect becomes a defective device and reduces the yield.

[0003] Therefore, in order to stabilize the yield of a production line at a high level, defects generated by dust, scratches and the like are found at an early stage, the causes thereof are identified, and effective measures are taken for production facilities and production processes. It is preferable to take it.

[0006] In recent years, there has been proposed an inspection apparatus which picks up an image of a semiconductor wafer surface, performs image processing based on the picked-up image, and automatically determines the type of defect occurring in the semiconductor wafer.

[0005] When a defect is detected, the inspection apparatus checks what the defect is, classifies the defect, and specifies the equipment or process that caused the defect. Further, the inspection device is a device using an optical microscope, so to speak, and identifies what the defect is by enlarging and viewing the defect.

[0006]

When a semiconductor wafer is inspected using the above-described inspection apparatus, the thickness of the film formed on the semiconductor wafer and the variation in autofocus are not within the depth of field. In such a case, the image is blurred. This inspection apparatus has a problem that, in an extreme case, a defect is erroneously detected due to blurring of an image. This is because an image blur is detected as a defect.

The present invention has been made in view of such circumstances, and has as its object to provide an inspection apparatus that can more accurately detect a defect to be inspected such as a semiconductor wafer.

[0008]

An inspection apparatus according to the present invention is an inspection apparatus for inspecting a defect generated in an inspection object based on an image of the inspection object. And an image processing means for processing the image of the inspection target imaged by the image imaging means. The inspection apparatus includes a pre-process in which the image processing unit detects a defect from the image of the inspection target captured by the image capturing unit,
It is characterized in that processing for correcting blurring of an image captured by the image capturing means is performed.

The inspection apparatus according to the present invention configured as described above performs a process of correcting blur of an image,
False detection can be reduced.

[0010]

Embodiments of the present invention will be described below in detail with reference to the drawings. Here, an example in which the present invention is applied to an inspection apparatus that inspects a semiconductor wafer will be specifically described. However, the present invention is not limited to this example, and an inspection of an inspection target having a fine shape is performed. It can be widely applied to an inspection apparatus for performing the above.

FIG. 1 shows an inspection apparatus to which the present invention is applied. The inspection apparatus 1 inspects a semiconductor wafer 5 on which a predetermined device pattern is formed, captures an image of the device pattern of the semiconductor wafer 5, and converts image data of the captured device pattern of the semiconductor wafer 5 into image data. The image data is corrected by performing a correction process as needed, and a defect extraction process is performed after correcting the image data. If a defect is found in the semiconductor wafer 5, the defect is examined and classified. I do. In addition, this inspection device 1
Captures an area having a defect and an area without a defect on the semiconductor wafer 5, and these images are referred to as a defect image and a reference image, respectively.

The inspection apparatus 1 includes an image pickup unit 2 as an image pickup unit, an image processing unit 3 as an image processing unit, and a defect detection and classification unit 4 as a defect detection and classification unit. Hereinafter, each part will be described in detail.

The image pickup section 2 includes an inspection stage 6 for supporting a semiconductor wafer 5 on which a predetermined device pattern is formed. The semiconductor wafer 5 is sucked on the inspection stage 6. The inspection stage 6 is connected to a control unit 7 that controls the image capturing unit 2, and moves the semiconductor wafer 5 to an appropriate position in a horizontal plane with respect to the surface of the semiconductor wafer 5 according to control from the control unit 7. Move to. Further, the inspection stage 6 includes a control unit 7
The semiconductor wafer 5 is moved in a direction perpendicular to the surface of the semiconductor wafer 5 in order to more accurately capture the defect image and the reference image of the semiconductor wafer 5 in accordance with the control from. The control unit 7 controls the inspection stage 6 and performs an auto-focus process for automatically adjusting an optical unit 9 that is an optical system for capturing a defect image and a reference image so as to have an appropriate distance.

The image pickup section 2 includes an illumination light source 8 for emitting illumination light for illuminating the semiconductor wafer 5. The illumination light source 8 emits illumination light for illuminating the semiconductor wafer 5. The illumination light source 8 is connected to a control unit 7 that controls the image capturing unit 2, and adjusts an output to be output under the control of the control unit 7. The illumination light source 8 includes, for example,
A light source capable of continuously oscillating a deep ultraviolet laser beam having a wavelength of 266 nm is used.

Further, the image pickup section 2 includes a semiconductor wafer 5
An optical unit 9 for imaging the device pattern of
As shown in FIG. 2, the optical unit 9 includes a CCD (charge coupled device) camera 11, which is a so-called charge coupled device, lenses 12, 13, 14, 15 and a half mirror 16. The above CCD camera 1
1 captures a defect image and a reference image in which the device pattern of the semiconductor wafer 5 illuminated by the illumination light from the illumination light source 8 is enlarged. The CCD camera 11 is connected to the image processing unit 3 and transfers the defect image and the reference image obtained by enlarging the captured device pattern to the image processing unit 3 as data.

The illumination light source 8 and the optical unit 9 are connected by an optical fiber 10 for transmitting illumination light. The illumination light source 8 transmits illumination light to the optical unit 9 through the optical fiber 10 and irradiates the semiconductor wafer 5 through the optical unit 9.

FIG. 2 shows the outline of the travel path of the illumination light inside the optical unit 9 described above. First, illumination light is emitted from the illumination light source 8 through the optical fiber 10. The illumination light that has entered the optical unit 9 through the optical fiber 10 transmits through the lens 12 and then transmits through the lens 13. Next, the illumination light is reflected by the half mirror 16,
The semiconductor wafer 5 is transmitted through the objective lens 14.
Is irradiated. The reflected light reflected by the semiconductor wafer 5 is
The light passes through the lens 14 and the half mirror 16. Further, the reflected light is transmitted to the CCD camera 11 by the lens 15.
The device pattern of the semiconductor wafer 5 is imaged on the chip surface.

As shown in FIG. 1, the image pickup section 2 includes a control section 7, which is connected to the inspection stage 6 and the illumination light source 8, and is connected to the inspection stage 6. And the illumination light source 8. The control unit 7 transmits a control signal to the inspection stage 6 and the illumination light source 8, and performs each control as described above.

An input device (not shown) including, for example, a pointing device such as a mouse and a keyboard is connected to the image pickup section 2. Instructions necessary for controlling the various devices 8 can be input to the control unit 7.

The image pickup unit 2 includes, for example, a CRT
A display device (not shown) including a display, a liquid crystal display, and the like is connected to the display device, and various conditions at the time of imaging a device pattern of the semiconductor wafer 5 can be displayed on the display device.

As shown in FIG. 3, the image processing unit 3 is a central processing unit (CPU) (central processing unit) 3.
1, an image data storage unit 32 for storing data of a defect image and a reference image obtained by enlarging a device pattern of the semiconductor wafer 5 captured by the CCD camera 11, and calculating the degree of blurring of the data of the defect image and the reference image A correction program storage unit 3 that stores a program for correcting an image having a smaller degree of blurring to have the same degree of blurring as an image having a larger degree of blurring.
3, a corrected image data storage unit 34 for storing data of the defect image and the reference image after the correction, an input interface 35 for controlling an input, and an output interface 36 for controlling an output.

The image processing unit 3 performs blurring of the defect image and the reference image as preprocessing for detecting and classifying defects from data of the defect image and the reference image obtained by enlarging the device pattern of the semiconductor wafer 5 captured by the CCD camera 11. Is calculated, the degree of blur of the defect image and the degree of blur of the reference image are compared, and correction is performed so that the image with the smaller degree of blur has the same degree of blur as the image with the larger degree of blur. Here, either one of the data of the defect image and the data of the reference image is subjected to blur correction, and the corrected image data and the other image data that has not been corrected are combined as corrected image data. I have.

The CPU 31 includes an image data storage unit 32,
The correction program storage unit 33, the correction image data storage unit 34, the input interface 35, and the output interface 36 are connected.

The CPU 31 acquires the data of the defect image and the reference image of the device pattern of the semiconductor wafer 5 captured by the CCD camera 11 via the input interface 35 and stores the data in the image data storage unit 32. Further, the CPU 31 stores the correction program storage unit 3
From step 3, the correction program is called, the degree of blur of the data of the defect image and the reference image is calculated, and the degree of blur of the defect image and the reference image is compared. An image processing operation is performed to make the same degree of blur as that of the larger image, and the corrected image data and the reference image data are stored in the corrected image data storage unit 34. And information is output via the output interface 36.

The image data storage section 32 stores the CCD camera 1
The data of the defect image and the reference image of the device pattern of the semiconductor wafer 5 captured in step 1 are stored. The image data storage unit 32 is, for example, a rewritable RAM.
(Random access memory).

The correction program storage unit 33 calculates the degree of blurring of the data of the defect image and the reference image, compares the degree of blurring of the defect image and the reference image, and determines the image having the smaller degree of blurring. A correction program for performing a correction to make the degree of blur the same as that of the image having a higher degree is stored. This correction program is CP
It is called according to the instruction of U31. The correction program storage unit 33 is, for example, a rewritable RAM.
Or non-rewritable ROM (read only memory)
Is used.

The corrected image data storage unit 34 stores the data of the defect image and the reference image whose degree of blur has been corrected by executing the above-described correction program.

Depending on the system configuration, the image data storage unit 32, the correction program storage unit 33, and the corrected image data storage unit 34 may be integrated and one storage unit may be used.

Here, an image processing method for calculating the degree of blur of the data of the above-described defect image and reference image and correcting the degree of blur will be described in detail below.

This image processing is executed by the CPU 31 based on the correction program. First, the image processing unit 3 sends a blur evaluation value, which is a value indicating the degree of blur of the image data stored in the image data storage unit 32, to the CPU 31.
Calculate using Here, the blur evaluation value is expressed using contrast or the like. For example, a difference between the brightness of each pixel and pixels adjacent to the pixel vertically and horizontally is calculated from the data of the defect image and the reference image. Is the maximum value of Also, other simple methods may be used.

Here, the image processing section 3 calculates the blur evaluation value of the data of the defect image and the reference image based on the result calculated by the CPU 31. For example, the CPU 31 executes the following calculation processing. Correction is performed so that the blur evaluation values of the defect image and the reference image are substantially the same.

First, the brightness of each pixel of the defect image to be inspected and the data of the reference image to be inspected for each pixel is compared with the brightness of this pixel and the pixels adjacent vertically and horizontally to determine the difference. This means that, as shown in FIG.
And the pixels E1, which are vertically and horizontally adjacent to this pixel,
The brightness of the pixel E0 and each pixel E
The difference between the brightness of E1, E2, E3, and E4 is calculated.

Next, this calculation process is performed for all the pixels of the image data, and the maximum value is obtained. However, at the end or the four corners of the image data, there is no adjacent pixel for all of the upper, lower, left, and right, so the brightness is compared only with the adjacent pixel to find this difference. Then, the maximum value of the difference is obtained for each of the defect image and the reference image, and this maximum value is used as the blur evaluation value.

Next, the magnitudes of the blur evaluation values of the defect image and the reference image data are compared. Then, the image data having a smaller degree of blur, that is, the larger blur evaluation value, is corrected so that the degree of blur is substantially equal to the image data having a larger degree of blur, that is, the smaller blur evaluation value. In other words, the image with the smaller degree of blur is intentionally blurred by image processing, and the degree of blur of the defect image and the reference image is made the same.

In this image processing, as shown in FIG. 5, for example, the brightness of the central pixel R0 having the data of the defect image or the reference image to be subjected to the image processing is set to 5 × 5 pixels R1 around the central pixel R0. By performing this calculation for all the pixels of the data of the defect image or the reference image, correction is performed to blur the entire image.

Specifically, the brightness of the pixel after the correction is represented by Pn
Assuming that the brightness of the pixel before correction is Pn, [Pn
2 = average of 5 × 5 pixels around Pn].
Here, n is a pixel number, and the pixel number is a number assigned to each pixel of the data of the defect image and the reference image, which are numerical values in a two-dimensional array.

In the above-mentioned correction, the area for calculating the average is 5 × 5 pixels around the area, but may be m × m pixels. Here, m is a natural number, and the value can be changed according to the degree of correction.
The inspection apparatus 1 according to the present invention can perform an effective correction by changing a region for calculating an average by adjusting m.

The inspection apparatus according to the present invention is not limited to the above-described image processing, but may be corrected by image processing in which the weights of the center pixel and the peripheral pixels for calculating the average are changed. .

Then, the CPU 31 causes the corrected image data storage unit 34 to store the data of the defective image and the reference image after the correction so that the blur evaluation values of the data of the defective image and the reference image are substantially the same.

The image processing unit 3 is connected to an input device (not shown) such as a pointing device such as a mouse or a keyboard, and the input device is connected to a defect taken from the CCD camera 11. An instruction necessary for correcting the degree of blur of the data of the image and the reference image can be input to the image processing unit 3.

The image processing unit 3 includes, for example, a CRT
A display device (not shown) including a display, a liquid crystal display, and the like is connected.
The processing result of the image transferred from the CD camera 11 can be displayed.

The image processing unit 3 as described above includes a CPU 3
The data of the corrected defect image and the reference image, which are the result of the image processing by the first and second image data, are retrieved from the corrected image data storage unit 34 and transferred to the defect detection and classification unit 4 via the output interface 36.

The defect detection / classification unit 4 is connected to a defect pattern database (not shown), calculates the data of the defect image and the reference image processed by the image processing unit 3, and extracts a defect. Here, the defect pattern database is a database of defect patterns obtained based on past knowledge.

The defect detection and classification unit 4 compares, for example, data of a defect image, which is an image of a device pattern having a defect processed by the image processing unit 3, with data of a reference image, which is an image of a correct device pattern. Then, a defect is extracted by taking the difference between them.

Next, the defect detection / classification unit 4 calls a defect pattern from the defect pattern database and searches for a pattern that matches the characteristic of the defect pattern extracted in the above processing. In addition, the defect detection and classification unit 4 classifies and outputs what is the defect due to the search result whose characteristics match those of the defect pattern database.

An input device (not shown), such as a pointing device such as a mouse or a keyboard, is connected to the defect detection / classification unit 4, and the defect detection / classification unit 4 is connected to the input device. Instructions necessary for control can be input to the defect detection / classification unit 4.

The defect detection and classification unit 4 includes, for example, C
A display device (not shown) including an RT display, a liquid crystal display, and the like is connected.
The result of the detection and classification performed by the defect detection and classification unit 4 can be displayed.

Here, for example, among the input devices and / or display devices connected to the image pickup unit 2, the image processing unit 3, and the defect detection / classification unit 4, those which can be shared are integrated into one. May be used.

The inspection apparatus 1 according to the present invention can reduce erroneous detection of defects and more accurately detect defects to be inspected such as the semiconductor wafer 5 by the above-described image correction processing. In addition, the inspection apparatus 1 adjusts the movement of the stage 6 by performing the calculation on the CPU 31 in the image processing unit 3 instead of the correction by the automatic focus function that operates the stage 6 by this processing. It can be adjusted quickly and individually.

In the inspection apparatus 1 according to the present invention,
When calculating the blur evaluation value, if the size of the image data is too large for the processing capability of the CPU 31, the sampling is performed to a size of the image data that can be processed at a sufficient speed with the processing capability of the CPU 31. May be.
This sampling is a method of extracting data at predetermined intervals in the vertical and horizontal directions of image data, and using the extracted data as an image again.

The inspection apparatus 1 performs the above-described sampling processing, thereby obtaining the CPU in the image processing unit 3.
Reference numeral 31 indicates that the blur evaluation value can be calculated within a predetermined time, the time required to calculate the blur evaluation value can be reduced as compared with the case where no sampling processing is performed, and the performance of the inspection apparatus 1 can be comprehensive. Can be improved.

Further, in the inspection apparatus 1 according to the present invention, even if the above-described processing for correcting the degree of blur in the image processing section 3 is always performed, there is no problem in defect detection itself, but the processing time increases. Therefore, as will be described in detail later, if the difference between the blur evaluation value of the data of the defect image before processing and the blur evaluation value of the data of the reference image does not exceed a predetermined reference value K, it is better to perform no processing. Can be expected to improve the overall performance of Note that the reference value K is a value determined in advance, and is determined by various conditions of the inspection apparatus 1.

Here, the flow of the operation of the inspection apparatus 1 described above is shown as a flowchart in FIGS. Hereinafter, a more detailed description will be given with reference to the flowcharts of FIGS.

First, in step S1, the inspection stage 6 is moved under the control of the control unit 7 in the image pickup unit 2, and a defect image of the semiconductor wafer 5 is picked up on the surface of the semiconductor wafer 5 in a horizontal plane. To place it in the right place.

Next, in step S2, under the control of the control unit 7, the inspection stage 6 is moved, and the semiconductor wafer 5 is moved in a direction perpendicular to the surface of the semiconductor wafer 5,
Adjust the auto focus function.

Next, in step S3, the optical unit 9
The device pattern of the semiconductor wafer 5 on the inspection stage 6 is imaged by the CCD camera 11 inside. At this time, under the control of the control unit 7, illumination light is emitted from the illumination light source 8, and the semiconductor wafer 5 is illuminated by the illumination light.

Next, in step S4, the defect image picked up by the CCD camera 11 is transferred from the CCD camera 11 to the image processing section 3 as two-dimensional array data.

Next, in step S5, the image processing section 3 sends the input interface 35 from the CCD camera 11.
The data of the defective image transferred via the
The image data is stored in the image data storage unit 32 in the inside.

Next, in step S6, the inspection stage 6 is moved under the control of the control unit 7 in the image pickup unit 2, and a reference image of the semiconductor wafer 5 is taken with respect to the surface of the semiconductor wafer 5 in a horizontal plane. To place it in the right place.

Next, in step S7, under the control of the control unit 7, the inspection stage 6 is moved, and the semiconductor wafer 5 is moved in a direction perpendicular to the surface of the semiconductor wafer 5,
Adjust the auto focus function.

Next, in step S8, the optical unit 9
The device pattern of the semiconductor wafer 5 on the inspection stage 6 is imaged by the CCD camera 11 inside. At this time, under the control of the control unit 7, illumination light is emitted from the illumination light source 8, and the semiconductor wafer 5 is illuminated by the illumination light.

Next, in step S9, the reference image captured by the CCD camera 11 is transferred from the CCD camera 11 to the image processing unit 3 as two-dimensional array data.

Next, in step S10, the image processing unit 3 sends the input interface 3
The data of the reference image transferred via the image processing unit 5 is stored in the image data storage unit 32 in the image processing unit 3.

Next, in step S11, the image processing unit 3
The stored image correction program is called from the correction program storage unit 33 according to a command from the CPU 31. Next, the process proceeds to step S12 in FIG.

In step S12, the amount of image data is compared with the processing capacity of the CPU 31 in the image processing unit 3 by executing the image correction program.
1 is used to calculate whether image processing can be performed at a sufficient speed. If the processing capacity of the CPU 31 is not sufficient, that is, if the image size is larger than the processing capacity of the CPU 31, the process proceeds to step S13. On the other hand, if the processing capacity of the CPU 31 is sufficient, that is, if the image size is smaller than the processing capacity of the CPU 31, the process proceeds to step S16.

In step S13, a sampling process for thinning out and calculating so as not to impair the features of the image is executed by the CPU.
This is executed at 31 to reduce the amount of image data. As a sampling method, for example, there is a method of extracting data at predetermined intervals in the vertical and horizontal directions of image data, and re-image the extracted data. Next, the process proceeds to step S14.

In step S14, the blur evaluation value Fd of the data of the sampled defect image is calculated. next,
Proceed to step S15.

In step S15, the blur evaluation value Fr of the data of the sampled reference image is calculated. next,
In step S16, which proceeds to step S18, a blur evaluation value Fd of the data of the defective image is calculated. Next, step S
Proceed to 17.

In step S17, a blur evaluation value Fr of the data of the reference image is calculated. Next, the process proceeds to step S18.

In step S18, the blur evaluation value of the defect image is compared with the blur evaluation value of the reference image.
If it is + K, the process proceeds to step S19. Fd ≦ Fr +
If it is K, the process proceeds to step S20.

In step S19, a calculation of [Pn2 = average value of 5 × 5 pixels around Pn] is performed on all pixel data of the defect image to blur the defect image. Next, the process proceeds to step S22 in FIG.

In step S20, the blur evaluation value of the defect image is compared with the blur evaluation value of the reference image.
If it is + K, the process proceeds to step S21. Fr ≦ Fd +
If it is K, the process proceeds to step S22 in FIG.

In step S21, the calculation is performed to set [Pn2 = average value of 5 × 5 pixels around Pn] for all the pixel data of the reference image, and the reference image is blurred. Next, the process proceeds to step S22 in FIG.

In step S22, the data of the defect image and the reference image after correction by the CPU 31 are stored in the corrected image data storage section 34, and the flow advances to step S23.

Next, in step S23, the data of the defect image and the reference image after the correction by the image processing are stored in the CPU 3.
1 is transmitted to the defect detection / classification unit 4 through the output interface 36.

Next, in step S24, the defect detection / classification unit 4 receives image data from the image processing unit 3 and extracts a defect by calculating the difference between the defect image and the reference image.
Perform defect detection.

Next, in step S25, the defect detection / classification unit 4 compares the detected defect pattern with the pattern in the defect database.

Next, in step S26, the defect detection / classification unit 4 classifies which type of defect the compared defect pattern is.

Next, in step S27, the result of classification by the defect detection / classification unit 4 is output to a display device.

According to the flow described above, the inspection apparatus 1 to which the present invention is applied improves the accuracy and speed of the defect detection / classification processing by correcting the degree of blurring of the image data and reducing erroneous detection of defects. be able to.

In the case where a color image is used for inspection, the blurring of each color of red, green, and blue does not change so much. Therefore, a sufficient effect can be obtained by performing a fixed correction instead of a correction for each color.

[0082]

As described above, the inspection apparatus according to the present invention corrects the degree of image blur according to the present invention when the defect cannot be properly detected because the captured image is blurred. I do. With this correction, erroneous detection due to blur can be reduced, and the performance of the inspection apparatus can be improved overall. In addition, since the correction by the image processing is performed in the processing computer, the time for automatically moving the inspection stage and performing the autofocus can be omitted,
Processing speed can be improved. Further, since the necessity of the processing is determined based on the required correction amount, the processing time can be reduced.

[Brief description of the drawings]

FIG. 1 is a schematic diagram of a configuration of an inspection apparatus to which the present invention is applied.

FIG. 2 is a schematic diagram showing an optical unit in an image pickup unit to which the present invention is applied.

FIG. 3 is a schematic diagram showing a configuration inside an image processing unit to which the present invention is applied.

FIG. 4 is a schematic diagram of a process of calculating a degree of blur of image data to which the present invention has been applied.

FIG. 5 is a schematic diagram of a process for blurring image data to which the present invention has been applied.

FIG. 6 is a flowchart showing a flow of processing to which the present invention is applied.

FIG. 7 is a flowchart showing a flow of a process to which the present invention is applied.

FIG. 8 is a flowchart showing a flow of processing to which the present invention is applied.

[Explanation of symbols]

2 image imaging section, 3 image processing section, 4 defect detection and classification section, 31 CPU, 32 image data storage section, 33 correction program storage section, 34 corrected image data storage section, 3
5 input interface, 36 output interface

Claims (4)

[Claims] 1. An inspection apparatus for inspecting a defect generated in an inspection target based on an image of the inspection target, wherein: an image imaging unit that captures an image of the inspection target; and an inspection target that is imaged by the image imaging unit. Image processing means for processing the image of the inspection target, wherein the image processing means performs, as preprocessing for detecting a defect from the image of the inspection target captured by the image capturing means, a blur of the image captured by the image capturing means. An inspection apparatus characterized by performing a process of correcting an error. 2. The image picked up by the image pick-up means is a defect image obtained by picking up an area where a defect has occurred in an inspection object, and a reference image picked up by picking up an area without a defect. The inspection device according to claim 1. 3. The image processing means judges the degree of blur of the defect image and the reference image of the inspection object imaged by the image imaging means, and determines the degree of blur with respect to the image having the smaller degree of blur. 3. The inspection apparatus according to claim 2, wherein a process of correcting the blur of the image is performed so that the degree of blur is substantially the same as that of the image having the larger image. 4. The image processing means determines the degree of blur of the image of the inspection target captured by the image capturing means, and determines the degree of blur of an image whose degree of blur exceeds a predetermined reference value. The inspection apparatus according to claim 1, wherein a correction process is performed.