JP7017916B2 - Board inspection equipment - Google Patents

Description

The present invention relates to a substrate inspection apparatus for inspecting a substrate.

Conventionally, an inspection device for inspecting a defect of a disk-shaped substrate has been widely used. The inspection device inspects the presence or absence of defects on the edge of the disk-shaped substrate and the surface on which a predetermined pattern is formed.

For example, in Patent Document 1, a wafer stage for rotating a wafer and an edge for inspecting the edge portion by irradiating the edge portion of the wafer with horizontal light when the wafer placed on the wafer stage rotates. An inspection unit and a moving device for moving the edge inspection unit along the optical axis of the horizontal light are provided, and when the center of rotation of the wafer deviates from the center of the wafer, the moving device moves the edge inspection unit. An optical inspection device for adjusting the depth of focus is disclosed.

Further, Patent Document 2 includes a holding table for holding the object to be inspected, a rotation mechanism for rotating the holding table, and an inspection unit including a first detection unit and a second detection unit. By rotating the holding table in the held state and linearly moving the inspection unit according to the linear region including the center of the object to be inspected, the first detection unit and the second detection unit 2 Since the type of inspection information is acquired, an inspection device capable of acquiring two types of inspection information at once while simplifying the movement of the inspection unit is disclosed.

Further, in Patent Document 3, when positioning a circular substrate, a non-contact type that non-contactly detects information indicating a change in the amount of displacement of the peripheral portion of the circular substrate from the rotation center during rotation of the turntable. A positioning device for detecting defects in the peripheral portion of such a circular substrate by using the output information from the detector is disclosed.

Japanese Unexamined Patent Publication No. 2013-93389 Japanese Unexamined Patent Publication No. 2017-116293 Japanese Unexamined Patent Publication No. 5-160245

On the other hand, the substrate inspection device for inspecting a disk-shaped substrate takes an image of the surface of a rotating disk-shaped substrate with a sensor, and inspects defects (dents, scratches, etc.) generated on the surface of the disk-shaped substrate using the captured image. However, if the center of rotation of the disk-shaped substrate deviates from the center of the disk-shaped substrate and eccentricity occurs, the position of the end face of the disk-shaped substrate changes in the radial direction with respect to the sensor field of view when the disk-shaped substrate rotates. do. That is, the relative positional relationship between the sensor and the surface of the disk-shaped substrate imaged by the sensor changes from directly above the end face of the disk-shaped substrate to diagonally above the disk-shaped substrate (hereinafter, simply referred to as parallax). .. Further, since the position of supporting the disk-shaped substrate is also biased due to the eccentricity, the disk-shaped substrate is deformed by its own weight and the disk-shaped substrate is slightly tilted.
In particular, when there is a limitation on the size of the device and it is unavoidable to take an image at a distance close to the disk-shaped substrate and a sufficient optical path cannot be secured, the influence of the parallax is large. Specifically, in the captured image captured by the substrate inspection device, the width of the polished inclined portion (so-called bevel portion) on the end side of the disk-shaped substrate is different. Also, a shadow on the background is reflected. In such a case, it may be difficult to distinguish between the shadow and the disk-shaped substrate, which hinders defect inspection using such an image.

However, all of the above-mentioned Patent Documents 1 to 3 require a complicated configuration using a sensor moving mechanism, an optical system for suppressing the influence of the parallax, and the like. Therefore, if there are restrictions on the size of the device or the location where the sensor is installed, the inventions of Patent Documents 1 to 3 cannot be applied, and this problem cannot be solved.

The present invention has been made in view of such circumstances, and an object of the present invention is accuracy even when a shadow due to parallax occurs because the center of rotation of the disk-shaped substrate deviates from the center of the disk-shaped substrate. The purpose is to provide a substrate inspection device capable of high-level defect inspection.

The substrate inspection apparatus according to the present invention is a substrate inspection apparatus that inspects the surface of a substrate using an image pickup sensor, and is a target area for the surface inspection with respect to an image taken on the surface of the edge of the substrate imaged by the image pickup sensor. It is characterized by including an exclusion processing unit that performs an exclusion process for excluding the surplus area excluding the above, and a detection unit that detects an abnormality in the target area using the image after the exclusion process.

In the present invention, when the surface of the substrate is inspected, the exclusion processing unit performs exclusion processing to eliminate the surplus region excluding the target region, and only the target region after the exclusion treatment is left as an image captured image. Is used by the detection unit to detect an abnormality in the target area.

The substrate inspection apparatus according to the present invention has a determination unit for determining a boundary between an inclined portion formed on the edge of the substrate and the surface based on the shading of the captured image, and the determination result of the determination unit is used. Based on this, the exclusion processing unit is characterized in that the exclusion processing of the surplus region including the inclined portion is performed.

In the present invention, the determination unit determines the boundary between the inclined portion and the surface based on the change in shading in the captured image. The exclusion processing unit performs exclusion processing of the surplus region defined by the boundary determined by the determination unit.

The substrate inspection apparatus according to the present invention is characterized in that the detection unit performs a process of emphasizing the difference in shading on the target region to detect the abnormality.

In the present invention, the detection unit performs a process of emphasizing the difference in shading on the target region to clarify the difference between the abnormal portion and the normal portion, and then detects the abnormality.

The substrate inspection apparatus according to the present invention is further provided with an edge inspection device for inspecting the edge of the substrate.

In the present invention, since the edge inspector is further provided, it is possible to inspect the surface of the edge of the substrate and inspect such an edge.

The surface inspection method according to the present invention is a surface inspection method for inspecting the surface of a substrate by a substrate inspection apparatus having an image pickup sensor. It is characterized by including a step of performing an exclusion process for excluding a surplus area excluding the target area of the surface inspection, and a step of detecting an abnormality in the target area using the image after the exclusion process.

The computer program according to the present invention performs an exclusion process for excluding a surplus area excluding the target area for surface inspection of the substrate from the captured image of the surface of the substrate edge imaged by the image pickup sensor, and after the exclusion process. A computer program characterized by causing a computer to execute a process of detecting an abnormality in the target area using an image.

In the present invention, at the time of surface inspection of the substrate, an exclusion process for excluding the surplus area excluding the target area is performed on the captured image, and only the target area after the exclusion process is left. Anomalies in the target area are detected using the captured image.

According to the present invention, the center of rotation of the disk-shaped substrate deviates from the center of the disk-shaped substrate, which causes eccentricity, and even if a shadow appears on the image taken by the disk-shaped substrate and the sensor due to this, the accuracy is high. High defect inspection is possible.

It is a block diagram which shows the main part structure of the alignment apparatus which concerns on this embodiment. It is sectional drawing which schematically illustrates the structure of the disk-shaped substrate W which concerns on this embodiment. It is a schematic plan view explaining the arrangement of the edge measuring device and the surface image pickup device in the alignment apparatus which concerns on this embodiment. It is a functional block diagram which shows the composition of the main part of the processing part of the alignment apparatus which concerns on this embodiment. It is a graph which plotted the distance data with respect to the rotation angle. It is a flowchart explaining the process with respect to the captured image in the alignment apparatus which concerns on this embodiment. It is a flowchart explaining the process with respect to the captured image in the alignment apparatus which concerns on this embodiment. It is a flowchart explaining the process with respect to the captured image in the alignment apparatus which concerns on this embodiment. It is a photograph showing before and after executing the exclusion process in the alignment apparatus which concerns on this embodiment. It is a photograph showing before and after executing the filter processing with respect to the target area in the alignment apparatus which concerns on this embodiment.

Hereinafter, the substrate inspection apparatus, the surface inspection method, and the computer program according to the present embodiment will be described in detail with reference to the drawings, taking as an example the case where the substrate inspection apparatus, the surface inspection method, and the computer program are applied to the so-called alignment apparatus.

FIG. 1 is a block diagram showing a main configuration of an alignment device 100 (board inspection device) according to the present embodiment. The alignment device 100 according to the present embodiment arranges the orientation, position, and the like of the disk-shaped substrate W, and detects defects at the edge of the disk-shaped substrate W. The alignment device 100 processes a defect detector 3, a control unit 1 that controls the defect detector 3, and a detection result by the defect detector 3 to detect defects on the edge and surface of the disk-shaped substrate W. A unit 10 is provided.

The defect detector 3 measures and images a defect at the edge of the disk-shaped substrate W. Specifically, the defect detector 3 includes an edge measuring device 4 (edge inspector) and a surface imager 5, measures notches and defects formed on the edges, and images the surface of the edges.

The disk-shaped substrate W is, for example, a semiconductor wafer, a glass substrate, or the like. FIG. 2 is a cross-sectional view schematically illustrating the configuration of the disk-shaped substrate W according to the present embodiment. The disk-shaped substrate W has two opposing flat surfaces. Hereinafter, of the two flat surfaces, the surface irradiated with the light from the edge measuring instrument 4 and the surface imager 5 is referred to as a surface W3.

A radial inclined portion is provided around the edge W2 of the disk-shaped substrate W between the edge W2 and the surface W3. Such an inclined portion is a so-called bevel W4. In other words, the disk-shaped substrate W is divided into a surface W3 and a bevel W4 with reference to the boundary W5.

The disk-shaped substrate W has, for example, a notch portion W1 such as an orientation flat and a notch in a part of the edge W2. The disk-shaped substrate W is placed on the turntable 6 and rotated, and at this time, the edge is measured by the edge measuring device 4 and the edge side of the surface W3 is imaged by the surface imager 5.

The defect detector 3 includes an electric motor 7 for rotating the turntable 6 on which the disk-shaped substrate W is placed, and the turntable 6 is driven by the electric motor 7.

FIG. 3 is a schematic plan view illustrating the arrangement of the edge measuring device 4 and the surface imager 5 in the alignment device 100 according to the present embodiment.
The edge measuring device 4 and the surface imager 5 are provided near the edge W2 of the disk-shaped substrate W, respectively. The edge measuring instrument 4 and the surface imager 5 are provided along the radial direction of the disk-shaped substrate W, and are arranged apart from each other in the circumferential direction of the disk-shaped substrate W. The edge measuring instrument 4 and the surface imager 5 are arranged apart from each other in the circumferential direction of the disk-shaped substrate W at a minimum necessary distance that is not affected by the illumination of each other.

The edge measuring instrument 4 for measuring the edge of the disk-shaped substrate W includes a light source 41, a lens 42, and a sensor 43. The light source 41 is, for example, a semiconductor laser, and the sensor 43 is, for example, an optical sensor. The output of the sensor 43 changes continuously while maintaining a specific relationship according to the amount of light received, and is output with respect to the amount of incident light called CCD (charge-coupled device) or PSD (Position Sensitive Detector: trade name). Is used that changes linearly. The sensor 43 emits light from the light source 41, is reduced by the disk-shaped substrate W directly underneath, and generates an output according to the amount of light that arrives. This output indicates the edge W2 of the disk-shaped substrate W. That is, based on the measurement result of the edge measuring device 4, the position and range of the notch portion W1 formed in the edge W2 can be detected, and defects such as burrs and chipping in the edge W2 can be detected.

The encoder 2 detects the amount of rotation of the electric motor 7 corresponding to the rotation angle of the turntable 6 and outputs a digital signal (rotation angle).

The surface imager 5 has an image sensor 51, an illumination 52, and a mirror 53. The image pickup sensor 51 is an image pickup sensor including an image pickup element such as a CCD or CMOS. The surface imager 5 has a half mirror (not shown) and employs so-called coaxial epi-illumination. That is, in the surface imager 5, the light from the illumination 52 is applied to the disk-shaped substrate W by the half mirror, and the reflected light reflected from the disk-shaped substrate W and the mirror 53 is used near the edge of the disk-shaped substrate W. The surface W3 of the image is imaged.

As described above, since the surface imager 5 employs coaxial epi-illumination, the incident light and the reflected light can be made coaxial as much as possible, and the influence of parallax and tilt due to the eccentricity of the disk-shaped substrate W is small. can do. The surface imager 5 sends the captured image to the processing unit 10.
Based on the image captured from the surface imager 5, it is possible to detect an abnormality such as a scratch or a dent on the surface W3 of the edge portion of the disk-shaped substrate W.

FIG. 4 is a functional block diagram showing a main configuration of the processing unit 10 of the alignment device 100 according to the present embodiment. The processing unit 10 includes a correction value calculation unit 11, a determination unit 12, a defect detection unit 13, an exclusion unit 14 (exclusion processing unit), a notch detection unit 15, a surface abnormality detection unit 16 (detection unit), a storage unit 17, and an output. It has a part 18 and the like.

In the processing unit 10, the output from the sensor 43 and the output from the encoder 2 are stored as a pair of data at each constant rotation angle of the turntable 6. In the processing unit 10, the output of the sensor 43 is converted into the distance from the rotation center of the disk-shaped substrate W to the edge W2. As a specific example of this, the distance data obtained by converting the output of the sensor 43 into the distance from the rotation center of the disk-shaped substrate W to the edge W2 and the rotation angle corresponding to the distance data are associated and stored in the storage unit 17. Will be done.
Further, in the processing unit 10, the image captured by the image pickup sensor 51 and the corresponding rotation angle are associated with each other and stored in the storage unit 17.

The storage unit 17 is composed of, for example, a non-volatile storage medium such as a flash memory, EEPROM (registered trademark), HDD, MRAM (magnetoresistive memory), FeRAM (ferroelectric memory), or OUM.
Further, the storage unit 17 stores the threshold value 1 and the threshold value 2 used in the defect detection unit 13 and the notch detection unit 15. Further, the storage unit 17 stores the threshold value used in the surface abnormality detection unit 16.

The correction value calculation unit 11 specifies the position and direction of the disk-shaped substrate W based on the measurement result of the edge measuring device 4, obtains the eccentricity amount, and calculates the correction value for correction to the desired position.

Specifically, the correction value calculation unit 11 generates an edge line representing the edge W2 of the disk-shaped substrate W based on the measurement result of the edge measuring device 4, and uses the generated edge line to determine the position of the disk-shaped substrate W and the position of the disk-shaped substrate W. The direction is specified, and the correction value is obtained based on the specified position and direction.

For example, the edge measuring instrument 4 sequentially measures the edge W2 every 0.036 degrees while rotating the disk-shaped substrate W once. That is, the rotation angle of 10,000 records and the distance data can be obtained in one rotation of the disk-shaped substrate W.

When the distance data thus obtained are continuously connected in ascending or descending order of the rotation angle, continuous data representing the edge W2 can be obtained. When the obtained data is plotted in relation to the rotation angle and the distance data, a curve as shown in FIG. 5 is obtained. In FIG. 5, the vertical axis shows the distance data, and the horizontal axis shows the rotation angle. Hereinafter, the curve shown in FIG. 5 or the continuous data related to the curve is referred to as an edge line.

When the center of rotation of the disk-shaped substrate W is deviated from the center of the disk-shaped substrate W, as shown in FIG. 5, the entire graph depicts, for example, a curved waveform with 360 degrees as one cycle. It becomes. In FIG. 5, L2 is a portion corresponding to the notch portion W1 of the disk-shaped substrate W. Further, L1 and L3 are portions corresponding to chipping of the disk-shaped substrate W. Further, L4 is a portion corresponding to the burr of the disk-shaped substrate W.

The correction value calculation unit 11 obtains the minimum rotation angle and the maximum rotation angle related to each of L1 to L4 from the edge line. The minimum rotation angle and the maximum rotation angle indicate the formation ranges of L1 to L4, respectively.

From the above processing results, the rotation center (position) and the notch portion W1 (direction) of the disk-shaped substrate W can be known. The correction value calculation unit 11 calculates a correction value for correcting the position and direction of the disk-shaped substrate W by using the distance data and the rotation angle corresponding to each distance data. That is, the correction value calculation unit 11 calculates the movement direction and the movement amount for moving the rotation center of the disk-shaped substrate W to the center of the disk-shaped substrate W. Such a calculation method is a known technique, and detailed description thereof will be omitted here. Hereinafter, the calculated movement direction and movement amount are referred to as correction values.

The correction value calculation unit 11 outputs the calculated correction value to the control unit 1 via the output unit 18. Further, the correction value calculation unit 11 may output the calculated correction value to, for example, a mechanism (not shown) for moving the position of the disk-shaped substrate W.

The notch detection unit 15 deletes the notch W1 such as L1 to L4 and the distance data related to the defect from the edge line to generate the notch W1 and the ideal edge line without defects. .. That is, such an ideal edge line is the distance from the rotation center of the disk-shaped substrate W, which is less affected by the notch portion W1 and the defective portion, to the edge W2.
The notch detection unit 15 obtains such an ideal edge line by using the correction value obtained by the correction value calculation unit 11. Such a method of obtaining is a known technique, and detailed description thereof will be omitted here.

The notch detection unit 15 includes distance data related to the edge line (hereinafter referred to as first distance data) and distance data related to the ideal edge line (hereinafter referred to as second distance data) at each rotation speed. Say.) And get the difference sequentially. As described above, since the ideal edge line is the distance data in the state without the notch portion W1 and the defect, the difference between the first distance data and the second distance data is in the radial direction of the disk-shaped substrate W. Represents the size of the notch W1 and the defect in the above.

The notch detection unit 15 detects the notch portion W1 by using the difference between the first distance data and the second distance data. For example, the notch detection unit 15 reads out the threshold value 1 stored in the storage unit 17, and cuts out a portion of the difference between the first distance data and the second distance data corresponding to a value larger than the threshold value 1. Detected as part W1. Thereby, the notch portion W1 (L2 in FIG. 5) of the disk-shaped substrate W can be detected, and the size of the notch portion W1 in the radial direction can also be detected.

Next, the defect detection unit 13 detects the defect of the edge W2 by using the difference between the first distance data and the second distance data. For example, the defect detection unit 13 reads a threshold value 2 smaller than the threshold value 1 from the storage unit 17, and defects a portion of the difference between the first distance data and the second distance data corresponding to a value larger than the threshold value 2. Detect as. As a result, defects (L1, L3, L4) at the edge W2 of the disk-shaped substrate W can be detected. The difference between the first distance data and the second distance data is determined as burrs or chipping depending on whether the value is positive or negative.

The determination unit 12 acquires an image of the surface W3 of the edge portion at each rotation angle captured by the surface imager 5 from the storage unit 17, and determines the boundary W5 between the bevel W4 and the surface W3. For example, the determination unit 12 makes such a determination based on the shading of the captured image.

The light reflected by the bevel W4 from the surface imager 5 cannot be collected by the surface imager 5 because the reflection angle is large. Therefore, in the captured image, the bevel W4 portion appears blackest, so that there is a large difference in shade between the surface W3 and the bevel W4.

Using such a fact, the surface imager 5 determines the position where the shade change is the largest in the shade change with respect to the captured image as the boundary W5. Further, the present invention is not limited to this, and the place where the peak is highest may be determined as the boundary W5 by differentiating the shading change, and the so-called subpixel processing may be performed on the differentiated result.

The exclusion unit 14 performs an exclusion process on the captured image to exclude a surplus region excluding the target region of the surface inspection of the disk-shaped substrate W.

In the alignment device 100 according to the present embodiment, a surface inspection for detecting dents, scratches, etc. on the surface W3 near the edge W2 of the disk-shaped substrate W is performed based on the image captured by the surface imager 5. A part of the background is reflected in the captured image in addition to the edge portion of the disk-shaped substrate W, but the target area of the surface inspection is a portion on the surface W3 side from the boundary W5. On the other hand, in the captured image, the surplus region is a region excluding the target region.

Based on the result of the determination unit 12, the exclusion unit 14 performs an exclusion process for excluding the portion (remainder region) excluding the target region on the surface W3 side from the boundary W5 for the captured image corresponding to each rotation angle. In the captured image subjected to such exclusion processing, only the target area of the surface inspection exists.

The surface abnormality detection unit 16 detects an abnormality in the target region by using the captured image after the exclusion process. The surface abnormality detection unit 16 performs, for example, a process of emphasizing the density difference, and detects a portion where the density (gradation) is equal to or less than a predetermined threshold value as an abnormality portion. The surface abnormality detecting unit 16 detects the range of the abnormal portion by, for example, a labeling process. Based on these detection results, the surface abnormality detecting unit 16 determines whether or not it is a defect.

The detection results by the defect detection unit 13, the notch detection unit 15, and the surface abnormality detection unit 16 are output to, for example, the control unit 1 via the output unit 18.

The control unit 1 includes a CPU, ROM, RAM, and the like. Basically fixed data among various control programs and parameters for calculation are stored in ROM in advance, and RAM temporarily stores the data and reads it out regardless of the storage order, storage position, etc. Is possible. Further, the RAM stores, for example, a program read from the ROM, various data generated by executing such a program, parameters that appropriately change during the execution, and the like.

The CPU loads and executes a control program stored in advance in the ROM on the RAM. For example, the CPU appropriately controls the drive of the turntable 6 (motor 7), the edge measuring device 4, and the surface imager 5. Further, the CPU integrates the detection results of the defect detection unit 13, the notch detection unit 15, and the surface abnormality detection unit 16 to determine the presence or absence of defects in the entire disk-shaped substrate W. Further, the CPU moves or rotates the disk-shaped substrate W based on the correction value calculated by the correction value calculation unit 11.

On the other hand, when the disk-shaped substrate W is placed on the turntable 6, the center of the disk-shaped substrate W and the center of the turntable 6 may not match. If the disk-shaped substrate W rotates as it is, the center of rotation of the disk-shaped substrate W and the center of the disk-shaped substrate W do not match, causing eccentricity and parallax.

When the amount of eccentricity is large, the position of supporting the disk-shaped substrate W on the turntable 6 is also biased, so that the disk-shaped substrate W may be deformed by the weight of the disk-shaped substrate W and the disk-shaped substrate W may be slightly tilted. In this case, the disk-shaped substrate W rotates while being tilted.
Further, if image pickup is performed by the surface imager 5 in a state where parallax is generated, the shadow of the disk-shaped substrate W is reflected in the captured image, which hinders the detection of defects based on the captured image.

The alignment device 100 according to the present embodiment is configured to solve such a problem. Hereinafter, it will be described in detail.

6 to 8 are flowcharts for explaining the processing for the captured image in the alignment device 100 according to the present embodiment. Hereinafter, for convenience of explanation, the disk-shaped substrate W is assumed to rotate by 0.036 °, and measurement is performed by the edge measuring device 4 and imaging by the surface imager 5 at each rotation angle of 0.036 °.

First, the control unit 1 gives an appropriate instruction to the electric motor 7 to start the rotation of the turntable 6 (step S101). At this time, the control unit 1 instructs the edge measuring device 4 to start the edge measurement, and tells the surface imager 5 the surface W3 of the edge portion of the disk-shaped substrate W (hereinafter, simply referred to as the surface W3 of the edge portion). Instruct to start imaging. In response to the instruction from the control unit 1, the edge measuring device 4 starts the edge measurement, and the surface imager 5 starts imaging the edge W2, the bevel W4, the boundary W5, and the surface W3 of the edge portion (step S102). ..

When one rotation of the disk-shaped substrate W is completed, the control unit 1 gives an appropriate instruction to the motor 7 to end the rotation of the turntable 6 (step S103).
After that, the processing unit 10 performs defect detection on the edge W2 and defect detection on the surface W3 of the edge portion (step S104).

First, defect detection at the edge W2 will be described with reference to FIG. 7.

While the disk-shaped substrate W makes one rotation, the edge line is generated by continuously connecting the results (distance data) measured by the edge measuring instrument 4 according to the ascending or descending order of the rotation angles (step S201). .. (See FIG. 5) The generation of the edge line has already been described, and detailed description thereof will be omitted here.

Next, the notch detection unit 15 deletes the notch portion W1 and the distance data corresponding to the defect from the edge line to generate an ideal edge line. Further, the notch detection unit 15 acquires the difference between the distance data (first distance data) related to the edge line and the distance data (second distance data) related to the ideal edge line at each rotation speed. do. Then, the notch detection unit 15 detects a portion of the difference between the first distance data and the second distance data corresponding to a value larger than the threshold value 1 as the notch portion W1 (step S202). The details of the cutout portion W1 have already been described, and detailed description thereof will be omitted here.

Next, the defect detection unit 13 detects a portion of the difference between the first distance data and the second distance data corresponding to a value smaller than the threshold value 1 and larger than the threshold value 2 as a defect of the edge W2 (step S203). ..

The correction value calculation unit 11 calculates a correction value for correcting the position and direction of the disk-shaped substrate W using the distance data and the rotation angle corresponding to each distance data (step S204). Such a calculation method is a known technique, and detailed description thereof will be omitted here.

Next, defect detection on the surface W3 of the edge portion will be described with reference to FIG.

The determination unit 12 determines the boundary W5 between the bevel W4 and the surface W3 with respect to the captured image at each rotation angle captured by the surface imager 5 (step S301). For example, the determination unit 12 determines the position having the largest change in shade as the boundary W5 based on the shade of the captured image.

Next, based on the result of the determination unit 12, the exclusion unit 14 performs an exclusion process for excluding the surplus region excluding the target region on the surface W3 side from the boundary W5 for the captured image corresponding to each rotation angle (). Step S302).

FIG. 9 is a photograph showing before and after executing the exclusion process in the alignment device 100 according to the present embodiment. FIG. 9A shows before the exclusion process is executed, and FIG. 9B shows after the exclusion process is executed. In FIG. 9A, the shadow of the disk-shaped substrate W is reflected due to the eccentricity (tilt) of the disk-shaped substrate W. On the other hand, in FIG. 9B, the surplus region including such a shadow is released, and only the target region for surface inspection exists.

After that, the surface abnormality detection unit 16 performs a filter process for emphasizing, for example, the density difference on the captured image after the exclusion process (step S303). The present invention is not limited to this, and the contrast conversion filter processing may be performed, or the density correction may be performed by using an integrated value obtained by adding the density differences.

FIG. 10 is a photograph showing before and after executing the filter processing on the target area in the alignment device 100 according to the present embodiment. FIG. 10A shows before the filter processing is executed, and FIG. 10B shows after the filter processing is executed. Compared with FIG. 10A, in FIG. 10B, the concentration difference between the abnormal portion and the normal portion in the target region is larger.

Further, the surface abnormality detecting unit 16 detects an abnormality on the surface W3 of the edge portion using the captured image subjected to the density difference enhancement processing as described above (step S304). The surface abnormality detecting unit 16 reads out the threshold value stored in the storage unit 17, detects the portion whose density (gradation) is equal to or lower than the threshold value as the abnormal portion, and detects the range of the abnormal portion by, for example, labeling processing. do. The surface abnormality detection unit 16 determines whether or not such an abnormality portion is a defect based on these detection results.

It will be described again with reference to FIG.
After the defect detection on the edge W2 and the defect detection on the surface W3 of the edge portion are performed (step S104), the processing unit 10 sends data related to these detection results to the control unit 1.

The control unit 1 integrates the data related to the detection result sent from the processing unit 10 (step S105). That is, the control unit 1 confirms the presence / absence of a defect on the edge W2 and the presence / absence of a defect on the surface W3 of the edge portion based on the data related to the detection result, and determines the presence / absence of a defect in the entire disk-shaped substrate W. ..

After that, the control unit 1 aligns the disk-shaped substrate W based on the correction value obtained by the processing unit 10. That is, the control unit 1 moves the disk-shaped substrate W in a specific direction based on the correction value, and corrects the position and direction (step S106).

Next, the control unit 1 outputs the data related to the alignment to the outside (step S107). For example, the control unit 1 outputs the data related to the alignment to the display unit, and the display unit displays such data to the user.

As described above, in the alignment device 100 according to the present embodiment, after excluding the surplus region excluding the target region of the surface inspection from the captured image, the abnormality on the surface W3 of the edge portion is inspected (surface inspection). Therefore, the accuracy of the surface inspection can be improved even when the disk-shaped substrate W is greatly eccentric. Further, since the range to be processed in such a captured image is reduced, the processing load can be reduced.

The correction value calculation unit 11, the determination unit 12, the defect detection unit 13, the exclusion unit 14, the notch detection unit 15, and the surface abnormality detection unit 16 may be configured by hardware logic, or the CPU may be predetermined. It may be constructed by software by executing the program of.

4 Edge measuring instrument 5 Boundary W
12 Judgment unit
14 Exclusion part 16 Surface abnormality detection part 51 Imaging sensor 100 Alignment device W Disc-shaped substrate W3 Surface W4 Bevel W5 Boundary

Claims (3)

In a substrate inspection device that inspects a disk-shaped substrate having a flat surface and an inclined portion formed on the veranda .
A surface imager that captures images of the surface and the inclined portion of the edge portion of the substrate, and
A processing unit that processes the image captured by the surface imager, and
A turntable for mounting the board and rotating the mounted board is provided.
The surface imager is
An image sensor placed above the substrate placed on the turntable, and an image pickup sensor.
Lighting placed above the board placed on the turntable and
It has a mirror that is arranged below the substrate placed on the turntable and reflects the light emitted from the illumination.
The processing unit
When the substrate is rotated by the turntable, the position where the contrast change is the largest is determined based on the shading of the captured image of the edge portion of the substrate and the inclined portion imaged by the image pickup sensor. A determination unit that determines the boundary between the inclined portion and the surface,
An exclusion processing unit that extracts an image of an inspection target area by performing an exclusion processing that excludes the edge side of the captured image from the boundary based on the determination result of the determination unit .
A substrate inspection apparatus having a surface abnormality detecting unit that performs a process of emphasizing a difference in shading on an image of the inspection target area and detects a portion having a density equal to or lower than a threshold value as an abnormal portion . Further equipped with an edge inspector for inspecting the abnormality of the edge of the substrate ,
The edge inspector has an irradiation means arranged above the substrate placed on the turntable.
The surface imager and the edge inspector are separated from each other in the circumferential direction of the substrate so that the irradiation from the illumination of the surface imager and the irradiation from the irradiation means of the edge inspector are not affected by each other. The substrate inspection apparatus according to claim 1, which is arranged in the same manner . The edge inspector is arranged below the substrate placed on the turntable, and after being irradiated by the irradiation means, the sensor is reduced by the substrate to generate an output according to the amount of received light received. Further have
Based on the output of the sensor acquired when the board is rotating by the turntable, information on the distance from the rotation center of the turntable to the edge of the board is calculated, and the calculated distance information is used. The substrate inspection apparatus according to claim 2, wherein the abnormality of the edge of the substrate is inspected based on the above.

Images