JP2005308727A - Inspection device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an inspection device capable of conducting accurate inspection by regulating brightness levels of a reference image and an inspected image to bring the both into proper levels. <P>SOLUTION: This inspection device is provided with a means 23 for calculating a brightness regulation value for regulating the brightness of the first image to a prescribed value, using image information of the first image, a storage means 25 for storing the calculated brightness regulation value, a means 22 for searching the first image having the image information corresponding to the second image different from the first image, the means 22 for reading out the brightness regulation value corresponding to the searched first image from the storage means, and the means 22 for regulating the brightness of the second image by the brightness regulation value read-out from the storage means. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an inspection apparatus, and more particularly to an inspection apparatus for performing an appearance inspection of a semiconductor wafer.

  As a method of inspecting the appearance of a semiconductor wafer, two originally two-dimensional images are picked up by optical means, and by comparing two obtained detection images, different portions in the detection images are obtained. A method of detecting as a defect is generally employed. As described above, conventionally, a difference image between comparison images is generally calculated, and a portion having a large difference value is detected as a defect.

  In the above-described appearance inspection, in order to perform image processing for detecting a pattern defect without being affected by the brightness or distortion of the image, it is essentially the same in the two detected images to be compared. There is known a technique for correcting the shading value of a power pattern and performing shading correction so as to make the shading difference small enough to be recognized as a normal part even if there is a shading difference in a portion having no defect (see Patent Document 1).

  In addition, the first inspection pattern is detected to obtain a first image of the first inspection pattern, the first image is stored, the second inspection pattern is detected, and the second inspection pattern is detected. A first image obtained by obtaining a second image of the pattern to be inspected, converting at least one of the gradations so that the brightness of the stored first image and the second image is substantially the same, and adjusting the brightness. There is also known a technique for inspecting a pattern by comparing the first image and the second image (see Patent Document 2).

As described above, erroneous detection can be prevented by adjusting the shade and brightness of the image. However, when inspecting an object to be inspected or when an inspector uses an inspection image group for checking operation information and inspection performance, not only the brightness of the two images is adjusted but also the two images are Appropriate brightness is required, and optimal inspection is possible by setting the two images to appropriate brightness.
Japanese Patent Laid-Open No. 10-253544 JP-A-11-304718

  The present invention is an inspection apparatus that inspects the state of a defect in an inspected image as compared with a reference image. By adjusting the brightness levels of the reference image and the inspected image so that both are appropriate, an accurate inspection An object of the present invention is to provide an inspection apparatus capable of performing the above.

  In order to solve the above problem, an inspection apparatus according to an aspect of the present invention provides a brightness adjustment value for adjusting the brightness of the first image to a desired value using image information of the first image. Means for calculating; storage means for storing the calculated brightness adjustment value; means for searching for a first image having image information corresponding to a second image different from the first image; Means for reading a brightness adjustment value corresponding to one image from the storage means, and means for adjusting the brightness of the second image by the brightness adjustment value read from the storage means. .

  By adjusting the luminance levels of the reference image and the image to be inspected so that both are appropriate, it is possible to provide an inspection apparatus capable of performing an accurate inspection.

Embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a diagram showing a schematic configuration of an inspection apparatus to which the present invention is applied. The inspection apparatus shown in FIG. 1 is applied to an optical instrument (inspection microscope).
The inspection apparatus according to FIG. 1 includes an inspection microscope 1 and an arithmetic processing device 2.
The inspection microscope 1 irradiates an observation semiconductor wafer 3 (hereinafter simply referred to as “wafer”) through an objective lens 12 by an epi-illumination projection tube 11. The reflected light is observed by the eyepiece 16 and is imaged by an image sensor 17 such as a CCD. An image signal captured by the image sensor 17 is output to the arithmetic processing unit 2. The arithmetic processing device 2 performs arithmetic processing on the image picked up by the image sensor 17 and executes brightness setting and the like based on the calculation result.

A schematic operation of the inspection apparatus according to FIG. 1 will be described.
The irradiation light from the light source 111 of the epi-illumination translucent tube 11 is reflected in the direction of the objective lens 12 by the polarization beam splitter 13 which is an optical path splitting element through the imaging lens 112 and the ND filter 113, and the objective lens 12. The wafer 3 is reached via

  The light incident on the wafer 3 is reflected by the wafer 3, passes through the polarizing beam splitter 13 again through the objective lens 12, and is visually observed by the observer with the eyepiece 16 through the relay lens 14 and the prism 15. Is done. In addition, an imaging element 17 for obtaining an observation image is disposed on the extension of the observation optical path O, and light passing through the polarization beam splitter 13 passes through the prism 15 or is removed from the observation optical path O. An image is picked up by the image sensor 17.

  The image output from the image sensor 17 is input to the image calculation unit 21 of the calculation processing device 2. The image calculation unit 21 calculates luminance information using a non-defective image (hereinafter referred to as “reference image”) captured by the image sensor 17. When the luminance adjustment is necessary for the reference image or the inspection image, the control unit 22 generates a voltage control signal for changing the transmittance of the ND filter 113 or the power supply voltage of the light source 111, and drives the drive motor ( Or the power source of the light source 111). For example, the transmittance of the ND filter 113 changes as the drive motor rotates by a predetermined angle based on the voltage control signal.

  On the other hand, the luminance adjustment value setting unit 23 sets a luminance adjustment value (here, a voltage control signal that changes the transmittance of the ND filter 113) so that the luminance adjustment is appropriate, and the luminance adjustment value storage unit 25 appropriately adjusts the luminance adjustment value. The brightness adjustment value at the time of being stored is stored. In this case, the luminance adjustment value storage unit 25 stores the luminance adjustment value together with the image if necessary.

  The control unit 22 receives luminance information output from the image calculation unit 21 and a storage signal (storage completion signal) from the luminance adjustment value storage unit 25, and controls the entire apparatus based on these signals. On the other hand, after changing the voltage based on the voltage control signal, the control unit 22 controls the inspection apparatus so as to acquire the reference image or the inspection image from the inspection microscope 1 again.

A specific process of the inspection apparatus according to the present embodiment configured as described above will be described.
FIG. 2 is a flowchart of the brightness adjustment process in the defect determination process of the inspection apparatus. In the present embodiment, the following process is basically executed in the brightness adjustment process in the defect determination process. First, an image of a target portion is captured from a reference pattern image having no defect, a luminance adjustment value applied to each image of the target portion is calculated, and a luminance adjustment value corresponding to the image of each target portion is stored. Next, the inspection image is captured, the attention portion of the reference pattern image corresponding to the attention portion of the inspection image is searched, and the luminance of the inspection image is corrected based on the luminance adjustment value of the attention portion of the reference pattern image. Then, defect determination is performed by comparing the corrected inspection image with the reference pattern image. Specifically, it is as follows.

First, an image of each target portion (hereinafter referred to as “reference pattern target image”) is acquired from the reference pattern image (step A1). Next, the luminance adjustment value setting unit 23 adjusts and optimizes the luminance of the reference pattern attention image 23, and sets the luminance adjustment value calculated at that time (step A2). Then, the calculated luminance adjustment value for the reference pattern attention image is stored in the luminance adjustment value storage unit 25 (step A3). Note that steps A1 to A3 are performed for a plurality of images of interest. Next, an image of the attention portion (hereinafter referred to as “inspection pattern attention image”) is acquired from the pattern of the inspection image (step A4). Then, the control unit 22 searches for the reference pattern attention image corresponding to the acquired inspection pattern attention image, and reads out the luminance adjustment value corresponding to each searched reference pattern attention image from the luminance adjustment value storage unit 25 (Step S22). A5) The luminance of the test pattern attention image is corrected according to the luminance adjustment value (step A6). Here, the luminance of the inspection pattern attention image may be corrected by adjusting the luminance adjustment value of the reference pattern attention image by the same procedure as the method of setting the luminance adjustment value of the reference pattern attention image (step A2). You may correct | amend by using the same value as a value as it is. If the same value is used as it is, step A3 and step A5 may be omitted. By the image comparison, the corrected inspection pattern image and the reference pattern image are compared, and defect determination is performed (step A7).
Once the luminance adjustment value is stored in the luminance adjustment storage unit 25 (step A3), in the subsequent defect determination processing flow of the inspection apparatus, it is determined whether the luminance adjustment value is stored in the luminance adjustment storage unit 25. If it is stored and stored, step A4 may be started.

  FIG. 3 is a flowchart showing details of a portion for setting the luminance adjustment value from the reference pattern attention image in step A2 in the flow of FIG. Here, as a value for adjusting the brightness, for example, a voltage control value for changing the transmittance of the ND filter 113 of the inspection microscope (the light amount of the light source itself may be adjusted) will be described. In this case, the shooting condition is changed. However, the luminance may be adjusted by changing the display condition without changing the shooting condition.

  First, a mask region is set so that an image of a defective portion positioned at the center of the visual field with respect to the target image is not captured. For example, the image of interest is divided into a plurality of blocks, and a central block where a defective portion is displayed is set as a mask region (step B1). The number of blocks in this case is not particularly limited. For example, the number of blocks is divided into 4 × 4 blocks, and the central four blocks are set as mask areas. An example of this division is shown in FIGS. 4B and 5B. FIG. 4 and FIG. 5 are diagrams for explaining image regions used in histogram calculation, which will be described in detail later. 4A shows a target image of the reference pattern, and FIG. 5A shows a target image of the inspection image in the case where the defect is located substantially at the center.

  Next, an attention image is acquired (or read out) (step B2). Subsequently, as shown in FIGS. 4C and 5C, the divided block set in step B1 is allocated to the target image, and the luminance information is obtained from the 12 blocks in the peripheral part other than the mask area in the central part. Then, a histogram of luminance information (that is, frequency with respect to luminance) is calculated (step B3). Here, the reason why the mask area in the central portion is excluded when obtaining the luminance information is as follows. When a defect on the wafer is imaged by the inspection microscope 1, the defect is positioned at the center of the visual field of the inspection microscope 1 (the optical axis of the objective lens 12). Therefore, if the central portion in the inspection image is set as a mask region, a histogram of luminance information can be calculated from image information other than the central portion that does not include a defect, without being affected by the defect, The brightness adjustment value for the inspection image can be set accurately. As the histogram, for example, a histogram as shown in FIG. 6 is obtained. FIG. 6 is a diagram showing examples of various histograms. In addition, since most of the defects are displayed in the central portion of the inspection image where the center of the visual field is the mask region, a circular or rectangular mask region larger than the defect can be set in the central portion of the inspection image. As described above, the mask area only needs to be set in an area where defects in the inspection image are displayed, and can be set at a position other than the center.

  Next, the maximum value of luminance and the total number of luminance frequencies (cumulative frequency) that are equal to or greater than the threshold value T_MAX (maximum luminance limit allowable as the luminance maximum value) are acquired from the calculated histogram (step B4).

  Next, it is checked whether or not the maximum luminance value is between the threshold value T_MAX and the threshold value T_MIN (minimum luminance limit allowable as the maximum luminance value) (step B5).

  In step B5, if the maximum luminance value is not between threshold value T_MAX and threshold value T_MIN (No in step B5), the maximum luminance value is greater than T_MAX, and the cumulative frequency at threshold value T_MAX is the threshold value. It is determined whether or not it is smaller than H_MAX (threshold value regarding cumulative frequency) (step B6). In FIG. 6, the threshold value H_MAX is set to a predetermined value so that the influence of noise can be ignored.

  In step B5, as shown in FIG. 6C, if the maximum value of the luminance is between the threshold value T_MAX and the threshold value T_MIN (Yes in step B5), the voltage control of the currently set ND filter 113 is performed. The value is set as a luminance adjustment value (that is, a value capable of setting the luminance of the target image to an appropriate luminance). (Step B7). In step B6, as shown in FIG. 6D, if the maximum luminance value is larger than T_MAX and the cumulative frequency at the threshold T_MAX is smaller than the threshold H_MAX (threshold for the cumulative frequency) ( As in the case of Yes in step B6), the voltage control value of the ND filter 113 that is currently set is set as the luminance adjustment value. (Step B7).

If No in step B6, control is performed to change the voltage control value of the ND filter 113 according to the current maximum luminance value (step B8). Specifically, it is as follows.
If the maximum value is equal to or less than T_MIN, the image is considered too dark as shown in FIG. 6A. Therefore, the voltage control value of the ND filter 113 is controlled (adjusted) so as to become brighter.
If the maximum value is equal to or higher than T_MAX and the cumulative frequency of luminance equal to or higher than T_MAX is equal to or higher than H_MAX, the image appears to be too bright as shown in FIG. Control (adjust) the control value.
Then, after adjusting the voltage control value of the ND filter 113, image acquisition of the same target portion is performed again.

  In this case, the brightness adjustment value may be expressed as, for example, a characteristic (linear format or nonlinear expression) when converting the brightness of the image. For example, the luminance characteristics are changed so that the luminance frequency indicating the bright portion and the luminance frequency indicating the dark portion are each equal to or less than a predetermined value (or so that the median value and the mode value of the histogram are within a predetermined range). Also good. In this case, adjustment is possible by changing the gain or changing the γ coefficient, for example. In this way, the brightness of the image can be changed without changing the shooting conditions, so that processing by software becomes possible.

  FIG. 7 is a diagram showing another adjustment method when setting the luminance adjustment value from the histogram. In the method shown in FIG. 7, only T_MAX is used as the threshold value. Also, it has an LUT (Look Up Table) by associating the voltage adjustment level with the luminance of the image.

  First, it is assumed that a histogram as shown in FIG. 7A is obtained when a histogram of an attention image is acquired. In this case, the maximum luminance value exceeds the threshold value T_MAX, and excessive light is incident on the image sensor. If the minimum luminance value in FIG. 7A is the luminance value a, the voltage adjustment level Va is set so that the minimum luminance value a becomes the luminance value 0 using the LUT shown in FIG. adjust.

  As a result of the voltage adjustment as described above, it is assumed that the histogram of the target image as shown in FIG. In the histogram shown in FIG. 7C, the maximum luminance value is less than the threshold value T_MAX. Therefore, the maximum luminance value b is obtained by adjusting the voltage adjustment level Vb at this time using the LUT as shown in FIG. 7D so that the maximum luminance value matches the threshold value T_MAX. It is made to coincide with the threshold value T_MAX. As a result, an image that is brighter than the image for calculating the histogram as shown in FIG.

  Finally, a histogram as shown in FIG. 7E is completed, and the voltage adjustment level at this time may be set as an appropriate luminance adjustment value. If a histogram as shown in FIG. 7C is obtained at the time of the first histogram calculation, the subsequent processing uses the LUT in FIG. 7D so that the maximum luminance becomes an appropriate value. If it is set to an appropriate value as shown in FIG. 7E at the initial stage, the voltage adjustment level at that time may be adopted as an appropriate luminance adjustment value.

  Further, when the histogram is FIG. 7 (a), adjustment is performed by the LUT of FIG. 7 (b), and when the histogram is FIG. 7 (e), the histogram is that of FIG. 7 (e). The voltage adjustment level may be set as an appropriate brightness adjustment value.

  Next, an inspection apparatus according to another embodiment of the present invention will be described with reference to FIGS.

  The wafer is placed on an imageable area of the microscope, and inspection is started. First, wafer alignment is performed. In general, when a wafer is placed in an imageable area, precise alignment is often not performed. Therefore, precise alignment is performed using alignment marks made on the wafer (see FIG. 9). By using two or more alignment marks and aligning them with the center position of the microscope, the positional and rotational deviations are corrected (step C1).

  Next, it is determined whether or not there is light amount data (luminance adjustment value) in the setting file (step C2). If there is no light quantity setting data, the first image (reference pattern attention image) is acquired, and the cell moves to a cell having no defect adjacent to the alignment mark. Here, whether or not there is a defect may be determined based on information from another defect inspection apparatus, or may be determined based on whether or not the luminance when an adjacent cell is imaged is within a set value range. . Then, a value input in advance as an intermediate luminance is set as the light amount setting data (step C3). If there is light quantity setting data in step C2, the data is read from the setting file and set (step C4). And a 1st image is imaged (step C5). It is determined whether the captured first image is captured with the optimum light amount (step C6). Here, the determination is made based on a predetermined range in which the total light amount is within a certain range or a part of the image is not less than a certain luminance value and below a certain luminance value. . If it is not within the range, the amount of light is adjusted and the first image is captured again (step C9). In step C6, when the total light amount is within the range, the light amount setting data for the first image used at that time is written in the setting file (brightness adjustment value storage unit 25) (step C7). Then, the second image (inspection pattern attention image) is imaged using the light amount setting data for the first image as it is (step C8). As described above, the adjustment of the light amount setting data (luminance adjustment value) of the first image is performed at the time of alignment, so that the stage moving time can be reduced. Further, since the light amount setting data of the first image is used as it is, it is not necessary to set the light amount setting data (brightness adjustment value) again when acquiring the second image (inspection pattern attention image). Therefore, the inspection time can be reduced.

  The following invention is extracted from each of the above embodiments. Note that the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. Further, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine suitably the component covering different embodiment.

  An inspection apparatus according to an embodiment of the present invention uses image information of a first image (for example, an image having no defect such as a reference image) to adjust the luminance of the first image to a desired value. Means for calculating a brightness adjustment value; storage means for storing the calculated brightness adjustment value; and first information having image information corresponding to a second image different from the first image (for example, an image to be inspected). Means for searching for an image; means for reading a brightness adjustment value corresponding to the searched first image from the storage means; and adjusting the brightness of the second image by the brightness adjustment value read from the storage means And a means for performing the above. In addition, it is preferable to further comprise means for performing an inspection using the first image and the second image whose brightness has been adjusted. In this way, in the inspection by image comparison, a luminance adjustment value for adjusting the luminance adjustment of one image to an appropriate luminance is obtained, and the luminance of the other image is adjusted by the already adjusted luminance adjustment value. Therefore, both images can be observed in accordance with an appropriate luminance to facilitate inspection. Therefore, the automatic inspection enables accurate inspection (that is, accurate defect detection).

  In the above inspection apparatus, the following embodiments are preferable. The following embodiments may be applied independently or may be applied in appropriate combination.

  (1) The image information used for calculating the brightness adjustment value is image information of a peripheral portion of the first image. Usually, the defect is often located at the center of the image. Therefore, by calculating the brightness adjustment value using the peripheral image information, the defect image can be adopted as the image for the first brightness adjustment, and the brightness adjustment value for the defect image is calculated. However, appropriate brightness adjustment can be performed without being adversely affected by defects.

  (2) The means for calculating the brightness adjustment value obtains the brightness of the image information of the first image, and determines the brightness adjustment value using statistical information based on a histogram using the brightness. Here, the statistical information means a maximum value, a minimum value, a mode value, a median value, and the like, and the brightness adjustment value can be optimized by using actual image information.

  (3) The luminance adjustment value is at least one of a value representing a characteristic for converting the gradation of an image and a value representing a control characteristic for adjusting a means for determining the brightness of the image. Regardless of whether the shooting conditions are changed (that is, the mechanism for determining brightness is adjusted by hardware) or the image after shooting is converted (that is, software is used for image processing) to adjust the brightness of the image. Since it is good, it is possible to obtain optimum shooting conditions by changing the shooting system. In addition, when image conversion is performed, luminance adjustment independent of the device configuration is possible. In the case where the brightness adjustment is performed in hardware, the voltage applied to the ND filter, the light source voltage, and other light amount adjustment mechanisms may be used.

  The present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the invention at the stage of implementation. Further, the above embodiments include inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements.

  In addition, for example, even if some structural requirements are deleted from all the structural requirements shown in each embodiment, the problem described in the column of the problem to be solved by the invention can be solved, and the effect described in the effect of the invention Can be obtained as an invention.

The figure which shows schematic structure of the test | inspection apparatus with which this invention is applied. The flowchart of the brightness adjustment process in the defect judgment process of an inspection apparatus. The flowchart which shows the detail of the part which sets a luminance adjustment value from the attention image of a reference pattern. The figure of description of the image field used by histogram calculation. The figure of description of the image field used by histogram calculation. The figure which shows the example of various histograms. The figure which shows the other adjustment method in the case of setting the luminance adjustment value from a histogram. The flowchart which shows operation | movement of the test | inspection apparatus which concerns on other embodiment of this invention. Explanatory drawing of operation | movement of the test | inspection apparatus which concerns on other embodiment of this invention.

Explanation of symbols

  O ... Observation optical path, 1 ... Inspection microscope, 2 ... Processing unit, 3 ... Semiconductor wafer, 11 ... Translucent tube for epi-illumination, 12 ... Objective lens, 13 ... Deflection beam splitter, 14 ... Relay lens, 15 ... Prism , 16 ... eyepiece, 17 ... image sensor, 21 ... image calculation unit, 22 ... control unit, 23 ... luminance adjustment value setting unit, 25 ... luminance adjustment value storage unit, 111 ... light source, 112 ... imaging lens, 113 ... ND filter.

Claims (7)

Means for calculating a luminance adjustment value for adjusting the luminance of the first image to a desired value using image information of the first image;
Storage means for storing the calculated brightness adjustment value;
Means for searching for a first image having image information corresponding to a second image different from the first image;
Means for reading out the brightness adjustment value corresponding to the searched first image from the storage means;
An inspection apparatus comprising: means for adjusting the brightness of the second image based on the brightness adjustment value read from the storage means. 2. The inspection apparatus according to claim 1, further comprising means for performing an inspection using the first image and the second image whose brightness has been adjusted. The inspection apparatus according to claim 1, wherein the image information used for calculating the brightness adjustment value is image information of a peripheral portion of the first image. 2. The inspection apparatus according to claim 1, wherein the means for calculating the brightness adjustment value obtains brightness of the image information of the first image, and uses the statistical information based on a histogram using the brightness to obtain the brightness adjustment value. An inspection apparatus characterized by determining. 2. The inspection apparatus according to claim 1, wherein the brightness adjustment value is at least one of a value representing a characteristic for converting a gradation of an image or a value representing a control characteristic for adjusting a means for determining the brightness of the image. Inspection device characterized by a value. 6. The inspection apparatus according to claim 1, wherein a brightness adjustment value by the means for adjusting the brightness of the second image is the same as the brightness adjustment value of the first image. Inspection device characterized by The inspection apparatus according to any one of claims 1 to 6, wherein the first image is an image adjacent to a mark for alignment.

Images