TW201514479A - Defect viewing device and defect viewing method - Google Patents

Abstract

Although highly sensitive at detecting some types of defects, a conventional technique has reduced sensitivity to other types of defects. The conventional technique also has drawbacks in that mechanical switching of filters increases the amount of time taken to acquire a plurality of images under different detection conditions. The present invention is a defect viewing device provided with SEM and optical microscopes, and a control unit, the defect viewing device characterized in that the optical microscope is provided with: a radiating system for radiating light to a sample; and a detection system having a spatial filter, a spatial shape of which can be electrically controlled, and a distributed polarizer, a polarization state of which can be electrically controlled, for detecting a signal based on light from the sample irradiated by the radiating system; and the control unit generates a synchronization signal (202), controls and electrically switches the polarization state of the distributed polarizer and the spatial shape of the spatial filter on the basis of the generated synchronization signal (207, 206), and processes an image detected by the detection system in a plurality of combinations of polarization states of the distributed polarizer and spatial shapes of the spatial filter (123).

Description

Defect observation device and method thereof

Regarding the defect observation device and the defect observation method.

The defect inspection of foreign matter defects and pattern defects on the semiconductor wafer in the semiconductor manufacturing process is performed by using the defect position detection of the visual inspection device and the defect observation using the defect observation device, and is refined based on the observation result of the defect. The procedure that should be dealt with. The progress of the miniaturization of the semiconductor pattern and the fine defects also affect the yield. Therefore, an SEM (Scanning Electron Microscope) is used for the observation device. Since the visual inspection device and the SEM-type observation device are different devices and there is a deviation of the table coordinates, only the defect position information detected by the visual inspection device is used, and the defect positioning system is performed in the field of view of the SEM-type observation device. difficult.

In particular, in the inspection device for a non-patterned wafer, in order to increase the throughput of the inspection, the spot size of the laser beam for dark field illumination on the surface of the semiconductor substrate is increased to scan the surface of the semiconductor substrate. Irradiation, so the accuracy of the position coordinates obtained from the position of the laser beam spot scanned on the surface of the semiconductor substrate is large. The error component. When the SEM is used to observe the defect in detail based on the positional information of the defect containing the large error component, it becomes difficult to store the defect in the field of view of the SEM which is observed at a magnification far higher than that of the optical foreign matter inspection apparatus. .

In order to solve this problem, Patent Document 1 (Japanese Laid-Open Patent Publication No. 2011-106974) discloses a method of performing dark-field optics mounted on an observation device when performing defect observation using a SEM-free pattern wafer. The microscope detects the position of the defect, and uses the detected position coordinates to perform imaging of the observed image of the SEM. In addition, in the method of detecting high-sensitivity defects on the unpatterned wafer, it has been disclosed that the position of the defect on the wafer is detected after the distributed polarizing element and the spatial filter are added to the detection light path of the dark-field microscope. method.

[Previous Technical Literature] [Patent Literature]

[Patent Document 1] Japanese Patent Laid-Open Publication No. 2011-106974

The defect observation using the SEM unpatterned wafer is expected to be high-sensitivity and high throughput for the defect type for re-detection using the defect positioning of the optical microscope.

Patent Document 1 discloses that a dark field light is mounted. The configuration of an electron microscope for observing the defect of the system is such that the distributed polarizing element and the spatial filter are disposed on the pupil plane of the detection light path.

However, in Patent Document 1, the detection of the defect only reveals the detection by a specially designated filter, and there is a problem in that a certain specified defect can be detected with high sensitivity, but the defect type of the other type will be reduce. Further, when the filter is switched and an image having different detection conditions is imaged, there is no disclosure of a method of simultaneously switching both the distributed polarization element and the spatial filter, and in addition, in the method of switching the mechanical filter disclosed. It is a problem that it takes time to obtain a plurality of images having different detection conditions.

An object of the present invention is to provide a defect detecting device and method and a defect observing device using the same, which can be used for visual inspection in a detailed observation using SEM for defects detected by a semiconductor wafer visual inspection device. The defect to be detected is not limited to the type thereof, and is highly sensitive and re-detected at a high speed, and the defect is surely dropped into the observation field of the SEM based on the re-detection position.

In order to solve the above problems, for example, a configuration described in the scope of the patent application is employed.

The present invention includes a plurality of means for solving the above problems, and the optical microscope includes an irradiation system and a detection system for irradiating light with a sample. The measurement system detects a signal based on light from the sample irradiated by the illumination system, a distributed polarization element that electrically controls the polarization state, and a spatial filter that controls the spatial shape of the electrical property. The control unit generates a synchronization signal, and electrically switches the polarization state of the distributed polarization element and the spatial shape of the spatial filter based on the generated synchronization signal, and performs polarization on the polarization states of the plurality of distributed polarization elements. The image detected by the aforementioned detection system is processed under the combination of the spatial shapes of the spatial filters.

According to the present invention, it is possible to provide a defect observation apparatus and a method thereof capable of detecting a defect position with high sensitivity and high speed.

101‧‧‧ wafer

102‧‧‧SEM

103‧‧‧Light microscope

104‧‧‧Workbench

105‧‧‧vacuum tank

106‧‧‧Control Department

107‧‧‧ Terminal

108‧‧‧ memory device

109‧‧‧Network

110‧‧‧Light source

111‧‧‧Vacuum sealed window

112‧‧‧Mirror

113‧‧‧Contact objective

114‧‧‧Vacuum sealed window

115‧‧‧ imaging optical system

116‧‧‧Photographic components

117‧‧‧Distributed polarizing elements

118‧‧‧ Spatial Filter

119‧‧‧Workbench control circuit

120‧‧‧SEM camera system control circuit

121‧‧‧Optical camera system control circuit

122‧‧‧External input and output I/F

123‧‧‧CPU

124‧‧‧ memory

125‧‧‧ busbar

201‧‧‧Information I/F

202‧‧‧Synchronous control circuit

203‧‧‧Synchronous signal generation circuit

204‧‧‧Image Information Memory Department

205‧‧‧Filter state control circuit

206‧‧‧Distributed polarizing element control circuit

207‧‧‧Spatial filter control circuit

208‧‧‧Internal busbar

Fig. 1 is a configuration diagram of a defect observation device according to the present invention.

2 is a diagram showing the detailed configuration of an optical imaging system control circuit 121 of the defect observation apparatus according to the present invention.

Fig. 3 is a flow chart showing an optical microscope image capturing step in the defect detecting method according to the present invention.

Fig. 4 is a timing chart of optical microscope image capturing in the defect observation method according to the present invention.

Fig. 5 is a view showing a detailed configuration of an optical microscope of a defect observation apparatus according to the present invention.

Fig. 6 is a flow chart showing the steps of calculating the defect coordinates in the defect observation method according to the present invention.

[Fig. 7] An overall flow chart of a defect observation method related to the present invention.

[Example 1]

Hereinafter, embodiments of the present invention will be described in detail using appropriate drawings.

Fig. 1 is a view showing the configuration of a defect observation apparatus according to the present invention.

The defect observation device of this embodiment is a device for observing defects on a wafer which is generated in the process of the semiconductor device.

101 is a wafer to be inspected. The 102-series is an electron microscope (hereinafter referred to as SEM) for observing the wafer 1 in detail, and the 103 is an optical microscope that optically detects defects on the wafer 1 to obtain defect position information. The 104 system can mount the wafer 1 table, and can move any place of the wafer 1 into the field of view of the SEM 102 and the optical microscope 103. In the 105-series vacuum chamber, the SEM 102, the stage 104, and the objective lens 113 of the optical microscope 103 are housed therein.

The inside of the optical microscope 103 will be described. 110 series illumination source. The laser light emitted by the illumination source 110 passes through the vacuum sealing window 111, and is reflected by the mirror 112 for controlling the illumination position, and is irradiated to an arbitrary position on the surface of the wafer 101. The 113 series is provided with an objective lens for illuminating the scattered light reflected from the sample 101. Through the objective lens 113 The light system is imaged on the imaging element 116 by the imaging optical system 115 through the vacuum sealing window 114. The imaging optical system 115 is provided with a distributed polarizing element 117 and a spatial filter 118 that electrically control the polarization state and the spatial shape.

The control unit 106 is composed of a table control circuit 119, an SEM imaging system control circuit 120, an optical system control circuit 121, an external I/F 122, a CPU 123, and a memory 124, and a table control circuit 119 to a memory. Each of the components 124 is connected to the bus bar 125 so that information can be input and received to each other. The table control circuit 119 controls the table 104, and the SEM camera system control circuit 120 controls the SEM 102 and detects the video signal to the memory 124. The optical system control circuit 121 performs control of the imaging element 116 of the optical microscope 103, the distributed polarization element 117, and the spatial filter 118, and memorizes the image signal obtained from the imaging element 116 to the memory 124. The external input/output I/F 122 is used to perform display information output to the terminal 107, information input from the terminal 107, and information input to the memory device 108, and the defect inspection device (not shown) is performed through the network 109. Information output from the upper management system, etc. The image data stored in the memory 124 is processed by the CPU 123.

In the defect observation apparatus configured as described above, the optical microscope 103 has a configuration in which the position of the defect on the wafer 101 is determined by the position information of the defect detected by the defect inspection device (not shown). The function of re-detection (hereinafter referred to as detection) is that the control unit 106 has a defect based on detection by the optical microscope 103. The function of the position correction means for correcting the position information of the defect by the position information, the SEM 102 has a function of observing the defect based on the defect position information corrected by the control unit 106. The image signal obtained from the optical microscope stored in the memory 124 is processed by the CPU 123 to detect the position of the defect, thereby correcting the position information of the defect outputted from the defect inspection device in the memory 124. The stage 104 has a configuration in which a defect that can be moved to be detected by the optical microscope 103 can be observed by the SEM 102.

Fig. 2 is a view showing the detailed configuration of the optical imaging system control circuit 121 of the defect observation apparatus according to the present invention.

The optical imaging system control circuit 121 includes a data I/F 201, a synchronization signal control circuit 202, a video information storage unit 204, a filter state control circuit 205, a distributed polarization element circuit 206, and a spatial filter control circuit 207. It is connected to the internal bus 208. The data I/F 201 is connected to the internal bus bar 208 and the bus bar 125 in the control unit 106, and performs data between the optical imaging system control circuit 121 and the other processing units of the control units 106 and 119 to 124. Grant. The synchronizing signal generated by the synchronizing signal generating circuit 203 in the synchronizing signal control circuit 202 is used by the synchronizing signal control circuit 202 for the trigger signal of the imaging start of the imaging element 116 or the image obtained from the imaging element 116. The trigger signal of the signal to the beginning of the memory of the image information storage unit 204. The distributed polarization element circuit 206 performs control of the distributed polarization element 117, and the spatial filter control circuit 207 controls the spatial filter 118. Filter state control circuit 205 for distribution The polarizing element control circuit 206 and the spatial filter control circuit 207 instruct the signals of the distributed polarization element 117 and the spatial filter 118 to be synchronized with the signals of the synchronous signal control circuit 202.

The operation of the circuit and the optical system shown in Fig. 2 will be described with reference to the processing flow of Fig. 3 and the timing chart of Fig. 4. Hereinafter, it is assumed that the filter state P refers to the polarization state of the distributed polarization element 117 or the information given to the distributed polarization element control circuit 206 in order to control the distributed polarization element 117 to the state, and the filter state S refers to the spatial filter 118. The state of the spatial shape of the filter or the information of the spatial filter control circuit 207 is given to control the spatial filter 118 to this state. In addition, (P _k , S _k ) (k=1, . . . , N) is used to indicate that the filter states P _k and S _{k are used} in a group of processed cases, and the filter state (P _{k )} , S _k ), refers to the data of the filter state P _k and the filter state S _k being one set.

Fig. 4 is a timing chart of optical microscope image capturing in the defect observation method of the present invention.

The synchronous signal shown in FIG. 4 is a signal generated by the synchronous signal generating circuit 203 of FIG. 2. The command signal is as follows: synchronized from the synchronous signal and outputted from the synchronous control circuit 202 to the filter state control circuit 205, image information memory. The portion 204 and the imaging element 116 are controlled for the processing steps shown in FIG. The state of polarization of the distributed polarizing element 117 shown in Fig. 4 indicates the state of the distributed polarizing element 117 at each time point, and this state is represented by P ₁ to P ₃ . The spatial shape of the spatial filter 118 shown in Fig. 4 indicates the state of the spatial filter 118 at each time point, and this state is represented by S ₁ to S ₃ . Further, the operation of the image pickup device 116 shown in FIG. 4 indicates the content of the captured image at each time, and the image is recorded as the image ₁ to the image _3, and the memory operation of the image information storage unit 204 is indicated by The content of the memory image at each time is represented by the image ₁ to the image ₃ . The image data captured by the image sensor 116 is transferred to the image information storage unit 204 and stored in synchronization with the next synchronization signal. The transfer of the image data shown in FIG. 4 to the memory 124 indicates the transfer timing of the image data from the video information storage unit 204 to the memory 124.

In FIG. 4, the polarization state of the distributed polarization element 117 is changed, the polarization state of the distributed polarization element 117, and the spatial shape of the spatial filter 118 are changed, and the imaging by the imaging element 118 is started, but the rising edge of the synchronization signal is from the filter state control circuit. 205 reads the filter state (P _k , S _k ), changes the polarization state of the distributed polarization element 117, and the spatial shape of the spatial filter 118, and may image the imaging element 116 from the falling edge of the synchronization signal. Thereby, it is expected that imaging is started after the state of the filter of the distributed polarization element 117 and the spatial filter 118 is stabilized at the imaging start time point of the imaging element 118.

Fig. 3 is a flow chart showing an optical microscope image capturing step in the defect detecting method of the present invention.

First, the filter state control circuit 205 memorizes the filter states (P _k , S _k ) (k = 1, ..., N) (S301). The filter state (P _k , S _k ) is transferred to the filter state control circuit 205, which is stored in the memory medium not shown in FIG. 1 and bonded to the pre-memory 124, or the memory device 108, or the network 109. memory.

Next, at time t ₁ (S302) shown in FIG. 4, the filter state (P _k , S _k ) is read from the filter state control circuit 205, and the filter state data P _{k is} output to the distributed polarization element control circuit 206. The filter state data S _{k is} output to the spatial filter control circuit 207, and the polarization state of the distributed polarization element 117 and the spatial shape of the spatial filter 118 are changed (S303). This change may be obtained by the start of imaging of the video 1 by the imaging element 116, and the state of the distributed polarization element 117 and the spatial filter 118 may be changed simultaneously or at different times.

Also at time t _1, by irradiation by light source 110 and the image pickup element 116 to start imaging a wafer (image ₁₎ 101, the image data of the captured image according to _a time t ₂ of the command signal to the image information memory Memory Section 204 (S304).

S303 and S304 are repeated for the portions of the filter state (P _k , S _k ) (k=1, . . . , N) (S305, S306). The repetition of this processing is as shown in Figure 4, synchronized with the command signal. FIG. 4 shows an example of N=3, but the N series is not limited to three.

After the image data of the filter state (P _k , S _k ) (k=1, . . . , N) is restored to the image information storage unit 204, the image data is transferred from the image information storage unit 204 to the memory 124. (S307). In FIG. 4, the transfer timing is displayed in the form of the image data stored in the image information storage unit 204 after being synchronized with the next synchronization signal, but the transfer start system is only the image data of the image information storage unit 204. After the memory, it can be any time, not limited to Figure 4. Further, the transfer time shown in FIG. 4 changes depending on the specifications of the transferred image data capacity, the internal bus bar 208, the bus bar 125, the CPU 123 of the control unit 106, the memory 124, and the like.

Finally, the image data stored in the memory 124 is processed by the CPU 123, and the defect position is specified (S308), and the defect position is written in the memory 124. When imaging with a defect of SEM, the CPU 123 reads the specially designated defect coordinates (position information) from the memory 124, converts it into a table coordinate, and gives the table coordinates to the table control circuit 119, so that The workbench can be moved to the corrected defect location. . The processing contents of S308 will be described later using FIG.

In the above, the processing shown in FIG. 3 is referred to as S300 as a whole, and is referred to in the processing flow shown in FIG.

The characteristics of the distributed polarization element 117 and the spatial filter 118 are determined in accordance with the type of defect of the object to be improved in the defect detection sensitivity, and are disclosed in Patent Document 1. For this reason, there is a problem that when an appropriate filter characteristic is set for the detection sensitivity of a certain defect type, the optimum detection sensitivity cannot be obtained for other types of defects. In order to solve this problem, the present method and apparatus are as follows: switching the characteristics of the distributed polarization element 117 and the spatial filter 118 to image an image, and performing defect detection using a plurality of images, thereby achieving acquisition without biasing the specification The sensitivity of the type of defect.

The change in the type of defect that needs to be considered from the viewpoint of the optical microscope is not the type of defect in the yield management as the type of defect expected to occur in the inspection program, but the defect that should be considered when testing with an optical microscope. The type refers to a change in the shape of the defect that is roughly distinguished by a shape such as a concave or convex shape, or a change that is substantially different depending on the optical characteristics of the defect. To this end, the number of filter states (P _k , S _k ) that should be imaged can be defined.

In an example of the electrically conductively controlled distributed polarizing element 117, there is a liquid crystal in which birefringence changes depending on an applied voltage. For a liquid crystal element composed of a plurality of pixels to which a voltage different in pixel can be applied, the applied voltage is controlled so that a distribution of a desired optical axis can be made in the filter face. Further, in terms of the spatially controllable spatial filter 118, a DMD (Digital Mirror Device) or the like is provided. The configuration of the optical microscope of Figs. 1 and 2 is simplified for the sake of explanation, but the optical element that reflects light by the DMD used as a spatial filter cannot transmit light. To this end, it is necessary to work on the optical path as shown in FIG.

Fig. 5 is a view showing the detailed configuration of an optical microscope of the defect observation apparatus according to the present invention.

In Fig. 5, the imaging optical system 115 is a lens 502 that images the pupil plane 501 of the objective lens 113, a mirror 503 that reflects the optical path to the spatial filter 118 of the DMD, a distributed polarizing element 119, and an image. The lens 504 is imaged. A distributed polarizing element 119 is disposed at a position 505 where the pupil plane 501 of the objective lens 113 is imaged. The spatial filter 118 is only required to have a spatial resolution required for the shape of the space to be controlled, and thus can be placed at a position where the position 505 imaged by the pupil plane 501 of the objective lens 113 is defocused on the optical path. In addition, as long as it is for borrowing The effect of detecting the sensitivity of the defect by the inter-filter 118 and the distributed polarization element 119 is not troublesome, and the spatial filter 118 may be disposed at the position 505 where the pupil plane is imaged, and the distributed polarization element 119 may be disposed on the pupil plane. In the vicinity of the light path of the position 505, both the spatial filter 118 and the distributed polarization element 119 are disposed in the vicinity of the position 505 where the pupil plane is formed on the light path.

Figure 6 is a flow chart showing the steps of calculating the defect coordinates in the defect observation method related to the present invention.

FIG. 6 illustrates a processing method for specifying a defect position by using N images of the filter state (P _k , S _k ) (k=1, . . . , N). Figure 6 is a detail of the processing of S308 of Figure 3. Therefore, the processing in Fig. 6 is all performed by the CPU 123.

The pixel average darkness value (average value _k ) of the image _k and the value (3σ _k ) of three times the standard deviation are obtained (S602).

Using the average value obtained from _{k, 3σ k} to generate a normalized image _k (S603). The α and β in the calculation formula shown in S603 are provided by a coefficient for controlling the calculation result in a range in which the pixel value of the normalized image _k can be obtained, and may be arbitrarily set.

S602 and S603 are repeated N times of the number of images to obtain a normalized image _k (k=1, . . . , N) (S605). At S606, the maximum pixel value of the same coordinate (i, j) of the N normalized images _{k is} the pixel value of the integrated image (i, j). The k specifying the filter state given to the maximum value is stored in the memory 124 as max_k. Recording the state of the filter in which the defect is most visible, so that the property of the defect corresponding to the filter, such as the unevenness, is exposed in the SEM observation image, and is used in the generation of the SEM observation image. For example, when an observation image is generated by a secondary electron image or a reflected electron image detected by SEM, the mixing ratio of each image is changed by a defect.

Next, the integrated image (i, j) is binarized according to the predetermined defect detection threshold TH to obtain a binary image (S607).

The label area of the largest area is detected from the label image tagged for the binary image (i, j), and the center of gravity of the largest area label is the defective coordinate (S608, S609).

The defect position detected as the last detected image coordinate is converted into a table coordinate (S610).

Figure 7 is an overall flow chart of the method of observing the defects associated with the present invention. The sequence of collecting SEM defect images using the corrected defect coordinates is illustrated by a flow chart.

First, the wafer to be observed is loaded on the stage 104 shown in Fig. 1 (S701).

Then, the defect coordinate data of the defect detected by the inspection device beforehand is read into the memory 124 through the external input/output I/F 122 of the overall control unit 106 (S702), and the defect of the M point as the observation target is selected therefrom ( S703). The selection of the defect is performed by the CPU 123 according to a preset program, and the operator can also select through the terminal 107.

The alignment of the wafer is then performed (S704). This is the case where the workbench 104 is moved based on the position of the defective coordinates described by the coordinates on the wafer, in order to create the position of the target defective coordinate. To the field of view of the SEM 102 and the center of the field of view of the optical microscope 103, the coordinates on the wafer are associated with the table coordinates using well-known positioning marks (alignment marks). This assignment result is stored in the memory 124 as alignment information.

Next, the defect positions are corrected for the defects 1 to M selected as the observation targets (S705, S708, and S709).

First, the defect m is moved to the field of view of the optical microscope 103 (S706). This movement is performed by calculating the coordinates of the workbench corresponding to the defect _m by the CPU 123 based on the defect coordinate data and the alignment information stored in the memory 124, thereby driving the workbench through the table control circuit 119. 104.

After the movement of the stage is completed, the position of the defect _m is specified by the process shown in Fig. 3 (S300), and the position of the specially designated defect is stored in the memory 124 as the corrected defect position _m (S707).

The sequence of S706, S300, and S707 above is performed for the defect m (m = 1, ..., M). According to the inspection apparatus, there is also means for outputting not only the detected defect position coordinates but also information on the characteristics of the defects. For example, if the defect information is known to be convex or concave in advance based on the feature information of the defect, the filter state may be changed in accordance with the defect. In order to achieve this, the filter state corresponding to the feature information of the defect is previously stored in the memory 124 in a table. Then, when the defect coordinate data of the defect detected by the inspection device is read into the memory 124, the feature information of the defect is also read, and the defect information is read by the CPU 123 according to the defect, and is stored in the memory 124. The table information is used to determine the filter state, and the filter state control circuit 205 transmits the filter state information. On the other hand, for all the defects, when the applied filter states (P _k , S _k ) (k=1, . . . , N) are the same, the process S301 shown in S300 of FIG. 3 is performed before S705. Processing is fine.

After obtaining the corrected defect position _{m of} all the defects m (m=1, . . . , M), the corrected defect position _m is read from the memory 124, and the position information is converted into the table coordinates as needed. The stage control circuit 119 is given so that the defect _{m is} sequentially moved to the field of view of the SEM 102 (S711), and the SEM image of the defect _m is imaged (S712, S713, S714). When imaging the SEM image of the defect _m , the SEM image of all the defects stored in the filter state of the memory 124 is imaged, and then the wafer is unloaded (S715), and the processing is terminated.

According to the above, the defect position can be detected with high sensitivity and high speed for all types of defects detected by the inspection device, and the optical detection system described above can be mounted on the SEM defect observation device, thereby detecting defects using optical detection. The SEM observation of the position allows the defect to fall within the observation field of the SEM, and the success rate of the automatic imaging of the SEM observation image of the defect detected by the inspection device is improved, and the processing amount of the automatic imaging by the defect of the SEM is also improved.

101‧‧‧ wafer

103‧‧‧Light microscope

110‧‧‧Light source

112‧‧‧Mirror

113‧‧‧Contact objective

115‧‧‧ imaging optical system

116‧‧‧Photographic components

117‧‧‧Distributed polarizing elements

118‧‧‧ Spatial Filter

121‧‧‧Optical camera system control circuit

125‧‧‧ busbar

201‧‧‧Information I/F

202‧‧‧Synchronous control circuit

203‧‧‧Synchronous signal generation circuit

204‧‧‧Image Information Memory Department

205‧‧‧Filter state control circuit

206‧‧‧Distributed polarizing element control circuit

207‧‧‧Spatial filter control circuit

208‧‧‧Internal busbar

Claims (12)

A defect observation apparatus comprising: an SEM, an optical microscope, and a control unit, wherein the optical microscope includes an illumination system that irradiates light to a sample; and light that is based on the sample that is irradiated by the illumination system The signal is detected, and the detection system of the spatial polarization filter for controlling the polarization state of the distributed polarization element and the electrical property for controlling the spatial shape; the control unit generates a synchronization signal (202), based on the generated signal a synchronization signal for electrically controlling the polarization state of the distributed polarization element and the spatial shape of the spatial filter (207, 206), for the polarization state of the plurality of distributed polarization elements and the space of the spatial filter The image detected by the aforementioned detection system under the combination of shapes is processed (123). The defect observation device of claim 1, wherein the image detected by the detection system is combined with a combination of a polarization state of the plurality of distributed polarization elements and a spatial shape of the spatial filter by the control unit Processed to detect the location of the defect. The defect observation device of claim 1, wherein the control unit generates a continuous synchronization signal at a fixed time interval, and continuously performs control of the distributed polarization element and the spatial filter with respect to the synchronization signal, and Detection of an image detected by the detection system under the combination of the polarization states of the plurality of distributed polarization elements and the spatial shape of the spatial filter. The defect observation device according to claim 1, wherein the control unit detects the second SEM by the combination of a polarization state of the plurality of distributed polarization elements and a spatial shape of the spatial filter. The mixing of the secondary electronic image and the reflected electronic image is changed. The defect observation device of claim 1, wherein the distributed polarization element is a liquid crystal filter, and the spatial filter coefficient is a micromirror device. The defect observation device of claim 1, wherein the control unit determines the state of the spatial filter based on the defect information of the defective device. A defect observation method using a defect observation device including an SEM, an optical microscope, and a control unit, the defect observation method comprising: an irradiation program for irradiating light to a sample; and light for the sample irradiated by the irradiation program The signal is detected, and the detection module of the spatial polarization filter for controlling the polarization state of the distributed polarization element and the electrical property for controlling the polarization state; and generating the synchronization signal (202), based on the generated synchronization signal And the polarization state of the distributed polarization element and the spatial shape of the spatial filter are controlled and electrically switched (207, 206), and the combination of the polarization state of the plurality of distributed polarization elements and the spatial shape of the spatial filter The image detected by the aforementioned detection system is used as a control program for processing (123). The defect observation method of claim 7, wherein the control program detects the image detected by the detection program under a combination of a polarization state of the plurality of distributed polarization elements and a spatial shape of the spatial filter. Processed to detect the location of the defect. The defect observation method of claim 7, wherein the control program generates a continuous synchronization signal at a fixed time interval, and continuously performs control of the distributed polarization element and the spatial filter with respect to the synchronization signal, And detecting the image detected by the detection procedure in combination with the polarization state of the plurality of distributed polarization elements and the spatial shape of the spatial filter. The defect observation method of claim 7, wherein the control program is based on a combination of a polarization state of the plurality of distributed polarization elements and a spatial shape of the spatial filter, and is detected by the SEM The mixing of the secondary electronic image and the reflected electronic image is changed. The defect observation method of claim 7, wherein the distributed polarization element is a liquid crystal filter, and the spatial filter coefficient is a micromirror device. The defect observation method of claim 7, wherein the control program determines the state of the spatial filter based on the defect information of the defective device.