WO2023002677A1 - 傾斜推定システム、傾斜推定方法、傾斜推定プログラム、半導体検査システム及び生体観察システム - Google Patents
傾斜推定システム、傾斜推定方法、傾斜推定プログラム、半導体検査システム及び生体観察システム Download PDFInfo
- Publication number
- WO2023002677A1 WO2023002677A1 PCT/JP2022/010674 JP2022010674W WO2023002677A1 WO 2023002677 A1 WO2023002677 A1 WO 2023002677A1 JP 2022010674 W JP2022010674 W JP 2022010674W WO 2023002677 A1 WO2023002677 A1 WO 2023002677A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- estimation
- image
- tilt
- focus
- focus position
- Prior art date
Links
- 238000007689 inspection Methods 0.000 title claims description 77
- 238000000034 method Methods 0.000 title claims description 40
- 239000004065 semiconductor Substances 0.000 title claims description 29
- 238000003384 imaging method Methods 0.000 claims abstract description 142
- 238000010801 machine learning Methods 0.000 claims abstract description 55
- 230000006870 function Effects 0.000 claims description 18
- 239000012472 biological sample Substances 0.000 claims description 9
- 230000004044 response Effects 0.000 claims 1
- 238000012549 training Methods 0.000 abstract description 4
- 238000004458 analytical method Methods 0.000 description 16
- 230000003287 optical effect Effects 0.000 description 16
- 210000002569 neuron Anatomy 0.000 description 12
- 238000013528 artificial neural network Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 7
- 239000000523 sample Substances 0.000 description 6
- 230000010365 information processing Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 4
- 230000005855 radiation Effects 0.000 description 4
- 238000005286 illumination Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000001427 coherent effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000020169 heat generation Effects 0.000 description 2
- 238000007654 immersion Methods 0.000 description 2
- 230000001678 irradiating effect Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 229910000530 Gallium indium arsenide Inorganic materials 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- WPYVAWXEWQSOGY-UHFFFAOYSA-N indium antimonide Chemical compound [Sb]#[In] WPYVAWXEWQSOGY-UHFFFAOYSA-N 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000004020 luminiscence type Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005498 polishing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B21/00—Microscopes
- G02B21/36—Microscopes arranged for photographic purposes or projection purposes or digital imaging or video purposes including associated control and data processing arrangements
- G02B21/365—Control or image processing arrangements for digital or video microscopes
- G02B21/367—Control or image processing arrangements for digital or video microscopes providing an output produced by processing a plurality of individual source images, e.g. image tiling, montage, composite images, depth sectioning, image comparison
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/26—Measuring arrangements characterised by the use of optical techniques for measuring angles or tapers; for testing the alignment of axes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B21/00—Microscopes
- G02B21/0004—Microscopes specially adapted for specific applications
- G02B21/0016—Technical microscopes, e.g. for inspection or measuring in industrial production processes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30148—Semiconductor; IC; Wafer
Definitions
- the present invention relates to an inclination estimation system, an inclination estimation method, an inclination estimation program, a semiconductor inspection system, and a living body observation system for estimating the inclination of an object captured in an image.
- Patent Literature 1 discloses estimating the tilt of a semiconductor device when imaging the semiconductor device for inspection of the semiconductor device. The estimated tilt is used to adjust the attitude of the semiconductor device.
- the inclination of the object to be imaged is estimated by an optical system including a dedicated lens (for example, a relay lens system), a scanner, and the like.
- Patent Document 1 When estimating the tilt of an imaging object using a special optical system as shown in Patent Document 1, a configuration (for example, a lens turret) and space for providing the special optical system within the apparatus are required. Moreover, the cost for preparing a special optical system is required. In addition, the method disclosed in Patent Document 1 requires processes such as turret switching, laser scanning, and alignment, and it takes a long time to estimate the tilt.
- An embodiment of the present invention has been made in view of the above, and includes an inclination estimation system, an inclination estimation method, and an inclination estimation program capable of estimating the inclination of an object to be imaged with a simple configuration and in a short time.
- An object of the present invention is to provide a semiconductor inspection system and a living body observation system related to them.
- a tilt estimation system is a tilt estimation system for estimating a tilt of an object captured in an image, and obtains an image of the object captured. Then, using an estimation target image acquisition means for acquiring an estimation target image, which is a plurality of partial images, from the image, and a feature amount output model for inputting information based on the image and outputting the feature amount of the image, estimation outputting a feature amount of each of the plurality of estimation target images from each of the plurality of estimation target images acquired by the target image acquiring means, and obtaining a focal position corresponding to each of the plurality of estimation target images from the output feature amount; and an inclination estimation means for estimating the inclination of an object captured in the image from the in-focus position corresponding to each of the plurality of estimation object images estimated by the focus position estimation means.
- the feature output model is generated by machine learning from a plurality of learning images associated with focus position information related to the focus position at the time of imaging, and is associated with two different learning images.
- the feature amounts of the two learning images are compared according to the focal position information, and machine learning is performed based on the comparison result.
- the tilt estimation system from each estimation target image that is a plurality of partial images of an image, a focal position at the time of focus corresponding to each of the plurality of estimation target images is estimated, is estimated. Therefore, if an image of the object to be imaged is obtained, the inclination of the object to be imaged can be estimated in a short period of time.
- the feature amount output model used for estimating the focal position can output a feature amount suitable for estimating the focal position, and by using this, it is possible to appropriately estimate the inclination of the object to be imaged. Therefore, according to the tilt estimation system according to the embodiment of the present invention, the tilt of the object to be imaged can be estimated with a simple configuration and in a short period of time.
- the focal position estimating means inputs the feature quantity output from the feature quantity output model, and estimates a plurality of estimation targets using the focal position estimating model for estimating the in-focus focal position corresponding to the image related to the feature quantity. It is assumed that the focus position at the time of focus corresponding to each image is estimated, and the focus position estimation model is generated by machine learning from the focus position information related to the focus position at the time of focus corresponding to each learning image. good too. According to this configuration, the focal position can be reliably and appropriately estimated by using the focal position estimation model. As a result, the tilt of the object to be imaged can be reliably and appropriately estimated.
- the tilt estimation system may further include control means for controlling the tilt of the object to be imaged based on the tilt of the object to be imaged estimated by the tilt estimation means. According to this configuration, it is possible to image the object to be imaged at an appropriate inclination.
- a semiconductor inspection system and a biological observation system can be configured to include the tilt estimation system described above. That is, a semiconductor inspection system according to an embodiment of the present invention includes the above tilt estimation system, a mounting section on which a semiconductor device is mounted as an object to be imaged according to the tilt estimation system, and an inspection section for testing the semiconductor device. And prepare. Further, a biological observation system according to an embodiment of the present invention includes the tilt estimation system described above, a mounting section on which a biological sample is mounted as an object to be imaged according to the tilt estimation system, and an observation section for observing the biological sample. And prepare.
- one embodiment of the present invention can be described as an invention of a tilt estimation system as described above, and can also be described as an invention of a tilt estimation method and a tilt estimation program as follows.
- a tilt estimation method is a tilt estimation method for estimating the tilt of an object captured in an image, in which an image of the object captured is acquired, and a plurality of An estimation target image acquisition step for acquiring an estimation target image that is a partial image of the target image acquired in the estimation target image acquisition step using a feature amount output model for inputting information based on the image and outputting the feature amount of the image a focus position estimation step of outputting a feature amount of each of the plurality of estimation target images from each of the plurality of estimation target images obtained, and estimating a focused position corresponding to each of the plurality of estimation target images from the output feature amount; and a tilt estimation step of estimating the tilt of the object captured in the image from the in-focus focus position corresponding to each of the plurality of estimation target images estimated in the focus position estimation step, and outputting a feature amount
- the model is generated by machine learning from a plurality of learning images associated with focal position information related to the focal position at the time of imaging, and the model is
- a plurality of estimation targets are estimated using a focus position estimation model for inputting a feature quantity output from a feature quantity output model and estimating a focus position at the time of focus corresponding to an image related to the feature quantity. It is assumed that the focus position at the time of focus corresponding to each image is estimated, and the focus position estimation model is generated by machine learning from the focus position information related to the focus position at the time of focus corresponding to each learning image. good too.
- the tilt estimation method may further include a control step of controlling the tilt of the imaging target during imaging based on the tilt of the imaging target estimated in the tilt estimation step.
- a tilt estimation program is a tilt estimation program that causes a computer to operate as a tilt estimation system for estimating the tilt of an object to be imaged in an image.
- an estimation target image acquiring means for acquiring an image in which is captured, and acquiring an estimation target image, which is a plurality of partial images, from the image; and a feature amount output for inputting information based on the image and outputting the feature amount of the image Using the model, outputting the feature amount of each of the plurality of estimation target images from each of the plurality of estimation target images acquired by the estimation target image acquisition means, and corresponding to each of the plurality of estimation target images from the output feature amount Focus position estimating means for estimating the focus position at the time of focus, and from the focus position at the time of focus corresponding to each of the plurality of estimation target images estimated by the focus position estimating means, the inclination of the imaging object captured in the image is calculated.
- the feature output model is generated by machine learning from a plurality of learning images associated with focus position information related to the focus position at the time of imaging, and two different learning The feature amounts of the two learning images are compared according to the focal position information associated with the training images, and machine learning is performed based on the comparison results.
- the focal position estimating means inputs the feature quantity output from the feature quantity output model, and estimates a plurality of estimation targets using the focal position estimating model for estimating the in-focus focal position corresponding to the image related to the feature quantity. It is assumed that the focus position at the time of focus corresponding to each image is estimated, and the focus position estimation model is generated by machine learning from the focus position information related to the focus position at the time of focus corresponding to each learning image. good too.
- the tilt estimation program may cause the computer to function also as control means for controlling the tilt of the imaging target during imaging based on the tilt of the imaging target estimated by the tilt estimating means.
- the tilt of the object it is possible to estimate the tilt of the object to be imaged with a simple configuration and in a short period of time.
- FIG. 4 is a diagram for explaining generation of a feature output model by machine learning
- FIG. 10 is a diagram showing an existing learned model used for generating a feature output model
- FIG. 4 is a diagram showing examples of a defocused image and a focused image
- FIG. 4 is a diagram showing an example of an estimation target image for an image
- FIG. 4 is a diagram for explaining estimation of the tilt of an imaging object
- FIG. 4 is a flowchart showing a feature output model generation method, which is a process executed by the feature output model generation system according to the embodiment of the present invention
- 4 is a flow chart showing a tilt estimation method, which is a process executed by the tilt estimation system according to the embodiment of the present invention
- 1 is a diagram showing a configuration of a feature output model generation program according to an embodiment of the present invention together with a recording medium
- FIG. FIG. 4 is a diagram showing the configuration of a tilt estimation program according to the embodiment of the present invention together with a recording medium
- FIG. 1 shows a computer 10 that is a tilt estimation system according to this embodiment.
- the computer 10 is a device (system) that performs information processing on images. Specifically, the computer 10 performs information processing on images captured by at least one of the inspection device 40 and the observation device 50 . Note that the computer 10 may perform information processing on images captured by devices other than the inspection device 40 and the observation device 50 . That is, the computer 10 may be applied to an imaging device other than the inspection device 40 and the observation device 50 .
- the inspection apparatus 40 is an apparatus that captures an image of a semiconductor device and inspects the semiconductor device based on the captured image. For example, the inspection apparatus 40 performs failure analysis of semiconductor devices.
- a semiconductor device to be inspected is, for example, a wafer on which ⁇ LEDs (Light Emitting Diodes) are spread. Failure analysis is performed, for example, by emission analysis, heat generation analysis, analysis using pattern images, or analysis using laser (OBIRCH, OBIC, DALS, etc.).
- Inspection device 40 may be a conventional inspection device. All of the configurations described below may be included in the conventional inspection apparatus.
- FIG. 2 shows, for example, a partial configuration of an inspection device 40 that performs luminescence analysis.
- the inspection device 40 includes a camera 41, a mounting section 42, a light source 43, an optical system 44, an objective lens 45, and a stage .
- the camera 41 is an imaging device that captures an image of a semiconductor device that is an imaging target. Camera 41 is, for example, an InGaAs camera.
- the mounting section 42 is configured to mount a semiconductor device, which is an object to be imaged.
- the standard sample 60 placed on the placement unit 42 corresponds to an image corresponding to a focal position, which will be described later, rather than an object to be inspected.
- the standard sample 60 is a sample with an artificial pattern (for example, a radial striped pattern as shown in FIG. 2).
- the placement section 42 is configured to be able to control the inclination (orientation) of the placed imaging target with respect to the imaging direction.
- the placement section 42 is configured such that the placement surface on which the object to be imaged is placed can be tilted with respect to the imaging direction. In this manner, the placement section 42 can perform tilt correction of the object to be imaged.
- the mounting section 42 a conventional one capable of controlling the inclination of the object to be imaged can be used.
- the inclination of the object to be imaged during imaging is controlled by the computer 10 as will be described later.
- the light source 43 is a device that outputs irradiation light that irradiates the object to be imaged.
- the light source 43 may output irradiation light of a specific wavelength (for example, a standard wavelength of 1100 nm and wavelengths different from the standard wavelength of 900 nm and 1300 nm). Also, the wavelength of the irradiation light may be switched, such as by preparing a plurality of light sources 43 .
- the optical system 44 is an optical system for irradiating a semiconductor device, which is an object to be imaged, with the irradiation light output from the light source 43 .
- the objective lens 45 is an objective lens used for imaging by the camera 41, and is, for example, a solid immersion lens (SIL).
- SIL solid immersion lens
- the stage 46 is a member for adjusting the focal position when the camera 41 takes an image.
- the stage 46 may be one that can move not only in the imaging direction (focus position direction, Z-axis direction) but also in any three-dimensional direction (that is, an XYZ stage).
- the focus position of the inspection device 40 during imaging is controlled by the computer 10 as will be described later.
- the inspection apparatus 40 includes an inspection unit that inspects a semiconductor device using an image obtained by the above configuration.
- an InSb camera may be used as an imaging device (camera 41) without illumination (light source 43).
- an incoherent light source or a coherent light source is used as the illumination (light source 43)
- a two-dimensional detector, or an optical scanning device and a photodiode are used as the imaging device.
- an incoherent light source or a coherent light source may be used as illumination (light source 43)
- an electrical characteristic acquisition device of a semiconductor device may be used as an imaging device.
- the observation device 50 is a device that captures an image of a biological sample placed on a slide glass and observes the captured image of the biological sample.
- the viewing device 50 may be a conventional viewing device.
- viewing device 50 is the conventional virtual slide scanner described above.
- the observation device 50 includes an imaging device that captures an image of a biological sample that is an imaging target, a mounting section that mounts the biological sample that is an imaging target, and an observation section that observes the biological sample based on the captured image. ing.
- the focus position of the observation device 50 during imaging is controlled by the computer 10 as will be described later.
- the mounting section of the observation device 50 is also configured to be able to control the inclination (orientation) of the mounted imaging target with respect to the imaging direction in the same manner as the mounting section 42 of the inspection device 40 described above.
- the state in which the object to be imaged is properly tilted means, for example, a state in which the surface of the object to be imaged is perpendicular to the imaging direction, that is, the object to be imaged is not tilted with respect to the imaging direction. state. This is for appropriately inspecting or observing an object to be imaged.
- the back surface of a semiconductor device which is the target of failure analysis, is polished during processing.
- the surface of the semiconductor device which is a sample, is warped and distorted due to distortion during polishing and stress caused by the process. If the semiconductor device is tilted, for example, contact with the solid immersion lens will cause a problem. Note that the tilt of the object to be imaged may be caused by a cause other than the above.
- FIG. 3 shows examples of captured images corresponding to focal positions. This image is obtained by imaging the standard sample 60 shown in FIG.
- FIG. 3(a) is an image captured at the focal position during focusing.
- FIG. 3B is an image when the focal position is 100 ⁇ m away from the in-focus focal position.
- FIG. 3(c) is an image when the focal position is further away than in the case of FIG. 3(b).
- FIG. 3(d) is an image when the focal position is 500 ⁇ m away from the in-focus focal position further than in the case of FIG. 3(c). That is, FIG. 3A is an example of a focused image, and FIGS. 3B to 3D are examples of defocused images.
- the computer 10 performs information processing so that the imaging by the inspection device 40 and the observation device 50 is performed with the object to be imaged tilted appropriately.
- the computer 10 may perform information processing so that imaging by the inspection device 40 and the observation device 50 is performed in a state in which the object to be imaged is in focus.
- the computer 10 uses a feature output model generation system 20 that generates a learned model by machine learning, and the learned model generated by the feature output model generation system 20 to enable the above imaging. and a tilt estimation system 30 that processes information for
- the feature amount output model generation system 20 is a system (apparatus) that inputs information based on an image and generates a feature amount output model that outputs the feature amount of the image, as will be described later in detail.
- the tilt estimation system 30 is a system (apparatus) for estimating the tilt of an object captured in an image, as will be described later in detail.
- the inspection device 40 and the observation device 50 are shown as devices that perform imaging, but devices other than the above may be used as long as they are devices (systems) that perform imaging of an object to be imaged.
- the computer 10 is, for example, a conventional computer including hardware such as a CPU (Central Processing Unit), memory, and communication modules.
- the computer 10 may be a computer system including multiple computers.
- the computer 10 may be configured by cloud computing. Each function of the computer 10, which will be described later, is exhibited by the operation of these constituent elements by a program or the like.
- the computer 10, the inspection device 40, and the observation device 50 are connected to each other so that information can be transmitted and received.
- the feature output model generation system 20 includes a learning image acquisition unit 21 , a feature output model generation unit 22 , and a focus position estimation model generation unit 23 .
- the learned models generated by the feature output model generation system 20 are two, a feature output model and a focus position estimation model.
- a feature value output model is a model that inputs information based on an image and outputs the feature value of the image.
- Images used for input to the feature quantity output model are partial images of the images captured by the inspection device 40 and the observation device 50 .
- the feature amount output from the feature amount output model is information indicating the feature of the input image.
- the feature reflects the focal position when the image was captured. That is, the feature output model is an optical model related to optical features.
- the feature amount is, for example, a vector with a preset number of dimensions (eg, 1024 dimensions). The feature amount is used as an input to the focus position estimation model as described later.
- the feature value output model is configured including, for example, a neural network.
- a neural network may be multi-layered. That is, the feature output model may be generated by deep learning.
- the neural network may be a convolutional neural network (CNN).
- the feature value output model has neurons for inputting image-based information to the input layer.
- the information input to the feature output model is the pixel value of each pixel of the image.
- the input layer is provided with as many neurons as there are pixels in the image, and each neuron receives the pixel value of the corresponding pixel.
- an image related to information input to the feature output model is an image of a preset size (for example, an image of 224 ⁇ 224 pixels).
- the information input to the feature output model may be other than the pixel value of each pixel as long as it is based on the image.
- the information may be a feature amount for input to a feature amount output model, which is obtained by performing preprocessing such as conventional image processing on an image in order to reduce the influence of the imaging environment. By performing such preprocessing, it is possible to improve the efficiency of machine learning and the accuracy of the generated feature value output model.
- the feature value output model is provided with neurons for outputting feature values in the output layer. For example, as many neurons as the number of dimensions of the vector of feature quantities are provided.
- the focus position estimation model is a model that inputs the feature quantity output from the feature quantity output model and estimates the focus position at the time of focusing corresponding to the image related to the feature quantity.
- the focus position estimation model uses information indicating the difference between the focus position when the image related to the input feature amount was captured and the focus position at the time of focusing as the estimation result of the focus position at the time of focus. Output.
- the difference is, for example, a value obtained by subtracting the focal length corresponding to the focal position when the image related to the feature amount was captured from the focal length corresponding to the focal position at the time of focusing.
- the output value is a value indicating the focal position when an image is captured in a coordinate system in which the position of the focal position at the time of focusing is 0.
- the focal position at the time of focusing is a focal position for capturing an image by focusing on the imaging object appearing in the image related to the input feature amount.
- a focused image can be captured by changing the focus position by the above difference from the focus position when the defocused image related to the input feature amount was captured.
- the above difference candidates may be set in advance, and the focus position estimation model may output a value indicating the degree to which the candidates are appropriate. For example, if the candidates for the difference are +50 ⁇ m, 0 ⁇ m, ⁇ 50 ⁇ m, ⁇ 100 ⁇ m, . For example, the candidate with the highest value is set as the above difference. Alternatively, the focal position estimation model may output the difference value itself.
- the focal position estimation model may output information indicating the in-focus focal position itself (for example, the focal length corresponding to the in-focus focal position).
- candidates for the in-focus focus position itself may be set in advance, and the focus position estimation model may output a value indicating the degree to which the candidates are appropriate.
- the focus position estimation model may output the value itself of the focus position at the time of focusing.
- the focus position estimation model is configured including, for example, a neural network.
- a neural network may be multi-layered. That is, the focal position estimation model may be generated by deep learning.
- the neural network may be a convolutional neural network (CNN).
- the focus position estimation model is provided with neurons for inputting feature values into the input layer.
- the input layer is provided with neurons corresponding to the neurons provided in the output layer of the feature output model. That is, the input layer is provided with as many neurons as the output layer of the feature output model.
- the focus position estimation model is provided with a neuron for outputting the estimation result of the focus position at the time of focusing as described above. For example, neurons for the number of candidates (when outputting a value for each candidate) or one neuron (when outputting the above difference or the focus position itself when in focus) are provided.
- the feature quantity output model and the focal position estimation model may be configured by something other than a neural network.
- the feature output model and focus position estimation model are expected to be used as program modules that are part of artificial intelligence software.
- the feature quantity output model and the focal position estimation model are used, for example, in a computer having a CPU and memory, and the CPU of the computer operates according to instructions from the model stored in the memory.
- the CPU of the computer operates to input information to the model, perform calculations according to the model, and output results from the model according to the instructions.
- the CPU of the computer inputs information to the input layer of the neural network according to the command, performs calculations based on parameters such as weighting coefficients that have been learned in the neural network, and outputs from the output layer of the neural network It works to output the result.
- the learning image acquisition unit 21 is a learning image acquisition unit that acquires a plurality of learning images associated with focus position information related to the focus position at the time of imaging.
- the learning image acquisition unit 21 acquires an image obtained by detecting radiation from the imaging target, an image obtained by detecting light from the imaging target when the imaging target is irradiated with light, or an image obtained by detecting light from the imaging target when the imaging target is irradiated with light. You may acquire the image which detected the electrical characteristic of the imaging target object as a learning image.
- the learning image acquisition unit 21 may acquire, as the learning image, an image obtained when the object to be imaged is irradiated with light of a specific wavelength.
- the learning image acquisition unit 21 acquires focus position information related to a focus position at the time of focusing corresponding to each learning image to be acquired.
- the learning image acquisition unit 21 acquires images captured by the inspection device 40 or the observation device 50 .
- This image shows an object to be imaged for use as a learning image.
- the imaging object for learning images may be, for example, the standard sample 60 shown in FIG.
- the object to be imaged for the learning image may be another object (for example, an object imaged by the inspection device 40 or the observation device 50).
- the learning image acquisition unit 21 cuts out a partial image from the acquired image 70 and uses it as a learning image 71 .
- the learning image 71 is an image used for machine learning to generate the two trained models described above. Therefore, the learning image acquiring unit 21 acquires a plurality of learning images 71 in a number sufficient to appropriately generate a trained model.
- the plurality of learning images 71 are made to include the defocused image. Also, there may be a plurality of defocused images, and the focal positions thereof may be various positions. That is, the focal lengths corresponding to those focal positions may be various distances. A focus image may be included in the plurality of learning images 71 .
- FIG. 4 shows an example in which there are three acquired images 70 . The vertical direction of the portion showing the image 70 in FIG. 4 is the imaging direction (focus position direction, Z-axis direction).
- the learning image 71 corresponds to an image used for input to the feature output model.
- the feature value output model inputs information based on a partial image of the image having the same size as the learning image 71 rather than the entire image captured by the inspection device 40 or the observation device 50 . Therefore, the learning image acquisition unit 21 cuts out from the acquired image 70 a learning image 71, which is a partial image of a preset size and used for inputting the feature output model.
- a position in the image 70 where the learning image 71 is cut out is a portion in which the object to be imaged is shown.
- the learning images 71 may include learning images 71 in which the imaging target is not shown.
- the position where the learning image 71 is cut out from the image 70 may be set in advance. Alternatively, the position where the image 70 is estimated to be captured by image recognition may be used as the position where the learning image 71 is cut out.
- the learning image acquisition unit 21 may cut out a plurality of learning images 71 from one image 70.
- the positions of the learning images 71 may overlap.
- imaging is performed and an image that is the basis of the learning image 71 is generated.
- the imaging position (XY) other than the imaging direction (Z-axis direction) is fixed, and multiple consecutive imagings with different focal positions are performed.
- the learning image acquisition unit 21 may acquire the image 70 from a device other than the inspection device 40 or the observation device 50.
- the learning images 71 are images obtained by detecting radiation from the imaging object (images used for light emission/heat generation analysis), and light from the imaging object when the imaging object is irradiated with light. It may be a detected image (image used for pattern analysis) or an image obtained by detecting the electrical characteristics of the imaging object when the imaging object is irradiated with light (image used for laser analysis). Further, the learning image 71 may be an image obtained when an object to be imaged is irradiated with light of a specific wavelength (for example, light of a specific wavelength used for inspection). These are the image types normally used by the inspection device 40 or the observation device 50 .
- a set of feature quantity output model and focal position estimation model to be generated corresponds to a specific image type.
- the wavelength of light used in light emission analysis varies depending on the driving voltage and design rules of the inspection device 40. Further, in an actual optical system, defocus occurs for each wavelength due to adjustment and the characteristics of optical elements. Moreover, the detection sensitivity may be maximized at a position different from the focal position for observing the pattern image (image).
- a feature amount output model for each wavelength may be generated using an image for each specific wavelength as described above. For example, an image for each wavelength (900 nm, 1300 nm) different from the standard wavelength (1100 nm) and the standard wavelength may be used to generate a feature output model for each wavelength.
- a set of feature quantity output model and focal position estimation model to be generated corresponds to a specific model of inspection device 40 or observation device 50 . That is, the feature quantity output model and the focal position estimation model reflect the features of the inspection device 40 or the observation device 50 .
- the focus position estimation model may be common to a plurality of image types or models.
- Each learning image 71 is associated with focal position information relating to the focal position when the learning image 71 was captured.
- the focal position information is, for example, information indicating the above focal position.
- the focal position information is information relating to the focal position, and may be any information other than the above as long as it can be used to generate the above-described trained model.
- the focal position information is obtained, for example, as information when an image is captured by the inspection device 40 or the observation device 50 .
- the learning image acquisition unit 21 receives and acquires an image associated with focal position information from the inspection device 40 or the observation device 50 .
- the learning image acquisition unit 21 acquires focus position information related to the focus position at the time of focusing corresponding to each learning image to be acquired.
- the in-focus position information is, for example, information indicating the focal position at the time of focusing.
- the in-focus position information is information related to the focal position at the time of focusing, and may be any information other than the above as long as it can be used to generate the learned model described above.
- the in-focus position information is obtained by a conventional method of measuring the focal position during focusing.
- the learning image acquisition unit 21 acquires the focus position information by accepting a user's input operation of the focus position information obtained by the measurement to the computer 10 .
- the learning image acquisition unit 21 outputs the acquired information to the feature output model generation unit 22 and the focus position estimation model generation unit 23 .
- the feature output model generation unit 22 is a feature output model generation unit that generates a feature output model by machine learning from the learning image 71 acquired by the learning image acquisition unit 21 .
- the feature amount output model generation unit 22 compares the feature amounts of the two learning images 71 according to the focus position information associated with the two different learning images 71, and performs machine learning based on the comparison result. I do.
- the feature output model generation unit 22 reduces the difference between the feature amounts of the two learning images 71 and
- machine learning may be performed so that the difference between the feature amounts of the two learning images 71 increases.
- the feature quantity output model generation unit 22 generates a feature quantity output model as follows.
- the feature output model generation unit 22 receives the learning image 71 and the focus position information related to the learning image 71 from the learning image acquisition unit 21 .
- the feature amount output model generation unit 22 uses two learning images 71 selected from the plurality of input learning images 71 as one set to perform machine learning for generating a feature amount output model.
- a set used for machine learning includes both a set of learning images 71 related to the same focus position and a set of learning images 71 related to different focus positions.
- the set of learning images 71 related to the same focal position may be the learning images 71 clipped from the same image 70 as shown in FIG. 4, for example.
- Selection of the set of learning images 71 may be performed by a preset method that satisfies the above conditions. Also, the set of learning images 71 may be selected from images 70 having the same imaging position (XY) other than the imaging direction (Z-axis direction).
- the feature output model generation unit 22 performs machine learning using information based on the selected set of learning images 71 as input to the feature output model. As shown in FIG. 4, when one set of learning images 71 is input to the feature amount output model 80, a feature amount is obtained for each of the learning images 71 as an output. In FIG. 4, the value of each element of the vector, which is the feature amount, is shown in a bar graph. At this time, the feature amount output model 80 to which one learning image 71 is input is set as a learning target, and the feature amount output model 80 to which the other learning image 71 is input is set as a comparison target. However, these feature quantity output models 80 are the same during learning.
- the feature quantity output model generation unit 22 refers to the focus position information, compares the two output feature quantities, and performs machine learning based on the comparison result. If the focal positions of the two learning images 71 indicated by the focal position information are the same focal position (that is, they are on the same plane), the feature amount output model generation unit 22 Machine learning is performed so that the difference in the feature amount becomes small. When the focus positions of the two learning images indicated by the focus position information are different from each other (that is, the Z positions are different), the feature quantity output model generation unit 22 calculates the feature quantity of the two learning images 71. Machine learning is performed so that the difference between Note that when the two learning images 71 are cut out from the same image, the focal positions of the two learning images 71 are the same focal position. Further, even when the focal positions of the two learning images 71 are close enough to be regarded as the same, the focal positions of the two learning images 71 may be regarded as the same focal position.
- the feature amounts of partial images cut out from images on the same focal plane are made to have a large correlation regardless of the cutout position.
- the correlation between the feature amounts of partial images cut out from images on different focal planes is reduced.
- the feature quantity output model generation unit 22 performs machine learning using the following loss_xy as a loss function.
- i (0 to n) is a suffix (subscript) indicating an element of the vector of feature amounts.
- the number of channels (number of dimensions) of the feature amount vector is n+1.
- F t0 to F tn are the values of the elements of the feature amount vector output from the feature amount output model 80 to be learned.
- F c0 to F cn are the values of the elements of the vector of feature quantities output from the feature quantity output model 80 to be compared.
- SD i is the standard deviation for element i of each feature.
- the feature quantity output model generation unit 22 performs machine learning using the following loss_z as a loss function. That is, the loss function in this case is the reciprocal of the loss function when the focus positions of the two learning images 71 are the same as each other.
- the machine learning itself based on the loss function that is, the updating of the parameters of the feature output model may be performed in the same manner as before. It should be noted that the loss function does not necessarily have to be the one described above, as long as it conforms to the criteria described above.
- the feature output model generation unit 22 repeatedly selects a set of learning images 71 and performs machine learning to generate a feature output model. For example, the feature quantity output model generation unit 22 repeats the above operations until the generation of the feature quantity output model converges based on a preset condition, or a preset specified number of times, in the same manner as in the conventional art, and outputs the feature quantity. Generate a model.
- the feature output model generation unit 22 may generate the feature output model using an existing learned model generated by machine learning.
- the existing learned model a model for inputting information based on an image is used, like the feature output model according to the present embodiment.
- an existing learned model having the same input as the feature output model according to this embodiment may be used.
- An existing trained model is, for example, a model for performing image recognition, specifically ResNet, VGG, Mobile Net, and the like. A part of the existing trained model is used to generate the feature output model. As shown in FIG. 5, the layer on the output side of the existing trained model 81 is deleted, and the part up to the intermediate layer of the existing trained model 81 is used to generate the feature quantity output model.
- the existing trained model 81 used to generate the feature output model may include all of the intermediate layers, or may include only part of the intermediate layers.
- the feature output model generation unit 22 inputs the above part of the existing learned model and uses it as a feature output model at the start of machine learning. That is, the feature quantity output model generation unit 22 performs fine tuning using the above part of the existing trained model as the initial parameters of the feature quantity output model. Also, a new output layer added to the output side of the above-mentioned part of the learned model may be used as the feature value output model at the start of machine learning. In addition, when adding a new output layer, a new intermediate layer is added between the above part of the output side of the learned model and the new output layer, and the feature value output at the start of machine learning It can be used as a model.
- the feature quantity output model generation unit 22 may generate a feature quantity output model without using an existing trained model. For example, as in conventional machine learning, a model using random values as initial parameters may be used as the feature output model at the start of machine learning.
- an existing trained model to generate a feature value output model has the following advantages. Learning time can be greatly shortened. It is possible to generate a feature amount output model with high accuracy even with a small number of learning images 71, that is, a feature amount output model capable of outputting a more appropriate feature amount.
- the existing trained models described above already acquire the ability to separate low-abstract features. Therefore, it is only necessary to perform learning focusing on features with a high degree of abstraction using the new learning image 71 .
- the feature output model generation unit 22 outputs the generated feature output model to the focus position estimation model generation unit 23 and the tilt estimation system 30 .
- the generated feature value output model may be used for purposes other than those in this embodiment.
- the feature output model generation unit 22 transmits or outputs the feature output model to another device or module that uses the feature output model.
- the feature output model generation unit 22 may store the generated feature output model in the computer 10 or other device so that it can be used by other devices or modules that use the feature output model.
- the focus position estimation model generation unit 23 is focus position estimation model generation means for generating a focus position estimation model by machine learning from the focus position information acquired by the learning image acquisition unit 21 .
- the focal position estimation model receives the feature amount output from the feature amount output model generated by the feature amount output model generation unit 22 as described above, and calculates the focus at the time of focusing corresponding to the image related to the feature amount. It estimates the position.
- the focus position estimation model generation unit 23 generates a focus position estimation model as follows.
- the focus position estimation model generation unit 23 inputs the learning image 71 and the focus position information related to the learning image 71 from the learning image acquisition unit 21 .
- the focus position estimation model generator 23 receives the feature output model from the feature output model generator 22 .
- the focus position estimation model generation unit 23 inputs information based on the learning image 71 to the feature amount output model, and acquires the feature amount of the learning image 71 output from the feature amount output model.
- the focus position estimation model generation unit 23 uses the acquired feature amount as an input to the focus position estimation model, and uses information based on the focus position information of the learning image 71 related to the input feature amount as an output of the focus position estimation model. do the learning.
- Information based on the focus position information is information corresponding to the output of the focus position estimation model.
- the information based on the focus position information is, for example, the value of the candidate corresponding to the focus position information is set to 1, and the focus position information is a value (one-hot vector) for each candidate, with the value of a candidate that does not correspond to 0 being set to 0. If the focus position estimation model outputs the above-mentioned difference or the value of the focus position itself, the information based on the focus position information is the focus position information itself or the value calculated from the focus position information. be.
- the focus position estimation model generation unit 23 generates information based on focus position information corresponding to the output of the focus position estimation model before performing machine learning.
- the machine learning itself that is, updating the parameters of the focus position estimation model, should be performed in the same way as before.
- the focal position estimation model generation unit 23 repeats the machine learning process until the generation of the focal position estimation model converges based on the preset conditions, or the preset number of times, in the same manner as in the conventional art, to obtain the focal position. Generate an inference model.
- the focus position estimation model generation unit 23 outputs the generated focus position estimation model to the tilt estimation system 30 .
- the generated focal position estimation model may be used for purposes other than those in this embodiment.
- the focus position estimation model generator 23 transmits or outputs the focus position estimation model to another device or module that uses the focus position estimation model.
- the focus position estimation model generator 23 may store the generated focus position estimation model in the computer 10 or other device so that it can be used by other devices or modules that use the focus position estimation model.
- the functions of the feature output model generation system 20 have been described above.
- the tilt estimation system 30 includes an estimation target image acquisition section 31 , a focus position estimation section 32 , a tilt estimation section 33 and a control section 34 .
- the tilt estimation system 30 estimates the tilt of the object captured in the image captured by the inspection device 40 or the observation device 50 . Therefore, the tilt estimation system 30 estimates the focal position when the imaging object is imaged by the inspection device 40 or the observation device 50 .
- the inspection device 40 or the observation device 50 picks up an image of the imaging target (imaging for inclination estimation).
- the focal position does not necessarily have to be the focal point on the object to be imaged, that is, the focal position at the time of focusing. Therefore, the image obtained by this imaging may be a defocused image.
- this image may be a defocused image as shown in FIG. 6(a).
- FIG. 6 is an image of a semiconductor device.
- the defocused image shown in FIG. 6A is an image obtained when the focal position at the time of imaging is +5 nm from the focal position at the time of focusing.
- the tilt estimation system 30 estimates focal positions at multiple positions in the image when in focus.
- the tilt estimation system 30 estimates the tilt of the object captured in the image from the estimated focal positions at the multiple positions of the image.
- an image in which the object to be imaged is in focus that is, a focused image can be obtained.
- a focused image as shown in FIG. 6B is obtained.
- the focused image shown in FIG. 6(b) corresponds to the defocused image shown in FIG. 6(a).
- the estimation target image acquisition unit 31 is an estimation target image acquisition unit that acquires an image of an imaging target and acquires an estimation target image, which is a plurality of partial images, from the image.
- the estimation target image is an image used for estimating the focal position at the time of focusing in the tilt estimation system 30 .
- the estimation target image is an image used for input to the feature value output model. That is, the estimation target image corresponds to the learning image 71 described above.
- the estimation target image acquisition unit 31 acquires an image captured by the inspection device 40 or the observation device 50 .
- the imaging by the inspection device 40 or the observation device 50 at this time is the above-described imaging for tilt estimation.
- the estimation target image acquiring unit 31 cuts out a plurality of partial images from the acquired image and uses them as estimation target images.
- the estimation target image acquisition unit 31 cuts out from the acquired image an estimation target image, which is a partial image of a preset size used for inputting the feature value output model.
- the position in the image where the estimation target image is cut out is the part where the imaging target is shown.
- a position in the image where the estimation target image is cut out may be set in advance.
- the estimation target image 91 When the estimation target image 91 is cut out as described above, the estimation target image 91 may be cut out from the entire image 90 if the imaging target is shown in the entire image 90, or the imaging target image may be is captured, the estimation target image 91 may be cut out from part of it. Further, the position where the image is estimated to be captured by performing image recognition on the image may be set as the position where the estimation target image is cut out.
- the tilt estimation system 30 is configured such that the position in the image where the estimation target image is cut out is also used for functions described later.
- the type of the estimation target image is the same type as the learning image described above.
- the estimated target image is an image obtained by detecting radiation from the imaging target, an image obtained by detecting light from the imaging target when the imaging target is irradiated with light, or an image obtained by detecting light from the imaging target when the imaging target is irradiated with light. It may be an image obtained by detecting the electrical characteristics of the object to be imaged.
- the estimation target image may be an image obtained by irradiating the object to be imaged with light of a specific wavelength (for example, light of a specific wavelength used for inspection).
- the estimation target image acquisition unit 31 outputs the acquired multiple estimation target images to the focus position estimation unit 32 .
- the focus position estimation unit 32 uses the feature amount output model to output the feature amount of each of the plurality of estimation target images from each of the plurality of estimation target images acquired by the estimation target image acquisition unit 31, and calculates the output feature Focus position estimating means for estimating a focus position at the time of focusing corresponding to each of a plurality of estimation target images from the quantity.
- the focal position estimation unit 32 may use the focal position estimation model to estimate the in-focus focal position corresponding to the estimation target image from the feature quantity output from the feature quantity output model.
- the focus position estimation unit 32 receives and stores the feature output model and the focus position estimation model generated by the feature output model generation system 20, and uses them for estimation.
- the focus position estimation unit 32 receives a plurality of estimation target images from the estimation target image acquisition unit 31 .
- the focal position estimation unit 32 inputs information based on the estimation target image to the feature amount output model, and acquires the feature amount of the estimation target image, which is the output from the feature amount output model.
- the focal position estimating unit 32 inputs the acquired feature amount to the focal position estimating model, and converts the information indicating the in-focus focal position corresponding to the estimation target image, which is output from the focal position estimating model, to the focal position. Get it as an estimation result.
- the focus position estimating unit 32 acquires information indicating the focus position at the time of focusing corresponding to each of the estimation target images for each of the plurality of estimation target images.
- the focal position estimating section 32 outputs to the tilt estimating section 33 information indicating the in-focus focal positions corresponding to the plurality of acquired estimation target images.
- the tilt estimating unit 33 is a tilt estimating unit that estimates the tilt of the object captured in the image from the in-focus focal positions corresponding to each of the plurality of estimation target images estimated by the focus position estimating unit 32 .
- the tilt estimator 33 estimates, for example, the tilt of the imaging object with respect to the imaging direction in the inspection device 40 or the observation device 50 .
- the tilt estimator 33 estimates the tilt of the imaging object as follows.
- the tilt estimating unit 33 receives from the focal position estimating unit 32 information indicating the in-focus focal position corresponding to each of the plurality of estimation target images.
- the position in the imaging direction corresponding to the focal position at the time of focusing corresponding to the estimation target image is assumed to be the position of the imaging target. Since the estimation target image is cut out from a plurality of positions in the image, it is possible to obtain the position of the imaging target in the imaging direction at each position of the estimation target image in the image.
- the tilt of the object to be imaged is estimated from the positions of the plurality of objects to be imaged.
- the tilt estimator 33 estimates the tilt of the imaging object for each coordinate axis on a plane perpendicular to the imaging direction.
- the tilt estimator 33 estimates the tilt of the imaging object for each of the X-axis and the Y-axis, which are two coordinate axes parallel to each side of the estimation target image 91 of the image 90 shown in FIG.
- the tilt estimating unit 33 calculates the tilt angle ⁇ 1 , and the tilt angle ⁇ 2 of the imaging object on the Y-axis with respect to the plane (hatched plane in the figure) perpendicular to the imaging direction (Z-axis direction), as shown in FIG . 8B.
- FIG. 8 shows information indicating the focused position estimated from each estimation target image 91 (specifically, the difference between the focused position when the image was captured and the focused position). (values such as +2, 0, -2, +3, 0 and -3).
- FIG. 8A shows an example in which the object to be imaged is tilted in the X-axis direction, specifically, the right side is lowered.
- FIG. 8B shows an example in which the object to be imaged is tilted in the Y-axis direction, specifically, the front side is lowered.
- the tilt estimator 33 calculates the angles ⁇ 1 and ⁇ 2 using the following equations.
- x is the length in the X - axis direction used for calculating the angle ⁇ 1 according to the position of the estimation target image in the image.
- z1 is the amount of deviation of the focal position at the time of focusing in the imaging direction (Z - axis direction) corresponding to x.
- y is the length in the Y - axis direction used for calculating the angle ⁇ 2 according to the position of the estimation target image in the image.
- z2 is the amount of deviation of the focal position at the time of focusing in the imaging direction (Z - axis direction) corresponding to y.
- x and y are determined based on the position of the estimation target image as shown in FIG.
- y is the Y-axis direction of two predetermined positions of the estimation target images P a and P c (for example, the central coordinate P center of the estimation target image) separated from each other in the Y-axis direction used for estimating the inclination; difference.
- z 1 and z 2 are calculated from the focal positions corresponding to the respective estimation target images.
- the estimation target image used for tilting the imaging target may be other than the above.
- the estimation target image used for the tilt of the object to be imaged may be estimated as follows. An image of an object to be imaged is divided into a plurality of regions each containing a plurality of estimation target images. For example, divide the image into 3 ⁇ 3 rectangular regions. outliers from the in -focus focus positions (Za1 , Za2 , Za3 , . Remove. Take the median value (Z median ) of the in-focus focal positions after removing the outliers. The median value (Z median ) and the estimation target image (P median ) corresponding to the median value (Z median ) are used as the in-focus focus position and the estimation target image representing the area (P a ) to estimate the tilt. .
- the tilt estimating unit 33 estimates the tilt of the object captured in the image from the in-focus focal positions corresponding to each of the plurality of estimation target images estimated by the focus position estimating unit 32, other than the above You may estimate a slope by the method of The tilt estimator 33 may also estimate angles other than the above-described angles ⁇ 1 and ⁇ 2 as the tilt of the object to be imaged.
- the tilt estimator 33 outputs information indicating the estimated tilt of the object to be imaged to the controller 34 .
- the tilt estimating unit 33 may also output to the control unit 34 information indicating the focal position at the time of focusing corresponding to the estimation target image used for estimating the tilt of the object to be imaged.
- the control unit 34 is control means for controlling the tilt of the object to be imaged based on the tilt of the object to be imaged estimated by the tilt estimation unit 33 . Based on the focal position estimated by the focal position estimating section 32, the control section 34 may control the focal position during imaging of the imaging target.
- the control unit 34 receives information indicating the tilt of the imaging object from the tilt estimating unit 33 . Further, the control unit 34 may input information indicating the focus position at the time of focusing corresponding to the estimation target image from the tilt estimation unit 33 .
- the control unit 34 controls the inspection device 40 or the observation device 50 so that the inclination of the imaging object indicated by the input information is eliminated during imaging. Specifically, the control unit 34 controls the inspection device 40 or the observation device 50 so as to tilt the object to be imaged opposite to the tilt of the object to be imaged indicated by the input information.
- the inspection device 40 or the observation device 50 that has received the control adjusts the inclination of the object to be imaged during imaging by, for example, operating the placement section.
- the control unit 34 controls tilt correction in the inspection device 40 or the observation device 50 .
- the image picked up by the inspection device 40 or the observation device 50 is an image in which the object to be picked up is properly tilted.
- control unit 34 controls the inspection device 40 or the observation device 50 so that the focal position during imaging becomes the in-focus focal position indicated by the input information.
- control unit 34 performs control so that the focus position at the time of focusing corresponds to a preset estimation target image among the plurality of estimation target images.
- the inspection device 40 or the observation device 50 that has received the control operates, for example, the stage 46 to adjust the focal position during imaging.
- the image captured by the inspection device 40 or the observation device 50 becomes a focused image.
- the control unit 34 controls autofocus in the inspection device 40 or the observation device 50 .
- the above is the configuration of the tilt estimation system 30 .
- FIG. 9 First, using the flowchart of FIG. 9, the process executed when generating the feature output model and the focal position estimation model, that is, the process executed by the feature output model generation system 20 according to the present embodiment. A feature output model generation method will be described.
- the learning image acquisition unit 21 acquires a plurality of learning images associated with focus position information related to the focus position at the time of imaging (S01, learning image acquisition step). Further, the learning image acquisition unit 21 acquires focus position information related to the focus position at the time of focusing corresponding to each learning image. Subsequently, the feature output model generation unit 22 generates a feature output model from the learning image by machine learning (S02, feature output model generation step). At this time, the feature amounts of the two learning images 71 are compared according to the focal position information associated with the two different learning images, and machine learning is performed based on the comparison results. Subsequently, the focus position estimation model generation unit 23 generates a focus position estimation model from the focus position information by machine learning (S03, focus position estimation model generation step).
- the generated feature quantity output model and focal position estimation model are output from the feature quantity output model generation system 20 to the tilt estimation system 30 .
- the feature quantity output model and the focus position estimation model are stored and used in the following processes.
- the above is the feature output model generation method, which is the process executed by the feature output model generation system 20 according to the present embodiment.
- the estimation target image acquisition unit 31 acquires an image of the imaging target, and acquires an estimation target image, which is a plurality of partial images, from the image (S11, estimation target image acquisition step ).
- the estimation target image is based on an image obtained by imaging for inclination estimation by the inspection device 40 or the observation device 50 .
- the feature quantity output model is used by the focus position estimation unit 32 to output the feature quantity of each of the plurality of estimation target images from each of the plurality of estimation target images.
- the focal position estimation unit 32 uses the focal position estimation model to estimate the in-focus focal position corresponding to each of the plurality of estimation target images from the feature amount of each of the plurality of estimation target images (S12, focus position estimation step).
- the tilt estimating unit 33 estimates the tilt of the object captured in the image from the in-focus focal position corresponding to each of the plurality of estimation target images (S13, tilt estimating step).
- the controller 34 controls the tilt of the object to be imaged by the inspection device 40 or the observation device 50 at the time of new imaging based on the estimated tilt of the object to be imaged (S14, control step).
- the image picked up by the inspection device 40 or the observation device 50 is an image in which the object to be picked up is properly tilted.
- the control unit 34 may control the focal position at the time of imaging of the object to be imaged by the inspection device 40 or the observation device 50 based on the estimated focal position. As a result, the image captured by the inspection device 40 or the observation device 50 becomes a focused image.
- the above is the tilt estimation method, which is the processing executed by the tilt estimation system 30 according to the present embodiment.
- a feature quantity output model that outputs the feature quantity of an image is generated by machine learning.
- the feature amounts of the two learning images are compared in accordance with the focal position information associated with the two different learning images, and machine learning is performed based on the comparison results.
- the machine learning for feature amount output model generation when two different learning images are related to the same focal position, the difference between the feature amounts of the two learning images is is smaller and two different learning images are associated with different focal positions, the difference between the feature amounts of the two learning images may be increased. According to this configuration, it is possible to reliably and appropriately generate a feature output model.
- the machine learning does not necessarily have to be performed as described above, and may be performed based on the result of comparing the feature amounts of the two learning images.
- the learning image and the estimation target image are images obtained by detecting radiation from the imaging target, images obtained by detecting light from the imaging target when the imaging target is irradiated with light, or It may be an image obtained by detecting the electrical characteristics of the object to be imaged when the object is irradiated with light. Furthermore, the learning image and the estimation target image may be images obtained when the imaging target is irradiated with light of a specific wavelength. According to these configurations, it is possible to generate an appropriate feature amount output model according to the type of image to be used, and to use the feature amount output model. However, the learning image and the estimation target image are not limited to those described above, and may be any image corresponding to the focal position.
- the feature output model generation system 20 may further include a focus position estimation model generation unit 23 that generates a focus position estimation model.
- a focus position estimation model generation unit 23 that generates a focus position estimation model.
- the feature output model generation system 20 does not have to include the focal position estimation model generation unit 23 . That is, the feature output model generation system 20 may be configured to generate only feature output models. Also, the generated feature amount output model may be used for purposes other than estimating the focal position at the time of focusing.
- the in-focus focus position corresponding to each of the plurality of estimation target images is estimated from each of the estimation target images, and the tilt of the imaging target is estimated. Presumed. Therefore, if an image of the object to be imaged is obtained, the inclination of the object to be imaged can be estimated in a short period of time.
- a configuration for example, a lens turret
- a space for providing a special optical system in the apparatus which were required when estimating the tilt of the imaging object using a special optical system, are not required. Also, the cost of preparing a special optical system is unnecessary.
- processes such as turret switching, laser scanning, and alignment are not required. Compared to that case, the estimation time of the inclination of the object to be imaged can be significantly shortened.
- the feature quantity output model used for estimating the focus position can output a feature quantity suitable for estimating the focus position, and by using this, it is possible to appropriately estimate the tilt of the imaging object. Therefore, according to the tilt estimation system of the present invention, the tilt of the object to be imaged can be estimated with a simple configuration and in a short period of time.
- the above-described feature quantity output model is used for estimating the focal position during focusing. Therefore, according to the present embodiment, it is possible to estimate the focus position at the time of focusing based on the image in a short preparation time. Further, according to the tilt estimation system 30 according to the present embodiment, it is possible to estimate the focal position at the time of focusing by imaging for tilt estimation once. Therefore, it is possible to estimate the focus position at the time of focusing more quickly than in the case of searching for the focus position at the time of focusing by performing imaging a plurality of times while changing the focus position. As a result, it is possible to quickly estimate the inclination of the object to be imaged.
- the focus position estimation model described above may be used to estimate the focus position during focusing. With this configuration, it is possible to reliably and appropriately estimate the focal position at the time of focusing. As a result, the tilt of the object to be imaged can be reliably and appropriately estimated. However, it is not necessary to use the focus position estimation model described above for estimating the focus position at the time of focusing, and the estimation may be performed from the feature quantity output from the feature quantity output model.
- the tilt estimation system 30 controls the tilt of the object to be imaged by the inspection device 40 or the observation device 50 based on the estimated tilt of the object to be imaged. It may be further provided. According to this configuration, the inspection device 40 or the observation device 50 can capture an image of the imaging target at an appropriate tilt. .
- the tilt estimation system 30 does not have to include the controller 34 . That is, the tilt estimation system 30 may be any system that estimates the tilt of the object to be imaged.
- the semiconductor inspection system according to this embodiment may be a system including the tilt estimation system 30 and the inspection device 40 .
- the biological observation system according to this embodiment may be a system including the tilt estimation system 30 and the observation device 50 .
- the computer 10 includes the feature quantity output model generation system 20 and the gradient estimation system 30.
- the feature quantity output model generation system 20 and the gradient estimation system 30 are independent. Each may be implemented.
- the feature output model generation program 200 is inserted into a computer and accessed, or stored in a program storage area 211 formed in a computer-readable recording medium 210 provided in the computer.
- the recording medium 210 may be a non-temporary recording medium.
- the feature output model generation program 200 comprises a learning image acquisition module 201 , a feature output model generation module 202 , and a focus position estimation model generation module 203 .
- a function realized by executing the learning image acquisition module 201, the feature output model generation module 202, and the focus position estimation model generation module 203 is the learning image acquisition of the feature output model generation system 20 described above.
- the functions are the same as those of the unit 21, the feature output model generation unit 22, and the focus position estimation model generation unit 23, respectively.
- the inclination estimation program 300 is stored in a program storage area 311 formed in a computer-readable recording medium 310 that is inserted into and accessed by a computer or that the computer has.
- the recording medium 310 may be a non-temporary recording medium. Note that the recording medium 310 may be the same as the recording medium 210 .
- the tilt estimation program 300 comprises an estimation target image acquisition module 301 , a focus position estimation module 302 , a tilt estimation module 303 and a control module 304 .
- Functions realized by executing the focus position estimation module 302, the tilt estimation module 303, and the control module 304 are the estimation target image acquisition unit 31, the focus position estimation unit 32, and the tilt estimation system 30 described above.
- the functions are the same as those of the tilt estimating section 33 and the control section 34, respectively.
- Part or all of the feature output model generation program 200 and the gradient estimation program 300 are transmitted via a transmission medium such as a communication line, and received and recorded (including installation) by another device. may be configured. Also, each module of the feature output model generation program 200 and the gradient estimation program 300 may be installed in one of a plurality of computers instead of one computer. In that case, the above-described series of processes are performed by the computer system comprising the plurality of computers.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Geometry (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Optics & Photonics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Testing Or Measuring Of Semiconductors Or The Like (AREA)
Abstract
Description
Claims (11)
- 画像に写った撮像対象物の傾斜を推定する傾斜推定システムであって、
撮像対象物が写った画像を取得して、当該画像から複数の部分画像である推定対象画像を取得する推定対象画像取得手段と、
画像に基づく情報を入力して当該画像の特徴量を出力する特徴量出力モデルを用いて、前記推定対象画像取得手段によって取得された複数の推定対象画像それぞれから複数の推定対象画像それぞれの特徴量を出力して、出力された特徴量から複数の推定対象画像それぞれに対応する合焦時の焦点位置を推定する焦点位置推定手段と、
前記焦点位置推定手段によって推定された複数の推定対象画像それぞれに対応する合焦時の焦点位置から、前記画像に写った撮像対象物の傾斜を推定する傾斜推定手段と、を備え、
前記特徴量出力モデルは、撮像時の焦点位置に係る焦点位置情報が対応付けられた複数の学習用画像から、機械学習によって生成され、互いに異なる2つの学習用画像に対応付けられた焦点位置情報に応じて当該2つの学習用画像の特徴量が比較されて、比較結果に基づいて機械学習が行われる傾斜推定システム。 - 前記焦点位置推定手段は、前記特徴量出力モデルから出力される特徴量を入力して、当該特徴量に係る画像に対応する合焦時の焦点位置を推定する焦点位置推定モデルを用いて複数の推定対象画像それぞれに対応する合焦時の焦点位置を推定し、
前記焦点位置推定モデルは、前記学習用画像それぞれに対応する合焦時の焦点位置に係る合焦位置情報から、機械学習によって生成される請求項1に記載の傾斜推定システム。 - 前記傾斜推定手段によって推定された撮像対象物の傾斜に基づいて、撮像対象物の撮像時の傾斜を制御する制御手段を更に備える請求項1又は2に記載の傾斜推定システム。
- 請求項1~3の何れか一項に記載の傾斜推定システムと、
前記傾斜推定システムに係る撮像対象物として半導体デバイスが載置される載置部と、
前記半導体デバイスを検査する検査部と、
を備える半導体検査システム。 - 請求項1~3の何れか一項に記載の傾斜推定システムと、
前記傾斜推定システムに係る撮像対象物として生体サンプルが載置される載置部と、
前記生体サンプルを観察する観察部と、
を備える生体観察システム。 - 画像に写った撮像対象物の傾斜を推定する傾斜推定方法であって、
撮像対象物が写った画像を取得して、当該画像から複数の部分画像である推定対象画像を取得する推定対象画像取得ステップと、
画像に基づく情報を入力して当該画像の特徴量を出力する特徴量出力モデルを用いて、前記推定対象画像取得ステップにおいて取得された複数の推定対象画像それぞれから複数の推定対象画像それぞれの特徴量を出力して、出力された特徴量から複数の推定対象画像それぞれに対応する合焦時の焦点位置を推定する焦点位置推定ステップと、
前記焦点位置推定ステップにおいて推定された複数の推定対象画像それぞれに対応する合焦時の焦点位置から、前記画像に写った撮像対象物の傾斜を推定する傾斜推定ステップと、を含み、
前記特徴量出力モデルは、撮像時の焦点位置に係る焦点位置情報が対応付けられた複数の学習用画像から、機械学習によって生成され、互いに異なる2つの学習用画像に対応付けられた焦点位置情報に応じて当該2つの学習用画像の特徴量が比較されて、比較結果に基づいて機械学習が行われる傾斜推定方法。 - 前記焦点位置推定ステップにおいて、前記特徴量出力モデルから出力される特徴量を入力して、当該特徴量に係る画像に対応する合焦時の焦点位置を推定する焦点位置推定モデルを用いて複数の推定対象画像それぞれに対応する合焦時の焦点位置を推定し、
前記焦点位置推定モデルは、前記学習用画像それぞれに対応する合焦時の焦点位置に係る合焦位置情報から、機械学習によって生成される請求項6に記載の傾斜推定方法。 - 前記傾斜推定ステップにおいて推定された撮像対象物の傾斜に基づいて、撮像対象物の撮像時の傾斜を制御する制御ステップを更に含む請求項6又は7に記載の傾斜推定方法。
- コンピュータを、画像に写った撮像対象物の傾斜を推定する傾斜推定システムとして動作させる傾斜推定プログラムであって、
当該コンピュータを、
撮像対象物が写った画像を取得して、当該画像から複数の部分画像である推定対象画像を取得する推定対象画像取得手段と、
画像に基づく情報を入力して当該画像の特徴量を出力する特徴量出力モデルを用いて、前記推定対象画像取得手段によって取得された複数の推定対象画像それぞれから複数の推定対象画像それぞれの特徴量を出力して、出力された特徴量から複数の推定対象画像それぞれに対応する合焦時の焦点位置を推定する焦点位置推定手段と、
前記焦点位置推定手段によって推定された複数の推定対象画像それぞれに対応する合焦時の焦点位置から、前記画像に写った撮像対象物の傾斜を推定する傾斜推定手段と、として機能させ、
前記特徴量出力モデルは、撮像時の焦点位置に係る焦点位置情報が対応付けられた複数の学習用画像から、機械学習によって生成され、互いに異なる2つの学習用画像に対応付けられた焦点位置情報に応じて当該2つの学習用画像の特徴量が比較されて、比較結果に基づいて機械学習が行われる傾斜推定プログラム。 - 前記焦点位置推定手段は、前記特徴量出力モデルから出力される特徴量を入力して、当該特徴量に係る画像に対応する合焦時の焦点位置を推定する焦点位置推定モデルを用いて複数の推定対象画像それぞれに対応する合焦時の焦点位置を推定し、
前記焦点位置推定モデルは、前記学習用画像それぞれに対応する合焦時の焦点位置に係る合焦位置情報から、機械学習によって生成される請求項9に記載の傾斜推定プログラム。 - 前記コンピュータを、
前記傾斜推定手段によって推定された撮像対象物の傾斜に基づいて、撮像対象物の撮像時の傾斜を制御する制御手段としても機能させる請求項9又は10に記載の傾斜推定プログラム。
Priority Applications (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US18/570,913 US20240282000A1 (en) | 2021-07-19 | 2022-03-10 | Incline estimation system, incline estimation method, incline estimation program, semiconductor inspection system, and organism observation system |
KR1020237041659A KR20240035945A (ko) | 2021-07-19 | 2022-03-10 | 경사 추정 시스템, 경사 추정 방법, 경사 추정 프로그램, 반도체 검사 시스템 및 생체 관찰 시스템 |
EP22845617.4A EP4336255A1 (en) | 2021-07-19 | 2022-03-10 | Incline estimation system, incline estimation method, incline estimation program, semiconductor inspection system, and organism observation system |
JP2022564614A JP7229439B1 (ja) | 2021-07-19 | 2022-03-10 | 傾斜推定システム、傾斜推定方法、傾斜推定プログラム、半導体検査システム及び生体観察システム |
CN202280050910.3A CN117677818A (zh) | 2021-07-19 | 2022-03-10 | 倾斜推定系统、倾斜推定方法、倾斜推定程序、半导体检查系统及生物体观察系统 |
JP2023020763A JP2023057119A (ja) | 2021-07-19 | 2023-02-14 | 傾斜推定システム、傾斜推定方法、傾斜推定プログラム、半導体検査システム及び生体観察システム |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2021118847 | 2021-07-19 | ||
JP2021-118847 | 2021-07-19 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2023002677A1 true WO2023002677A1 (ja) | 2023-01-26 |
Family
ID=84979881
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2022/010674 WO2023002677A1 (ja) | 2021-07-19 | 2022-03-10 | 傾斜推定システム、傾斜推定方法、傾斜推定プログラム、半導体検査システム及び生体観察システム |
Country Status (7)
Country | Link |
---|---|
US (1) | US20240282000A1 (ja) |
EP (1) | EP4336255A1 (ja) |
JP (2) | JP7229439B1 (ja) |
KR (1) | KR20240035945A (ja) |
CN (1) | CN117677818A (ja) |
TW (1) | TW202305659A (ja) |
WO (1) | WO2023002677A1 (ja) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP7426575B1 (ja) | 2023-07-13 | 2024-02-02 | ダイトロン株式会社 | 外観検査装置 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH11328408A (ja) * | 1998-05-12 | 1999-11-30 | Advantest Corp | データ処理装置および方法、情報記憶媒体 |
JP2012068188A (ja) * | 2010-09-27 | 2012-04-05 | Seiko Epson Corp | 調整方法 |
WO2017154895A1 (ja) | 2016-03-09 | 2017-09-14 | 浜松ホトニクス株式会社 | 測定装置、観察装置および測定方法 |
JP2021021637A (ja) * | 2019-07-29 | 2021-02-18 | コベルコ建機株式会社 | 建設機械の位置特定システム |
-
2022
- 2022-03-10 JP JP2022564614A patent/JP7229439B1/ja active Active
- 2022-03-10 KR KR1020237041659A patent/KR20240035945A/ko active Search and Examination
- 2022-03-10 US US18/570,913 patent/US20240282000A1/en active Pending
- 2022-03-10 CN CN202280050910.3A patent/CN117677818A/zh active Pending
- 2022-03-10 EP EP22845617.4A patent/EP4336255A1/en active Pending
- 2022-03-10 WO PCT/JP2022/010674 patent/WO2023002677A1/ja active Application Filing
- 2022-04-20 TW TW111114924A patent/TW202305659A/zh unknown
-
2023
- 2023-02-14 JP JP2023020763A patent/JP2023057119A/ja active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH11328408A (ja) * | 1998-05-12 | 1999-11-30 | Advantest Corp | データ処理装置および方法、情報記憶媒体 |
JP2012068188A (ja) * | 2010-09-27 | 2012-04-05 | Seiko Epson Corp | 調整方法 |
WO2017154895A1 (ja) | 2016-03-09 | 2017-09-14 | 浜松ホトニクス株式会社 | 測定装置、観察装置および測定方法 |
JP2021021637A (ja) * | 2019-07-29 | 2021-02-18 | コベルコ建機株式会社 | 建設機械の位置特定システム |
Also Published As
Publication number | Publication date |
---|---|
TW202305659A (zh) | 2023-02-01 |
EP4336255A1 (en) | 2024-03-13 |
JP7229439B1 (ja) | 2023-02-27 |
CN117677818A (zh) | 2024-03-08 |
JP2023057119A (ja) | 2023-04-20 |
KR20240035945A (ko) | 2024-03-19 |
JPWO2023002677A1 (ja) | 2023-01-26 |
US20240282000A1 (en) | 2024-08-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110270769B (zh) | 非破坏检测方法 | |
JP2012515434A (ja) | ウェーハを検査するためのシステム及び方法 | |
JP7229439B1 (ja) | 傾斜推定システム、傾斜推定方法、傾斜推定プログラム、半導体検査システム及び生体観察システム | |
EP3499178B1 (en) | Image processing system, image processing program, and image processing method | |
JP7205013B1 (ja) | 特徴量出力モデル生成システム、特徴量出力モデル生成方法、特徴量出力モデル生成プログラム及び特徴量出力モデル | |
US10475198B2 (en) | Microscope system and specimen observation method | |
CN110121629B (zh) | 借助角度选择的照射确定样本对象的布置 | |
JP7291303B1 (ja) | 焦点位置推定システム、焦点位置推定方法、焦点位置推定プログラム、半導体検査システム及び生体観察システム | |
WO2023002678A1 (ja) | 特徴量出力モデル生成システム、特徴量出力モデル生成方法、特徴量出力モデル生成プログラム及び特徴量出力モデル | |
JP2011133360A (ja) | 距離計測装置、距離計測方法、及びプログラム | |
KR102303073B1 (ko) | 다면 비전 검사 알고리즘 및 이를 이용한 시스템 | |
KR102191743B1 (ko) | 거리 측정 장치 | |
EP3499408B1 (en) | Image processing system and image processing program | |
CN108663370B (zh) | 端面检查装置及其聚焦图像数据获取方法 | |
US20240257490A1 (en) | Focal position estimation method, focal position estimation system, model generation method, model generation system, focal position estimation model, and non-transitory computer-readable storage medium | |
US6449388B1 (en) | Method for aligning an object with an image capture apparatus and a coupling eyepiece | |
JPH1184223A (ja) | 自動位置検出方法及び自動位置検出装置 | |
JP4874863B2 (ja) | Temのデフォーカス量測定方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
ENP | Entry into the national phase |
Ref document number: 2022564614 Country of ref document: JP Kind code of ref document: A |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22845617 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2022845617 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2022845617 Country of ref document: EP Effective date: 20231206 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 18570913 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202280050910.3 Country of ref document: CN |
|
NENP | Non-entry into the national phase |
Ref country code: DE |