JP5538179B2 - Image processing apparatus and image processing method - Google Patents

Description

  The present invention relates to an image processing apparatus and an image processing method for imaging an object such as a semiconductor wafer and measuring the position of the object from an obtained two-dimensional image.

  In a manufacturing process of a semiconductor device, a liquid crystal device, or the like, the substrate is moved between the semiconductor manufacturing devices in order to perform a plurality of processes on the substrate such as a wafer or glass. In each semiconductor manufacturing apparatus, for example, a substrate is fixed to a predetermined jig or processing position, and pattern exposure, plasma processing, sputtering, CVD, etching processing, or the like is performed. For this reason, the semiconductor manufacturing apparatus monitors the substrate transported in a required process so that it is correctly positioned at a predetermined position, for example, the mounting position or the processing position on the jig described above, and if necessary, the stop position or orientation of the substrate. Some have a mechanism for adjusting the deviation. For example, an imaging device and an image processing device are used for monitoring the position of the substrate.

  For example, Patent Document 1 proposes an image processing apparatus that observes a wafer as an object to be transferred with an imaging camera and measures the positional deviation of the wafer by image processing. In Patent Document 2, the two-dimensional image processing apparatus is used to observe three outer peripheral measurement points including notches on the outer periphery of the wafer to measure the two-dimensional position of the wafer and the rotational position of the wafer. Thus, the positional deviation of the wafer can be corrected.

JP 2009-88184 A Japanese Patent Laid-Open No. 9-186061

  As described above, an imaging device (camera) is provided at a predetermined position in a semiconductor manufacturing apparatus or the like in order to measure the position of a substrate (wafer) as a transfer object. For example, the image processing apparatus cuts out a part of the image of the target object that falls within the field of view of the imaging apparatus and can measure the contour of the target object or the position / shape of the target object included in the image, that is, the mark And the deviation from the reference position is measured based on this. The reason why the image processing device does not process the entire captured image but a part of it (measurement area) is to reduce the amount of data processing and shorten the measurement time, and to reduce noise on the image. This is due to reasons such as avoiding erroneous measurement due to noise. For this reason, a measurement region is set in advance in the captured image in image processing.

  By the way, when an imaging device (camera) is installed in a semiconductor manufacturing device or the like, a slight error in assembly may occur. As a result, even when the object (board) is correctly located at a predetermined position, the image of the object on the shooting screen may appear to be shifted from the planned position within the measurement area set on the screen. obtain. Therefore, when the imaging apparatus is installed in a semiconductor manufacturing apparatus or the like, it is necessary to appropriately set a measurement region on the imaging screen of the image processing apparatus in adjustment during assembly.

However, this adjustment work is unexpectedly troublesome because it is performed by measuring the dimensions of the image displayed on the screen.
For example, an image processing object such as a mark on the imaging screen causes an installation error when it is transported directly below the imaging device. Since the direction in which the installation error of the image processing object occurs is usually unpredictable, it is desirable that the reference position of the image processing object is at the center of the measurement region, assuming that it is shifted in all directions.
For this reason, the operator who sets the image processing apparatus measures the length from the edge of the measurement area to the image processing object on the imaging screen, and each edge of the image processing object and each edge of the measurement area. The position of the measurement region is adjusted so that the distance from the side is equal, that is, the image processing target is at the center of the measurement region.

  Since the operator repeats the length measurement and the measurement area position change to drive the position of the measurement area, it is unexpectedly troublesome. Furthermore, if the shape of the image processing object is a circle or a square like a general mark, the center position of the image processing object can be easily understood, which is relatively simple. For example, when an image processing target has a shape such as an arc, the center position is not easily known, and length measurement for determining the measurement region position becomes complicated and troublesome.

  Further, the setting adjustment of the measurement region to be subjected to the image processing in the shooting screen due to the position error of the image pickup apparatus is not limited to the semiconductor manufacturing apparatus, but the measurement system and the transport system using the image pickup apparatus and the image processing apparatus. Can also occur.

  Therefore, an object of the present invention is to facilitate the adjustment of the measurement range setting in the shooting screen in the image processing apparatus.

One aspect of the image processing apparatus of the present invention is an imaging means for obtaining a two-dimensional image by imaging at least one end of an object conveyed within the field of view of a camera, and a display means for displaying the two-dimensional image on a screen. Shape extracting means for extracting a contour line representing the contour of one end of the object from the obtained two-dimensional image, point position extracting means for extracting the positions of a plurality of points on the contour line, and the plurality of points. Representative position determining means for determining the position of a representative point representing the position of the contour line in the transport direction of the object on the two-dimensional image, and the vertical width n in the transport direction in the two-dimensional image, Measurement area setting means for setting a measurement area having a width m in the direction intersecting the conveyance direction and including the outline of the one end, and the measurement area setting means is configured to detect the normality of the conveyance direction from the position of the representative point. Conversely (n / 2) each Setting the vertical width, and wherein the.
An aspect of the image processing apparatus of the reference example includes an imaging unit that captures an object conveyed in a field of view to obtain a two-dimensional image, a display unit that displays the two-dimensional image on a screen, and the obtained two-dimensional Shape extracting means for extracting a contour line representing the contour of the object from the image, point position extracting means for extracting the positions of a plurality of points on the contour line, and the object on the two-dimensional image from the plurality of points Representative position determining means for determining the position of the representative point representing the position of the contour line in the transport direction, and at least the position of the representative point and the measurement set in advance in the transport direction of the object on the two-dimensional image Measurement area setting means for setting the measurement area including the contour line on the two-dimensional image obtained based on the width.
These means include those realized by execution of a control program and an image processing program by a computer. Moreover, what is implement | achieved by an electrical circuit, an apparatus, etc. is included.

  By adopting such a configuration, it is possible to automatically set an optimum measurement region for a captured image for reading the contour of the object.

  Preferably, the width of the measurement area in a direction intersecting the conveyance direction of the object on the screen is set in accordance with the size of the display area of the two-dimensional image on the screen.

  Preferably, the plurality of points include intersections of the contour line and the field frame in the coordinate system of the two-dimensional image.

  Preferably, the plurality of points include a point located at a convex portion or a concave portion of the contour line in the coordinate system of the two-dimensional image.

  Preferably, the position of the representative point includes an average position of the plurality of points in the transport direction of the object in the coordinate system of the two-dimensional image.

  Preferably, the position of the representative point includes an intermediate position between two points having a maximum separation distance in the conveyance direction of the object among the plurality of points in the coordinate system of the two-dimensional image.

  Preferably, the measurement region setting unit further sets a measurement region including the contour line on the two-dimensional image based on the conveyance error of the object.

In addition, an aspect of the image processing apparatus of the reference example is obtained by an imaging unit that captures an object conveyed in the field of view and obtains a two-dimensional image, and a display unit that displays the two-dimensional image on a screen. Shape extracting means for extracting the mark displayed on the object from the two-dimensional image, centroid position extracting means for extracting the position of the centroid of the mark in the coordinate system of the two-dimensional image, and at least the position of the centroid And a measurement area setting means for setting a measurement area including the barycentric point on the obtained image based on a measurement width set in advance in the conveyance direction of the object on the two-dimensional image.

  By adopting such a configuration, it becomes possible to automate the setting of the optimum measurement region for reading the (alignment) mark of the object.

  Preferably, the measurement area setting means further sets a measurement area including the mark on the two-dimensional image based on the conveyance error of the object.

According to another aspect of the image processing method of the present invention, there is provided an imaging process for obtaining a two-dimensional image by imaging at least one end of an object conveyed in the field of view of the camera, and the object from the obtained two-dimensional image. A shape extraction process for extracting a contour line representing the contour of one end of the image, a point position extraction process for extracting the positions of a plurality of points on the contour line, and transport of an object on the two-dimensional image from the plurality of points A representative position determination process for determining the position of a representative point representing the position of the contour line in the direction, and a vertical width n in the transport direction and a horizontal width m in the direction intersecting the transport direction in the two-dimensional image, A measurement area setting process for setting a measurement area including the outline of the one end, and the measurement area setting process has a vertical width of (n / 2) from the position of the representative point in the forward and reverse directions of the transport direction. Set .
In addition, one aspect of the image processing method of the reference example was obtained as an imaging process for capturing a two-dimensional image by capturing an object conveyed in the field of view, and a display process for displaying the two-dimensional image on the screen. A shape extracting process for extracting a contour line representing the contour of the object from a two-dimensional image, a point position extracting process for extracting the positions of a plurality of points on the contour line, and a plurality of points on the two-dimensional image from the plurality of points. A representative position determination process for determining the position of a representative point representing the position of the contour line in the conveyance direction of the object, and at least the position of the representative point and the conveyance direction of the object on the two-dimensional image are set in advance. A measurement region setting step of setting the measurement region including the contour line on the obtained two-dimensional image based on the measured width.

  By adopting such a configuration, it is possible to automatically set an optimum measurement region for a captured image for reading the contour of the object.

In addition, one aspect of the image processing method of the reference example was obtained as an imaging process for capturing a two-dimensional image by capturing an object conveyed in the field of view, and a display process for displaying the two-dimensional image on the screen. A shape extracting process for extracting the mark displayed on the object from the two-dimensional image, a centroid position extracting process for extracting a position of the centroid point of the mark in the coordinate system of the two-dimensional image, and at least the position of the centroid And a measurement area setting step for setting a measurement area including the barycentric point on the obtained image based on a measurement width preset in the conveyance direction of the object on the two-dimensional image.

  By adopting such a configuration, it becomes possible to automate the setting of the optimum measurement region for reading the (alignment) mark of the object.

  According to another aspect of the present invention, there is provided a semiconductor manufacturing apparatus including the above-described image processing apparatus.

  According to the present invention, the setting of the range to be measured in the image processing apparatus is automatically set to the optimum range as much as possible by the image processing apparatus.

It is explanatory drawing explaining the example of a semiconductor manufacturing apparatus provided with an imaging camera and an image processing apparatus. FIG. 2 is a block diagram illustrating a configuration example of an image processing apparatus. It is a flowchart explaining the procedure which sets a measurement area | region to the capture image of this invention. It is explanatory drawing explaining the example of the visual field of the imaging camera set on the object. It is explanatory drawing explaining the captured image when the position of an imaging camera has shifted | deviated to the vertical direction with respect to the target object (wafer). It is explanatory drawing explaining the example which displayed the captured image on the screen. It is explanatory drawing explaining the example by setting the measurement range optimal for a captured image. It is explanatory drawing explaining the example which sets a measurement area | region within a capture image. It is an explanatory view for explaining an example of determining the coordinates of P ₀ of the measurement area. It is explanatory drawing explaining the example of a captured image in case a camera position exists in the position close | similar to a target object. It is explanatory drawing explaining the example of a captured image in case a camera position exists in a position far from a target object. It is explanatory drawing explaining the capture image A of the alignment mark on a wafer. It is a flowchart explaining the procedure in the case of setting a measurement area | region to an alignment mark.

  Hereinafter, the present invention will be described through embodiments of the invention with reference to the drawings. However, the following embodiments do not limit the invention according to the claims, and are described in the embodiments. Not all combinations of features are essential to the solution of the invention. In the following embodiments, an example in which the image processing apparatus of the present invention is applied to a semiconductor manufacturing apparatus will be described as an example. However, the application of the image processing apparatus of the present invention is not limited to this, for example, workpiece transfer It can also be used for devices.

Example 1
In the following embodiments of the invention, for convenience of explanation, first, an outline of a semiconductor device using the image processing measure of the present invention will be described, and then the configuration of the image processing measure and setting of a measurement region in the image processing device will be described. explain.

(Semiconductor processing equipment)
FIG. 1 is a diagram schematically showing the configuration of an image processing apparatus and a semiconductor manufacturing apparatus using the same according to an embodiment of the present invention. As shown in FIG. 1, the image processing apparatus 1 measures the center position of the wafer 3 and the rotation position (rotation angle) of the wafer 3 with reference to the orientation flat part or the notch part in a single wafer transfer semiconductor manufacturing apparatus. . The semiconductor manufacturing apparatus includes a transfer chamber (transfer vacuum chamber) 4, a process chamber (processing chamber) 5, a load / unload device 6, and a control device 8.

  The transfer chamber 4 is provided with a robot arm (transfer mechanism) 9 for transferring the wafer 3, and the robot arm 9 is used to load / unload the wafer 3 into / from the process chamber 5 and load / unload apparatus 6. The wafer 3 is transferred between them. Although a plurality of process chambers 5 can be received, only one is shown here. In addition, a view port (not shown) for removing the inside of the camera is provided at the upper and lower portions of the transfer chamber 4.

  In the example of FIG. 1, the imaging camera 2 is attached to the outside of the transfer chamber 4 via a view port provided at the top of the transfer chamber 4. In addition, transmitted illumination, which will be described later with reference to FIG. 2, is attached to the outside of the transfer chamber 4 through a view port provided in the lower part of the transfer chamber 4. This transmitted illumination can illuminate a part of the outer periphery of the wafer 3 located in the visual field A of the imaging camera 2 from directly below through the lower viewport. Thereby, the imaging camera 2 can capture an image of a part of the outer periphery of the wafer 3 illuminated with transmission illumination from directly below from above, via the upper viewport.

  For the sake of simplicity, FIG. 1 shows a case where the transfer chamber 4 is provided with one imaging camera 2 and one transmitted illumination corresponding to the process chamber 5, but there are a plurality of imaging cameras not shown. A plurality of process chambers 5 and load / unload apparatuses 6 are provided.

The process chamber 5 is provided for each type of vacuum processing such as sputtering, PVD (Physical Vapor Deposition), etcher, and CVD, and the vacuum processing is performed on the wafer 3 carried in from the transfer chamber 4. The process chamber 5 is not provided with a view port for the reason of avoiding ambient light. The transfer chamber 4 and the process chamber 5 are in a vacuum state by an exhaust system (not shown).
The load / unload device 6 loads and unloads the wafer 3 between the vacuum transfer chamber 4 and the outside.

  The control device 8 controls a robot arm (transfer mechanism) 9 for transferring wafers to load / unload the wafers 3 into / from the process chamber 5 and transfer the wafers 3 to / from the load / unload device 6. .

(Image processing device)
FIG. 2 is a block diagram showing the configuration of the image processing apparatus 1 in FIG. In the figure, the image processing apparatus 1 includes a signal conversion unit 102, a captured image storage unit 104, a display image storage unit 106, a signal conversion unit 108, an input interface 110, a database unit 112, a communication control unit 114, a ROM unit 116, a main unit. The memory unit 118, the CPU unit 120, the display unit 130, the input unit 140, and the like are included.

  As shown in FIG. 2, the illumination light source 42 illuminates for photographing from below the wafer 3 through a view port (view window) (not shown) provided in the chamber. This illumination makes the outline of the wafer clear. The imaging camera 2 sets a visual field A at the edge portion of the wafer. Settings regarding the field of view A of the imaging camera 2 will be described later. The imaging camera 2 includes a two-dimensional sensor such as a CCD array, and sends an image of the wafer to the signal conversion unit 102 as an image signal. The signal conversion unit 12 converts the image signal into data of each pixel of the wafer image and sends it to the captured image storage unit 104. The captured image storage unit 104 is a memory that temporarily stores image data. The signal converter 102 and the captured image storage unit 104 can be provided on the imaging camera 2 side. The image data in the captured image storage unit 104 is sent to the main memory unit 118 and subjected to image processing by the CPU, and control parameters required by the control device 8 such as the position and inclination of the wafer edge are extracted.

  The display image storage unit 106 is a memory that temporarily stores image data to be displayed on the display unit 130 output from the main memory unit 118. The signal conversion unit 108 converts the image data into an image signal and outputs the image signal to the display unit 130. The display unit 130 is, for example, an LCD (liquid crystal) display that displays a two-dimensional image, but may be a plasma display or a CRT display, and more preferably has a so-called touch panel on the screen. The display image storage unit 106 and the signal conversion unit 108 can be provided on the display unit 130 side.

  An input unit 140 is disposed near the display unit 130. The input unit 140 is a keyboard (code input device) for inputting a command code, and a mouse (screen input device) for inputting a predetermined command by a combination of the position of a pointer displayed on the screen and display information (such as an icon) on the screen. ), A touch panel for inputting coordinates on the screen, a power switch, and the like.

  The interface 110 converts an operation command signal from the input unit 140 into various commands, input data, and the like and transmits the command to an input / output processing program (not shown).

  The database unit 112 includes the shape, contour shape, center position, approximate expression indicating the shape (contour), the shape of the alignment mark, the position of the feature point for shape recognition, Data such as the position of the center of gravity is stored in advance as a database. In addition, an image processing program and a subroutine program corresponding to various objects are stored. Wafers and substrates handled by semiconductor manufacturing apparatuses are known in advance in terms of overall shape and dimensions, marker positions, alignment mark shapes, etc. (design information).

  The communication control unit 114 performs data communication between the image processing device 1 and the control device 8 that is an external device or other devices (for example, a process control device, a server device, a network storage device, etc., not shown) connected to the network. I do. The ROM unit 116 continuously holds basic programs necessary for the operation of the image processing apparatus.

  The main memory unit 118 stores a CPU operating system, an image processing control program, a misregistration detection program, image data, and the like, and updates them as needed. The CPU unit 120 is an arithmetic processing unit of the computer system, and executes a program to perform image processing of a captured object and perform a predetermined output. The CPU unit 120 can execute a plurality of programs concurrently. In the illustrated example, there is one imaging camera 2, but image data of each camera provided at a plurality of locations can be processed. The result is sent to the control device 8 or the like and used for controlling the stop position of the robot arm 9 or the like.

(Measurement area setting for captured images)
FIG. 3 is a flowchart for describing processing for setting a measurement region in a captured image among image processing executed by the CPU unit 120 in the image processing apparatus 1.
As described above, a measurement region is set in the field of view of the imaging camera 2 after the imaging camera 2 is assembled in the semiconductor manufacturing apparatus. As described above, the measurement time can be shortened by setting a part of the camera image as an object of image processing. Further, by setting the target region to be narrow, it is possible to reduce noise in the image existing in the region and reduce erroneous measurement.

  As shown in FIG. 3, the CPU executing the control program executes this routine when receiving a command to execute the program for setting the measurement area in the captured image in a standby state or an interrupt process (step S100).

(1-1) Image Capture In this process, an imaging process for capturing a two-dimensional image by capturing an object conveyed in the field of view and a display process for displaying the two-dimensional image on the screen are performed (step S110). .
First, an image of the wafer 3 transferred into the field of view A of the imaging camera 2 by the robot arm 9 is captured (imaged) by the imaging camera 2 and is stored in the main memory unit 118 via the signal conversion unit 102 and the captured image storage unit 104. As a bitmap image. The captured image is sent to the display unit 130 via the display image storage unit 106 and the signal conversion unit 108, and a captured image (monitor image) is displayed. Note that the display unit 130 does not necessarily display the captured image itself. For example, the display unit 130 may display only the measurement region obtained as a result of performing the processing described later in the present embodiment. Needless to say, it can be changed as appropriate.

  FIG. 4 shows an example of two fields of view A and B of the imaging camera 2 set on the wafer 3. The center O (a, b) of the wafer 3 exists on the X and Y coordinates of the image bitmap, and the visual fields A and B are set so that the center O includes a straight line in the region. A field of view A ′ indicates a case where the setting of the field of view A of the imaging camera 2 is shifted.

  FIG. 5A shows an image example of the monitor screen when the position of the field of view A of the imaging camera 2 is not shifted from the wafer 3. FIG. 4B shows an example of an image A ′ (see FIG. 4) when the position of the visual field A of the imaging camera 2 is shifted in the Y-axis direction with respect to the wafer 3. The wafer 3 is relatively moved upward (Y direction) on the monitor screen.

  FIG. 6 shows an example of a captured image displayed on the screen of the display unit 130. The left area and the center area of the screen display a capture image (hereinafter also referred to as capture image A) of the wafer 3 that enters the field of view A of the imaging camera 2 selected in advance. In the right area of the screen, parameter display areas 132 for displaying the parameters for image processing of the selected imaging camera are provided for the number of imaging cameras 2. If the entire captured image A is set as a measurement target area, data processing takes time. Therefore, the purpose of this process is to set an optimal measurement region Q for a part of the captured image A as shown in FIG. 7 described later and perform measurement using image data in the region.

(1-2) Contour Calculation of Object In this process, a shape extraction process for extracting a contour line representing the contour of the object from the obtained two-dimensional image is performed (step S112).
An expression (approximate expression) representing the contour of the measurement object is obtained. As shown in FIG. 4, when the object is a wafer 3, the shape is circular. Contour processing for extracting a contour line from the bitmap image of the wafer 3 is performed, coordinates of arbitrary points constituting the contour line are obtained, and a true circle formula (= (x−a) ² + (y−b) ² = r ² ) to obtain an equation R (circular) representing the contour of the wafer 3.

(1-3) Optimal measurement region setting In this process, a point position extraction process for extracting the positions of a plurality of points on the contour line, and the contour line in the conveyance direction of the object on the two-dimensional image from the plurality of points. A measurement area including a contour line based on a representative position determination process for determining a position of a representative point representing the position, and at least a position of the representative point and a measurement width preset in the conveyance direction of the object on the two-dimensional image A measurement area setting process for setting the obtained image on the obtained two-dimensional image is performed (step S114).

As shown in FIG. 8, a rectangular measurement region Q defined by a horizontal width (the number of horizontal pixels) m × a vertical width (the number of vertical pixels) n is set in the captured image A. The horizontal width m in the X direction of the measurement area Q is, for example, the horizontal width value of the area of the captured image A in order to improve measurement accuracy. The vertical width n in the Y direction of the measurement region Q is preferably as small as possible for shortening the measurement time, for example. The minimum value of the vertical width n is determined in consideration of the conveyance accuracy in the Y direction of the object (wafer). Even if there is variation in the position of the object, the contour line (approximate circle) R of the object is set within the vertical width n. The measurement area Q can be represented by, for example, the coordinate position of the upper left corner P ₀ of the quadrangular area, the horizontal width m, and the vertical width n.

Next, the coordinate position P ₀ (x, y) of the corner portion P ₀ is obtained. As shown in FIG. 9, the coordinates of _three points P ₁ , P ₂ , P ₃ on the contour line R in the field of view A are obtained. The point P ₁ is set to the coordinates (a ₁ , b ₁ ) of the intersection of the straight line at the left end in the X direction of the measurement region Q and the approximate circle R. The point P ₂ is set to the coordinates (a ₂ , b ₂ ) of the intersection of the straight line at the right end in the X direction of the measurement region and the approximate circle R. The point P ₃ has coordinates (a ₃ , b ₃ ) of the intersection point P ₃ between the straight line X = a and the approximate circle R when the coordinates of the center O of the approximate circle R are (a, b) (see FIG. 4). Set to. Point P ₃ corresponds to a convex portion or a concave portion of the contour line.

In this case, the X position of the coordinate P _{0 at} the upper left of the measurement area is a ₁ . The y position of the coordinate P ₀ is calculated by a predetermined formula so as to satisfy the requirement for n. In one example, the point P _1, P _2, P each position in the Y-axis direction ₃ b _1, b _2, b intermediate value (P ₁ a plurality of points on the outline of the distribution of _3, P _2, P ₃ ) (Corresponding to the y value of the representative point P _R )) plus half (n / 2) of the vertical width n.

For example, an "average value of the point P _1, P _2" the average y-position of the half width (the representative point P corresponds to the y value of _R) (n / 2) was added to P ₀ and "point P ₃ ' The calculation is made according to the equation <[{(b ₁ + b ₂ ) / 2} + b ₃ ] / 2> + (n / 2). The formula is stored in advance in the database unit 112 corresponding to the object.

The horizontal width m of the measurement region is preferably set to a value equal to the maximum width in the captured image A when the approximate circle R intersects both side edges of the captured image A as shown in the drawing. This is because if m takes a value less than the maximum width, a part of the approximate circle R is out of the measurement region, which may cause a reduction in measurement accuracy. On the other hand, if the approximate circle R intersects the upper edge of the captured image A and does not intersect either or both of the side edges, between the two points where the approximate circle R and the captured image A intersect. It may be determined by adding the accuracy of conveyance of the object (wafer) in the X direction to the size of the width as necessary. Hereinafter, unless otherwise specified, the description will be limited to the case where the approximate circle R intersects both side edges of the captured image A.
(1) The measurement region Q in the case of the first example is defined as follows.
(a) Width m of measurement area: Pre-specified by worker (maximum capture area)
(b) Vertical width n of the measurement area: designated by the operator in advance (usually determined from the accuracy with which the transfer device installs the object (wafer) under the imaging camera 2)
(c) Upper left coordinate P ₀ (x, y) = (a ₁ , <[{(b ₁ + b ₂ ) / 2} + b ₃ ] / 2> + (n / 2)) of the measurement area

The definition of the optimum position for the measurement region Q may have various ways of thinking, for example, the following examples.
(2) If If point P ₁ and P ₂ of the second embodiment is b ₁ ≧ b _2, the y of the upper left coordinates P _0, the average value of b ₁ and b ₃ (the y value of the representative point P _R Equivalent)) plus half width (n / 2).
(a) Width m of measurement area: Pre-specified by worker (maximum capture area)
(b) Vertical width n of the measurement area: designated by the operator in advance (usually determined by the accuracy with which the apparatus places the wafer under the camera)
(c) Upper left coordinate P ₀ (x, y) = (a ₁ , {(b ₁ + b ₃ ) / 2} + (n / 2)) of the measurement area

(3) If the case point P ₁ and P ₂ of the third example is b ₁ ≦ b _2, the y of the upper left coordinates P _0, the average value of b ₂ and b ₃ (the y value of the representative point P _R Equivalent)) plus half width (n / 2).
(a) Width m of measurement area: Pre-specified by worker (maximum capture area)
(b) Vertical width n of the measurement area: designated by the operator in advance (usually determined by the accuracy with which the apparatus places the wafer under the camera)
(c) Upper left coordinate P ₀ (x, y) = (a ₁ , {(b ₂ + b ₃ ) / 2} + (n / 2)) of the measurement area

(4) a case y of the upper left coordinates P ₀ of the fourth example, the simple average value of the point P _1, P _2, Y coordinate values b1 of _{P 3, b 2, b 3} ( the y value of the representative point P _R Equivalent)) plus half width (n / 2).
(a) Width m of measurement area: Pre-specified by worker (maximum capture area)
(b) Vertical width n of the measurement area: designated by the operator in advance (usually determined by the accuracy with which the apparatus places the wafer under the camera)
(c) Upper left coordinate P ₀ (x, y) = (a ₁ , {(b ₁ + b ₂ + b ₃ ) / 3} + (n / 2)) of the measurement area
In this manner, up and down positions on the Y axis of the representative point P _R of points on the contour line (n / 2) the vertical width settings, X-axis direction of the imaging camera by setting the width m A measurement area Q is set in the captured image.

(1-4) Real Value Conversion Coefficient Calculation Next, returning to FIG. 3, the CPU calculates the real value conversion coefficient (step S116).
This is because when there is an error in the height position (distance between the wafer and the camera) of the imaging camera 2 existing above the wafer 3 (Z-axis direction) with respect to the wafer 3 existing in the XY plane, Since the size (size of one pixel) is different, this can be corrected.
FIG. 10 shows an example of imaging an object when the distance between the imaging camera 2 and the wafer 3 is relatively short. When the distance is short, the captured image becomes relatively large. FIG. 11 shows an example of imaging an object when the distance between the camera 2 and the wafer 3 is relatively long. The image becomes relatively small when the distance is long. If there are such errors (variations) in the Z direction in the imaging camera, when a plurality of imaging cameras are used, the size (length) on the object (wafer) corresponding to one pixel of the image differs depending on the imaging camera. It will be. In semiconductor manufacturing, a plurality of semiconductor manufacturing apparatuses using an imaging camera are used. In general, a plurality of imaging cameras are used in one semiconductor manufacturing apparatus. Therefore, it is necessary to be able to accurately measure the size of the object, the amount of displacement of the position of the object, and the like from the image of each imaging camera, and the actual value conversion coefficient of each imaging camera is calculated.

The actual value conversion coefficient is defined by a ratio k between the actual dimension r ₀ [mm] of the object and the dimension r [number of pixels] of the same object in the screen coordinate system (image). The unit of the actual dimension of the object is appropriately selected such as cm or μm.
The real value conversion coefficient k is calculated by k = (r ₀ / r) [mm / pix].
For example, in the case of a wafer, the actual dimension r ₀ (mm) of the object is known in advance such as a wafer radius of 150 mm (12 inches) or 100 mm (8 inches), and is stored in the database unit 112. Further, if necessary, a predetermined dimension of the object is measured and stored in the database unit 112. As the size of the object in the image, r of equation R obtained in the above-described contour calculation of the object can be used.

(1-5) Image processing parameter storage The CPU stores various parameters of the image processing determined in the above steps (step S118).
In FIG. 7, the CPU performs the above-described processing to cut out an optimum measurement region Q (upper left coordinate P ₀ , display region m × n) that includes the outline of the object from the captured image A of the imaging camera 2 at the approximate center. The state which displayed the said part on the indicator 130 is shown. The parameter display area 132 displays parameters such as the upper left coordinate P ₀ (x, y), the vertical width n, and the actual value conversion coefficient k of the optimum measurement area for each imaging camera. The worker confirms the display contents of the display unit 130, and if there is no problem, instructs the CPU to save image processing parameters. This parameter is stored in a nonvolatile manner in the ROM unit 116, the database unit 112, or the like. If there is a problem, steps S110 to S118 are repeated to set an optimal measurement region Q.

  Such parameter setting operation is performed for each imaging camera. After completion of each process, the CPU returns to the main program (step S120).

  For example, measurement of the contour position of the wafer by an image processing apparatus as described in JP-A-2009-88184 is performed by the measurement region Q set in this way.

  According to the first embodiment of the present invention, the measurement area Q can be easily set in the captured image of the observation object by the imaging camera, so that the burden on the worker is reduced and the condition is good.

(Example 2)
Next, an example in which the present invention is applied to alignment mark recognition will be described with reference to FIGS. In the first embodiment of the present invention described above, the case where the contour line of the object intersects the captured image A has been described. In this embodiment, the case where the contour line of the object does not intersect the captured image A is described. Both are common in that they relate to the setting of the optimum measurement region for reading the (alignment) mark of the object on the two-dimensional image.

FIG. 12 is an explanatory diagram illustrating an example in which an alignment mark formed on an object (wafer) is captured by an imaging camera. FIG. 13 is a flowchart for explaining the measurement region setting procedure for the alignment mark corresponding to FIG.
In the first embodiment of the present invention described above, the wafer before circuit pattern formation is targeted in the manufacturing process of the semiconductor device. However, the present invention is applicable to image processing of a wafer on which alignment marks are formed. The present invention can also be applied when image processing is performed using marks formed on a liquid crystal glass substrate or the like.

  As shown in FIG. 12, the image processing apparatus 1 captures an area including the alignment mark M of the object with the imaging camera 2 and captures it in the main memory. Further, the captured image A is displayed on the display unit 130. In this example, the alignment mark M is circular and the size (diameter) is a. As will be described later, the accuracy of installing the object in the field of view A of the image camera 2 is ± b [mm].

A measurement region (for example, a quadrangular region) Q is set including the alignment mark M substantially in the center. The coordinates (x, y), the horizontal width m, and the vertical width n of the upper left corner P ₀ of the measurement region Q are appropriately set as in the first embodiment. Thereby, the amount of data processing in measurement is reduced and the measurement time is shortened. In addition, by setting the target area to be narrow, the noise of the image existing in the area is reduced to reduce erroneous measurement.

  Next, an image processing procedure by the CPU will be described. As shown in FIG. 13, the CPU executing the control program executes this routine when receiving a command to execute a program for setting a measurement area in a captured image in a standby state or an interrupt process (step S200).

(2-1) Image Capture In this process, an imaging process for capturing a two-dimensional image by capturing an object conveyed in the field of view and a display process for displaying the two-dimensional image on the screen are performed (step S210). .
The CPU causes the imaging camera 2 to capture (capture) an image of the wafer 3 that has been transferred into the field of view A of the imaging camera 2 by the robot arm 9 and capture it as a bitmap image in the main memory unit 118. The captured image is also sent to the display unit 130, and a captured image (monitor image) is displayed.

(2-2) Mark Center of Gravity Calculation The CPU calculates the center of gravity position of the alignment mark M (step S212). In this process, a shape extraction process for extracting the mark displayed on the object from the two-dimensional image and a centroid position extraction process for extracting the position of the centroid point of the mark in the coordinate system of the two-dimensional image are performed (step S212). ).

For example, if the two-dimensional shape of the alignment mark M is a substantially perfect circle, a perfect circle approximation may be used. However, if the shape of the mark M is other than a perfect circle, for example, an unspecified shape including a polygon, it is convenient to use the centroid position calculation. Good. The center-of-gravity position calculation is performed, for example, by extracting pixel data that constitutes the area of the alignment mark M, and performing binarization processing on the luminance and color information of the pixel so that the mark M portion is black and the other portion (background) is white. Separate them so that Then, it can be obtained by calculating the center-of-gravity position P _g representing each pixel position of the black portion by formulas known centroid.
In order to further improve the accuracy of the centroid position, detailed edge position coordinates may be obtained from the luminance values of pixels near the edge of the alignment mark M by processing such as linear interpolation, and the centroid may be obtained.

(2-3) Calculation of Optimal Measurement Area The CPU calculates an optimal measurement area Q around the alignment mark M (step S214). In this process, a measurement region setting process is performed in which a measurement region including a centroid point is read and set on an image based on the position of the center of gravity and a measurement width set in advance in the conveyance direction of the object on the two-dimensional image ( Step S214).

Center-of-gravity position P _g of coordinates of the alignment mark M (x, y), the size of the alignment mark a [pix], accuracy ± b to install a measurement object to below the camera, and when, for example, the optimal measurement region It is expressed as follows.
(a) Width of measurement area m: m = a + 2b
(b) Vertical width n of measurement area: n = a + 2b
(c) of the center position P _g of the measuring area coordinates: (x, y)
(d) Coordinate of upper left P ₀ of measurement area: (x− (m / 2), y + (n / 2))
The CPU displays the measurement area Q thus determined in the capture screen A.

(2-4) Calculation of real value conversion coefficient The CPU calculates the real value conversion coefficient (step S216).
First, the size (image coordinate system) of the alignment mark M is obtained from the read image by image processing. For example, when the alignment mark M is set to black in the binarization processing of the image data, for example, the distance b [pix] between the leftmost black pixel and the rightmost black pixel corresponds to the size of the alignment mark. Then you can see. From this distance information and the known alignment mark size a [mm], the actual value conversion coefficient k can be obtained. The size a of the alignment mark M of the object is stored in advance in the database unit 112 as design information. The unit of the dimension of the alignment mark of the object is appropriately selected such as cm or μm.
Real value conversion coefficient k = a / b [mm / pix]

(2-5) Parameter storage of image processing apparatus The CPU stores various parameters of the image processing determined in the above steps (step S218). The worker confirms the display contents of the display unit 130, and if there is no problem, instructs the CPU to save image processing parameters. This parameter is stored in a nonvolatile manner in the ROM unit 116, the database unit 112, or the like. If there is a problem, steps S210 to S218 are repeated to set an optimal measurement region Q.

  Such parameter setting operation is performed for each imaging camera. After completion of each process, the CPU returns to the main program (step S220).

(Example 3)
(Method for obtaining the accuracy of placing the measurement object under the camera and method for obtaining the reference position of the measurement object)
In the first and second embodiments, the absolute position coordinates (reference position) where the target object should be set or the mark on the target position should be set with respect to the installation position when the measurement target is transported within the field of view of the camera. When the measurement object is actually installed or the error from the position coordinates where the mark is located is known, it is possible to automate the process including the process of obtaining the position error when the error is not known. . For example,

(1) Transfer the object from the position before transfer to the bottom of the camera, and place the object at the specified position.
(2) The position of the center of gravity of the alignment mark M of the object is measured (round circle approximate measurement) at this position.
(3) Transport the object to the position before transport (return it to the original position).

  Repeat the above steps (1) to (3) any number of times (the greater the number of samples, the better the accuracy), for example, about 100 times, and calculate the difference between the maximum and minimum values of the center of gravity obtained in (2) This is the accuracy when placing the measurement object under the camera. In order to give a margin to this value, it may be further increased by + α.

  Further, for example, with respect to the 100 center of gravity position coordinates obtained in (2), a position that becomes the center of gravity of each coordinate or a point (representative point) that is simply the average of each coordinate is calculated, thereby obtaining this as the reference position. The reference position can be used as the center position coordinate of the optimum measurement region as it is. Even if the outer shape of the image processing object is, for example, an arc shape, the present invention can be applied to various calculations by using the coordinates of the convex portion or concave portion as a reference.

  As described above, according to the embodiment of the present invention, since the image processing apparatus sets an optimal measurement region in the captured image A, the data processing amount in the image processing can be reduced to shorten the measurement time. Therefore, it is possible to reduce the labor of the worker while enjoying the advantages such as avoiding the erroneous measurement due to the noise by reducing the noise intake.

  The examples and application examples described through the embodiments of the present invention can be used in appropriate combination according to the application, or can be used with modifications or improvements, and the present invention is limited to the description of the above-described embodiments. It is not something. It is apparent from the description of the scope of claims that the embodiments added with such combinations or changes or improvements can be included in the technical scope of the present invention.

DESCRIPTION OF SYMBOLS 1 Image processing apparatus, 2 Imaging camera, 3 Wafer (object), 4 Transfer chamber, 5 Process chamber, 6 Load / unload apparatus, 8 Control apparatus, 102 Signal conversion part, 104 Captured image storage part, 106 Display image storage Unit, 108 signal conversion unit, 110 interface, 112 database unit, 114 communication control unit, 116 ROM unit, 118 main memory unit, 120 CPU unit, 130 display unit, 140 input unit

Claims (9)

Imaging means for obtaining a two-dimensional image by imaging at least one end of an object conveyed in the field of view of the camera ;
Display means for displaying the two-dimensional image on a screen;
Shape extraction means for extracting a contour line representing the contour of one end of the object from the obtained two-dimensional image;
Point position extracting means for extracting positions of a plurality of points on the contour line;
Representative position determining means for determining the position of a representative point representing the position of the contour line in the conveyance direction of the object on the two-dimensional image from the plurality of points;
A measurement area setting means for setting a measurement area including the outline of the one end of the vertical dimension n in the conveyance direction and a horizontal width m in the direction intersecting the conveyance direction in the two-dimensional image ;
The image processing apparatus, wherein the measurement region setting means sets a vertical width of (n / 2) from the position of the representative point in the forward and reverse directions of the transport direction . The width m of the measurement area is set corresponding to the size of the horizontal width of the display area of the two-dimensional image in the screen, the image processing apparatus according to claim 1.   The image processing apparatus according to claim 1, wherein the plurality of points include intersections of the contour line and the field frame in the coordinate system of the two-dimensional image.   The image processing apparatus according to claim 1, wherein the plurality of points include points located at a convex portion or a concave portion of the contour line in the coordinate system of the two-dimensional image.   5. The image processing apparatus according to claim 1, wherein the position of the representative point is an average position of the plurality of points in the transport direction of the object in the coordinate system of the two-dimensional image.   The position of the representative point is an intermediate position between two points having a maximum separation distance in the conveyance direction of the object among the plurality of points in the coordinate system of the two-dimensional image. An image processing apparatus according to 1.   The image processing apparatus according to claim 1, wherein the measurement area setting unit further sets a measurement area including the contour line on the two-dimensional image based on a conveyance error of the object. An imaging process for obtaining a two-dimensional image by imaging at least one end of an object conveyed within the field of view of the camera ;
The shape extraction step of the resulting et al two-dimensional image to extract a contour line representing the contour of one end portion of the object,
A point position extraction process for extracting positions of a plurality of points on the contour line;
A representative position determining process for determining the position of a representative point representing the position of the contour line in the conveyance direction of the object on the two-dimensional image from the plurality of points;
In the two-dimensional image, a measurement area setting process for setting a measurement area including the outline of the one end part of the vertical width n in the transport direction and the lateral width m in the direction intersecting the transport direction,
In the measurement area setting process, a vertical width of (n / 2) is set from the position of the representative point in the forward and reverse directions of the transport direction.
Image processing method. A semiconductor manufacturing apparatus including an image processing apparatus according to any one of claims 1 to 7.