JP2005069795A - Predetermined part registration method of dimension measuring apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a predetermined part registration method of a dimension measuring apparatus capable of registering a predetermined part of a measuring object by simple operation. <P>SOLUTION: The dimension measuring apparatus photographs the measuring object via an optical microscope, and measures the dimensions of a predetermined part of the measuring object from the obtained video signal. At at least two spots of the predetermined part of the measuring object, measuring windows are set respectively, and a measuring reference window having a predetermined relation with respect to the measuring windows is set. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an improvement in an apparatus for measuring a dimension of an object in a non-contact manner using a two-dimensional sensor such as a TV camera.

  As a minute dimension measuring method for measuring a line width of a minute wiring pattern or the like such as a semiconductor or a liquid crystal display device, a minute dimension measuring apparatus (see, for example, Patent Document 1) is known. As shown in FIG. 7, this minute dimension measuring apparatus projects the surface of an object to be measured 701, for example, a semiconductor element, onto an imaging unit 703 of a TV camera using an optical microscope 702. The projected spatial image of the surface of the object 701 to be measured is picked up by the image pickup unit of the TV camera 703, and the dimensions of the desired part of the object 701 to be measured are electrically measured by the dimension measurement processing unit 704, and the TV monitor 705 is used. There is one that displays an image of the object to be measured 701 and a dimension measurement value. Reference numeral 706 denotes a mouse for operation.

  Here, a method for measuring the line width of the wiring pattern or the like of the device under test 701 using the above-described minute dimension measuring apparatus will be described with reference to FIG. First, FIG. 8A shows a monitor image 801 of the surface of the measurement object 701 taken by the TV camera 703. Reference numeral 802 denotes a horizontal scanning line for scanning the monitor image 801, and reference numeral 803 denotes, for example, a wiring pattern on the surface of the measured object 701. In addition, the wiring pattern 803 has a high light reflectivity and is bright, and the periphery thereof has a low reflectivity and a dark state. The same applies to the following drawings. Li represents the i-th horizontal scanning line, and the luminance distribution on the i-th horizontal scanning line is shown in FIG. That is, FIG. 8B shows luminance-pixel characteristics, the horizontal axis indicates the number of pixels N in the horizontal direction of the monitor image 801, and the vertical axis indicates the luminance level.

  As a measurement principle of the wiring pattern 803, the dimension is obtained from the luminance-pixel characteristics. That is, the maximum luminance level 805 in the luminance distribution curve 804 in FIG. 8B is set to 100%, the minimum luminance level 806 is set to 0%, and the interval between the a-th pixel and the b-th pixel corresponding to the luminance level 807 of 50%. The position difference Nab is obtained. Next, the dimension X of the wiring pattern 803 of the object to be measured corresponding to the positional difference Nab from the measurement magnification of the optical microscope 702 at this time and the coefficient k determined by the object distance from the TV camera 703 to the object to be measured 701 is The following equation (1) is obtained.

X = k x Nab (1)
A method for measuring the wiring pattern 803 of the DUT 701 using such a principle will be described. As shown in FIG. 8C, first, a measurement reference window 808 (with its center coordinate axis being (0, 0)) and a left measurement window 809 (with its center coordinate axis being (Lx, Ly)). ) And the right measurement window 810 (the coordinate axis at the center is (Rx, Ry)) will be described. Note that the window means a rectangular range shown in FIG. The same applies to the following drawings. As shown in FIG. 8C, first, a wiring pattern 803 of an object 701 to be measured that is a known reference is imaged on a TV monitor 705, and a measurement reference window 808 is dragged with a mouse 706 while viewing the image, and measurement is performed. A reference window 808 is selected, and an image within the measurement reference window range is registered as a pattern matching image. Next, the left measurement window 809 and the right measurement window 810 are selected by dragging with the mouse 706 and registered as measurement ranges. Note that the measurement reference window 808, the left measurement window 809, and the right measurement window 810 are registered as predetermined positions with the measurement object that is a known reference.

  Next, the unknown object to be measured 701 is replaced, and the wiring pattern of the unknown object to be measured is imaged on the TV monitor 705 to detect a location that coincides with the registered measurement reference window 808 described above. Since this detection method uses a conventionally well-known pattern matching method, detailed description thereof is omitted. By this pattern matching method, the measurement reference window 808 is detected from the wiring pattern 803 of the unknown object 701 to be measured, and the left measurement window 809 and the right measurement window 810 that have already been registered are determined based on this. Then, the left contour is detected in the left measurement window 809 and the right contour is detected in the right measurement window 810.

  Each positional relationship at this time is registered as follows. That is, the center position of the measurement reference window 808 is set to (0, 0) coordinates, the center position of the left measurement window 809 is set to (Lx, Ly) coordinates, and the center position of the right measurement window 810 is set to (Rx, Ry) coordinates. register. Further, by registering the width of the measurement reference window 808 as KWx and KWy, and then registering the width of the left measurement window 809 as LWx and LWy and the width of the right measurement window 810 as RWx and RWy, the wiring of the unknown object 701 is measured. It is possible to measure the dimension of a predetermined position of the pattern 803.

  Here, the method of measuring dimensions will be described in more detail with reference to FIGS. First, FIG. 9 shows the luminance-pixel characteristics of the wiring pattern 803 of the device under test 701 displayed on the monitor image 801 of the TV monitor 705, and the maximum is within the range of the left measurement window 809 and the right measurement window 810. It is a figure explaining the case where there exists a brightness | luminance and minimum brightness | luminance. In the dimension measurement, as a process of the right measurement window 810, a luminance level of 50% is determined from a luminance level of 100% maximum luminance and 0% minimum luminance between Rx− (RWx / 2) and Rx + (RWx / 2). To do. The pixel position b having the luminance level of 50% is calculated by the dimension measurement calculation processing device 704. Next, as the processing of the left measurement window 809, a luminance level of 50% is determined from the luminance level of 100% maximum luminance and 0% minimum luminance between Lx− (LWx / 2) and Lx + (LWx / 2). Similarly, the pixel position a having a luminance level of 50% is calculated by the dimension measurement calculation processing device 704. A difference between these pixel positions a and b (position difference Nab between pixels) is obtained, and this position difference Nab is a coefficient determined by the measurement magnification of the optical microscope 702 and the object distance from the TV camera 703 to the object 701 to be measured. The predetermined dimension value X of the wiring pattern 803 of the object 701 to be measured can be obtained from the above equation (1) from k.

  10 shows the luminance-pixel characteristics of the wiring pattern 803 of the device under test 701 displayed on the monitor image 801 of the TV monitor 705 as in FIG. 9, but the left measurement window 809 and the right measurement window 810 are shown. If there is no maximum luminance within one of the ranges, for example, in FIG. 10, if there is no maximum luminance within the range of the left measurement window 809, the width LWx of the left measurement window 809 is automatically increased to maximize the luminance. A position 811 is obtained, and a luminance level of 50% is obtained from a luminance level of 100% maximum luminance and 0% minimum luminance.

  Similarly, in FIG. 11, when there is no maximum brightness within the range of either the left measurement window 809 or the right measurement window 810, the widths LWx and RWx of the left measurement window 809 and the right measurement window 810 are automatically expanded to obtain the maximum brightness. A position 811 is obtained, and a luminance level of 50% is obtained from a luminance level of 100% maximum luminance and 0% minimum luminance. By this method, the predetermined dimension value X of the wiring pattern 803 of the device under test 701 can be obtained. In FIGS. 10 and 11, for the sake of explanation, a portion with a low luminance level is indicated by Nab, which is different from FIG.

Japanese Examined Patent Publication No. 6-103168

  In the prior art, the position is registered on the assumption that there is a maximum luminance position or a minimum luminance position in or around the measurement range, and the size of a predetermined part is obtained by calculation. In this case, if the position of either the maximum brightness or the minimum brightness is outside the measurement range, the operator will be wondering where to put the measurement range, and elements that cannot determine the range will lead to instability of the measurement results. . In addition, there is a problem that it takes time to perform operations such as registering the measurement reference window, the left measurement window, and the right measurement window while viewing the image.

  An object of the present invention is to provide a predetermined part registration method for a dimension measuring apparatus capable of registering a predetermined part of an object to be measured with a simple operation.

  The predetermined part registration method of the dimension measuring apparatus of the present invention is the dimension measuring apparatus for imaging the object to be measured through an optical microscope and measuring the dimension of the predetermined part of the object to be measured from the obtained video signal. It is configured by setting measurement windows in at least two predetermined portions of the object to be measured, and setting a measurement reference window having a predetermined relationship with the measurement windows.

  In the predetermined part registration method of the dimension measuring apparatus of the present invention, the measurement window is set based on a differential curve of a luminance distribution characteristic curve of a predetermined part of the measurement object.

  In the predetermined part registration method of the dimension measuring apparatus of the present invention, the measurement reference window is a differential curve of a luminance distribution characteristic curve in a direction different from the luminance distribution characteristic curve of the measurement window and the predetermined part of the object to be measured. Set based on.

  The predetermined part registration method of the dimension measuring apparatus of the present invention automatically recognizes the measurement position, determines the maximum position and the minimum position of the brightness, and eliminates the determination error of the registrant by setting these as the maximum and minimum positions. In addition, registration can be performed simply by designating the left and right measurement positions while viewing the image, thereby shortening the operation.

  Specifically, the left measurement position and the right measurement position are specified while viewing the image, the range of the left measurement window and the right measurement window is determined by image processing, and the measurement reference window for the reference pattern is determined. Thus, the registration work can be performed with two mouse instruction operations, and the operation of dimension measurement can be shortened.

  FIG. 1 is a flowchart for explaining an embodiment of the present invention. An example of the dimension measuring apparatus used in the present invention is shown in FIG. In FIG. 6, the same components as those in FIG. Reference numeral 601 denotes an XY stage on which an object to be measured is mounted, and 602 denotes a drive unit of the XY stage 601. This minute dimension measuring apparatus projects the surface of an object to be measured 701, for example, a semiconductor element, onto an imaging unit of a TV camera 703 with an optical microscope 702. The projected aerial image of the surface of the measured object 701 is picked up by the image pickup unit of the TV camera 703, converted into a video signal, and supplied to the dimension measurement calculation processing device 704. The dimension measurement arithmetic processing unit 704 processes this video signal, electrically calculates the dimension of a desired portion of the object 701 to be measured, and displays an image of the object 701 and a dimension measurement value on the TV monitor 705.

  The predetermined part registration method of the present invention using the above-described minute dimension measuring apparatus will be described with reference to FIG. First, in step 101, the XY stage 601 on which the device under test 701 is mounted is driven so that, for example, the wiring pattern 112 of the semiconductor element for measuring the line width is displayed as the monitor image 111 on the screen of the TV monitor 705. Adjust.

  In step 102, the right position of the wiring pattern 112 is clicked with the mouse 706, and the approximate position of the right measurement window 113 is designated.

  In step 103, an automatic edge detection range is determined. A method for determining the automatic edge detection range will be described with reference to FIG. 2A shows the vicinity of the right measurement window 113 (not shown) and corresponds to the luminance distribution characteristic curve 201 (804 in FIG. 9) of the scanning line AA ′ on the click point P1 of the mouse 706. FIG. ). A curve 202 shows a differential curve of the luminance distribution characteristic curve 201. From this curve, a minimum value a, a point b where the slope is gentle from the minimum value a to the left side, and a point c where the slope is gentle from the minimum value a to the right side are shown. To detect. As a result, the width RWx of the right measurement window 113 is set from the point b to the point c. It should be noted that the points b and c where the inclination becomes gentle can be set as positions having a predetermined inclination, or can be experimentally determined in advance. Further, the width RWy of the right measurement window 113 is appropriately determined in advance.

  In step 104, the left position of the wiring pattern 112 is clicked with the mouse 706, and the approximate position of the left measurement window 114 is designated. At this time, the Y direction is the same as that of the right measurement window 113.

  In step 105, the automatic edge detection range is determined as in step 103. That is, FIG. 2B shows the vicinity of the left measurement window 114 (not shown), and corresponds to the luminance distribution characteristic curve 203 of the scanning line BB ′ on the click point P2 of the mouse 706 (corresponding to 804 in FIG. 9). ). Note that the scanning line B-B ′ is the same scanning line as the scanning line A-A ′. A curve 204 shows a differential curve of the luminance distribution characteristic curve 203. From this curve, a maximum value a, a point b where the slope is gentle from the maximum value a to the left side, and a point c where the slope is gentle from the maximum value a to the right side are shown. To detect. As a result, the width LWx of the right measurement window 114 is set from the point b to the point c. It should be noted that the points b and c where the inclination becomes gentle can be set as positions having a predetermined inclination, or can be experimentally determined in advance. In addition, the width LWy of the left measurement window 114 is appropriately determined in advance.

Step 106 determines an automatic reference position window 115. A method for determining the automatic reference position window 115 will be described with reference to FIG. In FIG. 3, a luminance distribution curve 301 is obtained on a straight line AA ′ in the Y direction passing through point b of the right measurement window 113. To obtain this luminance distribution curve 301,
Among the luminance values of each pixel of each scanning line of the monitor image 111, the luminance value of each pixel located on the straight line AA ′ passing through the point b of the right measurement position is obtained by the dimension measurement arithmetic processing unit 704, and the luminance distribution is obtained. A curve 301 is obtained.

  Next, the luminance distribution curve 301 is differentiated to obtain a derivative curve 302. Then, on the differential curve 302, the displacement points f, d, and e are searched on the peak 303 side from the point R (corresponding to the position where the right position of the wiring pattern 112 is clicked with the mouse 706 in step 102). In order to search for the displacement points f, d, and e, the maximum value d, the point f at which the slope gradually decreases from the maximum value d, and the point e at which the slope gradually decreases from the maximum value d are detected. Note that the points f and e where the inclination becomes gentle can be set as positions having a predetermined inclination, or can be experimentally determined in advance.

  As a result, a range 305 surrounded by straight lines E, F and b in the X direction passing through the points f and e and straight lines AA ′ and BB ′ in the Y direction passing through the point c is set to the automatic reference position window 115. And As a result, the automatic reference position window 115 is determined. If there is no displacement point on the peak 303 side on the differential curve 302, the displacement points g, h, i are searched for on the peak 304 side, and the straight line G in the X direction passing through the g point and i point as described above. An automatic reference position window (measurement reference window) 115 is defined as a range 306 surrounded by straight lines AA ′ and BB ′ in the Y direction passing through points I, b, and c.

  If a displacement point is not found by the above method, the right measurement window 113 is shifted to the right, a straight line AA ′ is placed on a straight line BB ′ passing through the point c, and the displacement point is similarly searched. .

  If there is no displacement point in this way, the right detection window 113 is replaced with the left measurement window 114 and the same detection is performed. As a result, if there is no displacement point, the automatic detection is abandoned and manual setting is performed by dragging the mouse.

  Step 107 is a step for performing correction and registration. FIG. 4A shows a flowchart of correction and registration. The measurement reference window 115, the right measurement window 113, and the left measurement window 114 on the monitor image 111 automatically obtained in steps 101 to 106 described above are checked, and if correction is necessary, the process proceeds from step 401 to step 402. As shown in FIG. 4B, each position is freely set with a mouse and registered in a registration step 403. If correction is not necessary for automatic setting, the process proceeds to registration step 403 as it is. In this registration step 403, the center position of the measurement reference window 115 is set to (0, 0) coordinates, then the center position of the left measurement window 114 is (Lx, Ly), and the center position of the right measurement window 113 is (Rx, Ly). Ry) and register. At the same time, the width of the measurement reference window 115 is registered as KWx and KWy, the width of the left measurement window 114 is registered as LWx and LWy, and the width of the right measurement window 113 is registered as RWx and RWy.

In step 108, for example, line width measurement of the wiring pattern 112 is performed using the registered measurement reference window 115, right measurement window 113, and left measurement window 114 as shown in FIG. That is,
(1) Press the measurement start button with the mouse.
(2) An optical microscope image of a part to be measured, for example, a semiconductor wiring pattern, is displayed in the monitor image 111. From the displayed wiring pattern image, for example, for example, a location having the highest correlation among the wiring patterns in the measurement reference window 115 is automatically recognized by a method such as pattern matching, and the center position of the measurement reference window 115 ( (0,0) coordinates are detected.
(3) As described above, the dimension measurement is performed as a process of the right measurement window 113 between Rx− (RWx / 2) and Rx + (RWx / 2), and the luminance level of the maximum luminance level 100% and the minimum luminance level 0%. To a pixel position b having a luminance level of 50%. Next, a pixel position a having a luminance level of 50% from the luminance level of the maximum luminance level 100% and the minimum luminance level 0% between Lx− (LWx / 2) and Lx + (LWx / 2) is detected, and the dimension Nab Get. As a result, the line width X of the wiring pattern 112 is measured from the equation (1).

  As described above, the present invention has been described in detail. However, the present invention is not limited to the embodiment of the predetermined part registration method of the dimension measuring apparatus described here, and other dimension measuring apparatuses and dimension measuring methods other than those described above. Needless to say, it can be widely applied. Further, although two measurement windows have been described as measurement windows, two or more measurement windows can be set as necessary.

It is a figure which shows the flowchart explaining operation | movement of one Example of this invention. It is a figure for demonstrating the principle of the setting of the measurement window of this invention. It is a figure for demonstrating the principle of the setting of the measurement reference | standard window of this invention. It is a figure for demonstrating correction and registration of this invention. It is the figure which applied this invention to the light-width measurement of a part of to-be-measured object. It is a block diagram which shows one Example of this invention. It is a block diagram which shows an example of the conventional dimension measuring apparatus. It is a figure for demonstrating the basic principle of the conventional dimension measuring apparatus. It is a figure for demonstrating the detailed measurement principle of the conventional dimension measuring apparatus. It is a figure for demonstrating the detailed measurement principle of the conventional dimension measuring apparatus. It is a figure for demonstrating the detailed measurement principle of the conventional dimension measuring apparatus.

Explanation of symbols

  111: monitor image, 112: wiring pattern, 113: right measurement window, 114: left measurement window, 115: measurement reference window, 301: luminance distribution curve, 302: differential curve, 303, 304: peak, 601: XY stage, 602: XY drive unit, 701: object to be measured, 702: optical microscope, 703: TV camera, 704: dimension measurement processing unit, 705: TV monitor, 706: mouse.

Claims (1)

In a dimension measuring apparatus that measures an object to be measured through an optical microscope and measures a dimension of a predetermined part of the object to be measured from an obtained video signal, at least two of the predetermined parts of the object to be measured are respectively provided A method for registering a predetermined part of a dimension measuring apparatus, comprising: setting a measurement window and setting a measurement reference window having a predetermined relationship with the measurement window.

Images