JP5236615B2 - Edge detection method, length measurement method, charged particle beam device - Google Patents

Description

  The present invention relates to a method for detecting an edge portion of a semiconductor pattern, a method for measuring a distance between semiconductor patterns, and a charged particle beam apparatus that executes these methods.

  When inspecting a semiconductor formation pattern of a semiconductor device, the semiconductor device is irradiated with charged particles to obtain an observation image, and the inspection is performed by analyzing the image. One of the inspection items of the semiconductor pattern is an inspection for measuring an interval between the semiconductor patterns. Patent Document 1 below describes a technique for measuring a length using a charged particle beam.

JP 2008-215824 A

  When measuring the interval between semiconductor patterns, the measurer determines the area to be measured, scans the area and detects the luminance change of the observation image, detects the edge portion of the semiconductor pattern, and the edge The interval is measured using the part. At this time, if the size of the pattern on the observation image and the size of the measurement target area are greatly different from each other, it is difficult to detect many miscellaneous signals such as noise on the scanning line and identify the edge portion of the pattern.

  For example, when the pattern shape largely changes in one wafer like a FEM (Focus Exposure Matrix) wafer, a situation occurs in which the semiconductor pattern width is too small with respect to the size of the measurement target region. In this case, the above situation is likely to occur. In addition, since the test semiconductor pattern used for the OPC (Optics Proximity Collection) design includes many line end patterns, the above situation is likely to occur.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a technique for dynamically setting the size of a measurement target region and detecting an edge portion of a semiconductor pattern with high accuracy. To do.

  In the edge portion detection method according to the present invention, a second image portion that is orthogonal to the edge portion to be detected is detected by scanning a larger image area including the edge portion of the semiconductor pattern in the observed image. The scanning range is set inside the second edge portion.

  According to the edge portion detection method of the present invention, the scan range size for detecting the edge portion can be set to a size slightly smaller than the width of the edge portion. As a result, it is difficult to generate a situation in which the pattern size and the scanning range size greatly deviate, and the edge portion detection accuracy can be improved.

It is a figure which shows the observation image 100 of the semiconductor pattern 110. FIG. It is a figure which shows a mode that the image brightness | luminance in the length measurement cursor 120 is scanned. It is a figure which shows the example whose size of a length measurement cursor is significantly larger than the size of the edge part of the semiconductor pattern 110. FIG. It is a figure which shows a mode that the image brightness | luminance in the length measurement cursor 120 of FIG. 3 is scanned. It is a figure which shows a mode that the FOV area | region containing an edge part was set. It is a figure which shows a mode that the inside of FOV is scanned. It is a figure which shows a mode that the inside of FOV is scanned in the reverse direction. It is a figure which shows a mode that the detection result of step 2-step 3 was plotted on the graph. It is a figure which shows a mode that the y coordinate and size of the length measurement cursor 120 are defined. It is a figure which shows a mode that the x coordinate and size of the length measurement cursor 120 were defined. It is a figure which shows a mode that the shape of the edge part detected in step 4 was approximated with the polynomial. It is a figure which shows a mode that the y coordinate and size of the length measurement cursor 120 were defined. It is a figure which shows the example whose x coordinate and size of the length measurement cursor 120 are improper. It is a figure which shows a mode that the x coordinate and size of the length measurement cursor 120 are defined. It is a figure explaining another method which detects an edge part in Step 6 of Embodiment 3. FIG. It is a figure which shows a mode that the two opposing length measurement cursors 120 have overlapped. It is a figure which shows the mode after adjusting the position and size of the length measurement cursor 120 again. It is a figure which shows a mode that the length measurement cursor 120 protrudes out of the observation image 100. FIG. It is a figure which shows the mode after adjusting the position and size of the length measurement cursor 120 again. FIG. 10 is a schematic diagram showing a configuration of a charged particle beam apparatus 200 according to an eighth embodiment.

  In the following, first, a general method for detecting an edge portion of a semiconductor line pattern will be described, and a situation in which the detection accuracy of the edge portion is reduced will be described. Then, the edge part detection method according to the present invention will be described.

  FIG. 1 is a view showing an observation image 100 of the semiconductor pattern 110. Here, an example in which the edge portions of two semiconductor line patterns are arranged to face each other is shown. In FIG. 1, an observation image 100 is obtained, for example, by irradiating a semiconductor device with a charged particle beam. An SEM (Scanning Electron Microscope) observation image corresponds to this. Any method may be used for obtaining the observation image, and an observation image other than the SEM observation image may be used.

  In FIG. 1, the distance between the semiconductor patterns 110 is a measurement target. In order to measure the interval between the semiconductor patterns 110, it is necessary to detect an edge portion where the semiconductor patterns 110 face each other. Therefore, the length measurement cursor 120 is placed in the vicinity of the edge portion on the observation image 100, and luminance scanning is performed.

  FIG. 2 is a diagram illustrating a state in which the image brightness in the length measurement cursor 120 is scanned. When the image luminance is acquired while changing the scanning position along the edge portion of the semiconductor pattern 110, the image luminance greatly changes at the edge portion. Based on this luminance change, the position of the edge portion can be detected.

  FIG. 3 is a diagram illustrating an example in which the size of the length measurement cursor is significantly larger than the size of the edge portion of the semiconductor pattern 110. In FIG. 3, the semiconductor pattern 110 has a narrower shape than that in FIG. 1, and the width of the edge portion is significantly narrower than the width of the length measurement cursor.

  FIG. 4 is a diagram illustrating a state in which the image luminance in the length measurement cursor 120 in FIG. 3 is scanned. When the width of the length measurement cursor 120 is significantly wider than the width of the edge portion of the semiconductor pattern 110 as shown in FIG. 3, there are many scanning lines that do not intersect the edge portion. In this case, a scanning line that does not intersect the edge portion detects a lot of noise on the observation image 100, and it becomes difficult to distinguish between the luminance of the edge portion and the luminance of the noise. Therefore, the detection accuracy of the edge part tends to be lowered.

  In order to prevent detection of noise as much as possible, it is desirable that the number of scanning lines that do not intersect the edge portion be as small as possible.

  For the reason described above, in order to accurately detect the edge portion of the semiconductor pattern 110, the size of the length measuring cursor 120 in the edge width direction (vertical direction in FIG. 1) may be set slightly smaller than the edge width. This is desirable.

<Embodiment 1>
Hereinafter, a method of measuring the length by setting the size of the length measurement cursor 120 in the edge width direction to a desirable size will be described in the following (Step 1) to (Step 6).

(Step 1: FOV setting)
FIG. 5 is a diagram illustrating a state in which an FOV (Field Of View) region including an edge portion is set. The FOV is an area set on the observation image 100 in order to measure the width of the edge portion of the semiconductor pattern 110. The FOV corresponds to “an image area including an edge portion” in the present invention. The FOV is set to a size and position that includes the edge portion of the semiconductor pattern 110 inside. The size of the FOV is set larger than the length measurement cursor 120. As a method for setting the size and position of the FOV, for example, the following can be considered.

(Step 1: FOV setting: Example 1)
While the measurer visually confirms the observation image 100 on the screen, the FOV is manually set at a suitable position and size including the edge portion of the semiconductor pattern 110.

(Step 1: FOV setting: Example 2)
An arithmetic unit or an external computer included in the charged particle beam apparatus (only the arithmetic unit will be described for convenience in the following description) acquires design data of the semiconductor pattern 110 in advance. The arithmetic device sets the edge portion defined on the design data at the center of the FOV, and the size of the FOV is sufficiently larger than the width on the design data of the semiconductor pattern 110 (for example, twice the width). .

(Step 2: Scan the FOV)
FIG. 6 is a diagram illustrating a state in which scanning is performed in the FOV. The arithmetic device sequentially scans the FOV set in step 1 in the coordinate axis direction (y-axis direction in FIG. 6) parallel to the edge portion of the semiconductor pattern 110. The scanning position in the x-axis direction is sequentially changed from the smaller x coordinate to the larger x coordinate. The scanning in the y-axis direction is from the larger y coordinate to the smaller one. The arithmetic device acquires an image luminance value on each scanning line.

(Step 3: Scan the FOV in the reverse direction)
FIG. 7 is a diagram illustrating a state where the FOV is scanned in the reverse direction. The arithmetic unit scans the FOV in the direction opposite to that in step 2. The scanning position in the x-axis direction is changed sequentially from the largest x coordinate to the smallest x coordinate. Scanning in the y-axis direction is from the smaller y coordinate to the larger y coordinate. The arithmetic device acquires an image luminance value on each scanning line.

(Step 4: Detect an edge part along the x-axis)
FIG. 8 is a diagram illustrating a state in which the detection results of Step 2 to Step 3 are plotted on a graph. The left half of FIG. 8 shows the detection result of Step 2, and the right half of FIG. The arithmetic unit plots the position where the peak of the luminance value is first detected on the scanning line on a graph as shown in FIG. It is not necessary to actually display a graph on the screen, and processing equivalent to plotting may be executed internally. The same applies to the subsequent steps and embodiments. The arithmetic unit determines that an edge portion along the x-axis exists in a portion where the plotted points are continuous.

(Step 4: Detect edge part along x-axis: Supplement)
Since the arithmetic unit plots the position where the peak of the luminance value is first detected on the scanning line as shown in FIG. 8, the noise is detected and plotted in the portion where the semiconductor pattern 110 does not exist. Since noise is present throughout the FOV, in a portion where the semiconductor pattern 110 does not exist, a peak is detected immediately after scanning is started. Therefore, the position where the peak is detected differs greatly before and after the scanning position in the x-axis direction passes through the edge portion along the y-axis. The arithmetic unit can detect the edge portion along the x-axis of the semiconductor pattern 110 by utilizing this fact.

(Step 5: Determine the y coordinate and size of the length measurement cursor 120)
FIG. 9 is a diagram illustrating how the y coordinate and size of the length measurement cursor 120 are determined. The arithmetic unit sets a position slightly inside the edge portion along the x-axis detected in step 4 for each edge portion. The arithmetic unit sets the center of the y coordinate of the two positions as the center y coordinate of the length measurement cursor 120. The interval between the two positions in the y-axis direction is the y-axis direction size of the length measurement cursor 120.

(Step 5: Determine y-coordinate and size of measuring cursor 120: Supplement)
The inside in this step means a direction closer to the inside of the semiconductor pattern 110 than the edge portion along the x axis. By setting the y-axis direction size of the length measurement cursor 120 with reference to the position inside the edge portion, the length measurement cursor 120 having a width slightly smaller than the width of the semiconductor pattern 110 in the y-axis direction can be generated. . In other words, the size of the length measurement cursor 120 in the y-axis direction and the width of the semiconductor pattern in the y-axis direction can be prevented from greatly differing.

(Step 6: Determine the x coordinate and size of the length measurement cursor 120)
FIG. 10 is a diagram illustrating a state in which the x coordinate and the size of the length measurement cursor 120 are determined. The arithmetic device determines the x coordinate and size of the length measurement cursor 120 based on the design data. For example, the x coordinate of the edge portion along the y-axis direction of the semiconductor pattern 110 on the design data is set as the center x coordinate of the length measurement cursor 120. The size in the x-axis direction is a constant multiple of the size y1 in the y-axis direction determined in step 5, for example, 2y1 which is twice as large.

  As described above, in the first embodiment, the method in which the arithmetic device automatically sets the position and size of the length measurement cursor 120 has been described.

  As described above, according to the first embodiment, the arithmetic device sets the FOV region including the edge portion along the y axis of the semiconductor pattern 110 on the observation image 100, and the inside of the FOV is along the y axis. And scanning to detect an edge portion along the x-axis. Based on the y-coordinate of the edge portion along the x-axis, the size of the length measurement cursor 120 in the y-axis direction can be automatically set to a width slightly smaller than the edge width along the y-axis. Therefore, the size of the length measurement cursor 120 in the y-axis direction and the edge width of the semiconductor pattern 110 in the y-axis direction do not greatly deviate, and the edge portion can be detected with high accuracy.

<Embodiment 2>
In the second embodiment of the present invention, a modification of step 5 described in the first embodiment will be described. Since other steps are the same as those in the first embodiment, only step 5 will be described below. In the second embodiment, step 5 is executed in two stages, step 5-1 and step 5-2.

(Step 5-1: Approximate the shape of the edge portion along the x-axis)
FIG. 11 is a diagram illustrating a state in which the shape of the edge portion along the x-axis detected in step 4 is approximated by a polynomial. The approximate expression used here is a polynomial having a shape in which two or more extreme values exist above and below the edge portion along the x-axis, and these extreme values sandwich the edge portion along the x-axis. Use. When two extreme values sandwich the edge along the x-axis, there are two extreme values in the left half and two in the right half of FIG. 11, so use a fifth order polynomial. Can do. The reason why the edge portion is sandwiched between extreme values will be described with reference to FIG.

(Step 5-2: Determine y-coordinate and size of measuring cursor 120)
FIG. 12 is a diagram illustrating a state in which the y coordinate and the size of the length measurement cursor 120 are determined. The arithmetic unit specifies two extreme points inside the edge portion along the x-axis direction of the semiconductor pattern 110 in the approximate expression determined in Step 5-1. When there are three or more corresponding extreme values, one extreme value corresponding to each of the edge portion detected in step 2 and the edge portion detected in step 3 is arbitrarily selected. The arithmetic unit sets the center of the y coordinates of the two extreme points as the center y coordinate of the length measurement cursor 120. The interval between the two extreme points in the y-axis direction is the size of the length measurement cursor 120 in the y-axis direction.

  As described above, according to the second embodiment, the same effect as in the first embodiment can be exhibited by approximating the edge portion along the x-axis of the semiconductor pattern with a polynomial.

<Embodiment 3>
In step 6 of the first and second embodiments, it has been described that the x coordinate of the length measurement cursor 120 is determined based on the design data. In the third embodiment of the present invention, the procedure for determining the x coordinate of the length measurement cursor 120 based on the scanning result in the FOV in step 6 will be described. Since steps other than step 6 are the same as those in the first and second embodiments, only step 6 will be described below.

  FIG. 13 is a diagram illustrating an example in which the x coordinate and the size of the length measurement cursor 120 are inappropriate. In FIG. 13, the length measurement cursor 120 enters inside the edge portion along the y-axis direction of the semiconductor pattern 110, and the edge portion cannot be scanned within the length measurement cursor 120.

  In order to avoid the situation shown in FIG. 13, it is necessary to appropriately set the x coordinate and the size of the length measurement cursor 120. However, when the x coordinate and the size of the length measurement cursor 120 are determined based on the design data, there is a possibility that the design data and the actual semiconductor pattern 110 are greatly deviated. In this case, the position of the length measurement cursor 120 deviates from the assumption, and the situation shown in FIG. 13 may occur. Therefore, in the third embodiment, the actual observation image 100 is used to determine the x coordinate and size of the length measurement cursor 120. Hereinafter, the content of step 6 in this Embodiment 3 is demonstrated.

(Step 6: Determine the x coordinate and size of the length measurement cursor 120)
FIG. 14 is a diagram showing how the x coordinate and the size of the length measurement cursor 120 are determined. As described with reference to FIG. 8 of the first embodiment, since the noise is picked up in the portion where the semiconductor pattern 110 on the scanning line does not exist, the position where the luminance peak is detected is greatly separated from the edge portion along the x axis. Therefore, when the luminance peak is plotted, the edge portion and the noise portion are discontinuous. The arithmetic unit can detect the x coordinate of the edge portion along the y-axis of the semiconductor pattern 110 by detecting the discontinuous portion. The arithmetic unit sets the x coordinate of the edge portion as the center x coordinate of the length measurement cursor 120. The size of the length measurement cursor 120 in the x-axis direction may be determined in the same manner as in the first and second embodiments.

(Step 6: Determine x-coordinate and size of measuring cursor 120: Supplement)
The process of determining the x coordinate of the edge portion may be performed only for either the edge portion along the x axis detected in step 2 or the edge portion along the x axis detected in step 3. In other words, it is only necessary to execute for either the left or right half of the plot result shown in FIG.

  As described above, according to the third embodiment, the x coordinate and the size of the length measurement cursor 120 can be determined based on the result of scanning the FOV on the observation image 100. That is, the x coordinate and the size of the length measurement cursor 120 are dynamically determined based on the actual measurement result. Therefore, even if the design data and the actual semiconductor pattern 110 are deviated, the x coordinate and the size of the length measuring cursor 120 can be appropriately determined, and the edge portion can be included in the length measuring cursor 120.

<Embodiment 4>
FIG. 15 is a diagram for explaining another method for detecting an edge portion in step 6 of the third embodiment. In FIG. 14, it has been described that when the luminance peak on the scanning line is plotted, the edge portion and the noise portion along the x-axis are discontinuous. Using this property, the edge portion can be detected statistically. The contents of step 6 in the fourth embodiment will be described below. Other steps are the same as those in the third embodiment.

(Step 6: Determine the x coordinate and size of the length measurement cursor 120)
The arithmetic unit obtains the y coordinate value of the luminance peak obtained by scanning the FOV, and plots the average value and the variance value as shown in FIG. An extreme value is obtained at a position corresponding to the x coordinate of the edge portion along the y axis due to the discontinuity of the luminance peaks of the edge portion and the noise portion along the x axis. The arithmetic unit can specify the x coordinate of the edge portion along the y axis by specifying the extreme value. The arithmetic unit sets the x coordinate of the edge portion as the center x coordinate of the length measurement cursor 120. The size of the length measurement cursor 120 in the x-axis direction may be determined in the same manner as in the first and second embodiments.

  As described above, according to the fourth embodiment, the x-coordinate of the edge portion along the y-axis is obtained by utilizing the property that the luminance peak on the scanning line is discontinuous between the edge portion along the x-axis and the noise portion. It can be detected statistically. Thereby, the same effect as Embodiment 3 can be exhibited.

<Embodiment 5>
In the fifth embodiment of the present invention, a processing procedure when the position and size of the length measurement cursor 120 are set and then overlapped with another length measurement cursor 120 will be described.

  FIG. 16 is a diagram illustrating a state in which two opposing length measurement cursors 120 overlap each other. In FIG. 16, each length measuring cursor 120 includes an edge portion of the semiconductor pattern 110 and is set to an appropriate position and size, but overlaps with another length measuring cursor 120 arranged oppositely.

  FIG. 17 is a diagram illustrating a state after the position and size of the length measurement cursor 120 are readjusted. The arithmetic device can determine whether or not the length measurement cursors 120 overlap each other based on the position and size of each length measurement cursor 120. When any length measuring cursor 120 overlaps, the arithmetic unit moves the position of each overlapping length measuring cursor 120 around the center of the overlapped portion to eliminate the overlapping portion. At this time, the size of the length measurement cursor 120 may be reset as appropriate.

  As described above, according to the fifth embodiment, the arithmetic device adjusts the position and size of the length measurement cursor 120 so that there is no portion where the length measurement cursor 120 overlaps. As a result, inconveniences such as detecting edge portions of other semiconductor patterns 110 that should not be detected can be avoided.

<Embodiment 6>
In the fifth embodiment of the present invention, a processing procedure when the length measurement cursor 120 protrudes out of the observation image 100 after setting the position and size of the length measurement cursor 120 will be described.

  FIG. 18 is a diagram illustrating a state in which the length measurement cursor 120 protrudes outside the observation image 100. In FIG. 18, each length measurement cursor 120 includes the edge portion of the semiconductor pattern 110 and is set to an appropriate position and size. However, since the size is larger than the size of the observation image 100, the observation image 100 is displayed. It protrudes to the outside.

  FIG. 19 is a diagram illustrating a state after the position and size of the length measurement cursor 120 are readjusted. The arithmetic device can determine whether or not each length measurement cursor 120 protrudes outside the observation image 100 based on the position and size of each length measurement cursor 120. When any one of the length measurement cursors 120 protrudes outside the observation image 100, the arithmetic device readjusts the size and position of each length measurement cursor 120 to eliminate the protruding portion.

  As described above, according to the sixth embodiment, the arithmetic device adjusts the position and size of the length measurement cursor 120 so that there is no portion where the length measurement cursor 120 protrudes outside the observation image 100. As a result, it is not necessary to perform unnecessary processing such as performing scanning processing in a region where the semiconductor pattern 110 does not exist on the observation image 100, and measurement errors caused by scanning outside the observation image 100 can be avoided. it can.

<Embodiment 7>
By the method described in the first to sixth embodiments, the arithmetic unit can detect the edge portion of the semiconductor pattern 110 with high accuracy. After detecting the edge portion, the arithmetic unit can also perform other measurements using the detection result. For example, a length measurement process for measuring an interval between the semiconductor patterns 110 facing each other can be executed.

<Eighth embodiment>
FIG. 20 is a schematic diagram showing a configuration of a charged particle beam apparatus 200 according to Embodiment 8 of the present invention. The charged particle beam device 200 according to the eighth embodiment includes an arithmetic device 270. The arithmetic device 270 controls the overall operation of the charged particle beam device 200. The arithmetic device 270 is at least one of the method for detecting the edge portion of the semiconductor pattern 110 described in any one of the first to sixth embodiments and the method for measuring the interval between the semiconductor patterns 110 described in the seventh embodiment. Do something. Hereinafter, each configuration of FIG. 20 will be described.

  The charged particle beam 212 emitted from the emitter 211 is focused by the electrostatic lenses 221 and 222 and irradiated onto the sample 213 on the sample stage 214. The irradiation position of the charged particle beam 212 on the sample 213 is adjusted by deflecting the charged particle beam 212 with the deflectors 231 and 232.

  The secondary electrons 215 generated from the sample 213 are detected by the secondary electron detector 240, and a secondary electron observation image in which the signal intensity corresponds to the deflection intensity is formed by the arithmetic unit 270. The user can confirm the measurement result on the screen while viewing the secondary electron observation image output from the arithmetic device 270 on a display device such as a display.

  The lens system 220 including the electrostatic lenses 221 and 222, the beam limiting aperture 223, and the aligner 224 is controlled by a lens system controller 250. The deflection system 230 including the deflectors 231 and 232 is controlled by the deflection system controller 260. In addition, the code | symbol of the box showing the driver of each part is abbreviate | omitted.

  The “charged particle irradiation unit” in the eighth embodiment corresponds to the emitter 211. The “image acquisition unit” corresponds to the secondary electron detector 240 and the arithmetic device 270.

  As described above, according to the eighth embodiment, the method for detecting the edge portion of the semiconductor pattern 110 described in any of the first to sixth embodiments and the semiconductor pattern 110 described in the seventh embodiment. The method for measuring the interval can be executed using the arithmetic device 270 provided in the charged particle beam device 200. Note that an external computer or the like may be used instead of the arithmetic unit 270.

  100: Observation image, 110: Semiconductor pattern, 120: Measuring cursor, 200: Charged particle beam device, 211: Emitter, 212: Charged particle beam, 213: Sample, 214: Sample stage, 215: Secondary electrons, 220: Lens system, 221 and 222: electrostatic lens, 223: beam limiting aperture, 224: aligner, 230: deflection system, 231 and 232: deflector, 240: secondary electron detector, 250: lens system controller, 260: Deflection system controller, 270: arithmetic unit.

Claims (8)

A method for detecting an edge portion of a semiconductor pattern of a semiconductor device,
Obtaining an observation image of the semiconductor device;
Identifying an image region that includes the edge portion of the observed image;
An orthogonal edge detection step of scanning the image region in a coordinate axis direction parallel to the edge portion on the observation image and detecting a second edge portion existing in a coordinate axis direction orthogonal to the edge portion of the semiconductor pattern;
A scanning range setting step of setting an end portion in a coordinate axis direction parallel to the edge portion of the scanning range for detecting the edge portion closer to the semiconductor pattern than the second edge portion;
Scanning the scanning range to detect the edge portion;
An edge portion detection method characterized by comprising: In the scanning range setting step,
Approximating the shape of the second edge portion using a polynomial having extreme values closer to the semiconductor pattern than the second edge portion,
The edge portion detection method according to claim 1, wherein a position corresponding to the extreme value is set as an end portion in a coordinate axis direction parallel to the edge portion of the scanning range. In the scanning range setting step,
The end in the coordinate axis direction orthogonal to the edge portion of the scanning range is set so that a portion where the detection result of the second edge portion is discontinuous is included in the scanning range. Item 2. The edge portion detection method according to Item 1. In the scanning range setting step,
A portion where the detection result of the second edge portion is discontinuous is set as a central coordinate in a coordinate axis direction orthogonal to the edge portion in the scanning range,
4. The edge portion according to claim 3, wherein the size in the coordinate axis direction orthogonal to the edge portion in the scanning range is set to a constant multiple of the size in the coordinate axis direction parallel to the edge portion in the scanning range. Detection method. In the scanning range setting step,
The edge part detection method according to claim 3, wherein, when the scanning range overlaps with a scanning range for detecting another edge part, the position or size of the scanning range is adjusted to avoid the overlap. . In the scanning range setting step,
The edge part detection method according to claim 3, wherein when the scanning range protrudes outside the observation image, the position or size of the scanning range is adjusted to avoid the protrusion. A method for measuring an interval between a plurality of semiconductor patterns of a semiconductor device,
Detecting edge portions of the plurality of semiconductor patterns facing each other using the edge portion detection method according to claim 1;
Measuring a distance between the plurality of semiconductor patterns using a detection result of the edge portion;
A length measuring method characterized by comprising: A charged particle irradiation unit that irradiates the measurement target with charged particles;
An image acquisition unit for acquiring an observation image of the measurement object;
An arithmetic unit for performing at least one of the edge portions detection method and claim 7 measurement method according to claim 1, wherein using the observation image,
A charged particle beam apparatus comprising:

Images