JP2001336916A - Method for measuring edge position - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for measuring an edge position capable of highly precisely measuring the edge position of a photographed object without damaging reproducibility even when a part of the photographed object is out of focus in a method for measuring the edge position of the photographed object from an image thereof taken by a television camera mounted to a microscope. SOLUTION: The image of the photographed object is filmed by the television camera at a predetermined sampling interval with moving the microscope in a perpendicular direction to an installing face of the photographed object at a predetermined moving speed. Differential image is calculated for an edge candidate point preliminary set for the image photographed and a region near the point, and each differential value is extracted for each pixel in the same position in the differential image. A peak value of the differential values is calculated for each pixel by performing a specified function approximation on the differential values, and a peak value image in a range of the differential image is obtained by composing the peak values of the differential values. The edge position of the photographed object is measured based on the peak value image.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a system for digitally processing an image using a computer or the like, and more particularly to an edge position measuring method capable of obtaining an edge position of an object with high accuracy.

[0002]

2. Description of the Related Art Various methods have been conventionally used for so-called sub-pixel processing for detecting an edge position of an object to be imaged with an accuracy equal to or smaller than a pixel. Specifically, 1
There are a method of obtaining a mode value of the second derivative, a method of obtaining a peak position of the first derivative by Gaussian approximation, a method of obtaining a zero cross point of the second derivative, and a method of obtaining a peak of a correlation value of template matching. In addition,
These methods are described in, for example, "Image Processing Industry Application Directory" (1994, Fuji Techno System, 10-13).
Pp.). All of these methods are based on the premise that the entire object to be imaged is photographed in a focused state, and there are relatively few problems as far as the method is concerned.

[0003]

However, in the above-described method, it is difficult to accurately calculate the edge position when the whole object is not at the in-focus position. The case where the entire object to be imaged is not at the in-focus position is, for example, a case where the object to be imaged W is installed inclined with respect to the optical axis of the microscope 1 as shown in FIG. In order to cope with this problem, it is common practice to always keep the objective lens and the object to be in focus by means such as auto-focusing. However, even in this method, when the lens magnification becomes high, In some cases, the edge or a part near the edge may not enter the focus position. In addition, it is difficult to accurately adjust the position between the object to be imaged and the objective lens to the in-focus position due to a problem such as mechanical rigidity.

The present invention provides an edge position measuring method capable of measuring an edge position of an object with high accuracy without impairing reproducibility even when a part of the object does not enter a focus position. The purpose is to provide.

[0005]

In order to achieve the above object, the present invention provides an edge position measuring method described below. That is, in the method of measuring the edge position from the image of the object captured by the television camera mounted on the microscope, first, the microscope is moved at a predetermined moving speed in a direction perpendicular to the installation surface of the object. While moving the camera, an image of the object to be picked up is taken by a television camera at every predetermined sampling interval, and a preset edge candidate point for a captured image obtained at each sampling interval and a differential image for a nearby area are set. Is calculated. At this stage, a plurality of differential images are calculated for each sampling interval, that is, for the distance between the microscope and the object.

Next, with respect to a plurality of differential images obtained corresponding to the distance between the microscope and the object, differential values are extracted for each pixel at the same position. Then, an approximate curve is calculated by applying a predetermined function approximation to the differential value group extracted for each pixel, and the peak value of the approximate curve is used as the peak value of the differential value for each pixel. By combining the peak values of the differential values obtained for each pixel, a peak value image in the area of the differential image is obtained. The peak value image obtained here is a differential image obtained assuming that each pixel is located at the in-focus position. Finally, based on this peak value image (for example, by performing sub-pixel processing on the peak value image),
The edge position of the object is measured.

According to the above-described method, there is a possibility that some of the pixels of the plurality of differential images obtained halfway may be at the in-focus position and some may not be at the in-focus position. In this case, an accurate edge position cannot be measured for a region including a pixel that is not at the in-focus position or a pixel in the vicinity thereof. However, regarding the differential value data included in a plurality of differential images obtained corresponding to the distance between the microscope and the object to be imaged, the differential value data is re-divided for each pixel at the same position, If the peak value is obtained for the differential value group re-sorted for each pixel, the peak value can be obtained regardless of whether the pixel is at the in-focus position. The peak value image obtained by synthesizing the peak values does not depend on the presence or absence of focusing. Therefore, if the edge position of the object is calculated based on the peak value image, the edge position of the object can be accurately determined even when a part of the object does not enter the focus position. It can be measured.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to the drawings. FIG. 2 is a schematic diagram of an apparatus for measuring an edge position of an object using an image processing technique for implementing an edge position detection method according to the present invention.
FIG. 3 shows an example of an edge of an object to be imaged and an image near the edge. 4A and 4B show the density values and the differential values of the image on aa 'in FIG. 3, respectively. FIG. 5 shows an example of the differential value of the image on the line aa ′ obtained when the distance between the microscope and the object to be imaged is changed, and FIG. This is a distribution of differential values, and FIG.
9 is a graph showing a relationship between a distance between a microscope and an object at an arbitrary pixel on a line and a differential value.

In FIG. 2 showing a schematic diagram of the edge position measuring device, 1 is a microscope for observing an object to be imaged, 2 is a television camera provided in the microscope 1, and 3 is movable to hold the microscope 1 Stage 4 is stage 3
5 is an encoder for detecting the position of the microscope 1, 6 is a frame memory connected to the television camera 2, 7 is connected to the frame memory 6, the drive motor 4 and the encoder 5, respectively. 8 is a monitor television connected to the frame memory 6, 9 is a work memory for storing data connected to the arithmetic and control unit 7, and W is an installation surface 10a of the measuring table 10.
Each of the imaging objects placed on the top is shown. In addition, since the stage 3 can be moved in the vertical direction with respect to the installation surface 10a of the measuring table 10, the stage 3
The microscope 1 held by the microscope gradually changes the image to be captured according to the movement of the stage 3. Further, the encoder 5 directly detects the position of the stage 3, but indirectly detects the position of the microscope 1 because the microscope 1 is held on the stage 3. .

Next, the procedure will be described with reference to FIG. 1 which is a flowchart showing the procedure of the edge position measuring method according to the present embodiment. Upon receiving an operation command from the outside, the microscope 1 moves to a predetermined search start position (step 11), so that the measurement of the edge position starts. Specifically, when an operation command from the outside is input to the arithmetic and control unit 7, the arithmetic and control unit 7 outputs a signal for rotating the drive motor 4 in a predetermined forward or reverse direction, and receives this signal. As a result, the drive motor 4 rotates, whereby the stage 3 operates, and the microscope 1 provided on the stage 3 moves to the search start position.

When the microscope 1 reaches the search start position,
As the drive motor 4 rotates, the microscope 1 starts moving toward the search end position at a predetermined moving speed (step 12). Then, the processing of steps 13 to 15 described below is performed at every preset sampling interval while the microscope 1 is moving from the search start position to the search end position. That is, the imaging object W is photographed by the television camera 2 at each sampling interval (step 13), and a region near a pixel set as an edge candidate point from the photographed image (for example, a rectangular region shown in FIG. 3). Is extracted (step 14), and a differentiated image is obtained by performing differential processing on the extracted image (step 15), and this is stored in the frame memory 6. Simultaneously with the storage of the differential image data in the frame memory 6,
The detection value of the encoder 5 is stored in the work memory 9 as the distance between the object W and the microscope 1.

After all, in the series of processing of steps 13 to 15, a plurality of data relating to the distance between the microscope 1 and the object W and the data of the differential image corresponding to each of the plurality of distance data Is obtained. In addition, regarding the setting of the edge candidate point, the edge candidate point may be a pixel near the edge position finally calculated. This may be dealt with by confirming the position of the object W and manually inputting the position (coordinates) of the edge candidate point. , The edge position measuring device may automatically set the position (coordinates) of the edge candidate point.

If the microscope 1 has reached the search end position, the process proceeds to step 17 (step 16Y). If the microscope 1 has not yet reached the search end position, the processes of steps 13 to 15 are repeated. (Step 1
6N).

Next, from the data of a plurality of differential images at each sampling interval obtained by the processing of the above-described steps 13 to 16, first, for one pixel, for each differential image obtained at each sampling interval, the microscope 1 Data relating to the distance between the object and the object W and a differential value corresponding to the distance data are extracted (steps 17 and 18). A plurality of data relating to the distance between the microscope 1 and the object W at one pixel and the differential values corresponding to each of the plurality of distance data are approximated by a quadratic function. The peak value, that is, the differential peak value is calculated, and the distance between the microscope 1 and the object W corresponding to the differential peak value and the differential peak value is calculated from the approximate curve, and these are stored (step 19). The processing of steps 17 to 19 is performed for all the pixels in the extracted image (step 2).
0), a differential peak value image is obtained (step 21). Finally, sub-pixel processing is performed on the differential peak value image based on the differential value to calculate an edge position (step 22).

Here, step 1 described above with reference to FIG.
The processing of 7 to 22 will be specifically described. For example, it is assumed that a differential value at an arbitrary pixel in a rectangular area shown in FIG. 3 at each sampling interval is extracted, and the function is approximated to calculate a differential peak value at the pixel. In FIG. 5 showing an example of the differential value of the image on the line aa ′ in FIG. 3 obtained when the distance between the microscope 1 and the object W is changed, FIG. Is the differential value distribution obtained in the above, but for each differential value -1, -2, -3, and -4 at the pixel,
(B) As shown in the figure, the peak value is obtained by approximating these with a quadratic function, and that value is used as the differential peak value in the pixel. This is performed for all the pixels in the rectangular area shown in FIG. 3 to obtain a differential peak value image. From the differential peak value image, a process of calculating an edge position using sub-pixel processing based on the differential value is performed.

The embodiment of the present invention has been described above. In the present embodiment, it is necessary to increase the calculation accuracy of the differential peak value in order to increase the calculation accuracy of the edge position. In order to improve the calculation accuracy of the differential peak value, the number of differential images acquired in the processing of steps 13 to 16 may be increased, which slows down the moving speed of the microscope 1 moving from the search start position to the search end position. Or shorten the sampling interval. However, it is necessary to consider that if the moving speed of the microscope 1 is reduced, the measurement time becomes longer, while if the sampling interval is shortened, a high-speed arithmetic unit and a large memory are required to process a large amount of data. There is. Therefore, it is necessary to set the moving speed and the sampling interval of the microscope 1 in consideration of the required edge position calculation accuracy, the measurement time, and the performance of the arithmetic device.

[0017]

As described above, according to the edge position measuring method of the present invention, the differential value data contained in a plurality of differential images obtained corresponding to the distance between the microscope and the object to be imaged is obtained by the differential operation. It is obtained by rearranging the value data for each pixel at the same position, obtaining the peak value of the differential value group re-classified for each pixel, and synthesizing the peak value for each pixel. Since the edge position of the object is calculated based on the peak value image, the edge position of the object can be measured with high accuracy even when a part of the object does not enter the focus position. I can do it. In addition, since the edge position can be measured irrespective of the position of the object and the microscope, sufficient measurement reproducibility can be ensured. In addition, since a conventional measuring device can be used, it can be applied at a relatively low cost.

[Brief description of the drawings]

FIG. 1 is a flowchart showing a procedure of an edge position measuring method according to an embodiment of the present invention.

FIG. 2 is an example of a schematic diagram of an edge position measuring mechanism for implementing the present invention.

FIG. 3 is an example of an edge candidate of an object to be imaged and a region in the vicinity thereof.

FIG. 4 is an example of density values and differential values of an image on aa ′ in FIG. 3;

FIG. 5 shows an example of a differential value of an image on the line aa ′ obtained when the distance between the microscope and the object to be imaged is changed, and FIG. FIG. 4B is a graph showing the relationship between the distance between the microscope and the object at any pixel on the line aa ′ and the differential value.

FIG. 6 is a conceptual diagram illustrating an example of a case where a part of an object to be imaged is not at a focus position.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Microscope 2 Television camera 3 Stage 4 Drive motor 5 Encoder 6 Frame memory 7 Arithmetic control unit 8 Monitor television 9 Work memory 10 Measurement table W Object to be imaged

Claims (1)

[Claims] 1. A method for measuring an edge position from an image of an object photographed by a television camera mounted on a microscope, the method comprising: moving the microscope in a direction perpendicular to an installation surface of the object. While moving at a speed, an image of the object to be picked up is taken by a television camera at a predetermined sampling interval, and a predetermined edge candidate point and a neighboring area thereof are differentiated with respect to the taken image obtained at the sampling interval. Calculating an image, extracting a differential value for each pixel at the same position from the differential image obtained at each sampling interval, and applying a predetermined function approximation to the differential value group extracted for each pixel Calculates the peak value of the differential value for each pixel, and combines the peak values of the differential value obtained for each pixel to obtain a peak value image in the area of the differential image. Determined, the edge position measurement method is characterized in that so as to measure an edge position of the object to be imaged on the basis of the peak value image.