JPH10206113A - Edge location detecting method and image measuring device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain resolving power finer than a pixel pitch by storing the same values corresponding to the detected signal levels of pixels corresponding to a predetermined number of memory groups provided for each pixel and detecting edge locations on the basis of the values obtained by averaging the values in memories arranged in order. SOLUTION: A CCD camera 13 captures images pixel by pixel for detection, and these output signals are inputted from the CCD camera 13 to an image processing part 2 by a unit of pixel. In the image processing part 2, the same value as the output signal of each of corresponding pixels from the CCD camera 13 is stored in each memory for processing high resolving power prepared for each one of a plurality of pixels. Average values of movement on these data is obtained by a predetermined expression. Each pixel is plurally divided corresponding to an average value of movement and binarized through the use of a threshold value to obtain edge locations.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an edge position detecting method for detecting an edge position on an image of an object to be measured, and an image measuring apparatus for displaying the object to be measured and detecting the edge position.

[0002]

2. Description of the Related Art In a conventional image measuring apparatus, the detection of the edge position of an object to be measured has been performed by a binarization process for each pixel on a screen. The edge position detecting method according to the prior art will be described with reference to FIGS. FIG.
3A is a plan view showing a screen of the image measuring device, and a measurement object 60 is displayed on a screen 50. FIG. Screen 50
Above, the measurement target 60 is displayed as a dark part 50a, and the part that is not the measurement target is displayed as a bright part 50b.
In the detection unit 70 in the screen 50 including the dark part 50a and the light part 50b, the edge of the measurement target 60 on the screen 50 for which the boundary 60a between the dark part 50a and the light part 50b indicated by a broken line in the drawing is to be obtained. It is. FIG. 1B shows an actual light intensity distribution corresponding to each pixel in the detection unit 70. From the light intensity distribution shown in FIG.
A method according to the related art for obtaining the position of the edge 60a on the screen 50 of 0 will be described with reference to FIG.

FIG. 5A shows an actual light intensity distribution in the detection unit 70 similar to FIG. 1B. FIG. 5B shows an output signal level from each pixel when an image having such a light intensity distribution is captured by a CCD camera, that is, an image signal. When this image signal is subjected to a binarization process using 50% as a threshold value, a binary image output as shown in FIG. 5C is obtained. Then, the boundary position between the fourth pixel and the fifth pixel whose value has changed from 0 to 1 is set as the position of the edge 60a.

[0004]

However, according to the above-mentioned prior art, the edge position of the object to be measured can be obtained only with the resolution of the pixel pitch. Therefore, the resolution is further improved in the image measuring device. It was not possible to obtain the performance that was performed, and the detection accuracy of the edge position could not be improved.

An object of the present invention is to provide an edge position detecting method and an image measuring device capable of solving the above-mentioned problems of the prior art and obtaining a resolution finer than the pixel pitch and accurately obtaining an edge position. Aim.

[0006]

In order to achieve the above object, an edge position detecting method according to the present invention is an edge position detecting method for detecting an edge position of an image captured by dividing the image into pixels. Providing a memory group consisting of a predetermined number of memories in association with each other, and storing the same value corresponding to the detected signal level of the pixel corresponding to each memory group in each memory of each memory group; And arranging the memory groups in order, and arranging the memory group in the order of the corresponding pixels; averaging the values of the memory arranged in the order; and determining the position of the edge based on the averaged value. Detecting.

According to the edge position detecting method of the present invention, the same resolution can be obtained as when each pixel is subdivided by a predetermined number in the memory, and the resolution is improved. Therefore, the number of memories in each memory group may be increased according to the degree to which the resolution is desired to be increased.

The number of divisions per pixel is determined according to the required resolution, and the number of divisions is set as a predetermined number, and a predetermined number of memories are prepared for each pixel unit. For example, if one pixel is 10 μm and it is desired to obtain a resolution of 1 μm, one pixel is set to 10 μm.
It is divided and 10 memories are prepared as one memory group for each pixel. Assuming that 20 pixels are used to detect the edge position, 20 memory groups are required,
20 × 10 = 200 memories are prepared. The detection signal data for each pixel is directly input to and stored in the ten memories of the one memory group, and similarly, the data is input to and stored in the ten memories of each memory group for each corresponding pixel unit. The memories included in the memory group are arranged in order, the memory groups are arranged in the order of the corresponding pixels, and then the stored memories are subjected to averaging processing for the number of divisions. As a result, an image signal with increased resolution is created on the memory. By performing binarization processing and differentiation processing on this image signal, the position of the edge can be detected.

[0009] Preferably, the averaging step is a step of setting the values of the memory arranged in the order as an average value of the values of the memories adjacent to each other by the same number on both sides of the memory.

Further, according to the image measuring apparatus of the present invention, there is provided an image measuring apparatus for displaying a detection signal from a camera for capturing and detecting an object to be measured by dividing it for each pixel as an image as an image. A plurality of memories that store the same value corresponding to the level by a predetermined number in correspondence with each of the pixels, a means for averaging the values stored in the memory, and, based on the averaged value, Means for detecting the position of the edge of the object to be measured. With this device, the above-described edge position detection method can be executed.

[0011]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 2 is a diagram schematically showing the image measuring device according to the present embodiment. The image measuring device includes a measuring microscope 1 and an image processing unit 2 including an image display unit 4. The measurement microscope 1 includes a support 3 on which an imaging optical system 6, an XY stage 5 on which a measurement object 7 can be moved during measurement and observation, and a transmission illumination device 8 are respectively supported. The imaging optical system 6 includes an optical system 9, an objective lens 12, and a CCD camera 1.
3 is provided. The CCD camera 13 divides and captures the object to be measured for each pixel and detects it. The enlarged image signal of the object to be measured 7 from the CCD camera 13 is sent to the camera image processing unit 2 and the screen of the image display unit 4 4a.

FIGS. 3A to 3D are diagrams for explaining an edge position detecting method in the present embodiment, and FIG. 4 is a flowchart for explaining a process for detecting the edge position. It is. FIG. 3A is a diagram showing a light intensity distribution on an actual image plane similar to FIG. 1B.
(B) is a diagram showing the level of an image signal which is an output signal from each pixel when this image is captured by the CCD camera 13.

An edge position detecting method according to the present embodiment will be described with reference to FIGS. First, CCD camera 1
3 captures and detects an image in pixel units, and outputs this signal from the camera 13 to the image processing unit 2 in pixel units (S
1). A pixel including an edge to be detected by the edge detection unit set on the screen 4a representing this image is selected (S2). 3A and 3B, the first to sixth pixels including an edge are selected, and FIG. 3B shows output signal data for each pixel. The same value of the output signal from the CCD camera 13 for each corresponding pixel is stored in each of the five high-resolution processing memories prepared for each of the first to sixth pixels in the image processing unit 2 ( S3). In this way, each data is stored in the 30 high-resolution processing memories. FIG. 3B shows the data stored in each memory in a state where each pixel is divided into five.

As described above, the output signal data X1, X2 are stored in the 30 memories corresponding to the first to sixth pixels.
, Xi, ..., X29, X30 (i = 1 to
30) are respectively stored. Each of these data (X
The moving average value (Ai) for i) is obtained by the following equation.

Ai = (Xi−2 + Xi−1 + Xi + Xi + 1 + Xi + 2) / 5 (1) Ai represents a moving average value of data Xi. Then, moving average processing is performed on each of the data X1 to X30 by moving i one by one in order from 1 to 30 to obtain moving average values A1 to A30 (S4). The moving average thus obtained is shown in FIG. 3C in a state where each pixel is divided into five corresponding to FIG. 3B. In the moving average processing, when the pixel data represents the minimum value or the maximum value in the pixel terminal, the moving average processing as described above is not performed, and the data stored in each memory is the minimum value or the maximum value. Leave as is. For example, FIG.
In the example of FIG. 3, the data of the first pixel and the second pixel represent the minimum value, and the data of the fifth pixel and the sixth pixel represent the maximum value.
(C) shows the data as it is.

When the binarization process is performed on the data shown in FIG. 3C with 50% as a threshold value, a distribution as shown in FIG. 3D is obtained. The fourth memory in the third pixel and 5
The value changes from 0 to 1 between the memories, and the center of the position corresponding to the fourth and fifth memories of the third pixel is the edge position to be obtained (S5).

As described above, in the example shown in FIG. 3, five memories are prepared for each pixel. Therefore, the same effect as when a pixel is reduced to one fifth is obtained, and the resolution is increased five times. improves. Thereby, in the image measuring device shown in FIG. 2, the detection accuracy of the edge position of the measurement object on the screen is improved, and accurate measurement can be performed. Thus, one pixel has N
When the memories correspond to each other, an image signal having N times resolution can be generated. Therefore, when it is desired to increase the resolution, the number N of memories corresponding to each pixel may be increased.

In this embodiment, the moving average processing is performed by the above equation (1). However, the present invention is not limited to this. For example, the moving average processing may be performed by the following equation.

Ai = (Xi−2 + Xi−1 + Xi + 1 + Xi + 2) / 4 (2) Further, the detection processing for obtaining the edge position is not limited to the binarization processing, and may be performed by, for example, differentiation processing. It can be carried out.

In this embodiment, the CCD camera divides an image into pixels and captures the pixels to output a detection signal. However, the present invention is not limited to this.
The image captured by V or the like may be divided and handled in the same manner as the image captured by the pixels of the CCD camera.

[0021]

According to the present invention, it is possible to provide an edge position detecting method and an image measuring device capable of obtaining a resolution finer than the pixel pitch and accurately determining an edge position.

[Brief description of the drawings]

FIG. 1A is a plan view showing a screen of an image measurement device, and FIG. 1B is a view showing an actual light intensity distribution corresponding to a pixel unit in a detection unit.

FIG. 2 is a schematic diagram showing an outline of an image measuring device according to an embodiment of the present invention.

FIG. 3 is a diagram ((a) to (d)) for describing an edge position detection method according to the present embodiment.

FIG. 4 is a flowchart illustrating a process for detecting an edge position according to the embodiment;

FIG. 5 is a diagram for explaining a method of detecting an edge position according to a conventional technique ((a), (b)).

[Explanation of symbols]

 Reference Signs List 1 Measurement microscope 13 CCD camera 2 Image processing unit 4a Screen 60a Edge of measurement object

Claims (3)

[Claims] 1. An edge position detection method for detecting an edge position of an image captured by dividing into pixels, wherein a memory group including a predetermined number of memories is provided for each of the pixels so as to correspond to each of the memory groups. Storing the same value according to the detected signal level of the pixel to be stored in each memory of each of the memory groups; and arranging each memory in each of the memory groups in order, and arranging the memory group in the order of the corresponding pixels. Averaging the values of the memory arranged in the order, and detecting a position of an edge based on the averaged value. 2. The method according to claim 1, wherein the averaging step is a step of taking the values of the memory arranged in the order as an average value of the values of the adjacent memory by the same number on both sides of the memory. 2. The edge position detection method according to 1. 3. An image measuring apparatus for displaying a detection signal from a camera for detecting and capturing an object to be measured by dividing it into pixels for each pixel as an image, wherein the same value corresponding to the level of the detection signal of each pixel is determined. A plurality of memories, each corresponding to each of the pixels, a unit for averaging the values stored in the memory, and a position of an edge of the measurement object based on the averaged value. An image measuring device, comprising: means for detecting.