JPS62215805A - Method and apparatus for detecting position of pattern - Google Patents

Abstract

PURPOSE:To make it possible to detect a pattern accurately at a high speed with high reliability, by determining the position of the pattern by calculating the overlapped area of the data of a pattern to be inspected and a mask pattern by high speed Fourier transformation data. CONSTITUTION:The data in the height direction corresponding to be position in an X-direction is inputted to a frame memory 10 from the pickup image signal of a pattern 3 by a TV camera 9. The time series data of the memory in the X-direction is subjected to Fourier transformation by a high speed Fourier transformation board 12 to be converted to frequency data which is, in turn, multiplied by mask pattern data converted to frequency data by the Fourier transformation of a data memory 11 by CPU 13 and the change in the overlapped area of the input pattern due to folding integration and a mask pattern is determined within a short calculation processing time. The position where the overlapped area obtained by the inverse Fourier transformation of the multiplication result becomes maximum comes to a detected pattern position to be positionally matched not relating to distortion due to disturbance. The same operation is also performed in a Y-direction and the detection of the pattern is accurately performed at a high speed with high reliability.

Description

[Detailed description of the invention] [Table of contents] - Overview - Field of industrial application - Prior art - Problems to be solved by the invention - Means for solving the problems - Effects - Examples - Effects of the invention [Summary] The change in the overlapping area when the image data of the pattern to be inspected is moved relative to the mask pattern data is calculated using FF.
A pattern detection method that detects a positional shift of a target pattern by calculating the position of the maximum area using T calculation.

[Industrial application field]

The present invention relates to a method and apparatus for detecting a pattern position in an automatic machine, such as a printed board pattern cutter or wire bonder, which positions a printed board according to a pattern and performs various processing.

[Conventional technology]

With the increasing density of printed circuit boards, patterns and pattern spacing have become finer, and it has become necessary to accurately detect the position of patterns in automatic machines that perform printed board assembly and inspection operations such as wire bonding and pattern cutting. .

The conventional method for detecting the position of patterns on printed circuit boards is to binarize the image information captured from a television camera to separate the target pattern and the background, and then extract the pattern features from the binary image to recognize the position. It was done.

[Problem that the invention seeks to solve]

In the conventional pattern position detection method, the entire image including the target pattern must be binarized, which requires a lot of processing time, and if the image fluctuates due to disturbance, it is necessary to binarize at an appropriate slice level. In some cases, accurate position detection was not possible.

The present invention has been made in view of the above-mentioned drawbacks of the prior art, and it is an object of the present invention to provide a pattern detection method with high processing speed, accuracy, and high reliability, and an apparatus for implementing the method.

[Means for solving problems]

In order to achieve this objective, in the present invention, image information of a pattern to be detected on a printed board is input to an arithmetic circuit, and when this is moved pixel by pixel with respect to a pre-input mask pattern, the area of the overlapping part is A pattern position detection method for detecting a pattern position from a position of maximum area by determining a change in the pattern position, the method comprising: (a) forming the image information; (b) A/D converting the image information; and (c)
) reading time series data in the X direction; (d) performing fast Fourier transform on the time series data; (e) performing multiplication by frequency series; (f) performing inverse fast Fourier transform on the multiplication results. (g) determining the peak position where the area is maximum; (h) determining the positional deviation by comparing the peak position and the position based on the mask pattern; The method is characterized by comprising a step of calculating an amount;

[For production]

An image of the target pattern is input to an arithmetic circuit using a television camera or the like, and the image is A/D converted to read data regarding the position of the target pattern. In order to find the change in the area of the overlapping portion when this inspection pattern data is moved pixel by pixel with respect to previously inputted mask pattern data, the inspection pattern data is subjected to fast Fourier transform and further multiplied by a frequency coefficient. The multiplication result is subjected to inverse fast Fourier transform to determine the area of the overlapping portion as a position function.

The peak position where the area is maximum is determined and compared with the position based on the mask pattern to detect the amount of positional deviation.

〔Example〕

FIG. 1 is a block diagram of an embodiment of a pattern position detection device for carrying out the method of the present invention. A printed board 2 is mounted on an XY table 1. A pattern 3 to be detected is formed on the printed board 2. Note that pattern 3 is drawn with its size exaggerated. A laser light source 4, a total reflection mirror 5, and a cylindrical lens 6 are provided to illuminate the pattern 3 with slit light. Laser light source 4
The empty light (for example, He-Ne laser light) is reflected by the total reflection mirror 5, and by passing through the cylindrical lens 6, it becomes a planar slit light 7 and forms a linear scan line (light cutting line) 8 in the irradiation area. . A television camera 9 is provided obliquely above the direction (Y direction) perpendicular to the direction of the scan line 8 (for example, the X direction). The television camera 9 is connected to a frame memory 10 for storing captured images, and this frame memory 10 is connected to a central control unit (CPU) 13. This CPU 13 is further connected to a data memory 11 storing mask pattern data, which is design value data for pattern 3, and a fast Fourier transform board (FFT board) 12. C
The PUI 3 is connected to a drive control circuit 14 of a processing mechanism (for example, a pattern cutter) for performing various processing on the printed board 2.

Next, the principle and operation of pattern detection by the pattern detection device described above will be explained.

The input image taken by the television camera 9 is shown in Fig. 2 (a
), data representing the pattern cross-sectional shape and in the height direction corresponding to the position in the X direction is stored in the frame memory 10. Therefore, data of a pattern shape corresponding to the position in the X direction can be read from the frame memory as time series data. The mask pattern (pattern based on the design data of pattern 3) with no positional deviation, which has been input into the data memory 11 in advance, forms an accurate rectangle at the predetermined design position, as shown by the solid line in FIG. 2(b). do. The pattern input to the frame memory 10 for such a mask pattern is displaced as shown by the dotted line in FIG. 2(b), and the waveform is distorted or distorted due to disturbance. Therefore, by fixing the position of one pattern and moving the other pattern pixel by pixel in the X-axis direction, we find the overlapping area g(t) of both waveforms at each position t and examine the change in area, as shown in Figure 2. A waveform like that shown in (c) is obtained. In practice, the detection pattern data of the dotted line is inverted to the negative position side and then moved pixel by pixel in the positive direction to find the overlapping area. Inlet cover pattern (dotted line) and mask pattern (solid line)
When they match, the overlapping area becomes maximum and that position is the position of the detection pattern to be aligned. Therefore, the second
The position of the pattern to be detected can be detected by finding the peak position tp of the overlapping area g(t) in Figure (c).

The input cover pattern in the X direction is X(t), and the mask pattern is h
(t), the change in the overlapping area can be calculated by the convolution integral expressed by .

This calculation can be performed by replacing it with frequency domain multiplication using the convolution theorem shown below, thereby reducing the calculation processing time.

Convolution theorem g(t)=h(t)*X(t) <H(f) XCf)
FIG. 3 shows a flowchart of a pattern position detection method using the above detection calculation principle.

Step a: Photograph the pattern with the television camera 9 and input the image information to the device.

Step b2: A/D convert the input image and write it into the frame memory IO.

Step C; The height of the image data relative to the position in the X direction is read as time-series data by the frame memory IO (determining X(t)).

Step d: FF the read time series data X(t)
The T calculation board 12 converts it into frequency data X (f). Mask pattern data H(f) is stored in the data memory 11 in advance.

Step e: Multiply the frequency data of the input cover pattern and the mask pattern to calculate H(f)X(f).

Step f: H(f)X(f)(7) Perform inverse FFT transform on the frequency function and return it to the position function g(t).

Step g: Calculate the peak position tp having the maximum area from the obtained g(t).

Step h: The difference between the peak position in the case where there is no positional deviation (in the case of two mask patterns) and the above-mentioned tp is determined and used as the positional deviation amount in the X direction.

Next, an image in the Y direction is input and the same steps as in the case of the X direction are repeated to obtain the amount of positional deviation in the Y direction. As described above, the X and Y coordinates of the target pattern are detected.

FIG. 4 shows the structure of another embodiment of a pattern position detecting device for carrying out the method of the present invention. In this embodiment, the entire printed board 1 including the pattern 3 is illuminated by a light source 15, and a television camera 9 photographs the pattern 3 from above.

Since pattern 3 has a larger amount of reflected light than the printed circuit board surface, the pattern can be identified as a difference in brightness.

Therefore, as shown in FIG. 5(a), a bright pattern corresponding section 3' corresponding to the pattern shape is input to the frame memory IO. For example, by scanning the dotted line portion of this frame memory, time series data corresponding to the position in the X direction as shown in FIG. 5(b) can be read. This corresponds to the data shown in FIG. 2(a) of the above embodiment. Using time series data based on this brightness information, pattern position detection is performed through the same steps as in the previous embodiment.

Another embodiment of the method of the invention is shown in FIG. In this embodiment, as shown in FIG. 6(a), position detection calculations are performed by obtaining time-series data for a plurality of scan lines (dotted lines) in the frame memory. This allows for even more accurate and reliable position detection. (b) is an example of time-series data regarding mask patterns. The scale represents the number of bits. Figure (c) shows the change in area when such time-series data is inverted, moved pixel by pixel, and superimposed on the original data. The bit of the peak position Lp is 240.

〔Effect of the invention〕

As explained above, in the pattern position detection method according to the present invention, since the pattern position is calculated based on the change in the overlapping area between the detection target pattern and the mask pattern, regardless of the distortion caused by the disturbance of the detected vs. Accurate position detection is performed. Further, since the overlap area is calculated by FFT transformation using the convolution theorem, the calculation processing time can be shortened.

[Brief explanation of drawings]

Fig. 1 is an explanatory diagram of the configuration of an apparatus for implementing the method of the present invention, Fig. 2 is an explanatory diagram of the principle of the method of the present invention, Fig. 3 is a flowchart of the method of the present invention, and Fig. 4 is a diagram for implementing the method of the present invention. FIG. 5 is an explanatory diagram of the input data of the embodiment of FIG. 4, and FIG. 6 is an explanatory diagram of another embodiment of the method of the present invention. 2...Printed board, 3...Pattern,
6... Cylindrical lens, 7... Slit light,
8...Scan line, 9...Television camera, 0...Frame memory, 11...Data memory, 12...High speed Fourier transform board, 13...CPU.

Claims (1)

[Claims] 1. Image information of a pattern to be detected on a printed board is input to an arithmetic circuit, and when this is moved pixel by pixel with respect to a pre-input mask pattern, changes in the area of the overlapping portion are calculated. A pattern position detection method for detecting a pattern position from a position of maximum area, comprising: (a) forming the image information; (b) A/D converting the image information; and (c) X direction. (d) performing fast Fourier transform on the time series data; (e) performing multiplication by frequency series; (f) performing inverse fast Fourier transform on the multiplication results to calculate the overlapping portions. (g) determining a peak position where the area is maximum; (h) calculating a positional shift amount by comparing the peak position with a position based on the mask pattern; A pattern position detection method comprising steps. 2. The change in area is calculated by a convolution integral, and the convolution integral is calculated by replacing the frequency domain multiplication with the fast Fourier transform.
Pattern position detection method described in section. 3. After detecting the position in the X direction, the same steps are repeated in the Y direction as well.
The pattern position detection method according to item 1 or 2. 4. The pattern position detection method according to any one of claims 1 to 3, wherein the image information is created for two or more scan lines in the same direction. 5. The image information forming step is performed by irradiating a planar slit light onto the pattern to be detected to form a linear irradiation section, and inputting data of the irradiation section into a frame memory via a television camera. A pattern position detection method according to any one of claims 1 to 4, characterized in that image information is formed according to the thickness of a pattern protruding onto a printed board surface. 6. The image information forming step is performed by irradiating the entire pattern to be detected on the printed board with light and inputting the data of the irradiated surface to a frame memory via a television camera, thereby forming a pattern on the printed board surface. A pattern position detection method according to any one of claims 1 to 4, characterized in that image information is formed according to a difference in the amount of reflected light. 7. A light irradiation means for irradiating the pattern to be detected on the printed board, a television camera for photographing the pattern, and arithmetic processing of the image inputted through the television camera based on mask pattern data inputted in advance. A pattern position detection device comprising an arithmetic circuit that calculates the area of an overlapping part when an image is moved in pixel units relative to a mask pattern, and detects the pattern position from the maximum value. In the above, the arithmetic circuit: A/D converts the image; reads time series data in the X direction; performs fast Fourier transform on the time series data; performs multiplication by a frequency series; performs inverse fast Fourier transform on the multiplication result to Find the area of the overlapping part as a position function;
A pattern position detecting device characterized in that the pattern position detecting device is configured to obtain a peak position where the area is maximum; and to calculate a positional deviation amount by comparing the peak position and a position based on the mask pattern. 8. The pattern position detection device according to claim 7, wherein the light irradiation means includes a laser light source and a cylindrical lens, and irradiates a slit light onto the pattern to form a linear irradiation section. . 9. Claim 7, wherein the light irradiation means comprises a light source that irradiates the entire pattern to be detected.
The pattern position detection device described in .