KR20100135101A - Hole cluster test method - Google Patents

Abstract

PURPOSE: A hole cluster test method is provided to analyze and process an image which photographs a hole cluster and measure sizes and locations of holes of the hole cluster, thereby testing whether the hole cluster is defective. CONSTITUTION: A test target including holes is photographed. A histogram of the photographed image is analyzed(S20). The histogram of the photographed image is controlled to easily process the photographed image. Fourier transform is performed in the image(S21). Band pass filtering is performed in the converted image to remove a noise component(S22). The size and locations of the holes are measured(S23).

Description

Hole cluster test method

The present invention relates to a method for inspecting a hole cluster, and more particularly, to analyze and process an image photographing a hole cluster to measure the size and position of each of the holes included in the hole cluster, and to measure each of the plurality of holes. The present invention relates to a method for inspecting whether a hole cluster is defective based on the size and position of the hole cluster.

Conventionally, the inspection of a hole cluster including a plurality of holes such as a printed circuit board (PCB) and a printed circuit board (PCB) inspection jig plate (JIG plate) is performed by a human eye using a microscope. Is done. Such inspection methods have reliability problems due to the condition of the inspector, fatigue, and the like, errors due to individual differences, problems in which real-time and full inspections are impossible, and manpower shortages due to avoiding visual inspection. In addition, the inspection method also has a problem of accuracy and speed reduction of inspection as the size of the jig plate hole is miniaturized.

An advanced test method for visual inspection is a method for hall cluster test using a vision test system. In the hall cluster inspection method using a vision inspection system, a test object is photographed using a camera and a hall cluster is inspected based on an image of the photographed hall cluster. The hole cluster inspection method, which is generally used, has a disadvantage in that an accurate inspection cannot be performed when the edge changes due to a change in the lighting environment.

Accordingly, the technical problem to be achieved by the present invention is to provide a method for inspecting a hole cluster that can obtain a uniform test result even in a non-constant lighting environment.

The hole cluster inspection method for solving the technical problem comprises the steps of photographing the inspection object including a plurality of holes; Analyzing a histogram of the photographed image and adjusting a histogram of the photographed image to facilitate processing of the photographed image; Performing Fourier transform on the histogram-adjusted image; Performing band pass filtering on the Fourier transformed image to remove noise components; And measuring sizes and positions of the plurality of holes based on a band pass filtered image, and determining whether the inspection object is defective based on a result of measuring the sizes and positions of the plurality of holes. The histogram analysis and adjustment of the photographed image may be performed on at least one of contrast, color, and saturation of the photographed image.

The analyzing the histogram of the photographed image divides the image information of the photographed image into a plurality of image information ranges, and corresponds to each of the plurality of image information ranges for the plurality of pixels of the photographed image. Calculating a probability of a pixel, wherein adjusting the histogram for the photographed image corresponds to each of the plurality of image information ranges for the plurality of pixels of the photographed image using a transform function. Summing the probabilities of the pixels.

Performing a Fourier transform on the histogram-adjusted image may include converting the histogram-adjusted image as a function of the spatial frequency and the vertical axis of the photographed image.

Performing band pass filtering on the Fourier transformed image to remove the noise component may be performed based on band elimination filtering on a high frequency component including noise generated by illumination used to photograph the inspection object. have.

Measuring the size and location of the plurality of holes based on the bandpass filtered image comprises detecting edges of the plurality of holes in the bandpass filtered image; Measuring the size of each of the plurality of holes based on a circumference of an edge of the detected plurality of holes; And defining a reference point in the captured image and measuring positions of the plurality of holes with respect to the reference point.

Detecting edges of the plurality of holes may include detecting edges of the plurality of holes by convolutional masking for the bandpass filtered image. In this case, the mask may be a mask based on a Prewitt filter.

The hole cluster test method according to an embodiment of the present invention may be implemented by executing a computer program for executing a booting method of the computer system stored in a computer-readable recording medium.

As described above, the hole cluster inspection method according to the present invention can detect the minute error as compared to the visual inspection, thereby ensuring the reliability of the inspection, and the high-speed inspection can reduce the time and cost, Even in the change of the lighting environment, it is possible to accurately determine whether or not the hall collective defects.

In order to fully understand the present invention, the operational advantages of the present invention, and the objects achieved by the practice of the present invention, reference should be made to the accompanying drawings which illustrate preferred embodiments of the present invention and the contents described in the accompanying drawings.

In the present specification, when one component 'transmits' data or a signal to another component, the component may directly transmit the data or signal to the other component, and at least one other component. Through this means that the data or signal can be transmitted to the other component.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Like reference numerals in the drawings denote like elements.

1 is a block diagram of a hole cluster test system 100 according to an embodiment of the present invention. Hole clusters include a printed circuit board (PCB) including a plurality of holes, a jig plate for testing a PCB, and the like. However, the scope of the present invention is not limited thereto.

Referring to FIG. 1, the hall cluster inspection system 100 includes a computer 110, a lighting device 120, a backlight 130, and a camera 140. The computer 110 processes the image photographed by the camera 140 to check whether the hole cluster is defective. The computer 110 is a system including an input device such as a keyboard and a display device such as a monitor, but may include a personal computer (PC), a notebook, and the like, but the scope of the present invention is not limited thereto.

The lighting device 120 and the backlight 130 provide illumination necessary for capturing the hole cluster as the inspection object. The camera 140 transmits the photographed image to the computer 110 by capturing a hole cluster on the backlight 130.

Figure 2 is a flow chart of the hole collective inspection method according to an embodiment of the present invention. Hereinafter, referring to FIGS. 1 and 2, the overall steps of the hole cluster inspection method using the hole cluster inspection system 100 will be described, and the operation of the hole cluster inspection system 100 in each step will be described in more detail. Take a look. Here, the hole community, that is, the inspection object may be a jig plate for inspecting whether or not the PCB is defective.

When an inspection object including a plurality of holes is photographed by using the camera 140, the computer 110 analyzes a histogram of the photographed image, and a histogram of the photographed image to facilitate processing of the photographed image. Adjust (S20). The computer 110 then performs a Fourier transform on the histogram-adjusted image to process the image in the frequency domain (S21), and band-passes the result of the Fourier transform on the histogram-adjusted image to remove noise components. Band pass filtering is performed (S22).

The above-described steps S20 to S22 are preprocessing steps for accurately determining sizes and positions of the plurality of holes included in the inspection object from the captured image. 3 is a view illustrating a change of an image according to a preprocessing process of a photographed image in the hole cluster inspection method according to an exemplary embodiment of the present invention. The image shown in FIG. 3 is a black and white image, and the black and white inversion of the black and white image is performed.

In FIG. 3, (a) is an image obtained by reversing an image photographed with respect to a test object including a plurality of holes, and (b) is a contrast histogram adjustment on the photographed image to facilitate processing of the captured image. One image is an inverted image, and (c) is an image in which the band pass filtering of the image on which the contrast histogram adjustment is performed is inverted.

Referring to FIG. 3, (b) the plurality of holes in the image appear clearer than (a) the plurality of holes in the image, which is a result of adjusting the contrast histogram for the captured image to facilitate image processing. to be. (c) The edge portion of the holes in the image is clearer than the edge portion in (b) image, which is the result of performing band pass filtering to remove noise components due to illumination included in the edge region. That is, the holes of the image that have undergone the preprocessing step are clearer than the holes of the image of the first object to be inspected, and the edges of the holes become clearer. This means that the preparation for the more accurate measurement of the position and size of the holes is done by going through the pretreatment step.

When the preprocessing step for the captured image is completed, the computer 110 measures the sizes and positions of the plurality of holes included in the inspection object based on the band pass filtered image (S23), and the sizes and positions of the plurality of holes. It is determined whether the inspection object is defective based on the measurement result (S24), and whether the inspection object is defective is displayed through the display device (S25).

Hereinafter, the operation of the hole cluster inspection system 100 for each of the steps of the hole cluster inspection method using the hole cluster inspection system 100 will be described in more detail.

The histogram analysis and adjustment step (S20) of the captured image may be performed on at least one of contrast, hue, and saturation of the captured image. Analyzing the histogram of the photographed image divides the image information of the photographed image (that is, at least one of contrast, hue, and saturation of the photographed image) into a plurality of image information ranges, and multiple pixels of the photographed image. Calculating a probability of a pixel corresponding to each of the plurality of image information ranges.

The adjusting of the histogram for the photographed image may be a step of adding a probability of a pixel corresponding to each of the plurality of image information ranges for the plurality of pixels of the photographed image by using a transform function.

Equation 1 represents a probability of a pixel corresponding to each of the plurality of image information ranges of the plurality of pixels of the photographed image, and Equation 2 represents a plurality of image information ranges of the plurality of pixels of the photographed image. It is the sum of the probabilities of the pixels corresponding to each.

Here, P _r denotes a probability of a pixel corresponding to each of the plurality of image information ranges, n _k denotes the number of pixels corresponding to the image information range, and n denotes the total number of pixels in the captured image. , L denotes the total number of the plurality of image information ranges.

Here, s denotes a conversion function, P _r denotes a probability of a pixel corresponding to each of the plurality of image information ranges, n _j denotes the number of pixels corresponding to the image information range, and n denotes a photographed position. It means the total number of pixels in the image, L means the total number of the plurality of image information ranges.

For example, the analyzing and adjusting the histogram of the captured image (S20) may be a step of analyzing the contrast histogram of the captured image and adjusting the contrast histogram of the captured image to facilitate processing of the captured image. Here, the contrast histogram is a data containing important information of the captured image. It is a tool used to show the contrast profile of the captured image. The contrast value of the captured image is the horizontal axis, and the vertical axis is the captured image. It may be implemented as a graph representing the frequency of the pixel corresponding to the contrast value of the horizontal axis within.

4A illustrates a histogram analysis result of the captured image, and FIG. 4B illustrates a histogram processing result of adjusting a business card with respect to the captured image. In FIG. 4A, the horizontal axis represents a range of contrast and the vertical axis represents a probability of a pixel corresponding to the range of contrast in the photographed image calculated by Equation 1. In FIG. 4B, the horizontal axis represents a range of contrast and the vertical axis represents a sum of probabilities of pixels corresponding to the range of contrast in the photographed image calculated by Equation 2.

Referring to FIG. 4A, in the case of the histogram of the image photographing the test object, the histogram is concentrated in a narrow contrast range (parts indicated by dashed lines), which makes it difficult to process the image for extracting the feature of the test object. The histogram of FIG. 4B is a histogram in which the portion indicated by the dotted line of FIG. 4A is widely extended along the horizontal axis by Equation 2, so that image processing for extracting the feature of the test object from the feature of the test object is easier.

Performing a Fourier transform on the histogram-adjusted image (S21) is performed to process the image in the frequency domain, and the histogram-adjusted image on the spatial frequency and vertical axis of the horizontal image of the captured image. Convert to a function. Equation 3 shows a Fourier transform equation for the image whose histogram is adjusted.

Here, H denotes a Fourier transform result, h denotes an image in which the histogram is adjusted, x denotes a horizontal axis of the image, y denotes a vertical axis of the image, and u denotes a spatial frequency with respect to the horizontal axis of the image. V means the spatial frequency of the horizontal axis of the image.

5 shows a result of performing Fourier transform on an image of which contrast is adjusted. Referring to FIG. 5, it can be seen that the photographed image of the object is converted into a high frequency region and a low frequency region by performing a Fast Fourier Transform (FFT). In the photographed image of the inspection object (eg, jig plate) including a plurality of holes, the edge portion of the plurality of holes includes a high frequency component including noise caused by illumination.

This high frequency component causes the edges of the plurality of holes to be unclear, leading to a decrease in the accuracy of the size and position measurement of the plurality of holes. Therefore, by removing the high frequency component included in the edge portion of the plurality of holes, the accuracy of the size and position measurement of the plurality of holes can be increased.

Performing band pass filtering on the Fourier transformed image (S22) is a step of removing a high frequency noise component included in the edge portion. However, if bandpass filtering is directly performed on an image, the detail of the image may be degraded. In order not to deteriorate the detail of the image, band pass filtering removes the noise component included in the edge portion of the band rejection filtering for high frequency components including noise generated by the lighting used for photographing an object. Can be performed based on this. Equation 4 is an equation for obtaining a transfer function for band pass filtering based on band elimination filtering.

Here, H _bp is a transfer function for band pass filtering and H _br is a transfer function for band rejection filtering.

6 illustrates a process of separating the high frequency region and the low frequency region from the image photographed by filtering. Referring to FIG. 6, in the fast Fourier transform result, an edge image (high frequency component) may be obtained by filtering in the form of a Laplacian distribution function, and the background (low frequency component) may be filtered in the form of a Gaussian distribution function. It can be seen that it can be obtained by. As described above, in the hole cluster inspection method according to the exemplary embodiment of the present invention, the Gaussian distribution function may be filtered based on band elimination filtering to remove high frequency noise components.

After the band pass filtering step S22 for removing the high frequency noise component is completed, the step S23 of measuring the size and location of the plurality of holes based on the band pass filtered image is performed. Detecting edges of the holes, measuring the size of each of the plurality of holes based on the circumference of the detected edges of the plurality of holes, and defining a reference point in the captured image and the location of the plurality of holes relative to the reference point It may include the step of measuring.

Detecting the edges of the plurality of holes may include detecting the edges of the plurality of holes by convolutional masking for the bandpass filtered image. The mask used to detect the edges of the plurality of holes may be a mask based on Prewitt filter. Edge detection using Prewitt filter-based mask is a method that detects the edge of an image by applying the horizontal filter and the vertical filter number sequentially, and is a technique well known to those skilled in the field of image processing. Detailed description will be omitted. (C) The image of FIG. 3 is an image including edges detected by a prewitt filter based mask on a bandpass filtered image, and (c) the edges of the plurality of holes in the image are histograms in the image of the inspection object. It is clearer than the edges of the holes detected in the conventional inspection system because they are obtained from the image from which the noise components are removed by bandpass filtering on the adjustment and Fourier transform results.

When edges of the plurality of holes are detected, the step of measuring the size of each of the plurality of holes is performed based on the circumference of the detected edges of the plurality of holes, and defining a reference point in the captured image and the position of the plurality of holes with respect to the reference point. The step of measuring is performed. The size and position of the plurality of holes is measured based on the edges of the holes which are clearer than the edges of the holes detected in the conventional inspection system, so it is more accurate than the value measured in the conventional inspection system, and more influenced by the noise caused by the illumination. Receive less.

7 is an image showing the center of each hole to measure the position of the holes. The center of each of the plurality of holes may also be measured by the detected edge, and the position of each of the plurality of holes may be defined as the coordinate value of the center of each of the plurality of holes with respect to the reference point.

When the size and position measuring step S23 of the plurality of holes is completed, a step S24 of determining whether or not the object to be inspected is performed is performed. Whether the inspection object is defective may include determining whether the inspection object is defective or not, based on whether the inspection object is defective or not, respectively.

Determining whether each of the holes is defective may include determining whether the size and location of each of the holes are within a predetermined range. For example, when the error for each of the size and position of the hole is 5% or less, the hole is determined to be normal, and when the error for each of the size and position of the hole is more than 5%, the hole may be determined to be defective.

The determining of whether the inspection object is defective may include determining whether the number of defective holes is within a predetermined range. For example, the inspection object may be determined to be normal when the number of defective holes is 5% or less, and the inspection object may be determined to be defective when the number of defective holes is more than 5%.

When the step S24 of determining whether or not the inspection object is defective is completed, an inspection result indicating whether the inspection object is defective is displayed (S25).

The hole collective inspection method according to an embodiment of the present invention may also be embodied as computer readable codes on a computer readable recording medium. That is, the hole cluster test method according to an embodiment of the present invention can be implemented by executing a computer program for executing the booting method of the computer system stored in a computer-readable recording medium.

The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. For example, the computer-readable recording medium may include a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and a program for performing the hole collective inspection method according to an embodiment of the present invention. The code may be transmitted in the form of a carrier wave (eg, transmission over the Internet).

The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. In addition, functional programs, codes, and code segments for implementing the hole cluster test method according to an embodiment of the present invention can be easily inferred by programmers in the art to which the present invention belongs.

Although the invention has been described with reference to one embodiment shown in the drawings, this is merely exemplary, and it will be understood by those skilled in the art that various modifications and equivalent other embodiments are possible. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

1 is a block diagram of a hole collective inspection system according to an embodiment of the present invention.

Figure 2 is a flow chart of the hole collective inspection method according to an embodiment of the present invention.

3 is a view illustrating a change of an image according to a preprocessing process of a photographed image in the hole cluster inspection method according to an exemplary embodiment of the present invention.

4A shows a histogram analysis result of the captured image.

4B shows a histogram processing result of adjusting a business card with respect to the captured image.

5 shows a result of performing Fourier transform on an image of which contrast is adjusted.

6 illustrates a process of separating the high frequency region and the low frequency region from the image photographed by filtering.

7 is an image showing the center of each hole to measure the position of the holes.

Claims (16)

Photographing an inspection object including a plurality of holes; Analyzing a histogram of the photographed image and adjusting a histogram of the photographed image to facilitate processing of the photographed image; Performing Fourier transform on the histogram-adjusted image; Performing band pass filtering on the Fourier transformed image to remove noise components; And And measuring the size and location of the plurality of holes based on a band pass filtered image, and determining whether the inspection object is defective based on a result of the size and location measurement of the plurality of holes. Way. The method of claim 1, wherein the analyzing and adjusting the histogram of the captured image And a method for inspecting a hole colony on at least one of contrast, color, and saturation of the captured image. The method of claim 2, wherein analyzing the histogram of the captured image Dividing image information of the photographed image into a plurality of image information ranges, and calculating a probability of a pixel corresponding to each of the plurality of image information ranges for a plurality of pixels of the photographed image; Adjusting the histogram for the captured image Summing a probability of a pixel corresponding to each of the plurality of image information ranges for the plurality of pixels of the photographed image using a transform function. The method of claim 1, wherein performing a Fourier transform on the histogram-adjusted image is performed. And converting the histogram-adjusted image into a function of a spatial frequency and a vertical axis of a horizontal axis of the captured image. The method of claim 1, wherein performing band pass filtering on a Fourier transformed image to remove the noise component And a method for inspecting a hole colony based on band-rejection filtering on a high frequency component including noise generated by illumination used for photographing the inspection object. The method of claim 1, wherein measuring the size and location of the plurality of holes based on the bandpass filtered image comprises: Detecting edges of the plurality of holes in the bandpass filtered image; Measuring the size of each of the plurality of holes based on a circumference of an edge of the detected plurality of holes; And And defining a reference point in the photographed image and measuring positions of the plurality of holes with respect to the reference point. The method of claim 6, wherein detecting edges of the plurality of holes comprises: Detecting edges of the plurality of holes by convolution a mask over the bandpass filtered image. The method of claim 7, wherein the mask is Prewitt filter based masking method for hole clusters. A computer readable recording medium storing computer readable codes for executing the hole colony testing method according to any one of claims 1 to 8. Photographing a JIG plate including a plurality of holes; Analyzing a contrast histogram of the photographed image and adjusting the contrast histogram of the photographed image to facilitate processing of the photographed image; Performing Fourier transform on the contrast-adjusted image; Performing band pass filtering on the Fourier transformed image to remove noise components; And And measuring the size and location of the plurality of holes based on a band pass filtered image, and determining whether the inspection object is defective based on a result of the size and location measurement of the plurality of holes. . The method of claim 10, wherein analyzing the contrast histogram of the captured image comprises: Dividing the contrast of the photographed image into a plurality of contrast ranges and calculating a probability of a pixel corresponding to each of the plurality of contrast ranges for the plurality of pixels of the photographed image, Adjusting the contrast histogram of the captured image Summing a probability of a pixel corresponding to each of the plurality of image information ranges for the plurality of pixels of the photographed image using a transform function. The method of claim 10, wherein performing band pass filtering on a Fourier transformed image to remove the noise component The jig plate inspection method is performed based on band-rejection filtering for noise including a high frequency component generated by illumination used in the jig plate photographing. The method of claim 10, wherein measuring the size and location of the plurality of holes based on the band pass filtered image comprises: Detecting edges of the plurality of holes in the bandpass filtered image; Measuring the size of each of the plurality of holes based on a circumference of an edge of the detected plurality of holes; And Defining a reference point in the photographed image and measuring a position of the plurality of holes with respect to the reference point. The method of claim 13, wherein detecting edges of the plurality of holes comprises: Detecting edges of the plurality of holes by convolution a mask over the bandpass filtered image. The method of claim 14, wherein the mask is Jig Platt inspection method based on Prewitt filter. A computer readable recording medium storing computer readable code for executing the jig plate inspection method according to any one of claims 10 to 15.

Images