JPH0429041A - Wiring pattern inspection device - Google Patents

Abstract

PURPOSE:To enable inspection which is free from wrong information and recognize the generation position of a defect by detecting a closed area outside the specific size of the through hole of a printed board from a binary image of through hole transmitted light. CONSTITUTION:First and second image input means 102 and 103 consist of CCD cameras, the means 102 detects light in a specific wavelength range which is reflected by the printed board 1, and the means 103 detects light in a wavelength range different from photodetection wavelength sensitivity; and respective density images are converted 104 and 105 into binary data with a constant threshold value to convert a conductor pattern into 1 or 0 and a base material and a through hole part into 0 or 1, and the closed area outside the specific size of the through hole is detected 106 in the binary image from the means 105 to indicate an inspection reference switching position to a defect detecting means 107. Then the means 107 applys a design rule to the binary pattern from the means 104 and while illegal reference line width is switched between the conductor part and land part, the features of defects such as improper line width, a disconnection, and a short circuit are detected.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a wiring pattern inspection device for inspecting wiring patterns on printed circuit boards for defects.

BACKGROUND OF THE INVENTION As the density of mounting electronic components on printed circuit boards increases, wiring patterns are becoming increasingly finer. Conventionally, defects in printed circuit boards, etc. have been visually inspected by humans, but as wiring patterns become more detailed, it has become difficult to continue inspection work for long periods of time while maintaining inspection accuracy, and automation of inspection has become necessary. is requested.

As a wiring pattern defect inspection method, G.L.
C Sants and Nie K Jain (J-L.C., 5
anz and A, ni, Jain, Mach
ine visionJ techniques for 1nspectt
on of printed wire -ing
boards and thick-film
circuits, J.

Opt, Soc, Amer, vol, 3.
No, 9. pp, 14661482, 5ept
(1986) have introduced many methods, and in particular, many methods classified as design rule methods or comparison methods have been proposed.

These methods have their advantages and disadvantages, but among them, the most promising and interesting method is Ju. Earl, Mandeville (J.
R, Mandeville, Novel method
d for anaIysis of pri
nted circuit images, I
B.M.J.

Res, Develop,, vol, 29. n
o, 1. pp, 349376, Jan,
1.979), which shrinks or expands binary image data and then thins it to detect defects in wiring patterns, which will be explained below as a conventional example. FIG. 9 shows the procedure of defect detection processing. Figures (a) to (d) show the missing defect detection procedure, and Figures (e) to (h) show the protrusion defect detection procedure.

First, a method for detecting missing defects will be explained with reference to the drawings. (a) shows a defect image, where point b and point 0 are detected as fatal defects due to insufficient line width and disconnection, and point a is not detected as a defect. In the first step (b), the connection at point b is cut off by shrinking (eroding) the image by a predetermined size. In the second step (c), the pattern is thinned to a width of one pixel. As the third step, in (d), a 3×3 local area (position indicated by the opening in the figure)
The connectivity of the thinned image is determined in , and point b and point 0 are detected as disconnections. It is also shown that the junction between the terminal part and the wiring pattern (the position indicated by a circle in the figure) can also be detected as a feature point by determining the connection of the 3 × 3 local area.
Ru.

Next, a method for detecting a protruding defect will be explained with reference to the drawings. (e) shows a defect image, where point b and point 0 are detected as fatal defects due to line width abnormalities and shorts, and a
Points shall not be detected as defects. As the first step (f), by expanding the image by a predetermined size, b
Generates a new connection between points. As a second step (gl
Now, thin the pattern to one pixel width. As the third step (in hl, the connectivity of the thinned image is determined in a 3×3 local area (position indicated by the opening in the figure), and point b and point 0 are detected as branch points, that is, shorts.The above steps Line width thickening, wire breakage, and short circuit can be detected by this method.

Problems to be Solved by the Invention A method has been described in which a binary image is contracted and expanded to enhance the characteristics of the defect and then thinned, and defects are detected by determining connectivity in a 3×3 local area. This method cleverly utilizes wiring pattern design rules and can reliably detect defects, so it can be said to be a promising method.

However, when inspecting a board with through-holes, the pattern width standards for the conductor part and the land part are different, so if they shrink to the same size, the pattern in the land part will be interrupted (・The line width may be mistakenly detected as insufficient at the land position). There was a possibility that it would be detected.

In view of the above problems, the present invention not only performs an inspection without causing false alarms even if there are through holes in a printed circuit board, but also recognizes whether defects such as line width violations, disconnections, or shorts occur in the conductor part or the land part. The present invention provides a wiring pattern inspection device that can perform various defect inspections.

Means for Solving the Problems In order to solve the above problems, the technical solution of the present invention is to illuminate a printed circuit board with illumination light in a predetermined wavelength range and photoelectrically convert the reflected light from the wiring pattern formed on the board. a first image input means for illuminating the printed circuit board with illumination light in a wavelength range separated from the illumination light from the back side of the printed circuit board, and photoelectrically converting the light passing through the through hole; a first binarization means that converts the grayscale image from the first image input means into a binary image; and a second binarization means that converts the grayscale image from the second image input means into a binary image. and through-hole area detection means for detecting a closed area outside a predetermined size of the through-hole from the binary image signal from the second binarization means, and the first binarization means and the through-hole area detection means. and defect detection means for distinguishing between lands and conductors from the binary image signal from the means and detecting the characteristics of the defects.

Function The present invention binarizes a grayscale image obtained by photoelectrically converting light passing through a through hole in a printed circuit board, and detects a closed area outside a predetermined size of the hole from the binary image of the through hole. Since the conductor part and the land part are distinguished from each other, and the standard of turn width is switched between the conductor part and the land part, line width violations will not be detected in the land part. Since it is possible to distinguish between whether a defect occurs in a land portion or a conductor portion, a more diverse range of defect inspections becomes possible.

EXAMPLES Hereinafter, examples of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram of a wiring/turn inspection device according to an embodiment of the present invention. In Figure 1, 10
1 is a printed circuit board, 102 is a first image input means for detecting reflected light from the printed circuit board 101, 1o3 is a second image input means for detecting light passing through a through hole of the printed circuit board 101, and 104 is the first image input means first binarization means for converting the grayscale image from the image input means into a binary image; 10;
5 is a second binarizing means for converting the grayscale image from the second image inputting means into a binary image; 1o6 is for determining a predetermined through hole using the binary image from the second binarizing means; Through-hole area detection means for detecting an area outside the size, 1
07 is a defect detection means that uses the binary images from the first binarization means and the through-hole area detection means to detect defects such as line width violations, disconnections, and shorts in the conductor portion and the land portion, respectively. , 114 is reflected illumination light that illuminates the surface of the substrate, 1
12 is a first wavelength filter, 110 is a diffused illumination device such as a ring-shaped light guide that guides reflected illumination light, 116 is illumination light illuminated from the back side of the substrate, and 113 is a wavelength filter different from that of the first wavelength filter. 111 is a light guide that guides the illumination light 115; 109 is an imaging lens; 108 is a second wavelength filter that directs light in two wavelength ranges that have passed through the first and second wavelength filters in two directions; A separating dichroic mirror is shown.

The operation of the above configuration will be explained below.

A diffuse illumination device 11 such as a ring-shaped light guide is applied to a wiring pattern formed on a printed circuit board 101 using light in a first wavelength range that has passed through a first wavelength filter 112.
Illumination at 0. At the same time, a second wavelength filter 113 is used to illuminate the printed circuit board 101 from the back side with light in a second wavelength range different from the diffused illumination. The reflected light from the illumination light in the first wavelength range and the illumination light in the second wavelength range passing through the through hole are transmitted to the first image input means 102 and the second image input by an imaging lens 109 and a dichroic mirror 108. Means 1
The images are respectively focused on o3. The first image input means 102 is a CC
With an imaging device such as a D camera (one-dimensional or two-dimensional),
The reflected light of the printed circuit board due to the light in the first wavelength range is detected. The second image input means 103 is similarly constituted by an imaging device such as a CCD camera, but unlike the light reception wavelength sensitivity of the first image input means 102, it detects light in the second wavelength range. The first binarization means 104 binarizes the grayscale image from the first image input means 102 using a fixed threshold value, and converts the conductor pattern into a 1 value and the base material and through-hole portion into a 002 value image.

The second binarization means 105 binarizes the grayscale image from the second image input means 103 using a fixed threshold value, and converts it into a binary image with 1 for the through-hole portion and 0 for the base material and conductor pattern. . The through-hole area detecting means 106 detects a closed area outside the predetermined size of the through-hole in the binary image from the second binarizing means 105, and instructs the defect detecting means 107 where to switch the inspection standard. The defect detection means 10 applies design rules to the binary pattern from the first binarization means 104 to detect defect characteristics such as violation of line width, disconnection, short circuit, etc. In response to the image signal of the through-hole area, defect characteristics are detected while switching the reference line width to be considered as a violation between the conductor part and the land part.

Next, the second embodiment of the through-hole area detection means will be explained.
This will be explained with reference to the figures and FIG.

FIG. 2(a) shows a procedure for scanning a binary image of a through hole obtained by the second binarizing means 105 with a scanning window,
FIG. 4B shows a processed image. FIG. 3 is a diagram showing reference pixel positions when performing calculations between pixels in the scanning window. The processing procedure for detecting a through-hole area will be explained below. In FIG. 2, an MxM scanning window 202 is shifted pixel by pixel in the main scanning direction on a binary image 201 of a through hole, dilation processing is performed by a logical operation described later, and the processing result is output to a position of interest 204 of an output image 203. do.

When the main scanning is completed, it moves one pixel in the sub-scanning direction and processes the next main scanning line. Since the window scanning process is a common technique, a description of its specific configuration will be omitted. The contents of the calculation in the scanning window 202 will be explained below. As shown in FIG. 3, reference pixels are arranged in a circle with the pixel of interest as the center and the radius is the number n of pixels to be expanded, and the output value is the value di(i,: 1 to N) using the following formula. + indicates logical sum.

n=do +dl+d2+-dN FIG. 3 shows an example where n=6, and n is set to a size that can accommodate the thinning pattern of the wiring pattern in the defect detection means 107. Through-hole image 206 of this result input
becomes an image 206 expanded in size n, and a signal is generated to notify the defect detection means 107 that it is a through-hole area.

Next, an embodiment of the defect detection means will be described with reference to FIG. FIG. 4 shows a specific example of the configuration of the defect detection means 107. In FIG. 4, 401 is an input terminal for the through-hole image from the through-hole detection means, and 402 is an input terminal for the binary image of the printed circuit board from the first binarization means 104. The binary image of the printed circuit board from the terminal 402 is thinned to one pixel width by the thinning means 403, and a thinned image 407 is output. The image signal from the terminal 402 is delayed by the number of main scanning lines required for line thinning in a delay memory 406, and then inputted to the length measuring means 409 together with a thinned image 407 as an original image 40B of the printed circuit board. When the target position is on the thinning pattern from the thinning image 407, the length measuring means 409 detects the original image 408 of the printed circuit board.
Measure the line width and output the measured length value. The measured length value is determined by a comparator 412 as a set value A410 or a set value B411.
When the measured length value is less than set value A or set value B, a line width violation is detected, and a line width violation signal is output from terminal 414. It is assumed that the set value A is set to the minimum line width of the wiring pattern other than the through-hole area, and the set value B is set to the minimum remaining width of the land. The set values A and B are controlled by a selector 412 by a through-hole area signal synchronized with the timing in the delay memory 404, and when it is a through-hole area, the set value B is input to the comparator 413, and when it is not, the set value A is input to the comparator 413. do. That is, the inspection standard is switched based on the through-hole area detection signal from the through-hole area detection means 106. Further, the branch/termination detection means 416 uses the thinned image 407 to detect branches and terminations that are characteristics of short circuits and disconnections in the wiring pattern. The configuration suppresses false alarms by masking branches that occur in the boundary area of the terminal 417.
Output from

Thinning means 403 and length measuring means 409 in FIG. 4 below.
A specific processing example of the branch/termination detecting means 416 will be explained with reference to FIGS. 6, 6, and 7. FIGS. FIG. The thinning process obtains the core line of the wiring pattern by repeating n times the process of erasing one pixel from the outside of the wiring turn.The general method of the thinning process is shown in Figures 5(a) to (5) below. e) will be explained using °C. A binary image is scanned with a 3 x 3 scanning window as shown in Fig. 6(a), and when the pixel of interest (center pixel of the window) is 1, it corresponds to the state of the pixels in the neighboring 8 pixels d I - d s. It is determined whether the pixel of interest should be converted to 0 (or erased) using a Lutsukua-Tsubuchiful (hereinafter abbreviated as LUT). The deletion determination of the pixel of interest is performed in four steps as shown in FIGS. 6(b) to 6(e). This is to prevent the disappearance of the 7% turn of even pixel width, and LUT(A) to L
Patterns for cutting from the top, bottom, left, and right are registered in the UT (D). Examples of patterns to be registered are shown in Fig. 6(b) to (
Shown in e). Figure 6(f) shows the operation of the thinning process at °C.
I will explain. Input image 5 from first binarization means 104
01 to 1 line delay memory 502 and 3x3 scanning nitrogen 60
3, transfer the data while taking timing with an image signal synchronization clock (not shown), and input the output of the 3×3 scanning window 603 to the LUT(A) 504.
It is determined whether or not to erase the pixel of interest. Similar processes are connected in cascade, and one pixel is thinned by performing four determination processes. This one time thinning processing block is 0
By connecting the wiring pattern in stages, a thinned image with a width of one pixel is generated from the wiring pattern.

Next, a specific processing example of the length measuring means 409 will be explained using FIG. FIG. 6(a) shows reference pixel positions for measuring the width of the wiring pattern from the original image 408 of the printed circuit board inputted to the length measuring means 406 and the thinned image 407 thereof. When the pixel of interest 609 is at the core line position of the thinned image, the number of pixels up to the boundary of the wiring pattern in the original image is measured in 8 directions from direction 1 (601) to direction 5 (eos), and The smallest value among the length measurement results of 8 to 8 is determined to be the length measurement value at the position of interest. 6th
Figure (b) is a diagram illustrating a length measurement calculation method using direction 2 (602) as an example. The number of pixels W is determined by the following formula based on the coordinates of the boundary points 611 and 612 of the wiring pattern in the XY coordinate system with the pixel of interest 610 as the origin.

Although the specific circuit configuration of the above calculation is omitted, (X1X2
) and (y+Y2) as reference addresses, this can be easily realized by preparing in advance the calculation result of the length measurement pixel number W in an LUT. Then, as the length measurement result of the target position, the width (number of pixels) of the wiring pattern in the direction substantially perpendicular to the core line of the thinned image is measured by selecting the minimum value of the length measurement pixel number in each direction.

Next, the specific processing of the branch/termination detection means 416 will be explained using FIG. That is, when the thinned image input to the branch/termination detection means 416 is scanned in a 3×3 window, and the pixel of interest (center position) is at the skeleton position, the thinned skeleton image is detected in the state of 8 neighboring pixels. Figure 7(a) shows a case where there are three or more black isolated areas in an area of 8 neighboring pixels, and Figure 7(b) shows a case where there are three or more black isolated areas in an area of 8 neighboring pixels. As a result of pattern determination, a branch, which is a characteristic of a short circuit, and a termination, which is a characteristic of a disconnection, are detected.

FIG. 8 shows an example of a defect detected by the defect detection means of FIG. 4. In FIG. 8, 801 shows a thinned image of the wiring pattern, and 802 and 803 show through-hole areas obtained by the through-hole area detection means 106. If the reference line width of the land portion is B, and the relationship between a and b in FIG. The position 806 is included in the through-hole area and is detected as a line width violation because it is compared with the reference line width B (・. Also, the position 806 is included in the through-hole area and is compared with the reference line width B. It is compared and detected as a line width violation.
9 and 810 indicate branches of the thinned image, and indicate that branches 807 and 808 occurring in through-hole areas 802 and 803 are not notified. 811.812.81
3 and 814 indicate the end of the thinned image due to wire breakage, and all of them are detected as violations.

More than the effects of the invention, the effects of the present invention include: Binarizing the image of transmitted light through the through-hole of a printed circuit board, and detecting a closed area obtained by expanding the binary image by a predetermined size as the through-hole area.
Inspection can be performed while switching between the reference line width of the through-hole area and the normal conductor section, allowing accurate line width inspection. Furthermore, when inspecting branches and terminations of core wires in wiring patterns, it is possible to control whether or not to detect defects depending on whether or not they occur in the through-hole area, making it possible to perform a variety of defect inspections.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a wiring pattern inspection apparatus according to an embodiment of the present invention, FIG. 2 is a diagram explaining the processing operation of the through-hole area detection means, and FIG. 3 is a scanning window of the through-hole area detection means. FIG. 4 is a block wiring diagram of the defect detection means, FIG. 5 is a diagram explaining the thinning processing procedure of the thinning means, and FIG. 6 is a diagram explaining the processing of the length measuring means. 7 is a diagram showing the judgment pattern of the branch/termination detection means, FIG. 8 is a diagram showing detected defects, and FIG. 9 is a diagram explaining the defect detection procedure in the conventional wiring pattern inspection device. It is a diagram. 10B... dichroic mirror, 109... imaging lens,
110... Ring-shaped light guide, 111... Light guide, 112... First wavelength filter, 113...
...Second wavelength filter, 114...Reflected illumination light, 1
15... Transmitted illumination light, 201... Output image of the second binarization means, 202... MXM scanning window, 203...
Output image of through-hole area detection means, 205... Through-hole image, 206... Expanded through-hole image, 401... Input terminal from W2 binarization means,
402...Input terminal from first binarization means, 407
. . . Thinned image of the wiring pattern, 408 . . . Original image of the wiring pattern, 414 . . . Output terminal for line width violation detection signal, 417 . Name of agent: Patent attorney Shigetaka Awano and one other person Figure 1 = 11+iL! Figure (a) (b) (C) (d) <e) LUT(A+ LUTfB+ LUT(C+ LUT(D+) Figure 6 Figure 9 (a)

Claims (3)

[Claims] (1) A first image input means for illuminating a printed circuit board having a through hole with illumination light in a predetermined wavelength band and photoelectrically converting light reflected from a wiring pattern formed on the circuit board; a second image input means for irradiating with illumination light in a wavelength range separated from the illumination light and photoelectrically converting the light passing through the through hole; and converting the grayscale image from the first image input means into a binary image. a first binarization means for
a second binarization means for converting the grayscale image from the second image input means into a binary image; and a second binarization means for converting the grayscale image from the second image input means into a binary image; through-hole area detection means for detecting an area; and two from the first binarization means and the through-hole area detection means.
A wiring pattern inspection device comprising defect detection means for distinguishing between lands and conductors from a value image signal and detecting characteristics of defects while switching inspection standards. (2) The defect detection means includes a thinning means for thinning the input image, and a length measuring means for measuring the width of the wiring pattern on the core line of the wiring pattern obtained by the thinning. Inspection is performed while switching the inspection reference value to be compared with the length measurement value by the length measuring means between the minimum line width of the normal conductor and the minimum remaining width of the land using the through hole area detection signal from the through hole area detection means of No. 1. Claim 1 characterized by
The wiring pattern inspection device described. (3) The defect detection means includes branch/termination detection means for detecting branches and terminations of the core wire of the wiring pattern obtained by the thinning means of claim 2, and The wiring pattern inspection device according to claim 1 or 2, wherein branching of the core wire occurring in a boundary area between a land portion and a conductor portion is not detected as a defect by the through-hole area detection signal. .