JP2005283583A - System for fault sensing prevention in visual inspection of film or tape shaped printed circuit board, and its processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a system for fault sensing prevention of an optical film and tape-shaped printed circuit board, and to provide its testing method. <P>SOLUTION: This automatic optical inspection system determines the presence of pattern defects in brightness of image data captured by the printed circuit board, by utilizing transmitted light and/or reflected light, through the 1st reference value of an illuminometer. Then it uses the 2nd brightness illuminometer reference value to determine whether the unit determined as defective actually has any pattern defect or non-defective foreign material. Thus, it uses subdivided reference values to determine the unit actually fault-sensed as defective due to non-defective foreign material to be acceptable, not the actual defective unit, preventing fault sensing to enhance inspection efficiency in the testing processes. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an automatic optical inspection system (Automated Optical Inspection system, AOI system), and more specifically, when a visual inspection of a printed circuit board (Printed Circuit Board, PC) in the form of a film and a tape is a defective product. The present invention relates to a system for preventing detected overdetection and a processing method thereof.

  Printed circuit boards, which are one of the main materials used in the manufacture of various semiconductor integrated circuits such as semiconductor devices such as driver integrated circuits (LCD Driver ICs) for liquid crystal display devices and memories, are films and tapes. (Tape) and the like.

  Examples of such a printed circuit board in the form of a film and a tape include a TAB (Tape Automatic Bonding) or COF (Chip On Film) substrate. A fine pattern is formed on the printed circuit board through a manufacturing process such as exposure and development in the form of a film or a tape substrate, and whether the pattern shape is good or bad is a very important factor for the productivity of the printed circuit board.

  In addition, when automatically inspecting the appearance of printed circuit boards in the form of films and tapes using light, over-detection that is determined to be defective due to foreign matters such as dust is acceptable as a product and automatic optics. It is an important factor that has an absolute influence on the reliability of the inspection system.

  In the industry that produces printed circuit boards in the form of films and tapes for manufacturing semiconductor integrated circuits, various defects such as merging, short-circuiting, protrusions, and dents in fine patterns of products have caused major problems. In fact, effective inspection of this is the key to productivity and quality control.

  For this purpose, one human inspection is gradually becoming difficult because the printed circuit board in the form of a film or tape is manufactured into an ultra-high fine pattern, and effective inspection becomes increasingly difficult. As a result, the demand for automatic inspection equipment by optical image acquisition has increased rapidly, making it an indispensable facility.

However, a general automatic inspection apparatus acquires an optical image and processes it by an image processing device and a computer to determine whether the pattern is good or bad, and there is a large overdetection phenomenon in which a non-defective product is judged as defective. Cause problems. This is because the manufactured product is automatically determined as defective and discarded, and the productivity is greatly reduced. This leads to a deterioration in profitability of manufacturing companies.

  1A to 1D are diagrams illustrating protrusions, alignments, or foreign matters determined from video data acquired using light transmitted through a printed circuit board in a film and tape form. 2A to 2B are diagrams illustrating protrusions or foreign matters determined from video data acquired using reflected light of a printed circuit board in a film and tape form.

  Referring to FIGS. 1A and 1D, image data 2 of a printed circuit board in the form of a film and a tape obtained by using a transmitted light by a general automatic optical inspection system (not shown) is a real lead 4. The image data 2 for the projection 8 and the merge line 10 and the projection 12 and the merge line 14 due to foreign matter or dust in FIGS. 1C and 2B are different in their illuminance meters. Here, the lead image 4 of the printed circuit board is displayed as black, and the space 6 is displayed as white.

  However, the protrusion 8 and the joining line 10 due to the actual manufacturing defect in FIGS. 1A and 1B, and the non-defective product that looks like the protrusion and the joining line shown in FIG. 1C and FIG. The sexual projections 12 and the merged line 14 are identified as the projections and merged lines that are projected from the lead 4 by the automatic optical inspection system, and are all processed as pattern defects.

  Referring to FIG. 2A, the image data acquired using the reflected light is a non-defective protrusion 18 that looks like an actual protrusion due to dust or foreign matter in FIG. 2B as a protrusion 16 due to an actual manufacturing defect. And are all treated as defective.

  As described above, the optical automatic inspection system introduced in the conventional printed circuit board manufacturing enterprises in the form of film and tape acquires images through transmitted light or reflected light. This is a method for judging good or bad. In such a system, when non-defective foreign matter and dust are present in the circuit, images such as those shown in FIGS. 1C, 1D, and 2B are obtained, causing over-detection. Therefore, the over-detection phenomenon that occurs during the inspection process of the printed circuit board in the form of film and tape causes a decrease in productivity and a cost increase.

  An object of the present invention is to solve the above-described problems, and prevents overdetection in which a non-defective product due to foreign matter or dust is determined as defective in a pattern to be inspected on a printed circuit board in the form of a film and a tape. It is to provide an inspection system and method thereof.

  Another object of the present invention is to solve the above-mentioned problems, and the optical brightness is used for visual inspection of printed circuit boards in the form of films and tapes using the characteristics of illuminometers. It is an object of the present invention to provide an inspection system and an inspection method for preventing an overdetection phenomenon that sometimes occurs.

  Another object of the present invention is to solve the above-mentioned problems, and in an automatic optical inspection system for printed circuit boards in the form of films and tapes, it is possible to prevent overdetection in which a non-defective product is determined as defective. It is to realize an apparatus and a method for improving productivity.

  According to an embodiment of the present invention for achieving the above object, an automatic optical inspection system for inspecting whether or not a circuit pattern of a plurality of printed circuit boards having circuit patterns and spaces is defective uses a light source. An automatic optical inspector that acquires image data corresponding to each printed circuit board, and the image data is received from the automatic optical inspector, and the brightness of each pixel unit of the image data is determined by the illuminometer. A control unit that determines the quality of the printed circuit board according to a first brightness level that distinguishes a pattern and the space, and the control unit includes the control unit that determines the quality of the printed circuit board according to the first brightness level. If the circuit pattern is determined as defective, the circuit pattern determined as defective using the second brightness level corresponding to the predetermined actual defect pattern is It is determined whether failure whether it is a time secondarily.

  With such a feature, if the light source is transmitted light, the control unit has the second brightness level larger than the first brightness level, and the brightness level of the circuit pattern determined as defective is the If it is between the first and second brightness levels, the circuit pattern is determined to be defective due to non-defective foreign matter, and the printed circuit board determined to be defective is determined to be non-defective. Here, if the brightness level of the circuit pattern determined to be defective is equal to or less than the second brightness level, the control unit determines the circuit pattern as an actual defect and prints the defect determined The circuit board is determined as defective. Here, if the brightness level of the circuit pattern determined as the failure is equal to or higher than the second brightness level, the control unit determines the circuit pattern as an actual failure and determines the failure as the failure. It is desirable to identify a printed circuit board as defective.

  According to these features, the control unit includes a computer system, and the computer system mutually transmits the video data and the secondary discrimination information through the automatic optical inspection device and a computer network.

  Due to these characteristics, the control unit includes a program for extracting a brightness level of the video data in units of pixels.

  According to another aspect of the present invention for achieving the above object, an automatic optical inspection system for inspecting whether or not a circuit pattern of a large number of printed circuit boards has a defect generates a transmitted light from a light source, An automatic optical inspection device that acquires image data corresponding to a printed circuit board, and the image data is received from the automatic optical inspection device, and the brightness of each pixel unit of the image data is used by a first reference value by an illuminometer. And a control unit that determines whether the printed circuit board is good or bad, and the control unit sets the brightness of the printed circuit board that has been determined as defective by the first reference value to a brightness that is less than the first reference value. The second reference value is used to secondarily determine whether the printed circuit board determined as defective is actually defective.

  Due to these characteristics, the control unit determines that the defect in the circuit pattern is a non-defective foreign substance if the brightness level of the circuit pattern of the printed circuit board determined as the defect is between the first and second reference values. And the printed circuit board determined as defective is determined as a non-defective product. Here, if the brightness level of the circuit pattern determined as the defect is equal to or less than the second reference value, the control unit determines the circuit pattern as an actual defect and the printed circuit determined as the defect. The substrate is determined as an actual defect.

  According to still another aspect of the present invention for achieving the above-described object, an automatic optical inspection system for inspecting whether or not a circuit pattern of a large number of printed circuit boards is defective may generate reflected light from a light source, An automatic optical inspection device for acquiring image data corresponding to the printed circuit board, and the image data is received from the automatic optical inspection device, and the brightness of each pixel unit of the image data is determined by a first reference value by an illuminometer. A control unit that determines whether the printed circuit board is good or bad, and the control unit determines that the printed circuit board that has been determined as defective by the first reference value has a brightness that is greater than the first reference value and that is a second reference of the illuminometer. Using the value, it is secondarily determined whether the printed circuit board determined to be defective is actually defective.

  Due to these characteristics, the control unit determines that the defect in the circuit pattern is a non-defective foreign substance if the brightness level of the circuit pattern of the printed circuit board determined as the defect is between the first and second reference values. And the printed circuit board determined as defective is determined as a non-defective product. The control unit determines the circuit pattern as an actual defect if the brightness level of the circuit pattern determined as the defect is equal to or higher than the second reference value, and determines the printed circuit board determined as the defect. It is determined as an actual defect.

  According to still another aspect of the present invention for achieving the above-described object, an over-detection for discriminating a non-defective product as a defect in an automatic optical inspection system for inspecting a circuit pattern of a printed circuit board in a film and tape form is prevented. An inspection method for obtaining the printed circuit board using a light source, obtaining video data corresponding to the printed circuit board, and converting the video data according to a brightness level in pixel units. Determining the quality of the printed circuit board based on a first reference value for identifying the printed circuit board, and determining whether the printed circuit board is defective based on the first reference value based on a second reference value of a brightness level corresponding to an actual defective pattern. Overdetection is prevented including the step of determining the actual defective state.

  According to such a feature, in the step of determining the defective state, when the light source is transmitted light, the second reference value has a brightness level smaller than the first reference value, and the circuit is determined to be defective. If the brightness level of the pattern is between the first and second reference values, the printed circuit board determined as defective is determined as a non-defective product due to non-defective foreign matter. The step of determining the defective state includes determining a defective printed circuit board if the brightness level of the circuit pattern of the printed circuit board determined to be defective by the first reference value is smaller than the second reference value. Is determined as an actual defect.

  According to such a feature, in the step of determining the defective state, when the light source is reflected light, the second reference value has a brightness level greater than the first reference value, and the circuit is determined to be defective. If the brightness level of the pattern is between the first and second reference values, the printed circuit board determined as defective is determined as a non-defective product due to non-defective foreign matter. In the step of determining the defective state, if the brightness level of the circuit pattern of the printed circuit board determined to be defective by the first reference value is greater than the second reference value, the printed circuit board determined to be defective is actually used. It is determined as a defect.

  According to the present invention, by using the subdivided reference threshold value, it is not actually a defective unit, but a unit that has been overdetected as a defect due to a foreign object is determined as a non-defective product, thereby preventing overdetection in the inspection process. Inspection efficiency and productivity can be improved.

  Embodiments of the present invention will be described below in detail with reference to the accompanying drawings.

  Specifically, FIG. 3 shows the configuration of an automatic optical inspection system according to the present invention.

  Referring to FIG. 3, the automatic optical inspection system 100 includes an automatic optical inspector (AOI) 102 and a computer system 170 capable of data communication with the AOI 102.

  The automatic optical inspection device (AOI 102) includes a vision inspection unit 120 and a defect processing unit 130 between an unwinding unit 110 and a winding unit 140, and various operations of these components. An AOI control unit 150 is further included to control the above.

  The unwinding unit 110 provides the vision inspection unit 120 with a printed circuit board in which a printed circuit board in the form of a film or a tape is continuously in a reel or roll form.

  The vision inspection unit 120 images the printed circuit board provided from the unwinding unit 110 and provides it to the AOI control unit 150. For example, the image data for the fine circuit pattern of the printed circuit board unit is acquired using a transmission light source or a reflection light source and at least one CCD camera, and the acquired image data is output to the AOI controller 150.

  The defect processing unit 130 performs punching or marking to display a defective printed circuit board that is finally determined to be defective as a result of the inspection, and outputs the result to the winding unit 140.

  The winding unit 140 receives and rewinds the printed circuit board that has been inspected from the vision inspection unit 120 through the defect processing unit 150.

  The AOI control unit 150 controls various operations of the AOI 102, receives the video data captured from the vision inspection unit 120, and the computer system through an interface (eg, LAN, RS-232C, etc.) for data communication. 170, and receives inspection information from the computer system 170 to determine whether it is acceptable or not, thereby driving the components 110, 120, 130, or 140 of the AOI 102.

  The computer system 170 includes components of a typical computer system, as shown in FIG. 4, as a main control unit for controlling various operations of the inspection process according to the present invention. The computer system 170 includes a control unit 172, a display unit 174, an input unit 176, an interface unit 178, and a storage unit 180. In addition, the computer system 170 is provided independently of the AOI 102 as shown in the drawing, but may be provided in a batch form integrated with the AOI 102.

  The control unit 172 includes, for example, a central processing unit, a control program, and the like, receives video data captured from the AOI 102, converts it into histogram data, and displays it. The operator uses the video data 182 and the histogram. Control to monitor the brightness level of the illuminometer. That is, the controller 172 is processed according to the operation procedure shown in FIG. 5 in order to improve the yield of optical inspection.

  The display unit 174 displays the video data 182 and displays histogram data on the illuminance meter in brightness in units of pixels.

  The input unit 176 outputs an input signal for selecting, inspecting, editing, and confirming the video data 182 in pixel units.

  The interface unit 178 accepts video data captured from the AOI 102 as a device for mutual data communication with the AOI 102 using, for example, a communication protocol such as LAN and RS-232, and sends information based on the inspection result to the AOI 102. Output.

  The storage unit 180 stores a histogram analysis program 186, video data 182 and an editing program 184 according to the present invention. The histogram analysis program 186 receives the video data 182 and converts the data into a histogram format, and outputs the brightness of each pixel for determining the quality of the pattern from the acquired video data to the illuminometer. In addition, the histogram analysis program 186 includes a second threshold value corresponding to the first reference value by the control unit 172 based on a threshold value based on the illuminance meter brightness level, that is, a first reference value and a light source, that is, transmitted light or reflected light. The quality of the pattern is determined from the video data using the reference value. At this time, the operator can check the brightness level of the illuminometer with respect to the specific portion of the displayed printed circuit board unit.

  Accordingly, the computer system 170 is an automatic optical inspection system that inspects a printed circuit board in the form of a film or a tape using an optical camera. The computer system 170 acquires an image using transmitted light or reflected light and is defined as a good product. Pass / fail is judged against the standard. At this time, the pattern component is recognized based on a specific threshold value (Threshold Value, for example, illuminance meter 120 level) that is divided into a pattern and a space by defective non-defective patterns or non-defective foreign matter and dust between patterns. become.

  However, as shown in FIGS. 1C and 1D, defects due to non-defective foreign substances are classified in the pattern area by a specific threshold value that divides the pattern and space. Displayed on an optical illuminometer (for example, 70 levels or more). As shown in FIGS. 1A and 1B, in the case of an actual pattern defect, it can be seen that the optical illuminometer (for example, 70 levels or less) is relatively different from the threshold value.

  Therefore, the present invention discriminates the quality by further subdividing the area divided into the pattern areas using the specific threshold value for dividing the pattern and the space as a reference value. That is, it is recognized as a pattern area that is lower than the threshold value, but the optical illuminometer determines that a relatively close fixed area is detected as a non-defective foreign substance and dust, and on the contrary, If the illuminance is similar to the optical illuminance of the actual pattern, it is regarded as an actual circuit failure.

  Further, it is obvious that the same principle can be applied to this portion even when reflected light is used. For example, the presence or absence of a pattern defect is determined using the first reference value of the illuminance meter 120 level, and the unit determined to be defective is determined again whether or not the second reference value illuminance meter level is 200 or more. If it is more than the above, it is determined as a defective unit such as a pattern protrusion or merge line. If the second reference value illuminometer is 200 levels or less, it is determined as a non-defective product due to non-defective foreign substances.

  As a result, by detecting only actual pattern defects through video data acquired using transmitted light and / or reflected light, it is possible to prevent overdetection in which an actual non-defective product is determined to be defective, thereby improving productivity.

  FIG. 5 is a flowchart showing an operation procedure of the automatic optical inspection system according to the embodiment of the present invention. This procedure is a program processed by the computer system 170, and this program is stored in the storage unit 180.

  Referring to FIG. 5, if the flexible printed circuit board is imaged from the AOI 102 in step S200, the computer system 170 acquires video data 182 through an image processing process in step S202. The video data acquired in step S204 is analyzed to determine the quality of the unit using the first reference value. That is, it is determined whether or not a defect has occurred in the unit pattern using the histogram type illuminometer with the first reference value of the specific threshold as a reference.

  Here, the histogram is a method of representing the distribution of light and dark values with respect to pixels in video data. In the normal histogram, the X-axis displays the brightness value, and the Y-axis displays the frequency number. The X-axis value that displays the light and dark numerical values in the strogram is usually divided into 0 to 255 levels. For example, if the histogram graph is close to the 0 level, the lightness is low, and the 255 level indicates that the lightness is high. Hereinafter, in the present invention, video data acquired through a histogram using transmitted light will be analyzed and described.

  Accordingly, in step S206, for example, the first reference value is determined using a reference value of the illuminometer 120 level to determine a primary pattern defect.

  Subsequently, the video data determined to be defective by the first reference value in step S208 indicates that the unit determined to be defective using the second reference value actually has a defective pattern, or overdetection due to foreign matter or the like. Is determined.

  If it is determined that the level is equal to or lower than the second reference value, the procedure proceeds to step S210 and is determined as an actual defective pattern such as merging and / or protrusion. Subsequently, in step S212, the corresponding unit is determined as defective.

  If the illuminance level exceeds the second reference value in step S208, the procedure proceeds to step S214 and is determined as a non-defective foreign substance, and the corresponding unit is determined as a non-defective product in step S216.

  As described above, the automatic optical inspection system of the present invention determines the quality of the secondary pattern of the unit determined to be the primary failure by using different reference values of the illuminometer.

  Next, FIGS. 6 to 8 show analysis screens using the histogram method according to the embodiment of the present invention. These screens are output to the display device 174 of the computer system 170, and are driven by a histogram analysis program 186 for analysis at the illuminometer level.

  Referring to FIG. 6, the screen 400 is divided into a lead 420 and a space 422 as a view showing a part of the video teeth 182 imaged from the automatic optical inspector 102, and an actual defect pattern 404 such as a protrusion or a joint line. The inspection result by the foreign matter 406 is output. On one side, a distribution diagram 410 of the illuminometer is shown using a histogram method.

  Referring to FIG. 7, the threshold screen is shown in units of pixels of video data based on a threshold in order to determine the quality of the pattern from the screen of FIG. 6. This shows the lead 420 and the space 422 separately according to the threshold value, and shows the inspection result of the protrusion, the joining line 404 'and the foreign matter 406'. Here, if a pixel at a specific position is selected, as shown in FIG. 8, the illuminance level of the region 412 including the corresponding pixel of protrusion, alignment, or foreign matter is output through the auxiliary window 430, and through this, the illuminance level on each pixel is output. The brightness can be confirmed by the illuminance level.

It is a figure which shows the protrusion, joining, or foreign material judged from the video data acquired using the transmitted light of the printed circuit board of a film and a tape form. It is a figure which shows the protrusion, joining, or foreign material judged from the video data acquired using the transmitted light of the printed circuit board of a film and a tape form. It is a figure which shows the protrusion, joining, or foreign material judged from the video data acquired using the transmitted light of the printed circuit board of a film and a tape form. It is a figure which shows the protrusion, joining, or foreign material judged from the video data acquired using the transmitted light of the printed circuit board of a film and a tape form. It is a figure which shows the processus | protrusion or foreign material judged from the video data acquired using the reflected light of a printed circuit board. It is a figure which shows the processus | protrusion or foreign material judged from the video data acquired using the reflected light of a printed circuit board. It is the block diagram which showed the structure of the automatic optical inspection system by embodiment of this invention. FIG. 4 is a block diagram illustrating a configuration of the computer system illustrated in FIG. 3. 3 is a flowchart showing an operation procedure of the automatic optical inspection system according to the embodiment of the present invention. It is a figure which shows the state for applying the histogram system by embodiment of this invention, and determining the defect of a pattern. It is a figure which shows the state for applying the histogram system by embodiment of this invention, and determining the defect of a pattern. It is a figure which shows the state for applying the histogram system by embodiment of this invention, and determining the defect of a pattern.

Explanation of symbols

100 Inspection system 102 AOI
DESCRIPTION OF SYMBOLS 110 Unwinding part 120 Vision inspection part 130 Defect processing part 140 Winding part 150 AOI control part 170 Computer system

Claims (18)

In an automatic optical inspection system for inspecting whether or not the circuit pattern is defective on a large number of printed circuit boards having circuit patterns and spaces,
An automatic optical inspection device that acquires image data corresponding to each of the printed circuit boards using a light source;
The printed circuit board is accepted or rejected by receiving the image data from the automatic optical inspection device, and determining the brightness of each pixel unit of the image data by a first brightness level that distinguishes the circuit pattern and the space by an illuminometer. And a control unit for determining
If the circuit pattern of the printed circuit board is determined to be defective according to the first brightness level, the controller uses the second brightness level corresponding to a predetermined actual defect pattern as the defect. An automatic optical inspection system that secondarily determines whether or not the determined circuit pattern is an actual defect. The controller is
If the light source is transmitted light, the second brightness level is less than the first brightness level;
If the brightness level of the circuit pattern determined as defective is between the first and second brightness levels, it is determined that the circuit pattern defect is due to a non-defective foreign substance, and the defect determination is performed. The automatic optical inspection system according to claim 1, wherein the printed circuit board is determined as a non-defective product. The controller is
If the brightness level of the circuit pattern determined as defective is equal to or lower than the second brightness level, the circuit pattern is determined as an actual defect and the printed circuit board determined as defective is determined as defective. The automatic optical inspection system according to claim 2, wherein: The controller is
If the light source is reflected light, the second brightness level is greater than the first brightness level;
If the brightness level of the circuit pattern determined as defective is between the first and second brightness levels, it is determined that the defect in the circuit pattern is due to a non-defective foreign substance, and the defect 2. The automatic optical inspection system according to claim 1, wherein the determined printed circuit board is determined as a non-defective product. The controller is
If the brightness level of the circuit pattern determined as defective is equal to or higher than the second brightness level, the circuit pattern is determined as an actual defect, and the printed circuit board determined as defective is determined as defective. The automatic optical inspection system according to claim 4, wherein: The control unit includes a computer system,
The automatic optical inspection system according to claim 1, wherein the computer system mutually transmits the video data and the secondary discrimination information through the automatic optical inspection device and a computer network. The controller is
The automatic optical inspection system according to claim 1, further comprising a program for extracting a brightness level in pixel units of the video data. In an automatic optical inspection system for inspecting whether or not a circuit pattern of a large number of printed circuit boards is defective,
An automatic optical inspection device that generates transmitted light from a light source and acquires image data corresponding to each of the printed circuit boards;
A controller that accepts the image data from the automatic optical inspection device, and determines the quality of the printed circuit board by using a first reference value by an illuminometer for brightness of each pixel of the image data;
The printed circuit board determined to be defective by using the second reference value of an illuminometer having a brightness smaller than the first reference value for the printed circuit board determined to be defective by the first reference value. An automatic optical inspection system characterized in that it is secondarily determined whether or not is actually defective. The controller is
If the brightness level of the circuit pattern of the printed circuit board determined as the defect is between the first and second reference values, it is determined that the defect in the circuit pattern is due to a non-defective foreign substance, 9. The automatic optical inspection system according to claim 8, wherein the printed circuit board determined as defective is determined as a non-defective product. The controller is
If the brightness level of the circuit pattern determined as defective is equal to or less than the second reference value, the circuit pattern is determined as an actual defect, and the printed circuit board determined as the defect is determined as an actual defect. The automatic optical inspection system according to claim 9. In an automatic optical inspection system for inspecting whether or not a circuit pattern of a large number of printed circuit boards is defective,
An automatic optical inspection device that generates reflected light from a light source to acquire image data corresponding to each of the printed circuit boards;
A controller that accepts the video data from the automatic optical inspection device, and determines the quality of the printed circuit board based on a first reference value by an illuminometer for brightness of each pixel unit of the video data;
The printed circuit board that has been determined to be defective by using the second reference value of an illuminometer that has a brightness greater than the first reference value for the printed circuit board that has been determined to be defective by the first reference value. An automatic optical inspection system characterized in that it is secondarily determined whether or not is actually defective. The controller is
If the brightness level of the circuit pattern of the printed circuit board determined as the defect is between the first and second reference values, it is determined that the defect in the circuit pattern is due to a non-defective foreign substance, and the defect The automatic optical inspection system according to claim 11, wherein the printed circuit board determined as is determined as a non-defective product. The controller is
If the brightness level of the circuit pattern determined as defective is equal to or higher than the second reference value, the circuit pattern is determined as an actual defect, and the printed circuit board determined as the defect is determined as an actual defect. The automatic optical inspection system according to claim 12. In an inspection method for preventing over-detection in which a non-defective product is determined as defective in an automatic optical inspection system that inspects a circuit pattern of a printed circuit board in a film and tape form,
Obtaining the printed circuit board using a light source;
Obtaining video data corresponding to the printed circuit board;
Determining the quality of the printed circuit board according to a first reference value for identifying the circuit pattern according to the brightness level in pixel units of the video data;
A step of determining an actual failure state of the printed circuit board determined to be defective based on the first reference value based on a second reference value that is a brightness level corresponding to an actual defect pattern, thereby preventing over-detection. Inspection method for an automatic optical inspection system characterized by the above. The step of determining the defective state includes
When the light source is transmitted light, the second reference value has a brightness level smaller than the first reference value;
If the brightness level of the circuit pattern determined as defective is between the first and second reference values, the printed circuit board determined as defective is determined as a non-defective product due to non-defective foreign matter. The inspection method of the automatic optical inspection system according to claim 14. The step of determining the defective state includes
If the brightness level of the circuit pattern of the printed circuit board determined as defective by the first reference value is smaller than the second reference value, the printed circuit board determined as defective is determined as an actual defect. The inspection method of the automatic optical inspection system according to claim 15, wherein The step of determining the defective state includes
When the light source is reflected light, the second reference value has a brightness level greater than the first reference value;
If the brightness level of the circuit pattern determined as defective is between the first and second reference values, the printed circuit board determined as defective is determined as a non-defective product due to non-defective foreign matter. The inspection method of the automatic optical inspection system according to claim 14. The step of determining the defective state includes
If the brightness level of the circuit pattern of the printed circuit board determined to be defective by the first reference value is larger than the second reference value, the printed circuit board determined to be defective is determined to be an actual defect. An inspection method for an automatic optical inspection system according to claim 17.

Images