JP2010117285A - Defect inspection device for substrate - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a defect inspection device for a substrate capable of reducing maximumly labor and a time for determining authenticity of defect candidates enumerated by an AOI (Automated Optical Inspection) device. <P>SOLUTION: A user selects one defect candidate based on characteristic information on each defect candidate displayed as a guide display, and observes an actual image of the defect candidate on a screen of an image display, to thereby enable discrimination whether the selected defect candidate is a pseudo defect or a true defect. Further, characteristic information generated and preserved by the AOI device includes a local domain image including each image of the defect candidates, and a defect candidate display means displays the local domain image corresponding to the defect candidate on the screen of the image display relative to each defect candidate. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention inspects defects in a display panel such as a liquid crystal display panel and a semiconductor substrate in which an integrated circuit is built using an optical automatic inspection apparatus (hereinafter referred to as “AOI (Automated Optical Inspection) apparatus”). The present invention relates to a substrate inspection apparatus.

  In recent years, in the field of manufacturing substrates for liquid crystal display panels, semiconductor integrated circuits, etc., it is more cost-effective to produce a modified substrate than to discard it, as the substrate becomes larger. It is becoming. For this reason, various defect inspection apparatuses and inspection result display apparatuses for inspecting defects on a substrate by an optical technique or an electric technique have been developed prior to repair processing (see, for example, Patent Documents 1 and 2).

  Since defects that are the targets of this type of defect inspection equipment are extremely fine, there is a limit to the reliability of the inspection results, and the inspection results indicate defects (hereinafter referred to as “defect candidates”). ) Generally includes a large number of non-defects (hereinafter referred to as “pseudo defects”). Therefore, it is required to efficiently identify such a pseudo defect and a true defect that needs to be corrected, and to correct the true defect reliably and quickly.

  Conventionally, a defect inspection of a substrate such as a liquid crystal panel is usually performed in two stages, a first stage inspection by a continuity inspection apparatus and a second stage inspection by visual inspection. Here, the visual inspection is usually performed by a technique in which each defect candidate obtained by the continuity inspection apparatus is electronically photographed with a microscope camera and the obtained image is observed on the screen of the image display. Is called. However, since visual inspection can cover only a field of view of about several hundred microns in one inspection, the time required for inspection becomes enormous as the substrate becomes larger, and there is a problem in work efficiency.

  Therefore, in recent years, instead of visual inspection, by introducing inspection using an AOI apparatus, this type of substrate inspection is performed in two stages: a first stage inspection using a continuity inspection apparatus and a second stage inspection using an AOI apparatus. It has been proposed to do.

  An example of the processing of the substrate repair apparatus incorporating the conventional AOI apparatus is shown in the flowchart of FIG. Although not shown in the drawings, this substrate repair apparatus irradiates a repair target portion on the substrate with a laser beam and repairs it, and includes a stage for mounting the substrate to be inspected, A microscope camera for photographing a circuit pattern carrying surface of a substrate to be inspected placed on the image processing apparatus, an image display functioning as a human machine interface, and a field scanning mechanism for scanning the field of view of the microscope camera on the circuit pattern carrying surface. It has.

  In the figure, when processing is started, the AOI apparatus is activated after waiting for an AOI command from the outside (step 401). Then, the AOI device is activated and a predetermined AOI operation is performed (step 402).

  The processing of the AOI device is shown in the flowchart of FIG. In the figure, in step 501, a peripheral image of a predetermined range including a designated position (defect candidate position) on the inspection target substrate obtained through the microscope camera is obtained and stored.

  In subsequent step 502, a small image including a circuit pattern serving as a normal reference is determined for each inspection region from the stored peripheral images according to a predetermined algorithm. In the subsequent step 503, a small image of the corresponding inspection target region is cut out from the acquired peripheral images.

  In subsequent step 504, the small image serving as the normal reference and the small image in the inspection target area are subjected to pattern matching. In the subsequent step 505, it is determined whether or not the collation does not coincide with a predetermined allowable range. Here, if it is determined that the two do not match (YES in step 505), it means that there is a high possibility that the small image to be inspected contains a defect. On the other hand, if it is determined that they match (NO in step 505), it means that there is a high possibility that the small image to be inspected does not contain a defect.

  If it is determined in step 505 that the two do not match (YES in step 505), the process proceeds to step 506. In step 506, based on the small image to be inspected that is not matched, feature information relating to the defect included therein is generated and stored in a predetermined memory. This feature information is used to create a defect candidate list described later.

  Thereafter, the above process (steps 501 to 506) is repeated, and the process ends after waiting for the inspection of the entire acquired peripheral image (step 507 YES). If the reference small image needs to be changed due to a pattern difference or the like in the acquired peripheral image (step 508 YES), a new reference small image is determined according to a predetermined algorithm (step 502). .

  As described above, the processing of the conventional AOI apparatus acquires the predetermined peripheral image for each of the designated defect candidates via the microscope camera, and for each inspection target region from the acquired peripheral images. A small image as a normal reference is determined, and the small image as a reference is compared with a small image to be inspected cut out from the acquired peripheral image, and a defect candidate is identified based on a mismatch between the two. Feature information is generated and stored.

  Returning to FIG. 7, in step 403, the inspection result, that is, the feature information of each defect candidate generated and stored is acquired from the AOI apparatus. In the next step 404, the defect feature list is displayed in a predetermined display format on the screen of the image display, using the acquired feature information as a guide display for assisting narrowing down defect candidates.

  An explanatory diagram showing an example of the defect candidate list is shown in FIG. In this example, the defect candidate list includes a defect No. For each defect specified in (2), its X coordinate, Y coordinate, defect size, and correlation rate with the reference image are described.

  Thereafter, the candidate selection operation by the operator is waited with the defect candidates still displayed (step 405). In this state, when one of the defect candidates is selected by the operator from the defect candidate list displayed on the screen (step 405 “Yes”), the XY coordinates of the selected defect candidate are read. As the target position, the visual field scanning mechanism is servo-controlled, so that the visual field of the microscope camera is automatically moved to the target position (step 406).

  In the subsequent step 407, the image of the field of view of the microscope camera (actual image) is displayed on the screen of the image display device after waiting for the field of view of the microscope camera to reach the target position. In the subsequent step 408, repair by the operator is performed. Wait for operation.

  In this state, the operator observes the actual image of the defect candidate displayed on the screen of the image display, so that it is a pseudo defect that does not require repair processing or a true defect that requires repair processing. Can be determined.

  If it is determined in step 408 that the repair operation by the operator is “present”, in step 409, a known laser repair process is performed in which laser irradiation is performed while aiming at a predetermined position in the field of view of the microscope camera.

  As described above, in the processing of the substrate repair apparatus in which the conventional AOI apparatus is incorporated, the user is 1 based on the feature information (defect candidate list) regarding each defect candidate displayed on the screen as a guide display. By selecting the defect candidate and observing the actual image of the defect candidate on the screen, it is possible to determine whether the selected defect candidate is a pseudo defect or a true defect.

This kind of AOI apparatus performs the true / false discrimination by observing the actual image in this way because of the determination principle that the defect recognition is performed based on the mismatch between the reference circuit pattern and the circuit pattern to be inspected. The defect candidates obtained must include not only true defects that are defective in circuit operation but also pseudo defects that are not problematic in circuit operation. Therefore, the more accurate the abnormality such as disconnection or foreign matter contamination is, the more the circuit pattern that does not have a problem in the circuit operation that the line width is slightly different or the entire pattern is distorted. This is because the possibility of being listed as a defect candidate increases.
JP 2006-133670 A JP-A-2005-24312

  However, the conventional guide display for assisting narrowing down defect candidates is to express the feature information of each defect candidate by numerical values such as coordinates, size, correlation rate, etc., as in the defect candidate list shown in FIG. Therefore, it is very difficult to accurately determine whether it is a pseudo defect or a true defect using only those display items. Eventually, all defect candidates or the majority of defect candidates are observed on the screen. There is a problem that it takes a lot of time and labor for authenticity determination.

  The present invention has been made paying attention to the above-mentioned problems, and the object of the present invention is to reduce as much as possible the labor and time for determining the authenticity of defect candidates listed in the AOI apparatus. An object of the present invention is to provide a defect inspection apparatus for a substrate that can be performed.

  Other objects and effects of the present invention will be easily understood by those skilled in the art by referring to the following description of the specification.

  It is considered that the above-mentioned “problem to solve the invention” can be solved by a substrate defect inspection apparatus having the following configuration.

  This substrate defect inspection apparatus functions as a stage for mounting a substrate to be inspected, a microscope camera for photographing a circuit pattern carrying surface of the inspection target substrate placed on the stage, and a human machine interface And a visual field scanning mechanism for scanning the visual field of the microscope camera on the circuit pattern carrying surface.

  The defect inspection apparatus for a substrate further includes an AOI apparatus, defect candidate display means, and actual image display means.

  The AOI apparatus scans the field of view of the microscope camera via the field-of-view scanning mechanism in response to a predetermined AOI command including the position designation, thereby allowing a predetermined peripheral range of the designated position on the circuit pattern carrying surface. And detecting one or more defect candidates by region-based pattern matching between the acquired image of the predetermined peripheral range and the unit region image serving as a normal reference, and feature information about each of the defect candidates is obtained. Generate and save.

  The defect candidate display means acquires feature information about each defect candidate generated and stored in the AOI device, and displays it as a guide display for supporting defect candidate narrowing down in a predetermined display format on the screen of the image display device. indicate.

  The real image display means moves the visual field of the camera through the visual field scanning mechanism by manual operation or automatic control according to the teaching of the target position, and displays the real image of the visual field of the camera of the image display. Display on the screen.

  Therefore, the user selects one defect candidate based on the feature information regarding each defect candidate displayed as the guide display, and observes the actual image of the defect candidate on the screen of the image display, It is possible to determine whether the selected defect candidate is a pseudo defect or a true defect.

  In the basic configuration described above, the feature information generated and stored by the AOI device includes a local area image including each image of the defect candidate, and the defect candidate display means is provided for each defect candidate. The local area image corresponding to the defect candidate is displayed on the screen of the image display.

  According to such a configuration, the operator observes the local area image corresponding to each defect candidate displayed on the screen as the guide display, and the reason why the image is listed as the defect candidate is the disconnection or An abnormality that does not lead to circuit malfunction such as contamination of the circuit pattern that is likely to occur due to a difference in line width or a circuit pattern that is likely to occur at the edge of the board, etc. Since it can be recognized with a certain degree of certainty because it exists, defect candidates that are clearly identified as not leading to defective circuit operation are excluded from the object of actual image observation by the AOI device. It is possible to reduce as much as possible the labor and time for determining the authenticity of the listed defect candidates.

  At this time, as to how the local area image is displayed on the defect candidate display means, the local area image is displayed on the display screen one by one so as to turn the page according to a predetermined increment operation. However, if a list of local area images corresponding to each of a plurality of defect candidates is displayed on the screen of the image display device, which defect candidate is likely to correspond to a true defect, Alternatively, it is possible to recognize at a glance which defect candidate is highly likely to correspond to a pseudo defect.

  In addition, when a predetermined user operation for performing laser repair is performed during display of the actual image by the actual image display means, a laser that performs laser repair processing by aiming at a predetermined position in the field of view of the camera You may make it further have a repair execution means.

  According to such a configuration, it is possible to improve the work efficiency in the substrate repair work by reducing as much as possible the labor and time for determining the authenticity of the defect candidates mentioned in the AOI apparatus.

  Furthermore, the above-described substrate defect inspection apparatus detects one or more defect candidates through circuit function inspection such as continuity inspection, and generates and stores approximate position information of the defect candidates; A defect candidate preliminary display means for acquiring approximate position information relating to each defect candidate generated and stored in the inspection apparatus and displaying it on the screen of the image display device, and the approximate position information via the actual image display means When a predetermined operation for AOI execution is performed while a real image is being observed, AOI command generation means for issuing an AOI command having the approximate position information as a designated position is further included. It may be activated by an AOI command.

  According to such a configuration, by providing a defect inspection based on the inspection result of the upper inspection apparatus before the defect inspection based on the inspection result of the AOI apparatus, it is difficult to determine the authenticity from the inspection result of the upper inspection apparatus. By adopting authenticity determination based on the inspection result of the AOI apparatus only for the candidates, the overall inspection efficiency can be improved.

  Note that examples of the inspection target of the defect inspection apparatus described above include a display panel such as a liquid crystal display panel or a semiconductor substrate on which an integrated circuit is built.

  According to the defect inspection apparatus for a substrate of the present invention, an operator observes a local area image corresponding to each defect candidate displayed on the screen as the guide display, so that the image is listed as a defect candidate. However, there is an abnormality that leads to circuit operation failure such as disconnection or foreign matter contamination, or circuit operation failure such as circuit pattern distortion that tends to occur at the edge of the board, etc. Because it is possible to recognize with certain certainty whether there is an abnormality that does not connect, defect candidates that are clearly identified as not leading to circuit malfunction will be removed from the target of actual image observation, It is possible to reduce as much as possible the labor and time for determining the authenticity of defect candidates listed in the AOI apparatus.

  Hereinafter, a preferred embodiment of a defect inspection apparatus for a substrate according to the present invention will be described in detail with reference to the accompanying drawings.

  A system configuration diagram of a repair apparatus for a liquid crystal substrate to which the present invention is applied is shown in FIG. 1, and a top view of a stage on which an optical unit is mounted is shown in FIG.

  As shown in these drawings, the repair device for a liquid crystal substrate is configured around a stage 1 on which a glass substrate (hereinafter referred to as “liquid crystal substrate”) 2 serving as a liquid crystal display panel is placed.

  The stage 1 has a rectangular shape in a top view, and front and rear direction guide members 1a and 1a are provided on the left and right side edges along the longitudinal direction, and the guide members 1a and 1a are provided with the guide members 1a and 1a. A portal beam member (hereinafter referred to as “gantry”) 1b is bridged in the left-right direction between two movable stands (not shown) that move in the front-rear direction.

  A left-right guide member (not shown) is provided on the gantry 1b, and the optical unit 3 that accommodates a laser-related device such as a laser oscillator and an imaging device such as an image sensor is provided in the left-right direction via the guide member. It is mounted so that it can move. Although not shown in the drawings, there are some repair apparatuses for substrates of the fourth generation or lower in which the optical system is fixed and the stage moves in the XY directions.

  The front projecting portion 4 of the optical unit 3 has a laser optical system for irradiating a circuit pattern carrying surface of the repair target substrate 2 located below the laser pattern for repair processing, and a circuit of the repair target substrate 2. A microscope optical system for enlarging and capturing an image of the pattern carrying surface, and an image sensor for converting the captured image into image data are accommodated. Therefore, a microscope camera for acquiring an image for visual observation (image data) is configured by the above-described microscope optical system, image sensor, and the like.

  An auxiliary unit 5 for AOI is attached to the side surface of the front protrusion 4 of the optical unit 3. In this AOI auxiliary unit 5, an AOI branching optical system for branching and capturing an image acquired from the microscope optical system housed in the front protrusion 4 of the optical unit 3, and the captured image are imaged. An image sensor or the like for converting to data is accommodated. Therefore, a microscope camera for acquiring an AOI image (image data) is configured by the above-described microscope optical system, AOI branching optical system, and image sensor. The AOI image acquisition optical system can be configured independently of the laser optical system.

  The above-mentioned front-rear direction guide members 1a and 1a and the left-right direction guide member are not shown in the above-mentioned field of view of the microscope camera for acquiring the image for visual observation and the field of view of the microscope camera for acquiring the image for AOI. By driving with a linear drive source, scanning is performed in the front-rear and left-right directions, so that the front-rear direction guide members 1a, 1a and the left-right direction guide members function as the visual field scanning mechanism of the present invention.

  An image (image data) acquired by the microscope camera in the AOI auxiliary unit 5 is sent to the image processing unit 6 where image processing for realizing the AOI function is executed. That is, in the image processing unit 6, as will be described in detail later, in conjunction with the scanning of the field of view of the microscope camera via the field of view scanning mechanism, an image of a predetermined peripheral range of a specified position on the circuit pattern carrying surface is obtained. At the same time, one or more defect candidates are detected by region-based pattern matching between the acquired image of the predetermined peripheral range and the unit region image as a normal reference, and feature information about each defect candidate is generated and stored To do.

  The personal computer 7 functions as a human machine interface, and includes an image display 7a configured with a liquid crystal display and an operation unit 7b configured with a keyboard, a mouse, and the like. The control of the visual field scanning mechanism, the control of the optical unit 3, the control of the image processing unit 6, and the like are mainly executed by the control unit 8.

  Next, on the premise of the system configuration of FIGS. 1 and 2, the processing of the substrate repair apparatus (processing of the control unit 8) using the inspection result of the host inspection apparatus (details will be described later) with reference to the flowchart of FIG. This will be described in detail.

  In the figure, when the process is started, in step 101, for example, in response to a start command operation from the operation unit 7b of the personal computer 7, the upper inspection apparatus activation process is executed to activate the upper inspection apparatus. . Here, although not shown, a host inspection apparatus that detects one or more defect candidates on a substrate through a circuit function inspection, such as a conventionally known continuity inspection, is employed.

  In the following step 102, the above-described upper inspection apparatus side processing is executed, one or more defect candidates on the inspection target substrate 2 are detected, and the respective approximate coordinates of these defect candidates are stored in the upper inspection apparatus. Generated and saved.

  In the subsequent step 103, the approximate coordinates of each defect candidate generated and stored in the upper inspection apparatus are read out from the upper inspection apparatus and acquired. In the next step 104, the approximate coordinates of the read defect candidates are displayed on the screen of the image display 7a of the personal computer in a predetermined display format as a guide display for assisting the defect candidate narrowing down. Thereafter, the display of the approximate coordinates is continued (step 104), and the operator waits for one of those defect candidates to be selected (step 105).

  In this state, when one of the defect candidates is selected by the operator (“Yes” at step 105), the approximate coordinates of the selected defect candidate are read, and the visual field scanning mechanism is servo-controlled using this as a target position. Thus, the field of view of the microscope camera is automatically moved to the target position (step 106).

  In the subsequent step 107, the microscope camera field of view waits for the target position to be reached, and an image (actual image) of the microscope camera field of view is displayed on the screen of the image display. In the subsequent step 108, the field of view by the operator is displayed. Wait for the fine adjustment operation.

  Here, since the defect determination accuracy in the defect inspection through the circuit function inspection such as the continuity inspection is generally set to about 80%, even if the approximate coordinates of the defect candidate are reached, the object is within the field of view of the microscope camera. In some cases, the defect to be found is not found. In such a case, the operator performs a predetermined visual field fine adjustment operation in the operation unit 7b.

  Then, the fine operation “Yes” is determined (Step 108 “Yes”). In Step 109, fine adjustment of the visual field according to the fine operation of the operator is performed through the visual field scanning mechanism, and in Step 110, Then, the image after the fine field of view adjustment is displayed again on the screen of the image display 7a.

  Here, if a target defect is found on the screen of the image display 7a, the operator performs a predetermined repair activation operation in the operation unit 7b. Then, in step 111, a determination is made that the repair operation is “present”, and subsequently, in step 112, a known laser repair process is performed in which laser irradiation is performed by aiming at a predetermined position in the field of view of the microscope camera, Defects on the substrate are repaired.

  Thereafter, after step 313 “No”, the process returns to step 104 again, the above processing (steps 101 to 112) is repeated, and the laser repair process is completed for all defect candidates displayed in the guide. Alternatively, after waiting for a predetermined end operation to be performed in the operation unit 7b (Step 313 “Yes”), the process ends.

  On the other hand, if the defect is still not found despite the fine adjustment of the field of view during the above processing (steps 109 and 110), the operator uses the operation unit 7b to perform a predetermined process. AOI start operation is performed.

  Then, in step 114, it is determined that the AOI apparatus activation operation is “present”, and then in step 115, the processing of the substrate repair apparatus using the inspection result of the host inspection apparatus is executed. If a defect is found, the defect is also repaired.

  On the other hand, in the monitor display in step 107 or the monitor display in step 110, when the defect candidate is recognized as a pseudo defect as a result of visual inspection, a predetermined skip operation is performed in the operation unit 7b, Wait for a predetermined time without doing anything.

  Then, the repair operation “None” (step 111) → AOI apparatus activation operation “None” (Step 114) → End operation “None” (Step 313) → Guide display (Step 104) proceeds, and defect candidates are guided again. While displaying, it will be in the state which waits for candidate selection operation.

  Next, repair processing based on the inspection result of the AOI apparatus, which is the main part of the present invention, will be described in detail with reference to the flowchart of FIG.

  In the figure, when processing is started, the AOI apparatus is activated after waiting for an AOI command from the outside (step 201). Then, a process corresponding to the AOI apparatus is activated, and a predetermined AOI operation is performed with the support of the image processing unit 6 (step 202).

  The processing of the AOI apparatus of the present invention is shown in the flowchart of FIG. In the figure, in step 301, a peripheral image of a predetermined range including a specified position (defect candidate position) on the inspection target substrate acquired through the microscope camera is acquired and stored.

  In the subsequent step 302, a small image including a circuit pattern serving as a normal reference is determined for each inspection region from the stored peripheral images according to a predetermined algorithm. In the subsequent step 303, a small image of the corresponding inspection target area is cut out from the acquired peripheral images.

  In the subsequent step 304, the small image serving as the normal reference is pattern-matched with the small image in the inspection target area. In the subsequent step 305, it is determined whether or not the collation does not coincide with a predetermined allowable range. Here, if it is determined that the two do not match (YES in step 305), it means that there is a high possibility that the small image to be inspected contains a defect. On the other hand, if it is determined that they match (NO in step 305), it means that there is a high possibility that the small image to be inspected does not contain a defect.

  If it is determined in step 305 that the two do not match (YES in step 305), the process proceeds to step 306. In step 306, based on the small image to be inspected that is not matched, feature information relating to the defect included therein is generated and stored in a predetermined memory. This feature information is used to create a defect candidate list described later.

  In the subsequent step 307, a local area image including each image of the defect candidate is acquired, and this is associated with the defect candidate in a predetermined memory and stored for reference. Here, the local region image to be stored for reference may be a small image of the inspection target region that is the target of verification, or a small image of a certain size that includes such a defect substantially at the center is separately generated. You may have done. Further, the resolution or definition of the local area image may be the same as that at the time of pattern matching, or may be coarser or finer than that. Further, the gradation of the local area image may be the same as that at the time of pattern matching, or may be obtained by performing appropriate binarization and emphasizing a defective portion. The contour of the local area image may be a square or a circle. The size or the number of pixels of the local area image may be determined in consideration of the size of the display area on the screen of the image display device.

  Thereafter, the above process (steps 301 to 307) is repeated, and the process ends after waiting for the inspection of the entire acquired peripheral image (step 308 YES). If the reference small image needs to be changed due to a pattern difference or the like in the acquired peripheral image (YES in step 309), a new reference small image is determined according to a predetermined algorithm (step 302). .

  As described above, the process of the AOI apparatus according to the present invention acquires the predetermined peripheral image for each of the designated defect candidates via the microscope camera, and each inspection target region from the acquired peripheral images. A small image as a normal reference is determined every time, the small image as a reference is compared with a small image to be inspected cut out from the acquired peripheral image, a defect candidate is identified by a mismatch between the two, and each defect candidate Feature information (feature information for creating a defect candidate list and a local area image including defect candidate images) is generated and stored.

  Returning to FIG. 4, in step 203, the inspection result from the AOI apparatus, that is, the feature information of each defect candidate generated and stored (local area image including feature information and defect candidate images for creating a defect candidate list) is stored. get. In subsequent step 204, the acquired feature information is displayed in a predetermined display format on the screen of the image display as a guide display for assisting in narrowing down defect candidates.

  An explanatory view showing an example of such a guide display is shown in FIG. As shown in FIG. 5A, in this example, the upper half of the screen of the image display 7a is a defect candidate list display area 31, and the lower half of the screen includes an image of a defect candidate. The display area 32 is a local area image (candidate image).

  In the display area 31, the defect candidate list described above with reference to FIG. 9 is displayed. In the display area 32, as shown in FIG. 6B, eight 16 candidate images are displayed. A list is displayed in each of the two horizontal columns together with a number indicating each defect candidate. In the illustrated example, each of the 16 candidate images 32a has a vertically long rectangular outline and corresponds to each repetition of the same circuit pattern existing in a predetermined region of the substrate. In FIG. 6B, 34 is a power supply electrode pattern, 35 is a GND electrode pattern, 36 is a GND pattern, and 33 is a defective portion.

  Thereafter, the candidate selection operation by the operator is waited while the defect candidates are still displayed (step 205). In this state, when one of the defect candidates is selected by the operator from the defect candidate list displayed on the screen (step 205 “Yes”), the XY coordinates of the selected defect candidate are read, The field of view scanning mechanism is servo-controlled as the target position, so that the field of view of the microscope camera is automatically moved to the target position (step 206).

  In the subsequent step 207, the microscope camera field of view waits for the target position to reach the target position, and the image of the microscope camera field of view (actual image) is displayed on the screen of the image display. Wait for operation.

  In this state, the operator observes the actual image of the defect candidate displayed on the screen of the image display, so that it is a pseudo defect that does not require repair processing or a true defect that requires repair processing. Can be determined.

  If it is determined in step 208 that the repair operation by the operator is “present”, in step 209, a known laser repair process is performed in which laser irradiation is performed while aiming at a predetermined position in the field of view of the microscope camera.

  As described above, in the substrate repair apparatus in which the AOI apparatus of the present invention is incorporated, as shown in FIG. 6, the operator applies each defect candidate displayed in the display area 32 on the screen as a guide display. By observing the corresponding candidate image (local region image) 32a, the reason why the image is listed as a defect candidate is because there is an abnormality that leads to circuit malfunction such as disconnection or contamination, or Clearly, it can be recognized with some certainty whether there is an abnormality that does not lead to circuit malfunction such as a difference in line width or circuit pattern distortion that tends to occur at the edge of the board, etc. With regard to defect candidates that are recognized as not leading to circuit malfunctions, by removing them from the target of actual image observation, it is possible to save time and effort for authenticating the defect candidates listed in the AOI device. During it can be reduced as much as possible, therefore it is possible to improve the work efficiency in the substrate repair operation.

  Further, in the above embodiment, a list of candidate images (local region images) 32a corresponding to each of a plurality of defect candidates is displayed in a list in the display area 32 on the screen. It is possible to recognize at a glance whether there is a high possibility of corresponding to a defect or which defect candidate is highly likely to correspond to a pseudo defect.

  Further, in the above embodiment, since the defect candidate list in which the feature information corresponding to each defect candidate is quantified is also displayed in the display area 31 on the screen, the list of candidate images 32a and the defect candidates are displayed. In combination with the list, it is possible to more reliably determine which defect candidates are highly likely to correspond to true defects or which defect candidates are likely to correspond to pseudo defects.

  Further, in the above embodiment, since the defect inspection based on the inspection result of the upper inspection apparatus is provided in the preceding stage of the defect inspection based on the inspection result of the AOI apparatus, it is true or false from the inspection result of the upper inspection apparatus. By adopting authenticity determination based on the inspection result of the AOI apparatus only for defect candidates that are difficult to determine, the overall inspection efficiency can be improved.

  The substrate defect inspection apparatus according to the present invention is applied to a laser repair apparatus for a liquid crystal display panel, for example, to reduce as much as possible the labor and time for determining the authenticity of defect candidates listed in the AOI apparatus. By doing so, the work efficiency in the substrate repair work can be improved.

It is a system block diagram of the repair apparatus for liquid crystal substrates to which this invention was applied. It is a top view of the stage in which an optical unit is mounted. It is a flowchart which shows the process of the board | substrate repair apparatus using the test result of a high-order inspection apparatus. It is a flowchart which shows the process of the board | substrate repair apparatus using the test result of an AOI apparatus. It is a flowchart which shows the process of the AOI apparatus of this invention. It is explanatory drawing which shows an example of a guide display. It is a flowchart which shows the process of the board | substrate repair apparatus using the test result of the conventional AOI apparatus. It is a flowchart which shows the process of the conventional AOI apparatus. It is explanatory drawing which shows an example of a defect candidate list | wrist.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Stage 1a Front-back direction guide 1b Gantry part 2 Liquid crystal substrate 3 Optical unit 4 Front protrusion part 5 AOI auxiliary unit 6 Image processing part 7 Personal computer 7a Image display 7b Operation part 8 Control part 31 Display area 32 candidate for defect candidate lists Display area for image list 32a Candidate image (local area image)
33 Defective part 34 Electrode pattern for power supply 35 Electrode pattern for GND 36 GND pattern

Claims (5)

A stage for placing a substrate to be inspected;
A microscope camera for photographing the circuit pattern carrying surface of the substrate to be inspected placed on the stage;
An image display that functions as a human-machine interface;
A field-of-view scanning mechanism for scanning the field of view of the microscope camera on the circuit pattern carrying surface;
In response to a predetermined AOI command including a position specification, the field of view of the specified position on the circuit pattern carrying surface is acquired by scanning the field of view of the microscope camera via the field-of-view scanning mechanism. At the same time, one or more defect candidates are detected by pattern matching by region between the acquired image of the predetermined peripheral range and the unit region image as a normal reference, and feature information relating to each of the defect candidates is generated and stored. An AOI device;
Defect candidate display means for acquiring feature information on each defect candidate generated and stored in the AOI apparatus and displaying it in a predetermined display format on the screen of the image display as a guide display for supporting defect candidate narrowing down When,
The visual field of the camera is moved through the visual field scanning mechanism by manual operation or automatic control according to the teaching of the target position, and an actual image of the visual field of the camera is displayed on the screen of the image display. Image display means,
Thereby, the operator selects one defect candidate based on the feature information regarding each defect candidate displayed as the guide display, and observes the actual image of the defect candidate on the screen of the image display, Whether the selected defect candidate is a pseudo defect or a true defect can be determined, and the feature information generated and stored by the AOI device includes a local area image including each image of the defect candidate. And the defect candidate display means displays, for each defect candidate, the local area image corresponding to the defect candidate on the screen of the image display device. apparatus.   2. The defect inspection apparatus for a substrate according to claim 1, wherein the defect candidate display means displays a list of local area images corresponding to each of a plurality of defect candidates on the screen of the image display.   Laser repair execution for executing laser repair processing by aiming at a predetermined position in the field of view of the camera when a predetermined user operation for laser repair execution is performed during display of the actual image by the actual image display means The defect inspection apparatus for a substrate according to claim 1, further comprising means. A high-order inspection device that detects one or more defect candidates through circuit function inspection such as continuity inspection, and generates and stores the approximate position information of the defect candidates;
Defect candidate preliminary display means for acquiring approximate position information about each defect candidate generated and stored in the host inspection apparatus and displaying it on the screen of the image display device;
An AOI command that issues an AOI command using the approximate position information as a designated position when a predetermined operation for AOI is performed while a real image related to the approximate position information is being observed via the actual image display means And generating means,
The said AOI apparatus is started by the said AOI instruction | command, The defect inspection apparatus of the board | substrate in any one of Claims 1-3 characterized by the above-mentioned.   5. The substrate defect inspection apparatus according to claim 1, wherein the substrate is a semiconductor substrate in which a display panel such as a liquid crystal display panel or an integrated circuit is built.

Images