JP2010151666A - Pattern inspection device and inspection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem wherein, though an image of a circuit pattern which is an inspection object is photographed in a defect inspection of the circuit pattern using image processing, in this case, if a camera is moved/stopped in each photographing, acquisition of image data requires much time. <P>SOLUTION: While moving a photographing camera at fixed speed relative to a substrate on which a plurality of inspection objects (circuit patterns) are arrayed, photographing is performed continuously on a visual field position determined beforehand. Therefor, moving speed, photographing timing or the like are determined beforehand from array information of the inspection objects and a hardware/software forced condition on the system side. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a circuit pattern inspection apparatus for automatically inspecting printed circuit board circuit patterns and LSI circuit pattern defects by image processing.

  The demand for a circuit pattern in which a conductor pattern is formed on an insulating member is rapidly increasing with the spread of electronic devices such as mobile phones. Since the circuit pattern affects the performance and durability of an electronic device using the circuit pattern, the circuit pattern is not only electrically connected as designed, but also requires a predetermined dimensional accuracy.

  Therefore, all the circuit patterns are inspected after being manufactured, and it is confirmed that a predetermined specification is satisfied. By the way, since the types and number of circuit patterns have become enormous, it is necessary to complete the inspection in as short a time as possible.

  Conventionally, several methods have been proposed for this inspection. A method called a line width sub-pixel measurement method is a method in which the line width is measured for every fixed length, the line width between them is determined from several measurement points before and after, and the quality with respect to the reference width is judged. This method can be inspected with a resolution of about 1/10 of the pattern pitch.

  The image comparison method is a method of obtaining a difference image between the binarized imaging data and the non-defective product data and obtaining a difference portion having a certain area or more as a defect. Since this method binarizes data, the calculation speed can be increased.

In this image comparison method, it is necessary to acquire image data of an inspection object. Therefore, the inspection apparatus has a two-dimensional imaging apparatus such as a CCD image sensor or an ITV area camera (Patent Document 1).
JP 63-19541 A

  The inspection object has a conductor portion that is a circuit pattern formed on a substrate. In the case of mass production, a plurality of circuit patterns are formed on the substrate. This is because many products can be manufactured by one process if formed on one wide substrate.

  On the other hand, when a circuit pattern is inspected for defects by image processing, it is often necessary to acquire images multiple times for one circuit pattern. The size to be determined as a defect becomes smaller as the pitch becomes finer. However, the optical system and the imaging device for capturing a circuit pattern with a single image and identifying the defect with sufficient resolution are too expensive. is there.

  Therefore, when a plurality of circuit patterns are formed on the substrate as described above, it is necessary to obtain necessary image data for each circuit pattern and perform defect inspection. However, in order to acquire a plurality of image data for each circuit pattern, it is necessary to photograph the circuit pattern by moving the field of view sequentially while moving the area camera, which requires a long inspection time. was there.

  The pattern inspection method of the present invention has been conceived to solve the above problems. That is, attention was paid to the fact that the tact time is increased when the entire image of a single circuit pattern is shot in multiple shots because the area camera is repeatedly moved and stopped for each shot. Therefore, the present invention provides an inspection apparatus and an inspection method for performing defect inspection by continuously capturing a plurality of circuit patterns formed on a substrate with an imaging area camera that sweeps in a certain direction.

More specifically,
A circuit pattern inspection apparatus in which a plurality of circuit patterns are formed on a substrate,
A stage for holding the substrate;
A Y-axis slider for moving the stage;
A θ-axis adjustment mechanism for rotating the substrate on the stage;
An area area camera disposed above the stage;
A strobe light source for irradiating light on the stage;
An X-axis slider for moving the area area camera in a direction perpendicular to the Y-axis slider;
Position information of a plurality of patterns on the substrate is input, a shooting plan is created based on the position information, and shooting is continuously performed with the area area camera while moving the Y-axis slider based on the shooting plan. Provided is a pattern inspection apparatus having a control device.

The present invention also provides:
A pattern inspection method for a plurality of circuit patterns formed on a substrate,
Creating a shooting plan in which the phase of the shooting field of view between the circuit patterns is equal from the position information of the circuit pattern on the substrate and the field of view information of the area camera;
Acquiring inspection reference image data based on the imaging plan from a good product;
Continuously acquiring image data of the circuit pattern based on the shooting plan;
The present invention provides a pattern inspection method, wherein the image data and the inspection reference image data are compared to detect a defect in the image data.

  In the present invention, the circuit pattern formed on the substrate is continuously photographed and the defect inspection is performed while sweeping the photographing area camera in a certain direction. Therefore, the area camera is repeatedly moved and stopped every photographing. This eliminates the waste of movement and improves the inspection speed.

  Embodiments of the present invention will be described below, but the present invention is not limited to these embodiments, and can be modified and changed within the scope of the gist of the present invention.

  FIG. 1 shows a schematic diagram of a pattern inspection apparatus 1 according to the present invention. A pattern inspection apparatus 1 according to the present invention includes a stage 2 on which an inspection object is arranged, an area camera 8 that images the inspection object on the stage, and a guide that moves the area camera in a direction transverse to the inspection object. Arm 4, a Y-axis slider 6 (see FIG. 2) that moves the stage relative to the area camera, and a θ-axis adjustment mechanism 10 that rotates the inspection object on the stage. The inspection object has a large number of circuit patterns 14 formed on the substrate 12. In this specification, the direction perpendicular to the arm 4 is referred to as the Y-axis direction, and the direction along the arm is referred to as the X-axis direction. The Y-axis direction is the moving direction when the area camera performs continuous shooting.

Further, FIG. 2 shows a configuration diagram of the pattern inspection apparatus 1 of the present invention. In addition to the configuration shown in FIG. 1, the pattern inspection apparatus 1 of the present invention includes a drive unit 16 that drives the Y-axis slider 6, an image processing unit 25 that processes video data from the area camera 8, and an image processing unit 25. A storage unit 26 that accumulates the data, a control unit 21 that controls the drive unit 16 and the image processing unit 25, and an instruction unit 30 that issues an instruction to the control unit 21. The control unit 21, the image processing unit 25, the drive unit 16, and the storage unit 26 are collectively referred to as a control device 20.

  Next, the pattern inspection apparatus 1 of the present invention will be described in detail with reference to FIG. 1 and FIG. What is to be inspected by the pattern inspection apparatus of the present invention is one in which a large number of circuit patterns are formed on the substrate 12. The individual circuit patterns are preferably the same, but may be different from each other. The substrate 12 is preferably made of glass, metal with a mirror finish, plastic, or the like. This is because the planarity of the substrate is required for forming the circuit pattern.

  The circuit pattern is a circuit pattern formed by copper plating on a heat-resistant film such as polyamide or polyimide. These are formed on a heat resistant film stuck on the substrate 12.

  Referring to FIG. 4A, the circuit pattern on the substrate 12 is accurately arranged at a predetermined position from the reference point or reference line on the substrate. Although the reference point is shown as the alignment mark 13 in FIG. 4A, the edge of the substrate 12 or the like may be used. This is because the pattern inspection apparatus of the present invention continuously acquires an image of a circuit pattern, and therefore the circuit pattern needs to be accurately arranged. Therefore, the size of each circuit pattern 14 and each arrangement interval are accurately determined.

  1 and 2 again, the substrate 12 on which the circuit pattern 14 is formed is held on the stage 2. The holding method is not particularly limited. However, the method of forming fine holes on the stage 2 and adsorbing and holding the substrate 12 with negative pressure therefrom is suitable when the substrate 12 is a highly brittle material such as glass.

  The substrate 12 has a θ-axis adjusting mechanism 10 for adjusting the position in the rotational direction with respect to the stage 2. This is because the substrate 12 needs to be accurately placed at a predetermined position on the stage in order for an area camera above the stage, which will be described later, to correctly acquire an image of the circuit pattern 14. 1 and 2, only one θ-axis adjusting mechanism 10 is shown, but a plurality of θ-axis adjusting mechanisms 10 may be arranged.

  A Y-axis slider 6 is disposed at the bottom of the stage 2. The Y-axis slider 6 moves the stage 2 along the groove 5. Specifically, a motor with a wheel attached, a linear motor, or the like can be suitably used. In particular, the pattern inspection apparatus of the present invention captures an image of the circuit pattern 12 while moving the stage 2 holding the substrate 12 at a predetermined speed in the Y-axis direction, so a drive mechanism having excellent acceleration / deceleration capability is suitable. Used for

  An area camera 8 is attached to the arm 4 so as to be movable along the arm 4 in the X-axis direction. More specifically, the area camera 8 is fixed to a moving gantry 9 that moves along the arm 4. The mechanism of the moving base 9 is not particularly limited as long as it can move with respect to the arm 4 while holding the area camera 8. However, since the amount of movement of the area camera 8 on the arm 4 affects the range of the field of view, the moving base 9 is required to be able to accurately control the moving distance.

The area camera 8 having an area sensor is preferably used. This is because the pattern inspection apparatus of the present invention creates image data to be inspected while photographing a circuit pattern on the substrate at a predetermined timing from a moving area camera. The field of view of the area camera (hereinafter simply referred to as “field of view” or “shooting field of view”) is determined by the size of the defect to be detected and the number of pixels of the area sensor. Since the data transfer rate of the area sensor is a constraint condition for increasing the imaging cycle, it is preferable that the transfer rate is as fast as possible. In order to photograph the circuit pattern from above, the lens is preferably of a telecentric type, but is not limited thereto. It is desirable that the shutter can perform exposure for a predetermined time by receiving a predetermined signal. Image data Vd captured by the area camera 8 is transferred to the image processing unit 25 as needed.

  A strobe 7 is disposed beside the lens of the area camera 8. The area camera 8 takes an image while moving on the substrate 12 at a predetermined speed. In order to determine the size of about 1 μm or 1 μm or less from image data taken while moving, it is necessary to prevent the flow of the image due to the moving speed, and it is necessary to sensitize the area sensor with light in a short time. It is. Accordingly, the strobe 7 emits light in a short time while the shutter is open in synchronization with a signal for opening the shutter of the area camera. Since the strobe 7 emits light for a short time, a light source with high illuminance is desirable. Furthermore, if the image data of the circuit pattern is shaded, the accuracy of defect inspection is reduced. Therefore, a strobe that emits light in at least four directions, preferably from the surroundings, is preferred.

  In addition, since the image is prevented from flowing by exposing only for the time of flash emission, it is preferable to cover the entire apparatus with a cover and shield it from light.

  FIG. 3 shows an example of a desirable strobe. FIG. 3 is a view of the lens 81 as seen from the area camera side. The strobe light emission source 71 is arranged in a circle around the lens 81 of the area camera. Each strobe light source 71 is connected to a power source 72 and emits light by a pulse voltage from the power source 72. The power source 72 can emit light for a predetermined time according to a light emission signal from the control device 20.

  The pattern inspection apparatus 1 of the present invention includes elements of an electronic control system (not shown in FIG. 1). FIG. 2 focuses on such elements. The control unit 21 controls the entire apparatus. Specifically, a drive unit 16 that drives the Y-axis slider 6 that moves the stage 2, a moving base 9 that moves while holding the area camera on the arm 4, and an image processing unit 25 that processes image data from the area camera. Connected and control these. In addition, the instruction unit 30 is also connected, and the entire apparatus is controlled by an instruction from the instruction unit 30.

  The drive unit 16 drives the Y-axis slider 6 that moves the stage 2. Specifically, by transmitting the drive signal Syd to the Y-axis slider 6, the Y-axis slider 6 performs acceleration / deceleration at a predetermined acceleration and moves at a predetermined speed. In addition, the drive unit 16 transmits a signal that informs the shutter timing of the area camera 8 and the movement instruction Slxd of the movable gantry 9 at a predetermined timing. These timings are instructed by the control unit 21. Note that the light emission timing of the strobe 7 may also be included in this instruction Slxd.

  In the present invention, since the area camera and the substrate to be inspected need only move relative to each other, the area camera side may be driven instead of the stage. However, in order to increase the moving speed over a short distance, it is necessary to apply a large acceleration. As a result, transient acceleration is applied to the area camera and the mobile gantry, and the risk of failure increases. Therefore, it is preferable to drive the stage side. In this specification, there may be a case where the area camera moves as if it is “the moving speed of the area camera”.

  The stage 2 is provided with a θ-axis adjusting mechanism 10 and is controlled so that the substrate 12 is placed at a predetermined position on the stage 2. The control amount is received from the control unit 21 as an instruction Sita.

The image processing unit 25 receives and processes the image data Vd from the area camera 8. As processing contents, an identification number indicating which image of which circuit pattern of which substrate is given to the sent image data, and stored in the storage unit 26. In addition, inspection reference image data having no defect may be stored in the storage unit 26 for each image, and defect inspection may be performed whenever the image data Vd is received. The image processing unit 25 may be composed of a plurality of MPUs (Micro Processor Units), and may receive the image data Vd and perform defect inspection based on the inspection reference image data in parallel.

  The image processing unit 25 operates in response to an instruction Cip from the control unit 21 and sends information to the control unit 21 by Sip. The information Sip from the image processing unit 25 can include the result of defect inspection for each circuit pattern on the substrate 12.

  The control unit 21, the image processing unit 25, the driving unit 16, and the storage unit 26 are included in the control device 20. The control device 20 can be configured in hardware or software. Here, the control unit, the image processing unit, and the drive unit have been described as being divided into three, but these may be configured as software or may be configured as hardware. Also, one MPU and software can be assigned to each. In addition to these, the control device 20 may have a connection terminal with an external line for transmitting and receiving data and a connection port with an external storage.

  The instruction unit 30 is a man-machine interface, and can specifically be configured from a display device and a keyboard. The instruction unit 30 instructs the control unit 21 with an instruction Ccom, and the control unit 21 transmits information to the instruction unit 30 with information Scom.

  Next, acquisition of image data of the pattern inspection apparatus 1 of the present invention will be described. Reference is made to FIG. FIG. 4B shows an enlarged part 50 of the circuit pattern on the substrate 12 in FIG. The area camera 8 has a field of view 51. Since the field of view 51 is smaller than the normal circuit pattern, in order to obtain image data of one circuit pattern, the image must be divided into a plurality of image data.

  In this way, when one object is made up of a plurality of pieces of image data, in order to enumerate the defects found in each piece of image data without excess or deficiency or duplication, it is only necessary to provide duplicate portions in each piece of image data. In addition, if an overlapping portion is provided, it is easy to obtain the positional relationship with the preceding and subsequent image data even if the shooting location is shifted, and this is useful for position correction at the time of defect inspection. This overlapping portion can be provided in each of the X-axis direction and the Y-axis direction.

  Here, the overlapping part is called a visual field overlapping part. Let fx and fy be the lengths in the X-axis direction and the Y-axis direction of the overlapping field of view. This visual field overlapping portion needs a fixed length in order to connect with adjacent images in image processing. That is, the minimum length is determined in advance.

  With reference to FIG. 4A, the movement of the area camera 8 will be described. The area camera 8 is initially placed at the start point 32 and takes a circuit pattern while securing a field overlap portion as described above while moving (33) in the Y-axis direction of the substrate 12. When the end point is reached, next, the substrate 12 is moved in the X-axis direction (34), and the image is taken again while moving in the substrate Y-axis direction at a constant speed (35). By repeating this, all the circuit patterns on the substrate are photographed to obtain image data.

  When the image data of the circuit pattern that is the object to be inspected is obtained while moving in a certain direction as described above, the number of times the area camera 8 is moved and stopped is reduced, and a lot of image data can be obtained in a short time.

  On the other hand, the image acquisition method of the pattern inspection apparatus of the present invention does not acquire image data for each circuit pattern, but continuously acquires image data. In addition, it is necessary to decide in advance at which timing the data is acquired.

  With reference to FIG. 5, items to be determined for shooting while moving will be described. FIG. 5 is a further enlarged view of FIG. It is assumed that the circuit pattern 14 a is a circuit pattern at the upper left of the substrate 12. The length in the Y-axis direction is A, and the interval between circuit patterns is C. The length of the circuit pattern in the X-axis direction is B, and the distance between the circuit patterns adjacent in the X-axis direction is D. The length of the field of view in the Y-axis direction is “a”, and the length in the X-axis direction is “b”. The area camera 8 will be described on the assumption that shooting starts from the upper left of the substrate 12.

  Since the area camera 8 is initially in a stationary state, an acceleration period is required to reach a constant speed. That is, the run-up period L is necessary before reaching the first visual field 53. Therefore, first, the area camera 8 is arranged at a position where the visual field 52 can be seen. This position is the start point 32 in FIG. The run-up period L is determined by the driving ability of the Y-axis slider 6 and the arrival speed, and can be determined if the camera moving speed Vcam is determined.

  The first visual field 53 is preferably set to the upper side in the Y-axis direction by 1 from the edge of the circuit pattern 14a because it is desirable to put the edge of the circuit pattern into the visual field. This l may also be determined in advance on the system side. The next field of view 54 is set while ensuring a field-of-view overlap with the field of view 53 by a distance fy. This visual field overlap portion fy has a minimum distance from the viewpoint of image processing. If the minimum distance can be secured, the circuit pattern length A and the gap C between the circuit patterns may be varied thereafter.

  It is desirable that the acquired image data is not necessarily image data of any circuit pattern or image data of any circuit pattern. This is because the processing becomes complicated if a plurality of circuit patterns are included in one image data. For example, if a part of the circuit patterns 14a and 14b is shown in the visual fields 55 and 56, the defect extracted from the image data may be recognized as a defect of both circuit patterns. Cannot inspect.

  In order to avoid such a situation, the first imaging field 53 of the circuit pattern 14a and the first imaging field 57 of the circuit pattern 14b are preferably arranged in the same phase. Here, the same phase means that the shift “l” between the circuit pattern and the photographing field of view is the same. Therefore, if the field of view overlaps fy between the first field of view 53 of the circuit pattern and the field of view 56 just before the first field of view 57 of the circuit pattern 14b, and the fields of view are arranged at equal intervals, the field 53 and the field of view are displayed. 57 can be arranged in the same phase.

  By determining such an imaging plan, the positional relationship between each circuit pattern and the field-of-view range captured by the area camera can be matched in all circuit patterns. For this reason, it is possible to inspect a circuit pattern having a common pattern among a plurality of circuit patterns existing on the substrate by using the common inspection reference image data. In particular, when all the circuit patterns are the same pattern, only one inspection reference image data to be compared can be inspected, all the circuit patterns can be inspected, and the inspection reference image data can be easily managed. Preparation time of the inspection reference image at the start of inspection can be shortened.

  More specifically, as the circuit pattern length A and the arrangement interval C, fy is adjusted so that n obtained by the following equation (1) becomes an integer. That is, the length fy of the overlapping field of view is set to be not less than the minimum distance of fy and n is an integer.

n = (A + C) / (a−fy) (1) Similarly, the field of view to be taken is determined while securing the field overlap portion fx in the X-axis direction. Also in this case, the visual field overlapping portion fx is set according to the distance between the circuit patterns adjacent in the X-axis direction and the length of the circuit pattern in the X-axis direction.

  When the position of the photographing field of view is determined, the shutter timing Ts and the strobe timing Tb can be obtained. In addition, the camera moving speed Vcam and the initial position can be determined by the constraint conditions from the system.

  FIG. 6 shows various timing charts. (A) is the shutter timing, (b) is the strobe timing, and (c) is the moving speed of the area camera. Each chart has time on the horizontal axis. In addition, the display shown in parentheses is a distance to be described. The area camera 8 accelerates from the initial speed zero at the time t = 0, and becomes Vcam at the time t1. The strobe timing tb1 is a point in the photographing visual field 53, and the strobe timing tb2 is a point in the photographing visual field 54. That is, the distance of (a-fy) is moved during this time. “A” was the length of the field of view in the Y-axis direction.

  On the other hand, the area camera 8 needs to open the shutter before the strobe emits light. Let ts be the shutter opening time. In addition, the image data Vd must be transferred to the image processing unit 25 before the next shutter is opened after shooting with the flash. This transfer time is assumed to be td. Since the shutter speed ts and the data transfer speed td are also determined by the hardware performance of the area camera 8, they are constraint conditions. Therefore, it is necessary to finish the shutter opening time ts and the data transfer time td during the distance (a-fy) in which the photographing field of view moves. Therefore, equation (2) is established. Note that 1 / (ts + td) is an imaging frame rate (the number of screens that can be imaged per second).

Vcam = (a−fy) / (ts + td) (2)
The maximum value of the camera moving speed Vcam is determined by the ability of the Y-axis slider 6 and the total weight of the stage 2 and the substrate 12. If the camera moving speed is smaller than the speed obtained by equation (2), the camera frame rate is lowered to cope with it. The maximum movement acceleration α of the area camera is determined by the ability of the Y-axis slider 6 and the total weight of the stage and the like. A time of Vcam / α is required until acceleration to the speed Vcam. Further, during this time, the lens moves by a distance of L1 = Vcam ² / (2α). Therefore, a distance longer than L1 is set as the run-up distance L between the first photographing visual field 53 and the start point 32.

  As described above, the overlap distance fy, the start point 32, the area camera moving speed Vcam, the shutter timing ts1, the strobe timing tb1, and the final timing of each timing can be obtained.

  Referring to FIG. 4A, regarding the X-axis direction, the stage stops once, and the area camera 8 moves (34) in the stopped state. The area camera movement distance 33 can be obtained only on the condition that the image data is not included and the condition that the visual field overlap portion fx is secured. This is because shooting is not performed while moving in the X-axis direction. In FIG. 5, this area camera moving distance is denoted by reference numeral 59.

  Next, a photographing plan created by the control unit 21 will be described. The control unit 21 creates an imaging plan based on the positional relationship information of the circuit pattern on the substrate 12 input from the instruction unit 30. The shooting plan is to determine the above parameters.

More specifically, the overlapping distances fy and fx, the position of the start point 32, and the area camera movement based on the size and arrangement interval of the circuit pattern on the substrate 12 and the constraint conditions that the system has in terms of hardware and image processing The speed Vcam, shutter timing ts1, strobe timing tb1, strobe interval (tb2-tb1), and the final timing of each timing can be obtained.

  By obtaining these values, it is possible to continuously shoot in the following procedure. First, the area camera is placed at the initial position before shooting is started. Next, start the area camera. The stage moves at a relative speed Vcam with respect to the area camera by the acceleration force of the Y-axis slider. Open the shutter at the first shutter timing. Then, exposure is performed by emitting a strobe at the first strobe timing tb1. Then, the shutter is opened every (ts + td) until it stops at the edge of the substrate on the stage, and the strobe is emitted every (a−fy) / Vcam.

  When the area camera stops, the area camera is moved in the X-axis direction by (b-fx), and shooting is performed while moving in the same manner in the Y-axis direction.

  Next, inspection using the pattern inspection apparatus 1 of the present invention will be described. In the inspection apparatus of the present invention, the captured image data is compared with inspection reference image data. For this reason, it is necessary to fetch inspection reference image data first.

  As the inspection reference image data, design data or the like can be used. However, it is preferable that the inspection reference image data is obtained by taking in the image data of a non-defective product among the inspection target substrates. In this case, in order to facilitate the image processing of the inspection reference image data and the data captured by the inspection, it is preferable to prepare image data continuously captured by the inspection and inspection reference image data having the same phase.

  FIG. 7 shows an example of acquiring inspection reference image data. It is assumed that the circuit pattern 15 on the substrate 12 is confirmed to be a good product. When acquiring the inspection reference image data from the circuit pattern 15, the visual field of the area camera is acquired with nine visual fields 1 to 9, respectively. The image data may be acquired by stopping the area camera for each field of view. This is because only image data of one circuit pattern is obtained.

  However, in order to match the phase with the image data obtained from the inspection object, it is preferable to match the deviation “1” from the edge and the visual field overlap portions fy and fx to the imaging plan.

  The field-of-view shooting order is also shown by arrows so as to be 1, 2, 3, 6, 5, 4, 7, 8, and 9 in FIG. 7, but any order may be used. However, it is necessary to record for each visual field which field of view the acquired inspection reference image data is. This is because the image data for each field of view of the inspection reference image data is used many times. Obtaining inspection reference image data in this way is called an inspection reference image data acquisition step.

  In FIG. 7, the circuit pattern produced on the substrate 12 is used as inspection reference image data. However, only one circuit pattern that is confirmed to be non-defective as a result of inspection separated from the substrate is set. May be.

  As described above, the pattern inspection apparatus of the present invention needs to acquire non-defective image data. Therefore, the more circuit patterns having the same shape are formed on one substrate, the more efficient inspection can be performed. This is because the step of taking good image data as inspection reference image data is also included in the tact time.

  FIG. 8 shows an inspection flow of the pattern inspection apparatus 1 of the present invention. When the inspection apparatus of the present invention is started (S100), substrate information is input (S102). The board information is the length and width of the circuit pattern on the board, the arrangement interval, and the like.

  Next, as described above, various parameters are obtained and a shooting plan is created (S104). Inspection reference image data is acquired (S105). The acquisition of the inspection reference image data may not be at this timing. However, inspection reference image data having the same phase as the image data obtained by the inspection can be obtained if the length of the deviation from the edge and the overlap of the field of view according to the imaging plan are determined. The obtained inspection reference image data is recorded in the storage unit.

  Next, initialization is performed (S106). Initialization places the area camera 8 at a predetermined position, and instructs the drive unit 16 about shutter timing, strobe timing, and area camera moving speed Vcam. Then, after determining the end (S108), shooting is started (S110). The end determination can be made based on whether image data of all circuit patterns on the substrate has been acquired.

  In the pattern inspection apparatus of the present invention, once the area camera starts traveling, it continues to photograph while traveling until it reaches the end of the substrate. Therefore, the image data is always acquired (S112) during that period. The image data Vd is transferred to the image processing unit 25 when the area camera is exposed and the shutter is closed.

  The image processing unit 25 assigns an identifier to the transmitted image data and records it in the storage unit. The identifier is a unique number of the circuit pattern and a divided piece number on the circuit pattern. For example, referring to FIG. 4B, it is assumed that the circuit patterns 14c and 14d are arranged in this order. Also, assume that one circuit pattern is covered with nine fields of view. Number them 1 through 9 respectively.

  Here, assuming that shooting is started from the upper left of the circuit pattern 14c, identifiers assigned to the respective image data are as follows as an example. First, (14c, 1), (14c, 2), (14c, 3), (14d, 1), (14d, 2), (14d, 3), ..., (14z, 3 ) Is added to the acquired image data. This is an identifier given to the image data taken during the camera movement (37) from the upper side to the lower side of the Y axis in FIG. In addition, the last circuit pattern was made into code | symbol "14z".

  Further, when the area camera moves (38) from the bottom to the top in the Y-axis direction, (14z, 6), (14d, 6), (14d, 5), (14d, 4) (14c, 6), (14c, 5), (14c, 4).

  Referring to FIG. 8 again, when the area camera reaches the end of the substrate and stops, the area camera is moved to the next row (S114). This is to prepare for the next shooting. Then, the process returns to the end determination again (S108).

  When all the circuit patterns have been imaged (Y branch of step S108), the end process is performed (S116). The termination processing includes processing such as returning the area camera to the standby position. Further, in this termination process, defect inspection based on each image data may be performed. When the end process is finished, the process ends.

The object of the present invention is to reduce the task time of defect inspection. Therefore, if image data can be acquired, defect inspection by image processing may be performed in parallel with imaging. Accordingly, when the first shooting is started (S110), the image processing can be started independently of shooting.

  FIG. 9 shows a flow of image processing. Here, the start (S130) may be the start of the inspection in FIG. 8 (step S100) or may be synchronized with the imaging start process (step S110). When image processing is started, it is determined whether there is image data in the storage unit 26 (S132). This is because the continuously captured image data is recorded in the storage unit. Then, it waits until image data is recorded.

  If there is image data (Y branch in step S132), an end determination is made (S134). The end determination may be made based on whether or not the image processing of all circuit patterns on the substrate has been completed.

  If the process continues, image processing is performed (S136). Here, image processing refers to defect inspection. Image processing is compared with inspection reference image data for the imaged part of the circuit pattern, resulting in shorts that connect separated wirings, missing parts of wiring, and parts of wiring that are larger than planned dimensions. Detect protrusions as defects.

  When the image processing is completed, the inspection results are totaled (S138). This is because the image processing is performed for each image data, and is performed every time the visual field of one circuit pattern is obtained, so that it is necessary to add up for each circuit pattern. At this time, the result may be displayed and transmitted to the instruction unit 30.

  When image processing is completed for all circuit patterns on the substrate (Y branch in S134), end processing (S140) is performed. The termination processing is displayed on the instruction unit 30 together with the number of defects and the defect locations for each circuit pattern. Of course, these data may be recorded.

  As described above, the pattern inspection apparatus according to the present invention continuously shoots while moving the substrate on the stage in one direction (Y-axis direction) with respect to the area camera. The image data acquisition time can be shortened.

  The present invention can be suitably used for automatic defect inspection of circuit patterns.

It is a figure which shows the external appearance of the pattern inspection apparatus of this invention. It is a figure which shows the structure of the pattern inspection apparatus of this invention. It is a figure which shows an example of the flash provided in the area camera of the present invention. It is a figure explaining the board | substrate and circuit pattern which are test | inspected with the pattern test | inspection apparatus of this invention. It is a figure explaining the relationship between a circuit pattern and a photography visual field. It is a timing chart of a shutter, strobe, and moving speed. It is a figure explaining the example of acquisition of inspection standard image data. It is a figure which shows the processing flow of the pattern inspection apparatus of this invention. It is a figure which shows the flow of the defect inspection of a pattern inspection apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Pattern inspection apparatus 2 Stage 4 Arm 5 Groove 6 Y-axis slider 7 Strobe 8 Area camera 10 θ-axis adjustment mechanism 12 Inspected object (substrate)
DESCRIPTION OF SYMBOLS 13 Alignment mark 14 Circuit pattern 16 Drive part 20 Control apparatus 21 Control part 25 Image processing part 26 Memory | storage part 30 Instruction part 53,54,55,56 Field of view 71 Strobe light emission source 72 Power supply 81 Lens

Claims (7)

A circuit pattern inspection apparatus in which a plurality of circuit patterns are formed on a substrate,
A stage for holding the substrate;
An area camera that images the substrate, and a strobe light source that emits light to the substrate;
A Y-axis slider for relatively moving the stage and the area camera;
A θ-axis adjustment mechanism for rotating the substrate on the stage;
An X-axis slider for moving the area camera in a direction perpendicular to the Y-axis slider;
Position information of a plurality of patterns on the substrate is input, a shooting plan is created based on the position information, and shooting is continuously performed by the area camera while moving the Y-axis slider based on the shooting plan. A pattern inspection apparatus having a control device that compares image data captured by the area camera with inspection reference image data to detect defects. 2. The imaging plan is an imaging plan in which a positional relationship between each circuit pattern and a visual field range captured by the area camera is the same for all the plurality of circuit patterns for a plurality of circuit patterns on the substrate. Pattern inspection device described in 1. The pattern inspection apparatus according to claim 1, wherein the photographing plan is a photographing plan in which an overlapping distance between continuous visual fields of the area cameras is equal. The shooting plan includes a circuit pattern length A in the Y-axis direction that is a moving direction during imaging of the area camera, an arrangement interval C of the circuit patterns, a length a of the area camera field of view in the Y-axis direction, The pattern inspection apparatus according to claim 1, wherein a length fy in the Y-axis direction of a field overlapping portion is an imaging plan that satisfies the expression (1).
n = (A + C) / (a-fy) where n is an integer (1) The pattern inspection apparatus according to claim 1, wherein the strobe light source is a strobe light source that irradiates light from at least four directions to a field of view of the area camera. A pattern inspection method for a plurality of circuit patterns formed on a substrate,
Creating a shooting plan in which the phase of the shooting field of view between the circuit patterns is equal from the position information of the circuit pattern on the substrate and the field of view information of the area camera;
Acquiring inspection reference image data based on the imaging plan from a good product;
Continuously acquiring image data of the circuit pattern based on the shooting plan;
A pattern inspection method comprising a step of comparing the image data with the inspection reference image data to detect a defect in the image data. The shooting plan includes a circuit pattern length A in the Y-axis direction that is a moving direction during imaging of the area camera, an arrangement interval C of the circuit patterns, a length a of the area camera field of view in the Y-axis direction, The pattern inspection method according to claim 6, wherein the length fy of the overlapping field of view in the Y-axis direction is an imaging plan that satisfies formula (1).
n = (A + C) / (a-fy) where n is an integer (1)