JP2014109511A - Positioning method, positioning device, and die bonder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a positioning device and a die bonder capable of minimizing failures of image recognition and performing a reliable positioning even when an object pattern is unstable or a retrieval object is tilted.SOLUTION: Illumination means 16 performs an initial illumination by either a bright field illumination or a dark field illumination on a retrieval object, selects one of plural kinds of matching methods in the state and sequentially performs it by processing means 17. First determination means 18 can determine whether or not recognition processing is successful. In control means 19, the recognition processing for the retrieval object is finished without performing a next matching method when the recognition processing is successful. When the recognition processing is unsuccessful with any kinds of matching methods, the control means 19 switches the initial illumination to a different field illumination and the processing means 17 sequentially performs a matching method selected by the selection means.

Description

  The present invention relates to a positioning method, a positioning device, and a die bonder.

  Conventionally, a die bonder is used when bonding a chip formed by dicing a wafer. That is, a wafer ring is set on the wafer table of the die bonder, and a plurality of chips are stuck on the wafer sheet stretched on the wafer ring. Then, the chip is picked up by a pick-up device and mounted on a substrate such as a lead frame.

  By the way, when picking up, it is necessary to align the chip. That is, this type of pickup device generally picks up a chip by sucking it with a suction collet. For this reason, if the chip is misaligned, it may not be possible to mount the chip at the proper position.

  Therefore, conventionally, it has been proposed to perform chip alignment using pattern matching. Pattern matching is a process of associating the same pattern that exists in a certain image with which part in another image, and a typical example is template matching. . That is, as shown in FIG. 7, a method in which a pattern called a template 51 expressed in units of pixels is prepared in advance, and this is overlapped with the image 52 while moving, and a pattern with high similarity is detected from the search target image. It is. There are various position detection devices using pattern matching as described in Patent Document 1, Patent Document 2, and the like.

JP 2008-283006 A JP 2006-41089 A

  However, in pattern matching, image recognition may fail if the recognition pattern is unstable or the search object is tilted. That is, depending on the quality of the workpiece (search object), there are cases where the pattern of the target is unstable or the shape changes.

  Therefore, in view of such circumstances, the present invention can reduce the failure of image recognition as much as possible even if the target pattern is unstable or the search target is inclined, and is reliable. An object of the present invention is to provide a positioning method, a positioning device, and a die bonder capable of performing high positioning.

  The positioning method of the present invention is a positioning method for specifying the position of a search object, and performs an initial illumination of either bright field illumination or dark field illumination on the search object, and in that state, a plurality of types of When the matching process is sequentially performed and the recognition process is completed without performing the next matching technique when the recognition process is successful, and the recognition process is not successful even if all types of matching techniques are performed, In this state, the initial illumination is switched to a different field of view illumination, and multiple types of matching methods are sequentially performed. When the recognition processing is successful, the recognition processing of the search object is completed without performing the next matching method. If the recognition process is not successful even if all kinds of methods are performed, this recognition target object recognition process is failed.

  Here, bright field illumination refers to illuminating a light beam illuminating a measurement object vertically along the center of the optical axis. Dark field illumination refers to irradiating a light beam illuminating a measurement object from an oblique direction, not the center of the optical axis. In other words, bright-field illumination is a type of illumination that directly observes the reflected or transmitted light, and the sample is placed in a bright background of the color of the illumination light (bulb color, warm white, etc.) Observe the light and look darker than the background. Dark field illumination is a type of illumination in which irradiated light is reflected or scattered to observe scattered light, and a bright sample floats on a dark background and is observed. For this reason, in the dark field illumination, it is possible to observe fine structures, scratches, defects, and the like that were not seen blurred in the bright field illumination.

  According to the positioning method of the present invention, first, a plurality of types of matching methods can be performed with either bright field illumination or dark field illumination. In this case, recognition can be successfully performed with either matching method. Thus, the search object can be positioned. Further, if the recognition is not successful with the plurality of types of matching methods, the plurality of types of matching methods can be performed again by switching to different illuminations. At this time, if the recognition is successful by any of the matching methods, the search object can be positioned.

  For this reason, in the positioning method of the present invention, various matching methods can be used, and an optimum matching method can be executed for various search objects. Depending on the search target, recognition is not successful in bright field illumination, but recognition is successful in dark field illumination. Conversely, recognition is not successful in dark field illumination, but recognition is successful in bright field illumination. Even in such a case, the matching method can be executed with both illuminations. Therefore, the probability of successful recognition is extremely high.

  Multiple types of matching methods include shape-based matching pattern matching that uses the shape pattern of the search target as a positioning model, maximum deformation matching with a large tolerance for shape change in this shape-based matching, and each pixel in the region. There is a normalized correlation matching that normalizes the luminance values to form a positioning model.

  Shape-based matching can be detected when the shape is projected even if the luminance of the illumination changes, but may not be detected when the shape changes greatly. Note that it is possible to detect a case where the entire shape has changed to the same magnification by scale matching.

  In the maximum deformation matching, it is possible to give a certain allowable range to each point on the edge of the model used for pattern matching, and it is possible to realize matching flexibly corresponding to various shape changes. That is, even if a part of the shape changes, it can be detected if it is within a certain allowable range.

  In the normalized correlation matching, when the edge of the search object does not appear clearly, the internal luminance value can be used. However, it cannot cope with nonlinear illumination changes, shielding, cracks, and the like.

  The initial illumination may be bright field illumination or dark field illumination. Moreover, it is preferable that the matching method to be selected is sequentially performed from a method with a short processing time. The processing time is shape base matching <maximum deformation matching <normalized correlation matching.

  The positioning device of the present invention is a positioning device that specifies the position of a search object, and is an illumination unit that can illuminate the search object with either bright field illumination or dark field illumination, and a plurality of types of matching. A processing unit that sequentially selects one matching method from the methods and performs a recognition process using the selected matching method; a first determination unit that determines whether the recognition process using the selected matching method is successful; When the recognition process is judged to have succeeded, the next matching method stops the recognition process, and when the recognition process is not successful even if all types of matching methods are performed, the illumination means is changed to a different field of view illumination. Switching, sequentially performing a plurality of types of matching methods again, and when the recognition processing is successful, the control means for ending the search object recognition processing without performing the next matching method; Second recognition means for determining that the recognition process for the search object is unsuccessful if the recognition process is not successful even after all types of matching techniques have been performed after the field-of-view illumination has been switched. Is.

  According to the positioning device of the present invention, the illumination unit performs initial illumination of either bright field illumination or dark field illumination on the search object, and in that state, one matching method is selected from a plurality of matching methods. Can be selected and sequentially performed by the processing means. And it can be judged by the 1st judgment means whether the recognition process was successful. The recognition process of the search object can be completed without performing the next matching method when the recognition process is successful in the control means. Further, the control means can switch the initial illumination to a different field illumination when the recognition process is not successful even if all types of matching methods are performed. After switching to the field illumination, the matching method selected by the selection unit can be sequentially performed by the processing unit. And it can be judged by the 1st judgment means whether the recognition process was successful. The recognition process of the search object can be completed without performing the next matching method when the recognition process is successful in the control means. If the recognition process is not successful even after all types of matching techniques have been performed after switching the field of view illumination once, the recognition process for the search object has failed in the second determination means. Judgment can be made. That is, according to the positioning device of the present invention, the positioning method of the present invention can be executed.

  The die bonder of the present invention includes the positioning device. For this reason, the positioning method of the present invention can be executed.

  In the present invention, the probability of successful recognition is extremely high, and even if the target pattern is unstable or tilted, it is easy to be recognized successfully, and high-accuracy positioning can be performed.

  If multiple types of matching methods are performed sequentially and if the recognition is successful, matching will end at that point. The processing time can be shortened and positioning with excellent work efficiency can be performed.

  In the die bonder of the present invention, highly accurate positioning can be performed, and a stable bonding process is possible.

It is a simplified lineblock diagram of an embodiment of a positioning device of the present invention. It is a principal part simplification figure of the die bonder of this invention. It is a flowchart figure in the case of bright field illumination which shows the positioning method of this invention. It is a flowchart figure in the case of dark field illumination which shows the positioning method of this invention. It is the schematic which shows the bonding process using the die bonder of this invention. It is a simplified perspective view which shows a wafer. It is explanatory drawing of pattern matching.

  Hereinafter, embodiments of the present invention will be described with reference to FIGS.

  FIG. 1 shows a positioning device, and FIG. 5 shows a bonding process of a die bonder using this positioning device. The die bonder is for piping up a chip 1 cut out from a wafer 9 (see FIG. 6) at a pickup position P and transferring (mounting) it to a bonding position Q of a substrate 2 such as a lead frame. As shown in FIG. 1, the wafer 9 is divided (divided) into a large number of chips 1 by a dicing process. Therefore, this chip 1 is arranged in a matrix as shown in FIG. 6 and is the search object W of the present invention.

  As shown in FIG. 5, the die bonder includes a collet (adsorption collet) 3. The collet 3 is moved by an unillustrated moving mechanism in the direction of arrow A and lowered in the direction of arrow B on the pickup position P, and in the direction of arrow C and lowered in the direction of arrow D on the bonding position Q. The reciprocation between the pickup position P and the bonding position Q in the directions of arrows E and F is possible. In the moving mechanism, the movement of the arrows A, B, C, D, E, and F is controlled by a control unit configured by, for example, a microcomputer. In addition, as a moving mechanism, it can comprise with various mechanisms, such as a cylinder mechanism, a ball screw mechanism, and a motor linear mechanism.

  The suction collet 3 is provided with a head (suction nozzle) 4 having an suction hole 8 opened on the lower surface thereof, and the chip 1 is vacuum-sucked through the suction hole 8, and the chip 1 is placed on the lower end surface (tip surface) of the head 4. Adsorbs. When this vacuum suction (evacuation) is released, the chip 1 is detached from the head 4.

  Further, the wafer 9 divided (divided) into a large number of chips 1 is placed on, for example, an XYθ table 5 (see FIG. 6), and the XYθ table 5 is provided with push-up means having push-up pins. That is, the push-up means pushes up the chip 1 to be picked up from below and makes it easy to peel from the adhesive sheet. In this state, the chip 1 is adsorbed to the adsorbing collet 3 that has been lowered.

  By the way, when the chip 1 is picked up at the pickup position P, the chip 1 needs to be in a normal posture. For this reason, the positioning device according to the present invention is used in this die bonder.

  FIG. 2 shows a die bonder including the positioning device of the present invention. This die bonder has a camera (CCD camera) 12 as an imaging means for observing the chip 1 and an illumination means 16 for illuminating the chip 1 (FIG. 1). Reference). The camera 12 and the illumination means 16 are controlled by the control unit 15. And this control part 15 comprises the positioning device of this invention.

  As shown in FIG. 1, the control unit 5 performs recognition processing using the illumination unit 16 that can illuminate the search object W (chip 1) and a matching method selected from a plurality of types of matching methods. A processing unit 17; a first determination unit 18 that determines whether or not the recognition process using the selected matching method has succeeded; a control unit 19 that performs illumination switching control of the illumination unit 16; And a second determination unit 20 that determines that the process has failed.

  That is, the illumination unit 16 includes a bright field illumination 22 and a dark field illumination 23 as shown in FIG. Bright field illumination refers to illuminating a light beam that illuminates a measurement object vertically along the center of the optical axis. Dark field illumination refers to irradiating a light beam illuminating a measurement object from an oblique direction, not the center of the optical axis. In other words, bright field illumination is a type of illumination that observes direct light reflected or transmitted by a sample, and is placed in a bright background of the color of the illumination light (bulb color, warm white, etc.). Observe the light and look darker than the background. Dark field illumination is a type of illumination in which irradiated light is reflected or scattered to observe scattered light, and a bright sample floats on a dark background and is observed. For this reason, in the dark field illumination, it is possible to observe fine structures, scratches, defects, and the like that were not seen blurred in the bright field illumination.

  A plurality of types of matching methods are input to the processing unit 17, and one matching method is sequentially performed under the control of the control unit 19. Matching methods include shape-based matching pattern matching that uses the shape pattern of the search object as a positioning model, maximum deformation matching with a large tolerance for shape change in this shape-based matching, and the luminance value at each pixel in the region. There is a normalized correlation matching that normalizes and forms a positioning model.

  Shape-based matching can be detected when the shape is projected even if the luminance of the illumination changes, but may not be detected when the shape changes greatly. Note that it is possible to detect a case where the entire shape has changed to the same magnification by scale matching.

  In the maximum deformation matching, it is possible to give a certain allowable range to each point on the edge of the model used for pattern matching, and it is possible to realize matching flexibly corresponding to various shape changes. That is, even if a part of the shape changes, it can be detected if it is within a certain allowable range.

  In the normalized correlation matching, when the edge of the search object does not appear clearly, the internal luminance value can be used. However, it cannot cope with nonlinear illumination changes, shielding, cracks, and the like.

  The processing time is shape base matching <maximum deformation matching <normalized correlation matching.

  The processing unit 17, the first determination unit 18, the control unit 19, the second determination unit 20, and the like are, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), or the like around a CPU (Central Processing Unit). It can be composed of microcomputers connected to each other via A storage device is connected to the microcomputer. The storage device stores determination criteria that are the determination criteria of the first determination means 18 and the second determination means 20, and various matching methods. The storage device can be composed of an HDD (Hard Disc Drive), a DVD (Digital Versatile Disk) drive, a CD-R (Compact Disc-Recordable) drive, an EEPROM (Electronically Erasable and Programmable Read Only Memory), or the like. The ROM stores programs executed by the CPU and data.

  Next, a positioning method using the positioning device configured as described above will be described with reference to FIGS. First, as bright field illumination (step S1), pattern matching (shape base matching) is performed (step S2). Then, it is determined whether or not the recognition by the pattern matching is successful (step S3). That is, if the position of the chip 1 to be picked up is normally detected, the recognition ends and the process ends. If the position of the chip 1 is not normally detected, the process proceeds to step S4.

  In step S4, maximum deformation matching is performed. Thereafter, the process proceeds to step S5 to determine whether the recognition is successful. That is, if the position of the chip 1 to be picked up is normally detected, the recognition ends and the process ends. If the position of the chip 1 is not normally detected, the process proceeds to step S6.

  In step S6, normalized correlation matching is performed. Thereafter, the process proceeds to step S7 to determine whether the recognition is successful. That is, if the position of the chip 1 to be picked up is normally detected, the recognition ends and the process ends. If the position of the chip 1 is not normally detected, the process proceeds to step S8 in FIG.

  In step S8, it is determined whether shape-based matching, maximum deformation matching, and normalized correlation matching have been executed. If these methods are executed, the process proceeds to step S9. If these methods are not executed, the process proceeds to step S2. In this determination, if any of the three matching methods has not been performed, it may be determined that the unmatched method is performed. That is, if all three matching methods are performed and the recognition is not successful, the process proceeds to step S9.

  In step S9, the illumination is switched from bright field illumination to dark field illumination. Then, it transfers to step S10 and performs pattern matching (shape base matching). Then, it is determined whether or not the recognition by the pattern matching is successful (step S11). That is, if the position of the chip 1 to be picked up is normally detected, the recognition ends and the process ends. If the position of the chip 1 is not normally detected, the process proceeds to step S12.

  In step S12, maximum deformation matching is performed. Thereafter, the process proceeds to step S13 to determine whether the recognition is successful. That is, if the position of the chip 1 to be picked up is normally detected, the recognition ends and the process ends. If the position of the chip 1 is not normally detected, the process proceeds to step S14.

  In step S14, normalized correlation matching is performed. Thereafter, the process proceeds to step S15 to determine whether the recognition is successful. That is, if the position of the chip 1 to be picked up is normally detected, the recognition ends and the process ends. If the position of the chip 1 is not normally detected, the process proceeds to step S16.

  In step S16, it is determined whether shape-based matching, maximum deformation matching, and normalized correlation matching have been executed. If these methods are executed, the process ends as unsuccessful recognition. If these methods are not executed, the process proceeds to step S10. In this determination, if any of the three matching methods has not been performed, it may be determined that the unmatched method is performed. That is, if all of the three matching methods are performed and the recognition is not successful, the recognition is unsuccessful.

  In the embodiment, first, a plurality of types of matching methods can be performed with bright field illumination. At this time, if the recognition is successful with any of the matching methods, the search object can be positioned. In addition, when the recognition is not successful by the plurality of types of matching methods, the plurality of types of matching methods can be performed again by switching to different illumination (dark field illumination). At this time, if the recognition is successful by any of the matching methods, the search object can be positioned.

  For this reason, in the positioning method of the present invention, various matching methods can be used, and an optimum matching method can be executed for various search objects W. Further, depending on the search object W, there are those that do not succeed in recognition with bright field illumination, but succeed in recognition with dark field illumination, and conversely do not succeed in recognition with dark field illumination, but do succeed in recognition with bright field illumination. Even in this case, the matching method can be executed with both illuminations. Therefore, the probability of successful recognition is extremely high. For this reason, even if the target pattern is unstable or tilted, it becomes easy to succeed in recognition, and high-accuracy positioning can be performed.

  If multiple types of matching methods are performed sequentially and if the recognition is successful, matching will end at that point. The processing time can be shortened and positioning with excellent work efficiency can be performed.

In the die bonder of the present invention, highly accurate positioning can be performed, and a stable bonding process is possible.

  As described above, the embodiment of the present invention has been described. However, the present invention is not limited to the above-described embodiment, and various modifications are possible. For example, in the above-described embodiment, after performing bright field illumination, dark field is performed. However, it may be switched to bright field illumination after performing dark field illumination. Further, the order of execution of the matching method is not limited to that of the above embodiment. Furthermore, other matching methods may be added as a matching method. That is, as other matching methods, there are a method in which a parameter (recognition parameter) is changed in shape-based matching, an edge matching for detecting an edge at a specific position of an object, and the like. If the recognition parameter is changed (changed), the pattern can be searched more finely. The change parameters include polarity inversion, weighting, minimum score, and the like. Edge detection can be used when the entire shape cannot be used as a positioning model or when a specific point is set as a positioning reference.

13 Illuminating device 16 Illuminating means 17 Processing means 18 First judging means 19 Control means 20 Second judging means 22 Bright field illumination 23 Dark field illumination W Search object

Claims (10)

A positioning method for specifying a position of a search object,
Perform initial illumination of either bright-field illumination or dark-field illumination on the search object, and sequentially perform multiple types of matching methods in that state, and perform the next matching method when the recognition process is successful. If the recognition process is not successful even after completing the recognition process of the search object and performing all types of matching methods, the initial illumination is switched to a different visual field illumination, and in that state, a plurality of types of matching methods are sequentially performed. When the recognition process is successful, the recognition process of the search object is terminated without performing the next matching method, and if the recognition process is not successful even if all types of matching methods are performed, the recognition of the search object is performed. A positioning method, characterized by processing failure.   The positioning method according to claim 1, wherein the plurality of types of matching methods include at least pattern matching of shape base matching using a shape pattern of a search object as a positioning model.   The plurality of types of matching methods include at least shape-based matching using a shape pattern of a search object as a positioning model and maximum deformation matching in which a large allowable amount for shape change is set in the shape-based matching. The positioning method according to claim 1.   The plurality of types of matching methods include at least shape-based matching that uses the shape pattern of the search target as a positioning model and normalized correlation matching that normalizes the luminance value at each pixel in the region to form a positioning model. The positioning method according to claim 1, wherein:   The multiple types of matching methods include shape-based matching that uses the shape pattern of the search object as a positioning model, maximum deformation matching with a large allowable amount for shape change in this shape-based matching, and each pixel in the region. The positioning method according to claim 1, further comprising at least normalized correlation matching that normalizes luminance values to form a positioning model.   6. The positioning method according to claim 1, wherein the initial illumination is bright field illumination.   The positioning method according to claim 1, wherein the initial illumination is dark field illumination.   The positioning method according to claim 1, wherein the matching method is sequentially performed from a method having a short processing time. A positioning device for specifying a position of a search object,
Illumination means that can illuminate the search object with either bright field illumination or dark field illumination, and one matching method is sequentially selected from multiple types of matching methods, and recognition processing by the selected matching method is performed. Processing means for performing
First determination means for determining whether or not the recognition processing with the selected matching method is successful;
When it is judged that the recognition process is successful, the recognition process in the next matching method is stopped, and when the recognition process is not successful even if all types of matching methods are performed, the illumination means is switched to a different visual field illumination. Then, a plurality of types of matching methods are sequentially performed again, and when the recognition processing is successful, the control means for ending the search object recognition processing without performing the next matching method,
And a second determination means for determining that the recognition processing for the search object is unsuccessful when the recognition processing is not successful even after all types of matching methods have been performed after the field-of-view illumination switching. A positioning device characterized by that.   A die bonder comprising the positioning device according to claim 9.

Images