JP3537679B2 - Method for determining parameters of automatic focusing device and storage medium storing the program - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for determining parameters of an automatic focusing apparatus and a storage medium storing a program thereof, and more particularly to a method of positioning a target component prior to a positioning process of a micro component in a semiconductor manufacturing line or the like. The present invention relates to a method for determining parameters of an automatic focusing device applied to an automatic focusing process executed for accurately grasping a shape and the like, and a storage medium storing a program thereof.

[0002]

2. Description of the Related Art In recent years, semiconductor devices have become highly integrated,
In a semiconductor manufacturing line, it is required to accurately and promptly handle a miniaturized object (such as a semiconductor chip). In order to respond to such a demand, it is necessary to accurately and quickly grasp the shape and the like of each minute component by an automatic focusing process.

[0003] The schematic structure of a conventional autofocus device is as follows.
This will be described with reference to the drawings. As shown in FIG. 6, the automatic focusing device is roughly divided into an XY stage 20 on which the object 10 is mounted, and an optimal focus position (focusing position) by moving in a direction perpendicular to the object 10. Is set in the optical barrel 30
A CCD camera 40 that converts an image captured through the optical barrel into an electric signal; and an image processing unit 50 that generates image information about the object 10 based on the electric signal output from the CCD camera 40. And a CPU, a memory, and the like, and outputs and displays image information obtained by the image processing unit 50 on the monitor 80 and the like, and controls the drive system 70 to arbitrarily set the XY stage 20 and the optical barrel unit 30. And a control unit 60 that adjusts the drive to the state described above.

Next, an automatic focusing process in the above-described automatic focusing device will be described with reference to the drawings. As shown in FIGS. 7 and 8, the automatic focus processing is roughly divided into a rough adjustment processing and a fine adjustment processing. First, as shown in FIG. 8, the depth of focus of the optical lens barrel 30 is changed at a preset sampling interval Δr, and a normal correlation value is indicated for each sampling position over a wide range A (indicated by X in FIG. 8). And a standard deviation value are calculated (S31).

Next, a point Pr showing the maximum value is extracted as an optimum point from the normal correlation values calculated by sampling by the coarse adjustment (S32). Next, with respect to the narrowed range B near the optimum point Pr, the focal depth of the optical barrel 30 is changed at an interval Δv smaller than the sampling interval Δr, and a normal correlation value (○ in FIG. Is calculated (S33).

Next, the point Pv showing the maximum value is extracted as a true optimum point from the normal correlation values calculated by the fine adjustment sampling (S34). Then, the depth of focus a indicating the true optimum point Pv is determined as the focus position (S3).
5) The optical lens barrel 30 is moved to the corresponding position. In addition,
In the automatic focus processing, when the standard deviation value or the luminance variance value is calculated to determine the in-focus position, unlike the normal correlation value described above, the optimum point is extracted by the minimum value in the coarse adjustment and the fine adjustment. Is done.

As described above, normally, when performing the automatic focusing process, a candidate point is set within a certain range at a predetermined sampling interval, and the feature value of the image at the candidate point is set.
With respect to the normal correlation value, the standard deviation value, and the like, a method is adopted in which the maximum or minimum depth of focus is determined as the optimal focus position (focus position). Then, in order to efficiently search for the in-focus position from a relatively wide adjustment range, first, the width between the candidate points, that is, the sampling range is roughly set to narrow the adjustment range by searching, and then the candidate point interval is further reduced. Is narrowed and a search is performed again, which is a two-stage adjustment process.

Therefore, the efficiency of the automatic focusing process can be improved by optimizing the balance of the sampling in the coarse adjustment and the fine adjustment.

[0009]

In the above-described automatic focus processing, usually, a method of empirically or recursively setting a sampling interval and a number of adjustment processing at the time of coarse adjustment, or an analytical calculation. By adopting a method of calculating an optimal value by using the method, the processing is optimized.

Here, a method of calculating the ideal value Δr of the sampling interval obtained by the analytical method will be described. First, the sum N of the number of times of coarse adjustment and the number of times of fine adjustment is obtained by equation (1), and the derivative of N is calculated as in equation (2).

[0011]

(Equation 1)

In the formula (2), by executing the automatic focusing process under the condition that N becomes the minimum, it is possible to execute the extraction of the optimum point accurately in a short time using the efficient number of candidate points. Therefore, when the condition of the expression (3) is satisfied,
Since N is minimized, the theoretical value Δr of the sampling interval applied to the coarse adjustment can be calculated as shown in Expression (4).

[0013]

(Equation 2)

However, in the above-described method of calculating the sampling interval, that is, the parameter, the ideal value calculated uniquely based on the two factors of the accuracy required for the automatic focusing process and the width of the adjustment range is used. This is effectively applied to an ideal normal distribution in which the curve shape of the normal correlation value shows a monotonic increase and a monotone decrease as shown in FIG. In the case of a complicated curve shape having the following, there is a problem that it is impossible to accurately grasp the maximum value and extract the optimum point.

Specifically, as shown in FIG. 9, in the case of a profile shape in which two maximum values of PL and PH exist in the normal correlation value due to the influence of peripheral patterns and the like, the sampling interval Δr applied to the coarse adjustment is used. The maximum value PH which becomes the maximum value at
, The maximum value PH is excluded from the search target, and a candidate point near the other maximum value PL is determined to be the maximum value.

In particular, as shown in FIG. 9, the curve shape of the normal correlation value calculated from an image of a micro component actually observed rarely shows an ideal normal distribution due to various causes. . Therefore, when the characteristic value of an image shows a complicated curve shape (profile shape) due to the coarse adjustment using the sampling interval (ideal value) Δr as described above, extraction of the optimum point by automatic focusing processing, that is, focusing There is a problem that the accuracy of determining the position is significantly reduced, the shape of a micro component or the like cannot be grasped, and the processing efficiency such as alignment is reduced.

The parameter setting in the automatic focusing process as described above is usually performed only once as an initial setting process before the operation of the automatic production apparatus, and the optimal value (focusing depth) of the once set focal depth is set. Position) was not changed during operation of the production equipment.
However, in mass production, even slight variations in the shape between the micro parts to be processed and the mounting status on the stage, etc., can easily have a large effect on the auto focus processing,
The focus position set in advance does not sufficiently cope with such variations, causing a problem that the production efficiency is reduced.

In particular, the maximum value of the distribution curve relating to the image feature amount as described above and the position of the depth of focus indicating the maximum value are extremely affected by individual slight variations even in the same type of component or product. It is desirable to always evaluate and verify the maximum value during operation. The present invention
In order to solve the above-described problems, the optimum focus position can be accurately and quickly searched and determined regardless of the profile shape of the feature amount of the processing target, and also during the operation of the automatic production device. It is an object of the present invention to provide a method for determining parameters of an automatic focusing apparatus capable of appropriately correcting a focus position and a storage medium storing a program thereof.

[0019]

In order to solve the above-mentioned problems, the invention according to claim 1 reduces the focus adjustment range by at least a two-stage focus adjustment process including a coarse adjustment and a fine adjustment. Extracting a specific information related to the image in the adjustment range, creating a distribution curve of the specific information, and a first sampling applied to the coarse adjustment. Calculating an interval, determining an extreme value present in the distribution curve based on the first sampling interval, and excluding the extreme value when the extreme value is determined to be present A second sampling interval defined by a threshold value set for the area is set, and a smaller one of the first and second sampling intervals is set to Applying the coarse adjustment and, if the second sampling interval is smaller than the first sampling interval, performing the coarse adjustment again by applying the second sampling interval. It is characterized by including.

According to a second aspect of the present invention, in the method for determining a parameter of the automatic focusing apparatus according to the first aspect, the distribution curve of the specific information is a distribution curve of a normal correlation value for an image in the adjustment range. Further, the extreme value is a maximum value of the normal correlation value curve. According to a third aspect of the present invention, in the parameter determination method of the automatic focusing apparatus according to the first aspect, the distribution curve of the search information is a distribution curve of a luminance variance value with respect to an image in the search range, and The extreme value is a minimum value of the brightness dispersion value curve.

According to a fourth aspect of the present invention, there is provided a parameter determining method for an automatic focusing apparatus for determining a focus position while narrowing a focus adjustment range by a two-step focus adjustment process including at least a coarse adjustment and a fine adjustment. A process of extracting specific information relating to the image in the adjustment range, creating a distribution curve of the specific information, and a first process applied to the coarse adjustment.
A process of calculating a sampling interval of, and a process of determining an extreme value present in the distribution curve based on the first sampling interval, and when it is determined that the extreme value exists, the extreme value is calculated. A process of setting a second sampling interval defined by a threshold value set in a region not included, and applying the smaller one of the first and second sampling intervals to the coarse adjustment And a process of executing the coarse adjustment again by applying the second sampling interval when the second sampling interval is smaller than the first sampling interval. It is characterized by being a medium.

That is, according to the present invention, by having the above-described series of processing procedures, the sampling interval (parameter) in the coarse adjustment is adjusted, and the extreme value at which the specific information (feature amount) of the image becomes maximum or minimum is determined. Since the adjustment range including the adjustment range is extracted, and the optimum point can be accurately determined from the adjustment range extracted in the fine adjustment, even for an image having a complicated profile shape, the focus position can be accurately and quickly determined. Can be determined.

In particular, the interval at which the feature amount of the image of the object is sampled is determined based on the presence or absence of an extreme value and a comparison result between a theoretical value (first sampling interval) and a correction value (second sampling interval). By adjusting and determining the parameters, the position at which the depth of focus is optimal can be set accurately. Then, by applying the present invention to a production device equipped with an automatic focusing device, the sampling interval in the coarse adjustment is adjusted as needed, and the automatic focusing process is executed at an accurate focusing position even for an individual object. Therefore, it is possible to shorten the processing time of the alignment processing and the like, improve the yield, and improve the production efficiency.

In addition, by incorporating the program relating to the method for determining the parameters of the automatic focusing apparatus into the control unit of the automatic focusing apparatus, the optimum focusing position can be determined accurately and quickly regardless of the shape of the object. In addition, since predetermined production processing such as alignment processing can be performed automatically and automatically, it can be easily applied at low cost to an existing autofocus apparatus or a production apparatus equipped with an autofocus apparatus. it can.

[0025]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS (Basic Concept) First, a basic concept of a method for determining parameters of an automatic focusing apparatus according to the present invention will be described with reference to the drawings. FIG. 1 is a flowchart showing the algorithm of the present invention, and FIG. 2 is a diagram showing the relationship between the shape of the evaluation value profile and the sampling interval. Hereinafter, description will be given with reference to FIG. 2 in accordance with the procedure of the flowchart shown in FIG.

First, an evaluation value profile is created for a micropart to be an object (S11). The evaluation value profile is a sampling interval at which the entire adjustment range is actually applied to the fine adjustment processing. An actual measurement value obtained by acquiring the data of the characteristic amount is stored in a memory of a computer, and is used as a profile.

More specifically, a sample image of a minute part to be subjected to the automatic focusing process is continuously obtained by finely changing the height of a CCD camera (optical barrel), and the characteristic amount of each image, that is, a normal value, is obtained. Calculate correlation value, standard deviation value, etc.
It is created by displaying as a dimensional data array (evaluation amount for height). Next, as shown in the above equations (1) to (3), the theoretical value Δr of the sampling interval applied to the coarse adjustment is analytically calculated as the number of candidate points at the time of coarse adjustment and the number of candidate points at the time of fine adjustment. (S12) so that the sum is minimized.

Then, it is determined from the evaluation value profile whether or not there is a local maximum value PL in the coarse adjustment using the sampling interval Δr (S13). If the maximum value does not exist within the predetermined adjustment range, it is determined that the sampling interval based on the theoretical value Δr is appropriate. On the other hand, when the maximum value PL exists, the threshold value TH that is equal to or greater than the relevant maximum value is set, and the intersection P between the evaluation value profile and the threshold value TH is set.
1. The width of the depth of focus determined from P2 is extracted and set to a new (corrected) sampling interval Δr ′ (S14).

Then, the magnitude relationship between the theoretical value Δr of the sampling interval and the correction value Δr ′ is compared. If the theoretical value Δr is smaller, the maximum value extracted by the theoretical value Δr is used as the evaluation value profile. Judging that it contains the maximum value that becomes the maximum value, as described in the related art, performs a fine adjustment by applying a sampling interval Δv narrower than the sampling interval Δr, and sets the maximum value that becomes the maximum value to: The in-focus position is determined by extracting it as a true optimum point (S15).

On the other hand, when the correction value Δr ′ is smaller, in addition to the maximum value PL extracted by the theoretical value Δr,
It is determined that the maximum value PH that is the maximum value of the evaluation value profile is included, and the correction value Δr ′ is newly adopted as a sampling interval to be applied to the coarse adjustment (S16). Repeat the algorithm. According to such an algorithm, a threshold value equal to or greater than the maximum value extracted in the coarse adjustment is set, and the depth of focus defined by the threshold value is set to a new (corrected) sampling interval. By comparing the theoretical value and the correction value of the preset sampling interval with each other, it is determined whether or not the maximum value extracted in the evaluation value profile is the maximum, and the maximum value that can be the optimum point is appropriately extracted. can do.

Therefore, the existence of the local maximum value which becomes the true optimum point which has been overlooked depending on the sampling interval based on the theoretical value can be accurately grasped at the stage of the coarse adjustment.
By the fine adjustment, the true optimum point can be reliably extracted to determine the focus position. By the way, in the above-described series of parameter determination processing, as shown in FIG. 3, a plurality of local maxima (P in FIG.
H, PH ′) are close to each other, the narrower sampling interval Δr is set within the range of the sampling interval Δr in the fine adjustment as described in the related art.
Since the search is performed with v and the true optimum point PH can be reliably extracted, it is not necessary to correct the sampling interval Δr applied to the coarse adjustment.

[0032]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment to which a method for determining parameters of an automatic focusing apparatus according to the present invention is applied will be described with reference to the drawings. The present embodiment relates to an automatic production apparatus having a focusing and positioning function by an automatic focusing process and capable of realizing mass production of minute parts and the like.

That is, as shown in FIG. 4, even if the products A, B, and C are of the same type, the above-described evaluation values are obtained due to slight differences in the shapes of the products, the mounting state on the stage, and the like. Increase or decrease of the maximum or minimum value in the profile,
In order to cause a change in the position of the focus advance, it is necessary to grasp the state of the extreme value for each individual product. FIG. 5 is a flowchart showing an algorithm for automatically correcting parameters related to the automatic focusing process while the automatic production device is operating.

First, for each product actually flowing in the production line, based on the above-described automatic focus processing,
The maximum value Pm 'is extracted (S21). Next, the maximum value Pm set as an initial value or the current value is compared with the maximum value Pm ′ extracted for the product actually flowing in the production line (S22).

When the actual maximum value Pm 'is larger than the currently set maximum value Pm, the sampling interval Δr at the time of the coarse adjustment is corrected based on the parameter determination method shown in FIG. (S23). On the other hand, when the actual maximum value Pm ′ is kept smaller than the currently set maximum value Pm, the automatic focus processing is performed for the next product using the currently set maximum value Pm as it is. Execute

The above-described series of algorithms is repeated for each product (S24), and a process of correcting the sampling interval Δr at the time of the coarse adjustment as needed by comparing the maximum value Pm ′ extracted for each product is executed. According to the automatic production equipment employing such an algorithm, the maximum value of the evaluation value profile caused by the variation of each product, which was conventionally difficult to sufficiently verify before the operation of the equipment, is changed during the operation. Verification in real time, the sampling interval at the time of coarse adjustment is corrected accordingly, and the optimal point can be accurately captured to maintain the in-focus position accurately, thereby shortening the processing time for the alignment process and reducing the yield. Can be improved.

As an application example of the present invention, an algorithm for calculating a normal correlation value as a feature amount of an image and performing an automatic focusing process has been described. However, the present invention is not limited to this. Needless to say, it may be a value, for example, a luminance dispersion value. In this case, in determining the optimum point, a process of extracting a minimum value is performed.

Incorporation of the logic relating to the automatic focus processing described above can be easily realized even in existing production equipment by storing the program of the parameter determination method of the present invention in a storage medium. Here, the storage medium is generally a floppy disk, a CD-
It refers to a portable type such as a ROM, a memory card, a ROM chip, and the like. However, a fixed storage device (hard disk) can also be removed from the system and transported, or provided with the above program installed in advance. ,
Such a hard disk can also be included in the storage medium. Further, when distributing a program via a network or a communication line, the storage device of the distribution source also includes:
Needless to say, it is included in the storage medium.

Therefore, the program stored in the above-described storage medium is loaded into a memory (including a virtual memory) of a hardware system in an executable form, and by executing the program, the recovery function and the reference function are executed. Contributes to the construction of a production management system that realizes

[0040]

As described above, according to the present invention,
By having the above-described series of processing procedures, the sampling interval in the coarse adjustment is adjusted, and the adjustment range including the extreme value at which the specific information of the image is the maximum or the minimum is extracted,
Further, since the optimum point can be accurately determined from the adjustment range extracted in the fine adjustment, the in-focus position can be accurately and quickly determined even for an image having a complicated profile shape.

In particular, by applying the present invention to a production apparatus equipped with an automatic focusing device, the sampling interval in the coarse adjustment is adjusted as needed, so that the individual objects can be automatically focused at an accurate focusing position. Since the processing is executed, shortening the processing time such as the alignment processing and improving the yield,
Production efficiency can be improved. In addition, by incorporating the program relating to the above-described method for determining the parameters of the automatic focusing apparatus into the control unit of the automatic focusing apparatus, the optimum focusing position can be determined accurately and quickly regardless of the shape of the object, and the alignment processing can be performed. , Etc., can be automatically and automatically executed in a software manner, so that it can be easily applied at low cost to an existing automatic focusing device or a production device equipped with the automatic focusing device.

[Brief description of the drawings]

FIG. 1 is a flowchart illustrating a basic concept of a method for determining parameters of an automatic focusing apparatus according to the present invention.

FIG. 2 is a diagram illustrating a relationship between a theoretical value of a sampling interval and a correction value.

FIG. 3 is a diagram showing another example when a plurality of local maxima exist.

FIG. 4 is a diagram showing a state of variation between products in automatic production.

FIG. 5 is a flowchart illustrating an embodiment of a method for determining a parameter of the automatic focusing apparatus according to the present invention.

FIG. 6 is a schematic configuration diagram of an automatic focusing device.

FIG. 7 is a flowchart showing a conventional automatic focusing process.

FIG. 8 is a diagram illustrating a process of searching for an optimum point by coarse adjustment and fine adjustment.

FIG. 9 is a diagram for explaining a problem of the related art.

[Explanation of symbols]

10 Objects 20 XY stage 30 Optical lens barrel 40 CCD camera 50 Image processing unit 60 control unit 70 drive system 80 monitors

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-3-50513 (JP, A) JP-A-10-48512 (JP, A) JP-A-1-309572 (JP, A) (58) Field (Int. Cl. ⁷ , DB name) G02B 7/28 G03B 13/00

Claims (4)

(57) [Claims] 1. A parameter determination method for an automatic focusing device that determines a focus position while narrowing a focus adjustment range by at least a two-stage focus adjustment process including a coarse adjustment and a fine adjustment. Extracting specific information and creating a distribution curve of the specific information; calculating a first sampling interval to be applied to the coarse adjustment; and presenting the distribution curve based on the first sampling interval. Determining an extreme value to be set, and when it is determined that the extreme value exists, setting a second sampling interval defined by a threshold value set in a region not including the extreme value, Applying the smaller one of the first and second sampling intervals to the coarse adjustment; and setting the second sampling interval to the first sampling interval. It is smaller than sampling interval, parameter determination method for an automatic focusing apparatus characterized by comprising a step of performing again the coarse adjustment by applying the second sampling interval. 2. A method according to claim 1, wherein the distribution curve of the specific information is a distribution curve of a normal correlation value for the image in the adjustment range, and the extreme value is a maximum value of the normal correlation value curve. A method for determining parameters of an automatic focusing apparatus according to claim 1. 3. A distribution curve of the search information is a distribution curve of a luminance variance value for an image in the search range, and the extreme value is a minimum value of the luminance variance value curve. A method for determining parameters of an automatic focusing apparatus according to claim 1. 4. A parameter determining method of an automatic focusing apparatus for determining a focus position while narrowing a focus adjustment range by at least a two-stage focus adjustment process including a coarse adjustment and a fine adjustment. A process for extracting specific information and creating a distribution curve of the specific information; a process for calculating a first sampling interval to be applied to the coarse adjustment; and a process for extracting the specific information from the distribution curve based on the first sampling interval. Processing to determine an extreme value to be performed, and when it is determined that the extreme value exists, a second sampling interval defined by a threshold value set in a region not including the extreme value is set; A process of applying the smaller one of the first and second sampling intervals to the coarse adjustment; and the second sampling interval being the first sampling interval. Is smaller than sampling interval, said second storage medium storing a program for executing the processing to be executed, a re the coarse adjustment by applying the sampling interval.