JP4416827B1 - Evaluation apparatus, calibration method, calibration program, and recording medium - Google Patents

Abstract

An evaluation apparatus capable of stable calibration while using an evaluation value calculated based on a captured image is provided.
[Solution]
The imaging device 110 repeats an imaging operation for imaging the specific evaluation object 200 from a plurality of imaging positions arranged in a predetermined pattern while shifting the entire imaging position. The control device 120 generates a high resolution image from each of a plurality of sets of captured image groups, and calculates an evaluation value based on each of the generated high resolution images. Then, the calculated evaluation value group is compared with the reference evaluation value group, and a coefficient to be multiplied by the evaluation value is set for calibration.
[Selection] Figure 1

Description

  The present invention relates to an evaluation apparatus that evaluates an evaluation object based on a captured image obtained by imaging the evaluation object. The present invention also relates to a calibration method for calibrating such an evaluation apparatus and a calibration program. The present invention relates to a recording medium on which such a calibration program is recorded.

  In order to evaluate a target object stably, at high speed and with high accuracy, development of an evaluation apparatus that automatically evaluates the target object based on a captured image has been underway. An imaging apparatus that supplies a captured image to such an evaluation apparatus usually includes a solid-state imaging device that functions as an area sensor and an optical system that forms an image of an evaluation object on the solid-state imaging device. The optical system consists of a lens and other optical elements, and includes a focus adjustment mechanism that adjusts the focal position of the lens, a work distance adjustment mechanism that adjusts the distance between the evaluation object and the lens, and light that adjusts the tilt of the optical axis of the optical system. It is adjusted to an appropriate state by various adjustment mechanisms such as an axis adjustment mechanism.

  However, the state of the optical system deteriorates with use over a long period of time. Factors that deteriorate the state of the optical system include various deviations of various adjustment mechanisms, positional deviations of the entire optical system, and changes in illuminance due to illumination deterioration. For this reason, it is not possible to properly evaluate the object based on the captured image obtained by the imaging device whose optical system state has deteriorated. That is, in order to perform appropriate evaluation at any time, periodic calibration (calibration) of the evaluation apparatus by readjustment of the optical system or the like is necessary.

Patent Document 1 discloses a technique for calibrating an area sensor using a reference pattern. Patent Document 2 discloses a technique for calibrating a line sensor using a special pattern.
JP-A-9-43292 (published February 14, 1997) JP 2004-28706 (published January 29, 2004)

  However, the conventional evaluation apparatus has a problem in that it requires physical adjustment of the optical system using various adjustment mechanisms every time calibration is performed, which requires a lot of time.

  In addition, when the imaging resolution is approximately the same as the size of the test pattern, the obtained captured image changes greatly only when the relative position between the imaging device and the test pattern is slightly shifted. Therefore, the evaluation value calculated based on the captured image also varies greatly. In other words, even if the same test pattern is an evaluation target, the evaluation value changes rapidly every time the evaluation is performed.

  Of course, if the resolution of the captured image to be referenced is increased, the variation in the evaluation value becomes smaller. However, as long as the evaluation value is calculated based on the captured image, such a variation in the evaluation value is unavoidable. That is, as long as the evaluation value calculated based on the captured image is used, it is difficult to perform stable calibration.

  In order to eliminate the influence of such variations, a method of calculating and averaging repeated evaluation values can be considered.

  However, if this is attempted using the technique described in Patent Document 1, it is necessary to repeat imaging by changing the reference pattern as many times as necessary, and the measurement time becomes enormous. Also, the measurement efficiency is poor and reproducibility is low. Patent Document 2 describes that position reproducibility confirmation is performed using a special pattern, but this is effective only for a line sensor that captures an image while the element moves, and is applied to an area sensor. It is not possible.

  The present invention has been made in view of the above problems, and an object thereof is to realize an evaluation apparatus capable of stable calibration while using an evaluation value calculated based on a captured image. .

In order to solve the above problems, an evaluation apparatus according to the present invention provides:
An evaluation device that calculates an evaluation value of an evaluation object based on a captured image obtained by imaging the evaluation object,
Control means for controlling the imaging apparatus to repeat an imaging operation of imaging a specific evaluation object from a plurality of imaging positions arranged in a predetermined pattern while shifting the entire plurality of imaging positions;
Generating means for generating a high-resolution image having a higher resolution than the imaging resolution of the imaging device from a group of captured images captured by the imaging device in each of repeated imaging operations;
Calculation means for calculating an evaluation value based on each of the high-resolution images generated by the generation means;
The coefficient by which the evaluation value is multiplied for calibration of the evaluation device is set to the ratio of the characteristic amount characterizing the distribution of the reference evaluation value group set in advance to the characteristic amount characterizing the distribution of the evaluation value group calculated by the calculating means. And setting means for setting.

  According to the above configuration, a plurality of sets of captured image groups obtained by imaging the specific evaluation object from a plurality of imaging positions are obtained, and a high resolution image is generated from each group. Since the evaluation value is calculated based on the high-resolution image, variation in the obtained evaluation value group is small as compared with the case where the evaluation value is calculated based on the captured image.

  Then, a coefficient for multiplying the evaluation value for calibration of the evaluation apparatus is set from the ratio between the characteristic amount representing the distribution of the obtained evaluation value group and the characteristic amount of the distribution of the reference evaluation value group set in advance. The For this reason, even if some variation remains in the evaluation value group calculated from the high resolution image, the influence of the variation hardly affects the coefficient. Therefore, there is an effect that stable calibration becomes possible.

  In addition, the captured image is obtained by an imaging apparatus controlled by the control means. That is, it is not necessary to manually reposition the specific evaluation object many times. For this reason, there is no possibility that the measurement time becomes enormous or the reproducibility is lowered.

  The control means controls, for example, a drive mechanism that drives an image pickup device included in the image pickup apparatus, and performs a drive operation for driving the image pickup device so as to go around the plurality of image pickup positions. The drive mechanism is controlled so as to be repeated while shifting the entire plurality of imaging positions.

In the evaluation apparatus according to the present invention,
The reference evaluation value group is preferably an evaluation value group calculated by the calculation means when the imaging apparatus is adjusted to an appropriate state.

  According to the above configuration, the evaluation value calculated when the imaging device is not adjusted to an appropriate state is changed to an evaluation value that would have been obtained if the imaging device was adjusted to an appropriate state. Can be calibrated.

  The reference evaluation value group may be an experimental value actually calculated by the calculation unit when the imaging device is adjusted to an appropriate state, or the imaging device is in an appropriate state. The theoretical value estimated by calculation may be the evaluation value that would have been calculated by the calculation means when the adjustment was made.

  Note that the appropriate state is obtained when, for example, the optical axis of the objective lens included in the imaging apparatus is orthogonal to the evaluation target surface of the evaluation target object, and incident light incident through the objective lens is captured. It is in a state of being focused on the light receiving surface of the image sensor provided in the apparatus.

  The feature amount is, for example, one of an average value, a maximum value, a minimum value, a variance, a standard deviation, or a combination thereof.

In order to solve the above problems, the calibration method according to the present invention is:
A calibration method for an evaluation apparatus that calculates an evaluation value of the evaluation object based on a captured image obtained by imaging the evaluation object,
A control step of controlling the imaging device to repeat an imaging operation of imaging a specific evaluation object from a plurality of imaging positions while shifting the entire plurality of imaging positions;
Generating a high-resolution image having a higher resolution than the imaging resolution of the imaging device from a group of captured images captured by the imaging device in each of the repeated imaging operations;
A calculation step of calculating an evaluation value based on each of the high-resolution images generated in the generation step;
The coefficient by which the evaluation value is multiplied for calibration of the evaluation apparatus is set to the ratio of the characteristic amount characterizing the distribution of the reference evaluation value group set in advance to the characteristic amount characterizing the distribution of the evaluation value group calculated in the calculation step. And a setting step for setting.

  According to said structure, there exists an effect similar to the said evaluation apparatus.

  The calibration method according to the present invention may be realized by a computer. In this case, a calibration program that realizes the calibration method in the computer by causing the computer to execute the steps described above, and a computer-readable recording medium that records the calibration program also fall within the scope of the present invention.

  As described above, the evaluation apparatus according to the present invention repeats the imaging operation of imaging a specific evaluation object from a plurality of imaging positions arranged in a predetermined pattern while shifting the entire plurality of imaging positions. A control unit that controls the image generation unit, a generation unit that generates a high-resolution image having a higher resolution than the imaging resolution of the imaging device from a group of captured images captured by the imaging device in each of repeated imaging operations, and the generation unit A feature that characterizes the distribution of evaluation value groups calculated by the calculating means by calculating means for calculating an evaluation value based on each of the generated high-resolution images, and a coefficient by which the evaluation value is multiplied for calibration of the evaluation device Setting means for setting a ratio of a feature quantity characterizing a distribution of a preset reference evaluation value group to a quantity.

  In addition, the calibration method according to the present invention was repeated with a control step of controlling the imaging device to repeat the imaging operation of imaging the specific evaluation object from a plurality of imaging positions while shifting the entire plurality of imaging positions. Based on each of the generation process of generating a high-resolution image having a higher resolution than the imaging resolution of the imaging apparatus from the group of captured images captured by the imaging apparatus in each imaging operation, and the high-resolution image generated in the generation process A reference evaluation value set in advance for a characteristic amount that characterizes the distribution of the evaluation value group calculated in the calculation step by a calculation step for calculating the evaluation value and a coefficient by which the evaluation value is multiplied for calibration of the evaluation device And a setting step for setting the ratio of the feature amount characterizing the distribution of the group.

  Therefore, there is an effect that stable calibration becomes possible.

  An embodiment of the present invention will be described below with reference to the drawings.

(Configuration of evaluation device)
First, the configuration of the evaluation apparatus 100 according to the present embodiment will be described with reference to FIG. FIG. 1 is a diagram illustrating a configuration of the evaluation apparatus 100. As illustrated in FIG. 1, the evaluation device 100 includes an imaging device 110, a control device 120, and a display device 130.

  The imaging device 110 is a means for imaging the evaluation object 200, and includes an optical system 111, a solid-state imaging device 112, and an actuator 113.

  The control device 120 is a means for controlling the imaging device 110 and evaluating the evaluation object 200 based on the captured image obtained by the imaging device 110. The control device 120 can be configured by, for example, a personal computer or a workstation.

  The display device 130 is a means for displaying the evaluation result obtained by the control device 120. The display device 130 can be configured by, for example, an LCD (Liquid Crystal Display) or a CRT (Cathode Ray Tube).

  The evaluation object 200 to be evaluated by the evaluation apparatus 100 may be anything as long as it is a flat object, for example, a liquid crystal panel.

(Details of imaging device)
Next, details of the imaging apparatus 110 included in the evaluation apparatus 100 will be described with reference to FIG. 1 again. As described above, the imaging device 110 includes the optical system 111, the solid-state imaging device 112, and the actuator 113.

  The optical system 111 is a means for forming an image of the evaluation object 200 on the light receiving unit of the solid-state image sensor 112. The optical system 111 is configured by a lens or other optical element.

  The solid-state image sensor 112 is a means for converting an image of the evaluation object 200 formed on the light receiving unit into an image signal. As the solid-state imaging device 112, for example, an area sensor type CCD image sensor, a CMOS image sensor, or the like can be used.

  Lenses and other optical elements that constitute the optical system 111 are attached to the inner wall of the housing of the imaging apparatus 110 via various adjustment mechanisms. On the other hand, the solid-state imaging element 112 is held by an actuator 113 fixed to the inner wall of the casing of the imaging device 110.

  The actuator 113 is a means for displacing the relative position of the solid-state image sensor 112 with respect to the evaluation object 200. More specifically, the solid-state image sensor 112 is two-dimensionally moved in a plane perpendicular to the optical axis connecting the lens of the optical system 111 and the evaluation object 200. However, the movement of the solid-state imaging device 112 by the actuator 113 is not limited to the two-dimensional movement in the plane, and may be a three-dimensional movement including rotation. As the actuator 113, a piezo actuator, a stepping motor, or the like can be used. Here, a piezo actuator is used.

(Details of control device)
The control device 120 calculates the evaluation value X based on the captured image acquired from the imaging device 110. However, instead of determining the quality of the evaluation object 200 based on the calculated evaluation value Y itself, the evaluation object 200 is based on an evaluation value Y ′ = αY obtained by multiplying the calculated evaluation value Y by a coefficient α. Judge the quality of the. The coefficient α is a calibration parameter for the evaluation apparatus 100. That is, even if the state of the optical system 111 deteriorates, the control device 120 can obtain a correct evaluation value Y ′ = αY by resetting the coefficient α. That is, the control device 120 can calibrate the evaluation device 100 without physically readjusting the optical system 111.

  Hereinafter, the function of the control device 120 will be described focusing on the setting method of the coefficient α. Note that the setting of the coefficient α described below may be performed irregularly (for example, when calibration is performed according to an operator's instruction) or periodically at a specific cycle (a specific cycle). The execution timing does not matter.

  FIG. 2 is a block diagram illustrating a configuration of the control device 120. The control device 120 includes an actuator control unit 121, an image data storage unit 122, a high resolution processing unit 123, an evaluation value calculation unit 124, a coefficient setting unit 125, as functional blocks involved in setting the coefficient α. A post-calibration evaluation value determination unit 126, a calibration information storage unit 127, and an optical system adjustment instruction unit 128 are provided. The setting of the coefficient α by the control device 120 is realized as follows by the cooperation of the above-described units.

<Imaging control process (control process)>
In order to set the coefficient α, first, an imaging process using a predetermined test pattern as the evaluation object 200 is performed. An example of the test pattern is shown in FIG. A simulated defect smaller than one pixel size of the solid-state imaging device 111 is provided at the center of the test pattern.

In the imaging process, M evaluation images 200 are imaged from M imaging positions (M is an integer equal to or greater than 2) arranged in a predetermined pattern, so that M captured images P ₁₁ , P ₁₂ ,. _A captured image group G ₁ consisting of _1M is obtained. This imaging operation is repeated N times if the entire M imaging positions are shifted without changing the arrangement pattern, and N sets of captured image groups G ₁ , G ₂ ,..., _GN are obtained (that is, total M × N). To obtain a single captured image). Note that the shift amount for shifting the entire imaging position is set to be smaller than the distance between the imaging positions. N sets of captured images obtained in the imaging step _{G 1, G 2, ...,} G N is stored in the image data storage unit 122 of the controller 120.

It should be noted that what kind of captured image should be used to configure the captured image group G _i , in other words, what pattern should be used for arranging the imaging position, is a resolution enhancement process executed in the next step It depends on the algorithm. For example, when an image shift method (pixel shift method) is used as an algorithm for high resolution processing, M images obtained by imaging a test pattern from each imaging position illustrated in FIGS. The captured image group G _i may be configured by the captured images P _i1 , P _i2 ,..., P _iM (here, M = 9).

  More specifically, imaging the evaluation object 200 from M imaging positions means that, as shown in FIG. 5, the direction of the light receiving surface is maintained in a direction orthogonal to the optical axis of the objective lens. In other words, the solid-state imaging device 112 circulates M imaging positions and images a test pattern at each imaging position. Note that the M imaging positions are all arranged in a virtual plane perpendicular to the optical axis of the objective lens.

  The imaging process as described above is mainly realized by the actuator control unit 121. The actuator control unit 121 controls the movement of the actuator 113 provided in the imaging device 110 by transmitting and receiving various control signals to and from the imaging device 110. More specifically, the solid-state image sensor 112 is moved as described above by controlling the motion start timing, motion speed, and motion distance of the actuator 113.

<High resolution process (generation process)>
Subsequent to the imaging process, a resolution increasing process is performed. In the resolution enhancement process, the resolution enhancement processing unit 123 reads the captured image group G ₁ = {P ₁₁ , P ₁₂ ,..., P _1M } from the image data storage unit 122 and reads the captured image group G _1. To generate a high-resolution image P ₁ by high-resolution processing. The same processing other captured image group G _2, G _3, ..., also repeated for G _N, the high resolution image of N sheets P _1, P _2, ..., and generates a P _N.

  Note that for the resolution enhancement processing executed by the resolution enhancement processing unit 123, an image shift method or a super-resolution method may be used. The image shift method is a method of mapping the luminance value of the low resolution image to the pixel of the high resolution image based on the position correspondence for each pixel of the plurality of low resolution images and the target high resolution image. The super-resolution method is a method for estimating a target high-resolution image from a plurality of low-resolution images. As a super-resolution algorithm, for example, an ML (Maximum-likelihood) method, a MAP (Maximum A Posterior) method, a POCS (Projection On to Convex Set) method, and the like are known.

High resolution image P ₁ of N frames generated in the high resolution step, P _2, ..., P _N and a high-resolution image P ₁ of N sheets, P _2, ..., in order to generate a P _N, imaging FIG. 6 shows M × N captured images P ₁₁ , P ₁₂ ,..., P _NM obtained in the process. In the example shown in FIG. 6, the image of simulated defect high resolution image P _1, P _2, ..., appearing as a low luminance region in each of the P _N.

<Evaluation value calculation process (calculation process)>
An evaluation value calculation step is performed following the high resolution step. In the evaluation value calculating step, the evaluation value calculation unit 124 obtains the high resolution image P1 from the high resolution processing section 123 calculates the evaluation value Y1 based on the acquired high-resolution image P _1. Further, the same processing other high-resolution image P _2, P _3, ..., also repeated for P _N, the evaluation value of the N Y _1, Y _2, ..., and calculates the Y _N. Specifically, the contrast volume is calculated as the evaluation value Y _i .

More specifically, the evaluation value calculation unit 124 calculates the contrast volume as follows. That is, first, the pixels constituting the high-resolution image P _i are classified into high-luminance pixels having a luminance higher than the threshold and low-luminance pixels having a luminance lower than the threshold. If the simulated defect included in the imaging target area is a black spot defect, the area composed of the low brightness pixel group can be regarded as a defect image, and the area composed of the high brightness pixel group can be regarded as the background. it can. Next, a luminance value difference from the average luminance value of the high luminance pixel group is calculated for each low luminance pixel. The luminance value difference calculated for each low luminance pixel is also called contrast. Finally, the contrast volume is obtained by adding the brightness value differences calculated for each low brightness pixel (in other words, integrating the contrast over the defect image).

  The evaluation value is not limited to the contrast volume, but the brightness (average brightness, minimum brightness, maximum brightness, or brightness of a specific pixel), brightness volume, defect size, or contrast (average contrast, maximum contrast, minimum contrast, Alternatively, the contrast of a specific pixel) may be used as the evaluation value.

<Coefficient setting process (setting process)>
A coefficient determination step is performed following the evaluation value calculation step. In the coefficient determination step, the coefficient setting unit 125 uses the evaluation value group {Y ₁ , Y ₂ ,..., Y _N } calculated in the evaluation value calculation step as the coefficient α by which the evaluation value X is multiplied for calibration. , And the reference evaluation value group {X ₁ , X ₂ ,..., X _N }. The set coefficient α is stored in the calibration information storage unit 127 as will be described later.

More specifically, the coefficient setting unit 125 determines the coefficient α as follows. That is, first, the average value X _{ave of} the reference evaluation value group {X ₁ , X ₂ ,..., X _N } and the evaluation value group {Y ₁ , Y _{2 ,.} } _Is calculated in accordance with the equations (1) and (2). And coefficient (alpha) is calculated according to (3) Formula. That is, the coefficient setting unit 125 applies the coefficient α multiplied by the evaluation value X for calibration to the reference evaluation value group {X ₁ for the average value Y _ave of the evaluation value group {Y ₁ , Y ₂ ,..., Y _N }. , X ₂ ,..., X _N } is set to the ratio value of the average value Xave. In other words, the average value αYave of the calibration values of the evaluation value group {Y ₁ , Y ₂ ,..., Y _N } calculated by the evaluation value calculation unit 124 and the reference evaluation value group {X ₁ , X ₂ ,. _The coefficient α is determined so that the average value X _{ave of N} } matches.

Incidentally, the coefficient alpha, evaluation value group _{{Y 1, Y 2, ...} , Y N} with respect to the average value Y _ave of the reference evaluation value group _{{X 1, X 2, ...} , X N} of the mean value X _ave of instead of matching the ratio X _ave / Y _ave, evaluation value group _{{Y 1, Y 2, ...} , Y N} to the maximum value Y _max of the reference evaluation value group _{{X 1, X 2, ...} , X N} the ratio X _max / Y _max of the maximum value X _max of or, evaluation value group _{{Y 1, Y 2, ...} , Y N} with respect to the minimum value Y _min of the reference evaluation value group _{{X 1, X 2, ...} , may be caused to coincide with the minimum value X _min of the ratio X _min / Y _min of X _N}. Further, the evaluation value group {Y ₁ , Y ₂ ,..., Y _N } and the reference evaluation value group {X ₁ , X ₂ ,..., X _N } with respect to the variance σY are made to coincide with the ratio σX / σY. May be. Instead of the variance, it may be made to coincide with the ratio of the standard deviation. In this way, if the coefficient α is set to the ratio of the feature quantity representing the distribution of the reference evaluation value group to the feature quantity representing the distribution of the evaluation value group {Y ₁ , Y ₂ ,..., Y _N }, the feature quantity is Regardless of what, the feature value of the evaluation value group obtained after calibration can be matched with the feature value of the evaluation value group that would be obtained when the imaging apparatus 110 is in an appropriate state.

<Post-processing process>
When the setting of the coefficient α is completed as described above, a post-processing step is performed. In the post-processing step, the post-calibration evaluation value determination unit 126 determines whether or not to perform calibration using the newly set coefficient α. When the calibration is performed, the post-calibration evaluation value determination unit 126 replaces the coefficient α stored in the calibration information storage unit 127 with the newly set coefficient α (therefore, the newly set coefficient α Will be calibrated using). When calibration is not performed, the optical system adjustment instruction unit 128 presents a message prompting the user to perform physical readjustment of the optical system (here, the message is presented by displaying the message on the display unit 140). You may also present by voice).

  The post-calibration evaluation value determination unit 126 determines whether or not to perform the calibration because the variation of the evaluation value becomes large due to the calibration using the newly set coefficient α, which adversely affects the inspection (detection result). This is to avoid a situation that impairs the stability of For convenience of explanation, in the following, a contrast volume is assumed as an evaluation value, and when the evaluation value after calibration exceeds a predetermined detection threshold Vth, it is determined that the evaluation object 200 is defective. And

The first situation to be avoided is that the defect on the test pattern is “defect” even if the presence or absence of the defect is determined based on any of the reference evaluation value groups {X ₁ , X ₂ ,..., X _N }. Is determined to be “no defect” when the presence or absence of a defect is determined based on any of the post-calibration evaluation value groups {αY ₁ , αY ₂ ,..., ΑY _N }. This is a situation where a determination result is obtained.

As shown in FIG. 7 (a), evaluation value group _{{Y 1, Y 2, ...} , Y N} average Y _ave reference evaluation value group _{{X 1, X 2, ...} , X N} Average When the value is below the value X _ave , α is set to a value larger than 1 according to the equation (3). Therefore, after calibration evaluation value group _{{αY 1, αY 2, ...} , αY N} variations of, and FIG. 7 (b) and as shown in FIG. 7 (c), the calibration before evaluation value group {Y _1, Y ₂ ,..., Y _N } is larger than the variation.

Even if the variation in the post-calibration evaluation value group {αY ₁ , αY ₂ ,..., ΑY _N } becomes large, as shown in FIG. 7B, the post-calibration evaluation value group {αY ₁ , αY ₂ ,. , ΑY _N } if the minimum value αY _min exceeds the detection threshold Vth, the above situation does not occur. However, as shown in FIG. 7 (c), as a result of an increase in variation due to calibration, the minimum value αY _{min of the} post-calibration evaluation value group {αY ₁ , αY ₂ ,..., ΑY _N } falls below the detection threshold Vth. Then, an erroneous determination result of “no defect” is obtained at least for the post-calibration evaluation value αY _min .

Therefore, the post-calibration evaluation value determination unit 126 defines the magnitude of variation of the evaluation value group {Y ₁ , Y ₂ ,..., Y _N } defined by the left side of the equation (4) by the right side of the equation (4). It is determined whether or not calibration is performed by comparing with a threshold value. More specifically, when the expression (4) is true, it is determined that “calibration is performed”, and when the expression (4) is false, it is determined that “calibration is not performed”. Avoid misjudgments.

  The reason why it is possible to avoid erroneous determination as described above by determining whether or not to perform calibration based on the true / false of the equation (4) is as follows.

First, with respect to the reference evaluation value group {X ₁ , X ₂ ,..., X _N } that is determined to be “defective” even if the presence / absence of a defect is determined based on any evaluation value, Equation (5) is It holds.

Now, evaluation value group _{{Y 1, Y 2, ...} , Y N} When was calibrated by using the coefficient α, and after calibration evaluation value group _{{αY 1, αY 2, ...} , αY N} mean value of the maximum The value and the minimum value are αY _ave , αY _max , and αY _min , respectively. Here, Y _ave , Y _max , and Y _min are the average value, maximum value, and minimum value of the evaluation value group {Y ₁ , Y ₂ ,..., Y _N }, respectively. At this time, equation (6) holds.

Here, after the calibration evaluation value group _{{αY 1, αY 2, ...} , αY N} for the reference evaluation value group _{{X 1, X 2, ...} , X N} like the, based on any of the evaluation value defect In order to determine that there is a defect even if the presence or absence is determined, it is necessary to satisfy equation (7).

If equation (7) is modified using equation (6), equation (4) is obtained. That is, if the expression (4) is true, the reference evaluation value group {X ₁ , regardless of whether or not there is a defect based on any of the post-calibration evaluation value groups {αY ₁ , αY ₂ ,..., ΑY _N }. , X ₂ ,..., X _N }, it is guaranteed that a “defect” determination result is obtained.

The second situation to be avoided is that even if the defect on the test pattern is determined based on any of the reference evaluation value groups {X ′ ₁ , X ′ ₂ ,..., X ′ _N }, Regardless of the defect for which the determination result “None” is obtained, as shown in FIG. 7D, any of the post-calibration evaluation value groups {αY ′ ₁ , αY ′ ₂ ,..., ΑY ′ _N } If the presence / absence of a defect is determined based on whether or not there is a defect, a determination result of “defect” is obtained. In such a case, as can be seen from FIG. 7D, the distribution range of the evaluation value group to be determined to be defective overlaps with the distribution range of the evaluation value group to be determined to be non-defective. Cannot be separated by the detection threshold. In other words, a pass / fail judgment gray zone is generated.

Therefore, the post-calibration evaluation value determination unit 126 determines the magnitude of the variation of the evaluation value group {Y ′ ₁ , Y ′ ₂ ,..., Y ′ _N } defined by the left side of the equation (8) as the right side of the equation (8). It is determined whether or not calibration is performed by comparing with the threshold value defined by. More specifically, when the expression (4) is true and the expression (8) is true, it is determined that “calibration is performed”, and the expression (4) is false or (8 ) When the expression is false, it is determined that “calibration is not performed”, thereby avoiding the erroneous determination as described above.

  The reason why the above erroneous determination can be avoided by determining whether or not to perform calibration based on the true / false of the equation (8) is as follows.

First, with respect to the reference evaluation value group {X ′ ₁ , X ′ ₂ ,..., X ′ _N } that is determined as “no defect” even if the presence or absence of a defect is determined based on any evaluation value, (9 ) Formula holds.

Now, evaluation value group _{{Y '1, Y' 2} , ..., Y 'N} When was calibrated by using the coefficient α and the calibration after the evaluation value set _{{αY' 1, αY '2} , ..., αY' N }, The average value, the maximum value, and the minimum value are αY _ave , αY _max , and αY _min , respectively. At this time, equation (10) holds.

Here, after the calibration evaluation value set _{{αY '1, αY' 2} , ..., αY 'N} for the reference evaluation value group _{{X' 1, X '2} , ..., X' N} similar to, any evaluation Even if the presence or absence of a defect is determined based on the value, it is necessary to satisfy the expression (11) in order to determine “no defect”.

If equation (11) is modified using equation (10), equation (8) is obtained. That is, if the equation (8) is true, the reference evaluation value group can be determined regardless of the presence or absence of the defect based on any of the evaluation value groups after calibration {αY ′ ₁ , αY ′ ₂ ,..., ΑY ′ _N }. Similar to the determination based on {X ′ ₁ , X ′ ₂ ,..., X ′ _N }, it is guaranteed that a determination result of “no defect” is obtained.

(Program and recording medium)
Finally, each block included in the control device 120 may be configured by hardware logic, or may be realized by software using a CPU as follows.

  That is, the control device 120 includes a CPU (central processing unit) that executes instructions of a calibration program that realizes each function, a ROM (read only memory) that stores the program, a RAM (random access memory) that expands the program, A storage device (recording medium) such as a memory for storing the program and various data is provided. An object of the present invention is to supply the control device 120 with a recording medium in which a program code (execution format program, intermediate code program, source program) of a calibration program, which is software that implements the functions described above, is recorded so as to be readable by a computer. However, this can also be achieved by the control device 120 reading and executing the program code recorded on the recording medium.

  Examples of the recording medium include tapes such as magnetic tapes and cassette tapes, magnetic disks such as floppy (registered trademark) disks / hard disks, and disks including optical disks such as CD-ROM / MO / MD / DVD / CD-R. Card system such as IC card, IC card (including memory card) / optical card, or semiconductor memory system such as mask ROM / EPROM / EEPROM / flash ROM.

  The control device 120 may be configured to be connectable to a communication network, and the program code may be supplied to the control device 120 via the communication network. The communication network is not particularly limited. For example, the Internet, intranet, extranet, LAN, ISDN, VAN, CATV communication network, virtual private network, telephone line network, mobile communication network, satellite communication. A net or the like is available. Further, the transmission medium constituting the communication network is not particularly limited. For example, even in the case of wired such as IEEE 1394, USB, power line carrier, cable TV line, telephone line, ADSL line, etc., infrared rays such as IrDA and remote control, Bluetooth ( (Registered trademark), 802.11 wireless, HDR, mobile phone network, satellite line, terrestrial digital network, and the like can also be used. The present invention can also be realized in the form of a computer data signal embedded in a carrier wave in which the program code is embodied by electronic transmission.

(Additional notes)
The present invention is not limited to the above-described embodiments, and various modifications can be made within the scope shown in the claims. That is, embodiments obtained by combining technical means appropriately modified within the scope of the claims are also included in the technical scope of the present invention.

  In addition, this invention can also be expressed as follows, for example.

  1. In an image inspection apparatus that inspects a flat object using a high-resolution image generated from a low-resolution image, a high-resolution image is obtained by using a plurality of images captured while shifting the imaging position with respect to the evaluation target pattern. Means for generating a plurality of the high-resolution images while shifting the imaging position, and measuring the evaluation value of the target pattern for each high-resolution image; variation from the evaluation value of the high-resolution image and a predetermined reference value A calibrating device comprising means for calculating a correction coefficient for calibrating the device in consideration of the above.

  2. In the image inspection apparatus, a plurality of low-resolution images obtained by capturing a plurality of images by shifting a relative position between an evaluation target pattern, a lens, and an imaging element by using a mechanically controlled drive unit by shifting by an arbitrary value. 2. The calibration apparatus according to 1, wherein a high-resolution image is generated by rearranging based on the shifted position.

  3. In the image inspection apparatus, a plurality of images are captured by shifting the relative positions of the pattern to be evaluated, the lens, and the image sensor with a movement amount equal to or less than the amount of positional deviation necessary for high resolution using a mechanically controlled drive unit. The calibration apparatus according to 1, wherein a plurality of high-resolution images having different imaging positions are generated.

  4). In the image inspection apparatus, an evaluation value is measured from a plurality of high-resolution images generated, and a correction coefficient is calculated by comparison with a reference value data group in consideration of variation in the evaluation value due to positional deviation. 1. The calibration apparatus according to 1.

  5. The reference value data group is obtained by measuring or calculating an evaluation value under optimum conditions for an optical system such as a focus, an optical axis, and a luminance in consideration of a positional deviation between a pattern to be evaluated, a lens, and an image sensor. 2. The calibration apparatus according to 1, wherein calculation is performed.

  6). The calibration apparatus is characterized in that the optimum condition of the optical system is an in-focus state where there is no deviation of the optical axis and the evaluation value of the evaluation target pattern can be detected with a sufficient size.

  7). After calibration using the correction coefficient, it is determined whether the variation in the evaluation value due to the correction coefficient is equal to or less than the variation of the predetermined reference value, and if the reference value is exceeded, the operator is prompted to perform physical adjustment. 2. The calibration device according to 1, wherein

  The present invention can be used to detect various defects in various evaluation objects. The type of defect may be anything that can be captured as an image in the captured image, and the type of the evaluation object may be anything as long as it can contain such a defect. In particular, it can be suitably used for detecting defects such as black spot defects and bright spot defects in display devices such as liquid crystal display panels.

BRIEF DESCRIPTION OF THE DRAWINGS It is an external view which shows embodiment of this invention and shows schematic structure of an evaluation apparatus. 1, showing an embodiment of the present invention, is a block diagram illustrating a configuration of a control device. FIG. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1, showing an embodiment of the present invention, is a diagram showing an example of a test pattern that becomes an evaluation object during calibration. FIG. 3 is a diagram illustrating an embodiment of the present invention and illustrating a group of captured images used for generating a high-resolution image. FIG. 2 is a diagram illustrating an embodiment of the present invention and a movement of a solid-state imaging device driven by an actuator. 1 illustrates an embodiment of the present invention, and illustrates N high-resolution images generated by high-resolution processing, and M × N captured images referred to generate these high-resolution images. FIG. It is a figure for demonstrating embodiment of this invention, and is a figure which shows the magnitude | size of the dispersion | variation in various evaluation values.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Evaluation apparatus 110 Imaging apparatus 111 Optical system 112 Solid-state image sensor (image sensor)
113 Actuator (drive mechanism)
120 Controller 121 Actuator Controller (Control Unit)
122 Image data storage unit 123 High resolution processing unit (generation unit)
124 evaluation value calculation unit (calculation means)
125 Coefficient setting unit (setting means)
126 Post-calibration evaluation value determination unit 127 Calibration information storage unit 128 Optical system adjustment instruction unit (presentation means)
130 Display device

Claims (10)

An evaluation device that calculates an evaluation value of an evaluation object based on a captured image obtained by imaging the evaluation object,
Control means for controlling the imaging apparatus to repeat an imaging operation of imaging a specific evaluation object from a plurality of imaging positions arranged in a predetermined pattern while shifting the entire plurality of imaging positions;
Generating means for generating a high-resolution image having a higher resolution than the imaging resolution of the imaging device from a group of captured images captured by the imaging device in each of repeated imaging operations;
Calculation means for calculating an evaluation value based on each of the high-resolution images generated by the generation means;
The coefficient by which the evaluation value is multiplied for calibration of the evaluation device is set to the ratio of the characteristic amount characterizing the distribution of the reference evaluation value group set in advance to the characteristic amount characterizing the distribution of the evaluation value group calculated by the calculating means. Setting means for setting,
An evaluation apparatus characterized by that. The control means controls a drive mechanism that drives an image pickup device included in the image pickup apparatus, and performs a drive operation for driving the image pickup device so as to go around the plurality of image pickup positions. Controlling the drive mechanism to repeat while shifting the entire position;
The evaluation apparatus according to claim 1, wherein: The reference evaluation value group is an evaluation value group calculated by the calculation means when the imaging device is adjusted to an appropriate state.
The evaluation apparatus according to claim 1, wherein: The appropriate state is that the optical axis of the objective lens included in the imaging device is orthogonal to the evaluation target surface of the evaluation target, and the incident light incident through the objective lens is provided in the imaging device. It is in a state of being focused on the light receiving surface of the image sensor that is
The evaluation apparatus according to claim 3. The feature amount is one of an average value, a maximum value, a minimum value, a variance, a standard deviation, or a combination thereof.
The evaluation apparatus according to any one of claims 1 to 4, wherein: The evaluation value is calibrated using the coefficient set by the setting means when at least one of the following expressions (A) and (B) is true : The evaluation apparatus according to 1 .

Where X _ave represents the average value of the reference evaluation value group, Y _ave represents the average value of the evaluation value group calculated by the calculation means , and Y _max represents the maximum value of the evaluation value group calculated by the calculation means. Y _min represents the minimum value of the evaluation value group calculated by the calculation means, and Vth represents a predetermined detection threshold value. When both the formula (A) and the formula (B) are false, a presentation means for presenting a message for prompting physical adjustment of the imaging device is provided.
The evaluation apparatus according to claim 6. A calibration method for an evaluation apparatus that calculates an evaluation value of the evaluation object based on a captured image obtained by imaging the evaluation object,
A control step of controlling the imaging device to repeat an imaging operation of imaging a specific evaluation object from a plurality of imaging positions arranged in a predetermined pattern while shifting the entire plurality of imaging positions;
Generating a high-resolution image having a higher resolution than the imaging resolution of the imaging device from a group of captured images captured by the imaging device in each of the repeated imaging operations;
A calculation step of calculating an evaluation value based on each of the high-resolution images generated in the generation step;
The coefficient by which the evaluation value is multiplied for calibration of the evaluation apparatus is set to the ratio of the characteristic amount characterizing the distribution of the reference evaluation value group set in advance to the characteristic amount characterizing the distribution of the evaluation value group calculated in the calculation step. A setting process to set,
A calibration method characterized by that.   A calibration program for causing a computer to execute the calibration method according to claim 8, wherein the calibration program causes the computer to execute the above steps.   A computer-readable recording medium on which the calibration program according to claim 9 is recorded.

Images