JP6874441B2 - Defect inspection methods, defect inspection programs, and defect inspection equipment - Google Patents

Description

The present invention relates to a defect inspection method, a defect inspection program, and a defect inspection apparatus.

In recent years, films with laminated coatings have been used in various devices. For example, there are various types such as a polarizing film, a color filter, a heat insulating film, a functional optical film for shading a specific wavelength (for example, ultraviolet rays), and a gas barrier film.

Conventionally, there are the following techniques as a defect inspection method for these films.

The technique of Patent Document 1 irradiates a film to be inspected with light, acquires image data using a plurality of optical systems such as specular reflection, refraction transmission, and direct transmission, and detects defects from the image data. Then, it is identified whether or not it can be detected by each optical system, and the type of defect is specified by the combination.

The technique of Patent Document 2 is that the object to be inspected is a transparent plate-like body, and specifically, the object to be inspected is glass. This technique uses the distance between the mirror image and the real image, which is reflected on the bottom surface of the glass by regular reflection, to determine the distance in the thickness direction of the defect, and compares the transmitted edge system with the bright and dark regions of the regular reflection to be inspected. Identify the type of defect inside the object.

The technique of Patent Document 3 visualizes irregularities on the surface of an object and air bubbles inside a transparent object as a pair of bright and dark parts and photographs them, and when the distance between the bright and dark parts of the obtained image is smaller than a predetermined value. , The bright and dark parts are judged to be uneven or air bubbles.

Japanese Unexamined Patent Publication No. 2012-167975 Japanese Unexamined Patent Publication No. 2010-48745 Japanese Unexamined Patent Publication No. 2004-361805

These prior arts disclose a method of identifying the type of defect by a plurality of detection means. However, it has not been possible to determine whether the detected defect is a defect that satisfies the required performance of the member to be inspected as a product or a defect that does not satisfy the required performance.

Therefore, an object of the present invention is to provide a defect inspection method capable of discriminating between a defect that satisfies the required performance of the member to be inspected as a product and a defect that does not satisfy the required performance. Another object of the present invention is to provide a defect inspection program capable of discriminating between a defect that satisfies the required performance of the member to be inspected as a product and a defect that does not satisfy the required performance. Further, another object of the present invention is to provide a defect inspection apparatus capable of discriminating between a defect that satisfies the required performance of the member to be inspected as a product and a defect that does not satisfy the required performance.

The above objectives are achieved by the following means.

( 1 ) The step (a) of irradiating the surface of the member to be inspected with light from a certain direction and acquiring image data obtained by photographing the surface.
The step (b) of obtaining the average luminance value from the image data and
From the image data, a step (c1) of determining a range of bright areas where the luminance value is equal to or greater than the first predetermined ratio from the average luminance value.
From the image data, a step (c2) of determining a range of a dark portion that is in contact with the range of the bright portion and whose luminance value is equal to or less than a second predetermined ratio from the average luminance value.
Range from said image data, the pale light portion luminance values have contact with at least one of the ranges has a first smaller than the predetermined ratio exceeds the average luminance value of the range and the dark portion of the range of the bright portion At the stage of determining (d1) and
From the image data, a range of a light dark portion that is in contact with at least one of the bright portion range and the dark portion range and whose brightness value exceeds the second predetermined ratio and is less than the average brightness value. The stage of discrimination (d2) and
Defect inspection method.

( 2 ) An evaluation formula is prepared in advance using the image data, the bright part, the dark part, the light bright part, and the feature amount obtained from the light dark part as explanatory variables, and the influence on the member to be inspected as the objective variable. Keep it
The image data acquired from the member to be inspected, the bright part, the dark part, the light bright part, and the feature amount obtained from the light dark part are applied to the evaluation formula, and the bright part, the dark part, and the light bright part are applied. The defect inspection method according to (1 ) above, further comprising a step (e) of evaluating the influence of the portion and the defect portion in which the light dark portion is detected on the member to be inspected.

( 3 ) The step (d1) of determining the range of the pale bright portion is a step of determining a range in which the luminance value is equal to or more than the third predetermined proportion and less than the first predetermined proportion than the average luminance value.
The step (d2) for discriminating the range of the light dark portion is a step for discriminating the range in which the luminance value exceeds the second predetermined proportion and is equal to or less than the fourth predetermined proportion from the average luminance value. ) Or (2) .
(4) The step (a) of irradiating the surface of the member to be inspected with light from a certain direction and acquiring image data obtained by photographing the surface.
The step (b) of obtaining the average luminance value from the image data and
From the image data, among the range of the bright part where the brightness value is equal to or more than the first predetermined ratio from the average brightness value and the range of the dark part where the brightness value is equal to or less than the second predetermined ratio from the average brightness value. The step (c) of determining at least one range and
When the dark portion is determined in the step (c), the range of the pale bright portion that is in contact with the range of the dark portion and whose luminance value exceeds the average luminance value and is less than the first predetermined ratio is the image. When the bright part is discriminated in the step (d1) of discrimination from the data or the step (c), it is in contact with the range of the bright part, and the brightness value exceeds the second predetermined ratio and is less than the average brightness value. At the stage (d2) of discriminating the range of the faint dark part, which is, from the image data,
Have,
The step (d1) of discriminating the range of the pale bright portion from the image data is a step of discriminating a range in which the brightness value is equal to or more than the third predetermined ratio and less than the first predetermined ratio from the average brightness value. ,
The step (d2) of discriminating the range of the light dark portion from the image data is a step of discriminating a range in which the luminance value exceeds the second predetermined proportion and is equal to or less than the fourth predetermined proportion from the average luminance value. , Defect inspection method.

(5) The defect inspection method according to any one of (1) to (4) above, wherein the member to be inspected is a gas barrier film having a plurality of layers.

(6) A program that causes a computer to execute the defect inspection method according to any one of (1) to (5) above.

( 7 ) A light irradiation unit that irradiates the surface of the member to be inspected with light from a certain direction,
An imaging unit that photographs the surface of the member to be inspected,
The average luminance value is obtained from the image data of the surface of the member to be inspected taken by the photographing unit, and the range of the bright portion where the luminance value is equal to or more than the first predetermined ratio from the average luminance value is determined from the image data. , The range of the dark part which is in contact with the range of the bright part and whose brightness value is equal to or less than the second predetermined ratio from the average brightness value is determined, and at least one of the range of the bright part and the range of the dark part is determined. The range of the light bright part which is in contact with the above and the brightness value exceeds the average brightness value and is less than the first predetermined ratio is determined, and at least one of the range of the bright part and the range of the dark part is determined. An evaluation unit that determines the range of a light dark portion that is in contact with and whose luminance value exceeds the second predetermined ratio and is less than the average luminance value.
Defect inspection equipment.

( 8 ) The evaluation unit evaluates the image data, the bright part, the dark part, the light bright part, and the feature amount obtained from the light dark part as explanatory variables, and the influence on the member to be inspected as the objective variable. Remember the formula in advance
The evaluation unit applies the image data acquired from the member to be inspected, the bright part, the dark part, the light bright part, and the feature amount obtained from the light dark part to the evaluation formula, and applies the bright part, the dark part, and the feature amount obtained from the light dark part to the evaluation formula. The defect inspection apparatus according to (7 ) above, which evaluates the influence of a dark portion, the pale bright portion, and a defective portion in which the pale dark portion is detected on the member to be inspected.

( 9 ) When determining the range of the pale bright portion, the evaluation unit sets a range in which the brightness value is higher than the average brightness value of the third predetermined ratio or more and less than the first predetermined ratio. Determined as the range of the part,
When determining the range of the light dark portion, the range in which the brightness value exceeds the second predetermined ratio and is equal to or less than the fourth predetermined ratio lower than the average brightness value is determined as the range of the light dark portion. 7) Or the defect inspection apparatus according to (8).
(10) A light irradiation unit that irradiates the surface of the member to be inspected with light from a certain direction,
An imaging unit that photographs the surface of the member to be inspected,
The average brightness value is obtained from the image data of the surface of the member to be inspected taken by the photographing unit, and the range and brightness of the bright part where the brightness value is equal to or more than the first predetermined ratio from the average brightness value from the image data. At least one of the dark areas whose values are equal to or less than the second predetermined ratio from the average luminance value is discriminated, and the determined range of the bright areas or the dark areas is the dark area. In this case, the range of the light bright portion which is in contact with the range of the dark portion and whose luminance value exceeds the average luminance value and is less than the first predetermined ratio is determined from the image data, or the bright portion of the bright portion. When the determined range of the range or the dark portion is the bright portion, the luminance value is in contact with the bright region and the luminance value exceeds the second predetermined ratio and is less than the average luminance value. An evaluation unit that determines the range of the faint dark area from the image data, and an evaluation unit.
Have,
When the evaluation unit determines the range of the pale bright portion from the image data, the evaluation unit determines the range in which the brightness value is higher than the average brightness value by a third predetermined ratio or more and less than the first predetermined ratio. Determined as the range of the bright part,
When the range of the light dark part is discriminated from the image data, the range in which the brightness value exceeds the second predetermined ratio and is equal to or less than the fourth predetermined ratio lower than the average brightness value is determined as the range of the light dark part. to, defect inspection apparatus.

(11) A light-shielding plate for adjusting the amount of light is provided either between the light irradiation unit and the member to be inspected or between the member to be inspected and the photographing unit.
The photographing unit has a line image sensor that photographs a portion of light that is not shielded by at least the light-shielding plate on the surface of the member to be inspected and is irradiated to the member to be inspected.
The defect inspection apparatus according to any one of (7) to (10) above, further comprising a moving portion for moving the member to be inspected relative to the light irradiation portion and the photographing portion.

According to the present invention, from an image taken by irradiating a member to be inspected with light from a certain direction, not only bright and dark areas, but also light bright areas that are brighter than the average luminance value but not as bright as the bright areas, and vice versa. It was decided to discriminate at least one range of a light dark part that is darker than the average brightness value but not as dark as the dark part. Thereby, it is possible to evaluate whether or not there is a defect that affects the required performance of the member to be inspected as a product depending on whether or not at least one of the light dark part and the light dark part is present.

It is a schematic bird's-eye view for demonstrating the main part of the defect inspection apparatus of embodiment. It is the schematic which shows the structure of the defect inspection apparatus of embodiment. It is a flowchart for demonstrating an inspection procedure. It is a drawing substitute cross-section SEM photograph for explaining a defect (non-defective product) which does not affect performance as a judgment criterion. It is a schematic view of the drawing substitute cross-section SEM photograph shown in FIG. It is a drawing substitute cross-section SEM photograph for explaining a defect (defective product) which affects performance as a judgment criterion. It is a schematic view of the drawing substitute cross-section SEM photograph shown in FIG. It is a drawing substitute surface enlarged photograph of a defective part judged as a non-defective product. It is a schematic diagram explaining the range which was determined as the bright part from the surface enlarged photograph shown in FIG. It is a schematic diagram explaining the range which was determined as a dark part from the surface enlarged photograph shown in FIG. It is a schematic diagram explaining the whole range which was determined as a bright part and a dark part from the surface enlarged photograph shown in FIG. It is a drawing substitute surface enlarged photograph of a defective part judged as a defective product. It is a schematic diagram explaining the range which was determined as the bright part from the surface enlarged photograph shown in FIG. It is a schematic diagram explaining the range determined as a dark part from the surface enlarged photograph shown in FIG. It is a schematic diagram explaining the range which was determined as the light bright part from the surface enlarged photograph shown in FIG. It is a schematic diagram explaining the range determined as a faint dark part from the surface enlarged photograph shown in FIG. It is a schematic diagram explaining the whole range which was determined as a bright part, a dark part, a light bright part, and a light dark part from the surface enlarged photograph shown in FIG.

Hereinafter, embodiments of the present invention will be described with reference to the attached drawings. The present invention is not limited to the following embodiments. In the description of the drawings, the same elements are designated by the same reference numerals, and duplicate description will be omitted. In addition, the dimensional ratios in the drawings are exaggerated for convenience of explanation and may differ from the actual ratios.

First, the configuration of the defect inspection device of the present embodiment will be described.

FIG. 1 is a schematic bird's-eye view for explaining a main part of the defect inspection device of the present embodiment, and FIG. 2 is a schematic view showing a configuration of the defect inspection device of the present embodiment.

The defect inspection device 100 of the present embodiment includes a light irradiation unit 101, an imaging unit 102, a support unit 103, and an evaluation unit 104.

The light irradiation unit 101 irradiates the member 200 to be inspected with rectangular light 110. For this purpose, the light irradiation unit 101 has a light source inside and has a rectangular opening. As the light source, for example, an LED (Light Emitting Diode) is preferable. When LEDs are used, a plurality of LEDs are arranged side by side in the rectangular longitudinal direction. Further, the LED used is preferable because the monochromatic LED has a peak wavelength. Above all, the resolution can be increased by using a blue (peak wavelength 460 nm) LED having a wavelength shorter than that of red (peak wavelength 660 nm) or green (peak wavelength 555 nm). Further, for example, it is preferable to arrange a light diffusing plate (not shown) in front of the light irradiation direction so that the amount of light is uniform. When a plurality of LEDs are arranged in one direction, a difference in the amount of light (unevenness in the amount of light) may occur between the area directly under the LED and the area under the LED depending on the arrangement interval. By installing a light diffusing plate, such unevenness in the amount of light can be prevented. Of course, the light diffuser is unnecessary if it does not affect the shooting.

The light irradiation unit 101 has a light shielding plate 108 for adjusting the amount of light on the front surface in the light irradiation direction. The light-shielding plate 108 is movable, and the amount of irradiation light is adjusted by moving the light-shielding plate 108 in a direction orthogonal to the rectangular longitudinal direction with respect to the rectangular light 110. Light that is not shielded by the light-shielding plate 108 (hereinafter referred to as one-sided light-shielding light 111) reaches the surface of the member 200 to be inspected. The one-sided light-shielding light 111 is also irradiated on the surface of the member 200 to be inspected in a rectangular shape and a narrow width (direction orthogonal to the longitudinal direction of the rectangular shape).

The light-shielding plate 108 may be located between the light irradiation unit 101 and the surface of the member to be inspected 200 (on the optical path), but it is preferably closer to the member to be inspected 200. The one closer to the member to be inspected 200 is prevented from being blurred at the edge of the one-sided light-shielding light 111 that hits the surface of the member to be inspected 200 or being incident on the photographing unit 102 due to diffused light such as diffracted light by the light-shielding plate 108. be able to. The light-shielding plate 108 may not be provided. In that case, for example, an LED whose amount of light can be adjusted may be used, or a lens system (collimator lens or the like) may be used to irradiate parallel light.

The photographing unit 102 photographs the surface of the member 200 to be inspected, which is irradiated with the light blocking light 111 on one side. The photographing unit 102 is preferably an imaging device having a line image sensor. By using the line image sensor, it is possible to prevent diffusely reflected light other than specularly reflected light from entering the surface of the member 200 to be inspected. A condenser lens (not shown) is provided in front of the line image sensor, and is adjusted so as to focus on a portion of the surface of the member 200 to be inspected that is irradiated with one-sided light-shielding light 111. The line image sensor may be for monochrome or for color (a color filter is provided on the front surface of the line image sensor), but in the present embodiment, it is not necessary to acquire a color image. Therefore, in terms of cost, a monochrome line image sensor is suitable.

A lens (not shown) is arranged in front of the line image sensor so that a range longer than the actual size in the pixel arrangement direction of the line sensor can be photographed. The photographing range is at least a portion of the surface of the member 200 to be inspected that is exposed to one-sided light-shielding light 111. Therefore, the photographing unit 102 is arranged so that the arrangement direction of the pixels of the line image sensor is along the direction of the edge of the one-sided light-shielding light 111.

The light irradiation unit 101 and the photographing unit 102 are arranged so that the photographing unit 102 efficiently photographs only the specularly reflected light from the surface of the member 200 to be inspected of the one-sided light shielding light 111 and does not pick up the diffusely reflected light as much as possible. To do. Therefore, as shown in FIG. 2, the arrangement of the light irradiation unit 101 and the imaging unit 102 has the same opening angle with respect to the perpendicular line h from the surface of the member 200 to be inspected. In FIG. 2, assuming that the opening angle of the light irradiation unit 101 with respect to the perpendicular line h is α and the opening angle of the photographing unit 102 with respect to the perpendicular line h is β, α = β is preferably 2 to 85 degrees, preferably 5 to 60 degrees. It is more preferable to do so. By setting such an angle range, it becomes easy to capture a light bright part and a light dark part, which will be described later.

The support portion 103 is, for example, a moving stage on which the member 200 to be inspected is placed and moved. In the case of illustration, since the member 200 to be inspected is larger than the width of the photographing area (the direction in which the pixels of the line image sensor are arranged), for example, the XY stage can be photographed by scanning the entire surface of the member 200 to be inspected. Is used. However, if the member 200 to be inspected is smaller than the width of the imaging region (the direction in which the pixels of the line image sensor are arranged), the stage may be a stage in which the member 200 to be inspected can be moved in a certain direction.

Instead of such a moving stage, the support portion 103 may be fixed by simply placing the member 200 to be inspected. In that case, the light irradiation unit 101 and the photographing unit 102 are moved. In order to move the light irradiation unit 101 and the imaging unit 102, for example, the light irradiation unit 101 and the imaging unit 102 are moved by using a robot, or the light irradiation unit 101 and the imaging unit 102 are suspended and moved on the rail. The member 200 to be inspected may be relatively movable with respect to the light irradiation unit 101 and the photographing unit 102, such as by using a rail moving mechanism. The moving direction is a direction that intersects the longitudinal direction of the line image sensor of the photographing unit 102, and is preferably a direction that is orthogonal to the longitudinal direction. As a result, it is possible to reliably photograph the portion irradiated with the light including the edge of the one-sided light-shielding light 111, and to photograph all the surfaces to be inspected of the member 200 to be inspected.

In this way, the stage, the robot, the rail moving mechanism, and the like, which are members that move the member to be inspected 200 relative to the light irradiation unit 101 and the photographing unit 102, are moving units.

The evaluation unit 104 evaluates the presence or absence of defects in the member 200 to be inspected and the degree of influence thereof from the image data obtained by photographing the member 200 to be inspected. The evaluation unit 104 is a computer, and for example, a so-called personal computer can be used. Therefore, the evaluation unit 104 has a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive) as a storage device, and other external storage, as in a normal personal computer. It has a device and the like. Further, it may be connected to a network so that image data can be stored in an external server or taken out from the server for evaluation.

The image data obtained by the photographing unit 102 is line image data (one-dimensional image data). Therefore, the evaluation unit 104 synthesizes the line image data into the two-dimensional image data. Then, from the synthesized two-dimensional image data, the presence or absence of defects in the member to be inspected 200 and the degree of influence thereof are evaluated using an evaluation formula stored in advance.

The evaluation unit 104 also controls the movement of the support unit 103 in order to acquire (photograph) the image data. That is, when the XY stage is used, the XY stage is moved so that the entire surface of the member 200 to be inspected can be scanned and moved with respect to the light irradiation unit 101 and the imaging unit 102 for imaging. The same applies to robots and rail movement mechanisms. The member 200 to be inspected may be moved to the light irradiation unit 101 and the imaging unit 102 by a control device different from the evaluation unit 104.

Next, the inspection procedure by the defect inspection apparatus 100 will be described. FIG. 3 is a flowchart for explaining the inspection procedure. Here, it is described that the evaluation unit 104 (computer) controls all operations, but as described above, some operations may be performed by another control device. Further, the shooting may be performed separately. When shooting is performed separately, for example, the image data obtained by shooting may be stored in the server, and the evaluation unit may acquire the image data in order to evaluate the defect.

Defect inspection starts from the initial setting (S1 to S2). The initial setting is a process for adjusting the amount of light to match the defect inspection.

First, the amount of light of the light irradiation unit 101 is confirmed (S1). To confirm the amount of light, the light-shielding plate 108 is fully opened, the member 200 to be inspected is irradiated with light, and the specularly reflected light is photographed by the photographing unit 102. The shooting at this time may be for one line. Then, it is sufficient that the brightness value of the captured image data is equal to or higher than the maximum gradation value. However, if it is less than the maximum gradation value, the amount of light becomes insufficient. When the amount of light is insufficient, it is preferable to increase the amount of light by replacing the light source of the light irradiation unit 101. However, the maximum gradation value may not necessarily be reached as long as the bright part, the dark part, the light bright part, and the light dark part, which will be described later, can be identified.

Subsequently, the amount of incident light of the photographing unit 102 is adjusted (S2). To do this, first, the reflected light is photographed without being shielded by the light-shielding plate 108. At the same time, the evaluation unit 104 acquires image data from the photographing unit 102 and calculates an average luminance value. Then, the amount of light of the light irradiation unit 101 is adjusted so that the average brightness value is half of the maximum brightness value (for example, if the maximum gradation value is 255, the average gradation value is 128). Subsequently, the light-shielding plate 108 is gradually moved to adjust the average brightness value halved so as to be uniformly about 1/10 (about 12 to 13 as an average gradation value) as a whole, and the light-shielding plate 108 Fix the position of. In this state, the amount of light of the light irradiation unit 101 is adjusted again so that the average luminance value becomes half of the maximum luminance value (128 in average gradation value).

It is desirable to determine how much light is shielded by the light-shielding plate 108 (for example, the average brightness value halved is further reduced to 1/10 or 1/8) depending on the inspection object.

The light-shielding plate 108 may be moved manually, or may be moved according to an instruction from the evaluation unit 104 if the light-shielding plate 108 can be moved by driving a motor or the like.

By adjusting the amount of light by the light-shielding plate 108 by the processing of S1 and S2, when a line image sensor without an aperture is used, the photographing unit 102 does not saturate when capturing a bright part. Then, by setting the average luminance value of the specularly reflected light from the surface of the member 200 to be the center value of the gradation that can be photographed by the photographing unit 102, the bright part and the dark part, which will be described later, and the lighter bright part and the lighter part are lightened. Make sure to capture the dark areas. Of course, if the photographing unit 102 has an aperture, the aperture may be used for adjustment.

It is preferable that this initial setting is reset every time the member 200 to be inspected changes. However, if the type of the member 200 to be inspected is the same and the reflectance of the surface does not change, the same setting may be used as it is. On the contrary, even if the type of the member 200 to be inspected is the same, if the surface reflectance is different, it is preferable to reset the initial setting.

Next, the evaluation unit 104 moves the stage to irradiate the light 110 from the light irradiation unit 101, and the surface of the member 200 to be inspected is photographed by the imaging unit 102 to acquire line image data (S11). The acquired line image data is sequentially combined with the two-dimensional image data (S12). The image data may be acquired by scanning the entire surface of the member to be inspected 200, or may be acquired by scanning only a specific inspection range.

Subsequently, the evaluation unit 104 obtains an average luminance value from the obtained two-dimensional image data (hereinafter, simply referred to as image data) (S13). As the average brightness value, the brightness value of the entire image data may be simply arithmetically averaged.

Subsequently, the evaluation unit 104 determines from the image data the range of the bright portion whose brightness value is equal to or greater than the first predetermined ratio from the average brightness value (S14). The first predetermined ratio is, for example, 115% with respect to the average brightness value. Therefore, the range in which the brightness value is 115% or more of the average brightness value is defined as the bright part. This first predetermined ratio is one of the scales for recognizing defects. The value of the first predetermined ratio is an arbitrary value, and it is preferable to determine a value that can be easily detected as a defect by an experiment or the like.

Subsequently, the evaluation unit 104 determines from the image data the range of the dark portion which is in contact with the range of the bright portion and whose luminance value is equal to or less than the second predetermined ratio from the average luminance value (S15). The second predetermined ratio is, for example, 90% with respect to the average brightness value. Therefore, the range in which the brightness value is 90% or less of the average brightness value is defined as the dark portion. This second predetermined ratio is one of the scales for recognizing as a defect. The value of the predetermined ratio of the dark part is an arbitrary value, and it is preferable to determine a value that can be easily detected as a defect by an experiment or the like.

Note that S14 and S15 may be in the reverse order. In that case, after the range of the dark part is determined, the range of the bright part in contact with the range of the dark part is determined.

Subsequently, the evaluation unit 104 is in contact with at least one of the bright area and the dark area from the image data, and the luminance value is lighter than the average luminance value and less than the first predetermined ratio. The range of the bright part is determined (S16). In order to more clearly discriminate the faint bright portion, it is preferable to discriminate the range of the third predetermined ratio or more and less than the first predetermined ratio with respect to the average luminance value. For example, the range of the brightness value of 105% or more and less than 115% is determined. If this only exceeds the average luminance value, the one having a gradation value higher than the average luminance value by only one gradation will be included. This is because the portion where the reflected light is high due to the undulation of the surface of the member to be inspected is recognized as a faint bright portion, and these may become noise.

Subsequently, the evaluation unit 104 is in contact with at least one of the bright area and the dark area from the image data, and the luminance value exceeds the second predetermined ratio and becomes less than the average luminance value. The range of the faint dark part is determined (S17). In order to more clearly discriminate the faint dark portion, it is preferable to discriminate the range in which the second predetermined ratio is exceeded and the average brightness value is equal to or less than the fourth predetermined ratio. For example, the range in which the luminance value exceeds 90% and becomes 97% or less is determined. If this is less than the average luminance value, the one having a gradation value lower than the average luminance value by only one gradation will be included. This is because the portion where the reflected light generated by the undulation of the surface of the member to be inspected is low is recognized as a faint dark portion, and these may become noise.

It should be noted that both the light bright part and the light dark part need to be in contact with at least one of the bright part and the dark part. This is because there is a high possibility that a light bright part or a light dark part that is separated from both a bright part and a dark part is not a defect in the first place. S16 and S17 may be in the reverse order.

Subsequently, the evaluation unit 104 obtains the feature amount required for the evaluation formula (S18). Here, the evaluation formula (formula (1) below) obtained in the examples described later is used.

Performance impact = 0.194 x minimum brightness value in dark area + 0.008 x total area -0.734 x total height + 0.018 x bright area area -0.007 x dark area area -0.714 x Dark area width-20 ... (1)
In the formula, the total area = the area of the bright part + the area of the dark part + the area of the light bright part + the area of the light dark part, the height is the ferret diameter in the light irradiation direction, and the total height is the bright part and the dark part. , The height of the total range of the light light part and the light dark part. The width is the diameter of the ferret only in the dark part in the direction orthogonal to the light irradiation direction, and the width of the dark part is the width of only the dark part.

Therefore, since the feature amount obtained in S16 is the feature amount required for the equation (1), it becomes the minimum luminance value in the dark part, the whole area, the whole height, the area of the bright part, the area of the dark part, and the width of the dark part. ..

Such an evaluation formula may be other than the formula (1), and is preferably created according to the member 200 to be inspected. A statistical method is used to create the evaluation formula. For example, an analysis method based on Mahalanobis distance (multiple regression analysis) can be used. An evaluation formula (multiple regression formula) is created with the degree of performance influence as the objective variable and various values obtained from the image data as explanatory variables. The degree of performance influence may be, for example, an alternative value of a non-defective product or a defective product of the member 200 to be inspected, or a certain scale (for example, 3 grades, 5 grades, 10 grades from non-defective products to defective products). It may be sorted, or if there is an evaluation value, the numerical value etc.) may be used. On the other hand, the explanatory variables are the feature amount obtained from the image data and the feature amount obtained from the bright part, the dark part, the light bright part, and the light dark part obtained in S12 to 15. From the image data, the maximum brightness value (usually the maximum brightness value in the bright part), the minimum brightness value (usually the minimum brightness value in the dark part), the average brightness value, the average brightness value dispersion, and the contrast can be obtained. .. From bright areas, dark areas, light bright areas, and light dark areas, each area and total area, height and total height of each part, width and total width of each part, circularity of each part and total circularity. , The number of Eulers in each part and the total number of Eulers are obtained. These can be used as explanatory variables of the evaluation formula (multiple regression equation).

Subsequently, the evaluation unit 104 evaluates the member to be inspected 200 using the value obtained in S18 and the evaluation formula stored in advance (S19). As the evaluation formula, for example, the above formula (1) is used as described above. After that, the evaluation unit 104 displays the evaluation result (S20) and ends the process.

By detecting a light bright part or a light dark part in contact with a bright part or a dark part by such a treatment, a defect in which such a light bright part and / or a light dark part is present can be detected as a product in the member to be inspected. It becomes clear that it affects the required performance of. In particular, by using an evaluation formula, it is possible to quantify how much a defect affects the required performance.

Hereinafter, the present invention will be further described with reference to Examples.

The high barrier film was evaluated using the defect inspection apparatus having the same configuration as that of the above-described embodiment.

The device configuration is as follows.

The light irradiation unit 101 is a bar-shaped lighting device using a blue LED. The photographing unit 102 is a line image sensor having a resolution of 3 μm and a gradation range of 0 to 255. The installation angle of the light irradiation unit 101 and the imaging unit 102 is α = β = 10 degrees.

The position of the light-shielding plate 108 was adjusted so that the specularly reflected light from the light irradiation unit 101 was photographed and the gradation value was 128. As a result, the area of the one-sided light-shielding light 111 of the light irradiation unit 101 became 9/10 from the area at the time of maximum opening.

The member 200 to be inspected is a high barrier film. The high barrier film is one of the flexible barrier films composed of a plurality of layers. In particular, a film having a high function of shielding water vapor and air (particularly oxygen) is called a high barrier film.

In order to obtain the evaluation formula, 2000 defects that may affect the performance of the product are photographed by the apparatus of this embodiment, and each defect is photographed by the cross-sectional SEM to actually perform the product as a high barrier film. It was visually determined whether or not the defect affects (becomes a defective product). Here, a defect that does not affect the performance of the product is regarded as a non-defective product, and a defect that affects the performance of the product is regarded as a defective product.

FIG. 4 is a drawing-substituting cross-sectional SEM photograph for explaining a defect (non-defective product) that does not affect the performance as a criterion. FIG. 5 is a schematic view of a drawing-substituting cross-sectional SEM photograph shown in FIG.

As shown in FIGS. 4 and 5, in the case of non-defective products, foreign matter does not destroy the CHC (Clear Hard Coat) layer (transparent hard coat layer) of the high barrier film at the defective portion. Therefore, defects such as gas permeation do not occur. In the cross-section SEM photograph of FIG. 4, a plurality of mountain-shaped objects can be seen, but these are scratches at the time of cross-section creation and do not originally exist in the high barrier film which is the member to be inspected.

FIG. 6 is a drawing-substituting cross-sectional SEM photograph for explaining defects (defective products) that affect performance as a determination criterion. FIG. 7 is a schematic view of a drawing-substituting cross-sectional SEM photograph shown in FIG.

As shown in FIGS. 6 and 7, in the case of a defective product, a foreign substance breaks the CHC layer of the high barrier film at the convex portion. For this reason, defects such as gas permeation occur.

A judgment formula was created from the photograph of the surface of the high barrier film and the judgment result by the cross-sectional SEM. The judgment formula was created by using the general analysis method of multivariate analysis and discriminant analysis of software JUSE-StatWorks manufactured by Nikkagiken Co., Ltd. The evaluation formula shown in the above equation (1) was obtained by using the determination result of the cross-section SEM as the objective variable and various combinations of the values obtained from the image data taken by the defect inspection apparatus 100 as explanatory variables.

The bright part, the dark part, the light bright part, and the light dark part for obtaining the explanatory variables were discriminated by the following ratios. The range where the brightness value is 115% or more with respect to the average brightness value is the bright part, the range where the brightness value is 90% or less with respect to the average brightness value is the dark part, and the range is 105% or more and less than 115% with respect to the average brightness value. The range of the brightness value was defined as a light bright part, and the range of a brightness value of more than 90% and 97% or less with respect to the average brightness value was defined as a light dark part. Incidentally, as a result of the judgment of the cross-sectional SEM, the number of defects (defective products) affecting the product was 12 out of 2000 points.

As a result, the evaluation formula of Eq. (1) already described was obtained.

Then, each value used as an explanatory variable of the evaluation formula was obtained from the image data obtained by photographing a high barrier film of unknown whether it was a non-defective product or a defective product, and substituted into the evaluation formula of the formula (1). Those having a positive (+) degree of performance influence, which is the objective variable of the obtained evaluation formula, were judged to be non-defective products, and those having a negative (-) were judged to be defective products. The determination results are shown in Table 1.

In this way, it is possible to select a non-defective product or a defective product based on the degree of performance influence obtained from the evaluation formula. In addition, it becomes possible to know how much it affects from the value of the degree of performance influence.

Furthermore, the surfaces of the defective sample 1 and the defective sample 2 were observed.

FIG. 8 is a drawing-substituting surface enlarged photograph of a defective portion determined to be a non-defective product. Further, FIG. 9 is a schematic view illustrating a range determined to be a bright portion from the surface enlarged photograph shown in FIG. FIG. 10 is a schematic view illustrating a range determined to be a dark portion from the enlarged surface photograph shown in FIG. Further, FIG. 11 is a schematic view illustrating the entire range determined to be a bright portion and a dark portion from the surface enlarged photograph shown in FIG.

In the defect that is a non-defective product, as shown in FIGS. 8 and 9, a bright portion 211 is present on the surface 201. Further, as shown in FIGS. 8 and 10, there is a dark portion 212 on the surface 201. However, there are no pale bright areas and pale dark areas. The entire area where the bright part 211 and the dark part 212 are connected has a range as shown in FIG.

FIG. 12 is a drawing-substituting surface enlarged photograph of a defective portion determined to be a defective product. Further, FIG. 13 is a schematic view illustrating a range determined to be a bright portion from the surface enlarged photograph shown in FIG. FIG. 14 is a schematic view illustrating a range determined to be a dark portion from the surface enlarged photograph shown in FIG. FIG. 15 is a schematic view illustrating a range determined to be a pale bright portion from the surface enlarged photograph shown in FIG. FIG. 16 is a schematic view illustrating a range determined to be a faint dark portion from the surface enlarged photograph shown in FIG. Further, FIG. 17 is a schematic view illustrating the entire range determined to be a bright part, a dark part, a light bright part, and a light dark part from the surface enlarged photograph shown in FIG.

In the defective defect, as shown in FIGS. 12 and 3, a bright portion 211 is present on the surface 201. Further, as shown in FIGS. 12 and 14, there is a dark portion 212 on the surface 201. Further, as shown in FIGS. 12 and 15, there is a pale bright portion 213. This pale bright part 213 is in close contact with the dark part 212. Further, the pale bright portion 213 is in contact with the bright portion 211, albeit slightly (see FIG. 17). Further, as shown in FIGS. 12 and 16, there is a faint dark portion 214. This faint dark portion 214 is in close contact with the bright portion 211. Further, the faint dark portion 214 is in contact with the dark portion 212, albeit slightly (see FIG. 17).

The entire area where the bright portion 211, the dark portion 212, the light bright portion 213, and the light dark portion 214 are connected is as shown in FIG.

As described above, from the results of the examples, it was found that when the member to be inspected has a defect that affects the required performance as a product, a light bright portion 213 and a light dark portion 214 are detected in the defect. From this, it is possible to determine whether or not the defect has an effect on the product only by whether or not at least one of the pale bright portion 213 and the pale dark portion 214 can be detected (discriminated). That is, in the embodiment described above, the evaluation unit may determine that there is a defect when at least one of the light light portion 213 and the light dark portion 214 is present. In this case, the process determines that there is a defect when at least one of the light light portion 213 and the light dark portion 214 is determined in S16 and 17 instead of S18 and 19 described above.

Furthermore, by using the evaluation formula obtained in advance, the degree of influence on the performance of the product can be known as a numerical value.

According to the embodiments and examples described above, the following effects are obtained.

In the embodiment and the embodiment, the light portion 211 and the dark portion are in contact with each other from the image data obtained by irradiating the member 200 to be inspected with light from a certain direction and photographing the specularly reflected light from the member 200 to be inspected. It was decided to discriminate 212, and further discriminate the light light portion 213 and the light dark portion 214 in contact with at least one of the bright portion 211 and the dark portion 212. When the image data has at least one of a light light portion 213 and a light dark portion 214, it can be determined that the portion is a defect affecting the product. Therefore, when a defect is detected in the member 200 to be inspected, it can be determined whether or not the defect does not affect the required performance, so that the yield can be expected to be improved.

Further, since the light is irradiated from a certain direction, the portion exposed to the light is the bright portion 211 and the shadow portion is the dark portion 212 in the defective portion. Therefore, the defect can be found only by discriminating at least one of the bright portion 211 and the dark portion 212. Similarly, a portion that is exposed to light but has little reflection is a pale bright portion 213, and a portion that is less exposed to light is a pale dark portion 214. Therefore, by detecting at least one of the light light portion 213 and the light dark portion 214, whether or not the defect which is the bright portion 211 or the dark portion 212 affects the required performance of the member to be inspected as a product. Can be determined.

In addition, each range of the bright part 211, the dark part 212, the light light part 213, and the light dark part 214 is discriminated, necessary features are extracted from these, and an evaluation formula is created by using an analysis method such as multiple regression calculation. By doing so, it is possible to more strictly evaluate the effect of defects on the product. For example, when the member 200 to be inspected is a member having a plurality of layers, it has been possible to detect foreign matter contamination and pinholes, but it is not known which layer the member 200 is generated in, and how much it affects it. In such a case, by creating an evaluation formula in advance, it is possible to know the degree of performance influence on the product from the subtle difference in the feature amount obtained from the image data. Regarding this degree of performance influence, in the above-mentioned embodiment, if this value is positive (+), it is determined as a good product, and if it is negative (-), it is determined as a defective product. It is possible not only to make an alternative judgment but also to make various judgments from the numerical value of the degree of performance influence. For example, even if the degree of performance influence is a positive (+) value, if the value is small, it is determined to be defective with a margin. Of course, the value to be considered as defective is arbitrary.

Although the embodiments and examples to which the present invention is applied have been described above, the present invention is not limited to these embodiments and examples.

For example, the light-shielding plate 108 may be installed between the surface of the member 200 to be inspected and the photographing unit 102 (the light-shielding plate 108 is the light that the reflected light from the surface of the member 200 to be inspected reaches the photographing unit 102. Install on the street). By arranging the light-shielding plate 108 at such a position, it is possible to prevent diffused reflection from the surface of the member 200 to be inspected from being incident on the photographing unit 102.

Further, the photographing unit 102 may use, for example, an area image sensor (a so-called still camera may be used) instead of the line image sensor. When the area image sensor is used, even if the light-shielding plate 108 is used, light other than specular reflection may be picked up. Therefore, if there are many components other than specular reflection, the brightness component corresponding to diffusely reflected light is cut from the image data by using a high-pass filter (cutting less than a certain brightness value), and then the bright part 211. , It is preferable to distinguish the dark part 212, the light light part 213, and the light dark part 214.

In addition, lighting that illuminates the entire area to be inspected may be used. In this case, when the illumination is a point light source and diffused light, the irradiation direction of the light is slightly different depending on the position in the inspection area. Therefore, it is preferable to use a surface light source or a collimator lens so that parallel light can be irradiated so that the light is irradiated from a certain direction at any position in the entire inspection area.

Furthermore, it is possible to combine surface irradiation of collimated light with an area image sensor.

Further, as the member 200 to be inspected, a high barrier film was used in the examples, but the present invention is not limited to this, and various gas barrier films, polarizing films, color filters, heat insulating films, shading of specific wavelengths (for example, ultraviolet rays), etc. are performed. It can be used for inspection of various members such as functional optical films and glass surfaces. In particular, it is suitable for defect inspection due to foreign matter contamination or pinholes in a film composed of a plurality of layers.

In addition, the present invention can be modified in various ways based on the configurations described in the claims, and these are also within the scope of the present invention.

100 defect inspection equipment,
101 light irradiation unit,
102 Shooting section,
103 Support,
104 Evaluation Department,
108 shading plate,
110 light,
111 One-sided shading light,
200 members to be inspected,
211 Akibe,
212 Dark part,
213 Pale bright part,
214 Light dark areas.

Claims (11)

The step (a) of irradiating the surface of the member to be inspected with light from a certain direction and acquiring image data obtained by photographing the surface, and
The step (b) of obtaining the average luminance value from the image data and
From the image data, a step (c1) of determining a range of bright areas where the luminance value is equal to or greater than the first predetermined ratio from the average luminance value.
From the image data, a step (c2) of determining a range of a dark portion that is in contact with the range of the bright portion and whose luminance value is equal to or less than a second predetermined ratio from the average luminance value.
Range from said image data, the pale light portion luminance values have contact with at least one of the ranges has a first smaller than the predetermined ratio exceeds the average luminance value of the range and the dark portion of the range of the bright portion At the stage of determining (d1) and
From the image data, a range of a light dark portion that is in contact with at least one of the bright portion range and the dark portion range and whose brightness value exceeds the second predetermined ratio and is less than the average brightness value. The stage of discrimination (d2) and
Defect inspection method. An evaluation formula is prepared in advance using the image data, the bright part, the dark part, the light bright part, and the feature amount obtained from the light dark part as explanatory variables, and the influence on the member to be inspected as the objective variable. ,
The image data acquired from the member to be inspected, the bright part, the dark part, the light bright part, and the feature amount obtained from the light dark part are applied to the evaluation formula, and the bright part, the dark part, and the light bright part are applied. parts, and the light dark portion detected defect portion further has a step (e) assessing the impact of the the inspection member, a defect inspection method according to claim 1. The step (d1) of discriminating the range of the light bright portion is a step of discriminating the range in which the luminance value is equal to or more than the third predetermined proportion and less than the first predetermined proportion than the average luminance value.
The step (d2) for determining the range of the light dark portion is a step for determining a range in which the luminance value exceeds the second predetermined proportion and is equal to or less than the fourth predetermined proportion from the average luminance value. or second defect inspection method according to. The step (a) of irradiating the surface of the member to be inspected with light from a certain direction and acquiring image data obtained by photographing the surface, and
The step (b) of obtaining the average luminance value from the image data and
From the image data, among the range of the bright part where the brightness value is equal to or more than the first predetermined ratio from the average brightness value and the range of the dark part where the brightness value is equal to or less than the second predetermined ratio from the average brightness value. The step (c) of determining at least one range and
When the dark portion is determined in the step (c), the range of the pale bright portion that is in contact with the range of the dark portion and whose luminance value exceeds the average luminance value and is less than the first predetermined ratio is the image. When the bright part is discriminated in the step (d1) of discrimination from the data or the step (c), it is in contact with the range of the bright part, and the brightness value exceeds the second predetermined ratio and is less than the average brightness value. At the stage (d2) of discriminating the range of the faint dark part, which is, from the image data,
Have,
The step (d1) of discriminating the range of the pale bright portion from the image data is a step of discriminating a range in which the brightness value is equal to or more than the third predetermined ratio and less than the first predetermined ratio from the average brightness value. ,
The step (d2) of discriminating the range of the light dark portion from the image data is a step of discriminating a range in which the luminance value exceeds the second predetermined proportion and is equal to or less than the fourth predetermined proportion from the average luminance value. , Defect inspection method. The defect inspection method according to any one of claims 1 to 4, wherein the member to be inspected is a gas barrier film having a plurality of layers. A program that causes a computer to execute the defect inspection method according to any one of claims 1 to 5. A light irradiation part that irradiates the surface of the member to be inspected with light from a certain direction,
An imaging unit that photographs the surface of the member to be inspected,
The average luminance value is obtained from the image data of the surface of the member to be inspected taken by the photographing unit, and the range of the bright portion where the luminance value is equal to or more than the first predetermined ratio from the average luminance value is determined from the image data. , The range of the dark part which is in contact with the range of the bright part and whose brightness value is equal to or less than the second predetermined ratio from the average brightness value is determined, and at least one of the range of the bright part and the range of the dark part is determined. The range of the light bright part which is in contact with the above and the brightness value exceeds the average brightness value and is less than the first predetermined ratio is determined, and at least one of the range of the bright part and the range of the dark part is determined. An evaluation unit that determines the range of a light dark portion that is in contact with and whose luminance value exceeds the second predetermined ratio and is less than the average luminance value.
Defect inspection equipment. The evaluation unit uses the image data, the bright part, the dark part, the light bright part, and the feature amount obtained from the light dark part as explanatory variables, and sets the evaluation formula in advance with the influence on the member to be inspected as the objective variable. I remember
The evaluation unit applies the image data acquired from the member to be inspected, the bright part, the dark part, the light bright part, and the feature amount obtained from the light dark part to the evaluation formula, and applies the bright part, the dark part, and the feature amount obtained from the light dark part to the evaluation formula. The defect inspection apparatus according to claim 7 , wherein the dark part, the light bright part, and the defect part in which the light dark part is detected evaluate the influence on the member to be inspected. When the evaluation unit determines the range of the light bright portion, the range of the light bright portion is a range in which the brightness value is higher than the average brightness value and is equal to or higher than the third predetermined ratio and less than the first predetermined ratio. And determine
A claim that, when determining the range of the light dark portion, the range in which the luminance value exceeds the second predetermined ratio and is equal to or less than the fourth predetermined ratio lower than the average brightness value is determined as the range of the pale dark portion. 7. The defect inspection apparatus according to 7. A light irradiation part that irradiates the surface of the member to be inspected with light from a certain direction,
An imaging unit that photographs the surface of the member to be inspected,
The average brightness value is obtained from the image data of the surface of the member to be inspected taken by the photographing unit, and the range and brightness of the bright part where the brightness value is equal to or more than the first predetermined ratio from the average brightness value from the image data. At least one of the dark areas whose values are equal to or less than the second predetermined ratio from the average luminance value is discriminated, and the determined range of the bright areas or the dark areas is the dark area. In this case, the range of the light bright portion which is in contact with the range of the dark portion and whose luminance value exceeds the average luminance value and is less than the first predetermined ratio is determined from the image data, or the bright portion of the bright portion. When the determined range of the range or the dark portion is the bright portion, the luminance value is in contact with the bright region and the luminance value exceeds the second predetermined ratio and is less than the average luminance value. An evaluation unit that determines the range of the faint dark area from the image data, and an evaluation unit.
Have,
When the evaluation unit determines the range of the pale bright portion from the image data, the evaluation unit determines the range in which the brightness value is higher than the average brightness value by a third predetermined ratio or more and less than the first predetermined ratio. Determined as the range of the bright part,
When the range of the light dark part is discriminated from the image data, the range in which the brightness value exceeds the second predetermined ratio and is equal to or less than the fourth predetermined ratio lower than the average brightness value is determined as the range of the light dark part. to, defect inspection apparatus. A light-shielding plate for adjusting the amount of light is provided either between the light irradiation unit and the member to be inspected or between the member to be inspected and the photographing unit.
The photographing unit has a line image sensor that photographs a portion of light that is not shielded by at least the light-shielding plate on the surface of the member to be inspected and is irradiated to the member to be inspected.
The defect inspection apparatus according to any one of claims 7 to 10, further comprising a moving portion for moving the member to be inspected relative to the light irradiation portion and the photographing portion.

Images