JP2002107305A - Device and method for inspecting birefringent object to be inspected - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an inspecting device and method, capable of making an object having birefringence to be measured, for example, the surface of a transparent film or a film package and the state of the package sharp, and performing accurate inspections. SOLUTION: The device for inspecting the birefringent object to be inspected is provided with an irradiating device 12 for irradiating the birefringent object to be inspected 1 with light polarized in a prescribed polarization direction, an image pickup element 22 for detecting light from the object to be inspected 1 to acquire color information each on a plurality of parts of the object to be inspected 1, an analyzer 21 arranged between the object to be inspected 1 and the image pickup element 22, and a classifying mechanism for classifying the plurality of parts into a plurality of regions, on the basis of the distance between each color information acquired at the image pickup element 22 and reference information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus and a method for inspecting a birefringent test object, and more particularly to a method for inspecting a birefringent test object such as a transparent film or a film package or a state inside a package. The present invention relates to an inspection apparatus and an inspection method capable of inspecting a birefringence inspection target.

[0002]

2. Description of the Related Art Conventionally, in the inspection of a transparent film or a film package, as shown in FIG. Then, the shadow or the like of the fold formed by the illumination is visually observed with the naked eye 5 or observed using a camera.

[0003]

However, according to the conventional method as described above, the state of the surface to be observed of the film or the film package, for example, the fold to be seen is not always clearly shown, so that the accurate Sometimes it was difficult to observe.

Accordingly, the present invention provides an inspection apparatus and an inspection method capable of accurately and accurately inspecting an object to be inspected having birefringence, for example, the surface of a transparent film or a film package or the state of the package. The purpose is.

[0005]

In order to achieve the above object, a birefringent inspection object inspection apparatus according to the first aspect of the present invention has a birefringence as shown in FIGS. An irradiation device 12 for irradiating a test object 1 having a property with light polarized in a predetermined polarization direction; an imaging device for detecting light from the test object 1 and acquiring color information of each of a plurality of portions of the test object 1 Element 22; inspection object 1 and image sensor 22
And a classification mechanism for classifying the plurality of locations into a plurality of regions based on the distance of each piece of color information obtained by the image sensor 22 from the reference information.

Here, the color information is information that can be digitized,
The digitized reference information and each color information can be expressed by numerically expressing the closeness and distance between each other. The numerical proximity and the magnitude of the numerical value are the distances between the color information. The color information that can be quantified is typically hue H, intensity I, and saturation S, but may be, for example, the intensity of three primary colors of red R, green G, and blue B light. The distance is not limited to a three-dimensional distance.
It may be a dimensional or one-dimensional distance.

According to a second aspect of the present invention, in the inspection apparatus for inspecting an inspection object having a birefringence according to the first aspect, the shape of the area is represented by coordinates based on the coordinates of the plurality of classified locations. And a calculating mechanism 41d for calculating parameters for specifying the shape.

The parameters are, for example, the area of the region, the perimeter,
It is at least one of the moment, the center of gravity, and the like in addition to the circularity. For example, the parameters of the non-defective inspection object are determined, the allowable range of each parameter is determined, and it is determined whether the non-defective or non-defective product is within the allowable range.

In the inspection apparatus for inspecting a birefringent inspection object according to the first or second aspect, furthermore, as shown in claim 3 and, for example, as shown in FIG. An HIS conversion mechanism 41b for converting information into hue, intensity, and saturation, respectively; the classification mechanism includes, for each of a plurality of predetermined regions of the inspection object 1, an average value and a standard deviation of hue, and an average value of intensity. It has a reference information storage unit 41f for obtaining and storing an average value and a standard deviation of standard deviation and saturation, and stores the hue, intensity, and saturation of the plurality of locations of the inspection object 1 in the reference information storage unit 41f. The distance is obtained based on the average value and the standard deviation of the corresponding hue, intensity, and saturation, and the plurality of locations are classified into an area that gives the shortest distance.

In order to achieve the above object, a method of inspecting an inspection object having birefringence according to the invention according to claim 4 is, for example, as shown in FIG. An irradiation step S121 of irradiating light polarized in a polarization direction; an imaging step S of detecting light from the inspection object 1 and acquiring color information of each of a plurality of portions of the inspection object 1
123a and S123; imaging steps S123a and S123
At this time, a light detection step S122 for detecting light from the inspection object 1; and a plurality of the plurality of locations are determined based on a distance from the reference information of each color information acquired in the imaging steps S123a and S123. And a classification step S125 for classifying the area.

In order to achieve the above object, a method of inspecting a birefringent test object having birefringence according to the invention according to claim 5 is, for example, as shown in FIGS. A first irradiation step S101 of irradiating a non-defective inspection object with light polarized in a predetermined polarization direction, the method including: detecting light from the non-defective inspection object and performing a plurality of predetermined inspections on the non-defective inspection object. First imaging step S103 for acquiring color information for a plurality of locations for each area in the area
a, S103; a first light detection step S102 for detecting light from the non-defective inspection object at the time of the first image pickup steps S103a, S103; and a first image pickup step S103.
a, a reference information obtaining step S104 for obtaining an average value and a standard deviation of each of the color information obtained in S103 for each predetermined region; and a predetermined polarization for the inspection object 1 to be determined having birefringence. A second irradiation step S121 of irradiating light polarized in a direction; a step of detecting light from the inspection target 1 to be determined and acquiring color information of each of a plurality of portions of the inspection target 1 to be determined. 2 imaging process S123a, S
123; a second light for detecting the light from the inspection object 1 to be determined in the second imaging steps S123a and S123.
The second imaging step S123a, S
A classification step S125 of classifying the plurality of places into the predetermined areas based on the distances of the respective pieces of color information obtained in 123 from the reference information for each of the predetermined areas.

Further, in the inspection method of an inspection object having birefringence according to claim 4 or 5, the shape of the plurality of predetermined regions is represented by coordinates based on the coordinates of the plurality of classified locations. And a calculating step S129 for calculating a parameter specifying the shape.

Further, the method for inspecting an inspection object having birefringence includes an HIS conversion step of converting the color information at the plurality of locations into hue, intensity, and saturation, respectively; A reference value storing step S104 for storing an average value and a standard deviation of hues at each of a plurality of locations in a predetermined region of the inspection object, an average value and a standard deviation of intensity, and an average value and a standard deviation of saturation; Based on the hue, intensity, and saturation of the plurality of locations of the inspection object to be determined, and the corresponding hue, intensity, and saturation average and standard deviation stored in the reference value storage step, the distance , And a shortest distance determining step S125 of obtaining an area giving the shortest distance.

In the above-described method for inspecting an inspection object having birefringence, the predetermined polarization direction and the polarization direction of the analysis may be set in a relationship of orthogonal Nicols.

[0015]

Embodiments of the present invention will be described below with reference to the drawings. In the drawings, the same or corresponding members are denoted by the same or similar reference numerals, and duplicate description is omitted.

FIG. 1 shows a state of a transparent film and a film package as a test object having birefringence. Here, the term “transparent” includes a case where the film placed on the other side is translucent to the extent that the object placed on the other side can be visually identified when viewed from one side of the film. (A) of the figure is a plan view showing a sheet-like transparent film spread out, and (b) and (c) are plan views showing a folded state of a film package wrapped with the transparent film as shown in (a). is there.

A transparent film for packaging is generally manufactured by forming a polymer material such as polypropylene or polyethylene into a sheet by stretching it in a substantially constant direction. Such a transparent film has birefringence, and when polarized light enters these films, the polarization state of the emitted light changes.

Since the refractive index of the transparent film varies depending on the wavelength of the incident light, the amount of change in the polarization state of the light emitted from the film depends on the wavelength of the light. Here, the light emitted from the film is such that the light incident on the package is first reflected on the surface of the film, and secondly is incident on the film and reflected on the inner surface of the film, and the light inside the film is again reflected. Thirdly, those which have passed through the film, and thirdly, those which have also passed through the inner surface of the film and have been reflected by the groundwork and have passed through the inside of the film again and have come out.

The first light (surface-reflected light) is a component of the light and does not pass through the film, so that the state of polarization does not change. In other words, the first light does not include information such as the overlap of the film packages, but merely information on the surface of the film.

Since the second light and the third light pass through the inside of the film, the state of polarized light changes due to the state of the film package overlapping. However, since the second light has not reached the base, the information of the base is not included, and only the state of the film is included.

On the other hand, since the third light has reached the base, it contains information on the state of the base together with the state of the film.

Further, the amount of change in the polarization state of the transmitted light is
It is proportional to the thickness of the film. Therefore, a thick portion of the film or a portion that is thick as a result of the overlap has a large change in polarization.

Referring to FIG. 15, the principle of the present invention will be described in the case of a transparent film having birefringence such as a film made of an organic polymer material produced by stretching. (A) shows how light enters and exits the film 1a, and (b) shows the axis of birefringence of the incident light and the outgoing light inside the film 1a, the O axis (the axis in the ordinary ray direction). ), The E axis (the axis in the extraordinary ray direction), and the incident light that is linearly polarized light form 45 °.

The incident light on the film 1a is linearly polarized at 45 ° with respect to the birefringent axis of the film 1a as shown in FIG. Here, assuming that K is a spatial frequency, the complex amplitude distribution Ei of linearly polarized light transmitted through the optical path length X is generally expressed by the following equation. Ei = Asin (K ・ X)

Ei is a component in two directions as follows:
(E-axis direction component) and Eie (E-axis direction component). Eio = Asin (Ko · X) / √2 Eie = Asin (Ke · X) / √2 where Ko = 2πNo / λ, Ke = 2πNe / λ, No is the refractive index of the ordinary ray component, and Ne is Ne This is the refractive index of the extraordinary ray component.

Therefore, the polarization state of light that enters from the surface of the film 1a, is reflected inside the film 1a, and then exits again from the surface of the film 1a is expressed by the following equation.
Here, assuming that the light vertically enters and exits perpendicularly to the film 1a having the thickness d, since the light reciprocates, X = 2d (thickness), so that Ero = Asin (Ko · X) / √ 2 Ere = Asin (Ke · X) / √2 = Asin (2Ko · d + 2 (Ke−Ko) d) / √2 = Asin (2Ko · d + θ) / √2 Here, it is assumed that θ = 2 (Ke−Ko) d. The light is assumed to be perpendicularly incident on the film 1a having a thickness d and emitted perpendicularly. However, even if the light is obliquely incident, if the oblique optical path between the surface and the inner surface on the opposite side is d, the above equation can be obtained. Applicable as it is.

Next, the amount of light I detected when the analyzer is placed in a crossed Nicols state is expressed by the following equation. I = ((Ero−Ere) / √2) ² = 1 / 4A ² (sin (2Ko · d) −sin (2Ko · d + θ)) ² = A ² cos ² (2Ko · d + θ / 2) sin ² (θ / 2) That is, because of the birefringence, the amount of light detected in the crossed Nicols state changes, and the refractive index also has wavelength dependence. Differences, sheet thickness unevenness, etc. are observed as color differences.

Here, when a birefringent plate is arranged in the optical path, the birefringence of light can be changed by the birefringent plate, and when the birefringent plate is rotated around the optical axis, the rotation angle changes. Since the birefringence can be changed, the effective spatial frequency K and the phase difference θ of the combined birefringent plate and transparent film change. Therefore,
As will be described later with reference to FIG. 12, the birefringent plate can change the coloring of light passing through the analyzer.

FIG. 1A is a plan view seen from one side of the unfolded film. In the process of manufacturing the film, the film is stretched in a direction indicated by a double-sided arrow pointing up and down in the figure. I have. In such a film sheet, there is an axis of birefringence in a stretching direction (direction indicated by p in the drawing) and a direction perpendicular to the drawing direction (direction indicated by q in the drawing).

When such a transparent film sheet is used for wrapping tobacco or caramel, a fold as shown in (b) and (c) is formed on one rectangular surface of the wrapped rectangular parallelepiped package. Can overlap.

The direction of stretching of the folded transparent film is, as indicated by the double-sided arrows in FIG. 3B, the direction of the long side (horizontal direction in the figure) or the direction of the short side (horizontal direction in FIG. (Vertical direction in the figure), or both directions of the long side direction and the short side direction. This means that there is a place where the axial directions of ordinary light and extraordinary light in birefringence are switched. The crossing of the double-headed arrow in the figure indicates that the stretching directions are mixed due to the overlapping of the transparent films.

As shown in (c), the number of overlapping transparent films differs for each region. This is equivalent to the difference in the thickness of the transparent film. Therefore, the amount of change in polarization differs in each region. In the drawing, the numbers 1 to 5 in each area indicate the number of overlapping sheets.

With reference to FIG. 2, a first example of an apparatus used to acquire the color information of each area as described above will be described. In the figure, a white light source 11 is disposed obliquely above the surface of the film package 1, and a polarizer 12 as an illumination-side polarization filter is disposed between the white light source 11 and the surface of the film package 1. Optical axis AX1 of polarizer 12
Crosses the surface of the film package 1 obliquely.

An analyzer 21 serving as a light-receiving-side polarization filter is arranged in a direction in which light from the light source 11 via the polarizer 12 is regularly reflected on the surface of the film package 1. Analyzer 2
The first optical axis AX2 is in the regular reflection direction of the optical axis AX1 with respect to the surface of the film package 1. A color CCD camera 22 as an image sensor is disposed at a position where regular reflection light from the light source 11 is received via the analyzer 21. In the figure, the polarizer 12 and the analyzer 21 are configured to be rotatable around the optical axis by rotating mechanisms 31 and 32, respectively, which will be described later in detail as a fourth example device.

For example, it is assumed that the light source 11 is a white light source having a wide wavelength range as shown in FIG. That is, wavelength 4
It contains visible light from 00 to 750 nm and includes a wide range of wavelengths. This becomes light linearly polarized via the polarizer 12.

FIG. 4 shows an example of an image when the transparent film package is irradiated by the above-described irradiation device.
On the surface of the film package 1, FIG.
As shown in (c), since the amount of change in polarization differs for each wavelength in each region bounded by the fold, light having a wavelength that can pass through the analyzer 21 is limited. Therefore, color C
The image obtained by the CD camera 22 is, as shown in FIG.
Each area is colored differently. Each area is, for example,
It has colors such as red, yellow, purple, peach, blue, and green.

Since each region is observed in different colors, it is possible to determine the fold of the package, the overlapping state of the films, and the like based on the color. Here, in order to automate the inspection of the quality of the package, it is convenient that the fold to be inspected is represented by coordinates and the shape of the fold can be compared with the shape of a good product.

Referring to the partial plan view of the film package (a) of FIG. 5 and the three-dimensional coordinate system display diagram (b) of the color information, the boundary of each area, which is the fold of the package, is represented by coordinates. An embodiment of the inspection method will be described. FIG.
Attention is focused on two of the plurality of regions, and these are defined as region 1 and region 2. First, with respect to a typical non-defective package having a normal crease, color information of a plurality of points is obtained for each area having the crease as a boundary line, and the color information of each point is used as a hue (hue) and an intensity (intensity). , Saturation (s
conversion) into three elements. These elements can be represented numerically. Next, for each of the regions 1 and 2, from the measured values in each region, the average hue value MH1,
MH2, average intensity values MI1, MI2, average saturation value MS
1, MS2, hue standard deviations σH1, σH2, intensity standard deviations σI1, σI2, saturation standard deviations σS1, σS
2 is calculated. These standard values and standard deviations serve as reference information. For example, the hue reference value of the area 1 is MH1.

Next, for a non-defective package,
As shown in (a), the color information of the point X on the surface to be inspected,
The hue XH, intensity XI, and saturation XS are measured. Then (b)
, Distances Y1 and Y2 between the point X and the average value of each area on the three-dimensional coordinates are obtained.

Hereinafter, a method of obtaining the distance will be described using equations. The pixel of the image sensor that detects each point can be specified by coordinates on the image sensor. The area to which the pixel belongs is determined in the following manner. The H direction component of the distance Y1 from the reference value of the area 1 to the point X, that is, the hue component is YH1,
If the I-direction component, that is, the intensity component is YI1, and the S-direction component, that is, the saturation component, is YS1, the following equation is established.

YH1 = | (XH-MH1) / σH1 | YS1 = | (XS-MS1) / σS1 | YI1 = | (XI-MI1) / σI1 | The distance components YH1, YS1, and YI1 are dimensionless.

Next, the distance between the two points (reference value and point X) is calculated. Examples of the distance that can be used in the embodiment of the present invention include a Euclidean distance, an urban distance, and a Mahalanobis distance. Of these, a typical Euclidean distance Y1
Can be calculated by the following equation.

Y1 = (YH1 ² + YS1 ² + YI1 ² ) ^1/2

As described above, for the region 1 and the region 2,
The distances Y1 and Y2 are calculated and compared. Y1 <Y
In the case of 2, the point X is determined to belong to the area 1 and is classified. Here, the case of two regions has been described, but if there are five regions for non-defective products, the distances Y1, Y2, Y3, Y4, and Y5 from the reference value to the point X are calculated for each of these regions, and the minimum is calculated. Classify into the area that gives the distance.

When the city distance is used, the distance Y1 from the point X to the reference value of the area 1 can be calculated by the following equation. Y1 = YH1 + YS1 + YI1 As in the case of the Euclidean distance, if Y1 <Y2,
It is determined that the point X belongs to the area 1.

Depending on the material of the film, the method of folding the package, etc., it may be possible to perform a more appropriate classification by using the city area distance than the Pythagorean distance. In some cases, the Mahaviras distance is optimal. An appropriate distance may be selected and used according to the situation.

In the above case, the pixels may be forcibly classified into any one of the regions depending on the length of the distance, and it is determined that the region is too unknown for a pixel whose distance from the reference value of each region is too large. Is also good. For this purpose, a threshold may be set for the distance. When the threshold value is exceeded, it is determined that it is unknown.

Further, isolated points are removed. An isolated point is a single point or a small number of points, and is a point determined to belong to an area different from the area to which surrounding pixels belong.

In this way, by classifying an arbitrary point on the surface of a non-defective product into each region, a boundary line between the regions is represented by coordinates. After the coordinates are determined, parameters such as the area, perimeter, circularity, moment, and center of gravity are calculated for each of the non-defective regions. The parameter can be used without being limited to the above as long as it is an attribute of the graphic. The circularity is a value obtained by dividing the calculated area by the area of a circle having the calculated perimeter, and becomes 1 when the shape of the region is a perfect circle, and approaches 0 as the region is farther from the perfect circle.

The above measurement is performed for a plurality of non-defective products.
For the plurality of non-defective products, calculate an average value and a standard deviation of the respective parameters, and use these as reference parameters.
It is stored in the image processing device. In addition, a threshold that can be determined as a non-defective product is set for each parameter. Alternatively, a threshold value is set for a composite distance in which each parameter is a distance in each dimension.

Next, similar color information measurement and parameter calculation were performed for an arbitrary point on the surface of the inspection object, and the distance between each obtained parameter and the reference parameter of each area was described by HIS. The pass / fail of the package is determined based on whether the parameter of each area exceeds the set threshold value.

The above method will be described with reference to the flowcharts of FIGS. FIG. 6 shows a learning process for determining and storing reference information and reference parameters. The flowchart of FIG. 6A shows a method of calculating the average value and the standard deviation of the color information for each of the non-defective areas. First, a non-defective product, for example, a non-defective film package is prepared and irradiated with polarized light (step S101 (hereinafter, “step” is abbreviated and simply referred to as S101)). As a result of the irradiation of the polarized light, the light reflected from the non-defective product is detected (S102).

Next, a two-dimensional image of the non-defective film package is obtained by forming an image of the detected light (S103a). Color information is acquired for each of the plurality of points of the acquired two-dimensional image in association with the coordinates one by one (S103). When the color information is obtained for a plurality of points in this way, as described above, the surface of a good product can be identified as a plurality of regions according to the obtained color information. Since it is a non-defective product, since the area bounded by the fold is known, the average value and the standard deviation can be obtained for the color information, for example, the hue, the intensity, and the saturation for each area. These are stored as reference information (S104). Since color information is almost constant for non-defective products, it is sufficient to obtain the average value and the standard deviation for one non-defective product.
Is performed, and all the averages are obtained in S104,
It is preferable because a more stable average value and standard deviation can be obtained. Alternatively, S101 to S104 may be performed on a plurality of non-defective products to obtain an average value and a standard deviation, and these may be further averaged.

Next, a method for obtaining reference parameters for non-defective products will be described with reference to the flowchart of FIG. A plurality of good products are prepared, and first, one of them is irradiated with polarized light in the same manner as in S101 (S105). The light reflected from the non-defective product as a result of the irradiation of the polarized light is analyzed (S106). Next, a two-dimensional image of a non-defective film package is obtained by forming an image of the detected light (S107a). Color information is acquired for each of the plurality of points of the acquired two-dimensional image in correspondence with the coordinates one by one (S107). At this time,
The images acquired in S105 to S107 include S101
Those obtained in steps S103 to S103 may be used. With respect to the color information about a plurality of points acquired in this way (S107), the distance from the reference information obtained and stored in S104 is calculated (S108).

The distance of each point is calculated with respect to the reference value for each area. The points are classified into the area giving the minimum distance (S109). As a result of classifying many points on the surface, the shape of each area, that is, the boundary line of each area can be represented by coordinates (S110).

Since the shape is represented by coordinates, parameters as attributes of such a shape can be calculated (S11).
1). As described above, the parameters are, for example, the area, the perimeter, the circularity, the moment, and the center of gravity.

The parameters are obtained for all the non-defective products prepared (S112, S113), and the average value and the standard deviation of the parameters of a plurality of non-defective products are obtained (S114). As for the center of gravity, an average position of the center of gravity may be obtained. These are stored in the image processing apparatus as reference parameters.

At this time, it is preferable to set a threshold value as to how far the distance from each reference parameter can be determined as a non-defective product. The above is the learning process.

In the embodiment of the method described above, a two-dimensional image is first acquired (S107a), and the steps following color information acquisition (S107) are performed for each point while moving points on the image (S107a). Is polarized light irradiation for each point (S1
05), photometry (S106), two-dimensional image acquisition (S107)
a), color information acquisition (S107) and a series of steps may be repeated while moving points. The same applies to the case (a).

Referring to FIG. 7, a method of inspecting an actual inspection object which is not necessarily a non-defective product will be described. In the drawing, the inspection object is irradiated with polarized light (S121), and the reflected light is analyzed (S12).
2) Acquiring a two-dimensional image (S123a) and acquiring color information of an arbitrary point (S123) are similar to the case of obtaining the reference value.

The distance between the acquired color information of the arbitrary point and the reference information which is obtained and stored for the non-defective product is calculated (S124). The distance is calculated from the average value and the standard deviation of non-defective products in each area. An arbitrary point is classified into an area that gives the shortest distance among the calculated distances (S1).
25). It is determined whether or not the entire inspection target surface of the inspection target has been covered (S126). If not, the process moves to the next point (S127), and the same measurement is performed on that point to classify the area.

Although the analysis is performed during the imaging step, in the present embodiment, the analysis is performed before the two-dimensional image acquisition S123a, and the detected light is received by the imaging device to obtain an image. Form. After that, the color information is obtained one after another by moving a point on the image once obtained (S127) (S12).
3).

Although not shown in FIG. 7, as another embodiment, polarized light irradiation (S121), light analysis (S122),
The color information acquisition (S123), the distance calculation (S124), and the area classification (S125) are performed as a continuous process. After the point is moved (S127), the process returns to the polarized light irradiation process (S121). It may be repeated twice. At this time, two-dimensional image acquisition, color information acquisition, and other steps may be repeated while the polarized light is kept irradiated and the analysis is continuously performed. Performing the analysis during the imaging process includes such a case.

When all the points are classified, the inspection target surface is classified into one of the areas. As a result, the boundary of the area can be represented by coordinates (S128). That is, the shape of the area is represented by coordinates. For the shape represented by coordinates, parameters can be calculated (S129).

Next, for example, when the area, perimeter, and moment are used as target parameters, a three-dimensional distance is calculated from the difference between the reference parameter and each parameter in the same manner as in the case of hue, intensity, and saturation. If the distance is smaller than the threshold, it is determined to be good, and if the distance is larger, it is determined to be bad (S131). The non-defective products are compared with the reference parameters obtained and stored (S130), and the quality is determined (S131).

In the above embodiment, the color information is the hue, the intensity, and the saturation. However, R (red), G (green), and B (blue), that is, three primary colors of light may be selected. At this time, the intensity of each primary color may be used as the numerical value.

Although the distance has been described using a three-dimensional distance,
Depending on the properties of the inspection target, the distance may be two-dimensional, for example, using the hue H and the intensity I, or may be one-dimensional, for example, selecting only the hue as the distance. Alternatively, a threshold is set for each parameter, and 1 is set for each parameter.
A dimensional comparison is made, and for example, if two out of three persons do not exceed the threshold value, it is possible to pass. Further, when four or more independent variables are obtained, a distance in a space of four or more dimensions may be used.

According to the above-described embodiment, the shape of each region becomes clear, and parameters for the shape can be obtained. Therefore, comparison with non-defective products can be easily performed. Therefore, the quality of the inspection object can be reliably determined.

Referring to FIG. 8, an example of the structure of the image processing apparatus used in the embodiment of the present invention will be described. The image processing device 41 includes a control unit 41A and a storage unit 41B. The control unit 41A includes an image processing module 41a, an HIS conversion module 41b, a distance calculation module 41c, a parameter calculation module 41d, and a comparison module 41e. The storage unit 41B includes a reference information storage unit 41f and a reference parameter storage unit 41g.

The image processing device 41 includes an image pickup device 2 of a camera.
The image processing module 41a receives a signal including the color information transmitted from the image processing module 2 and performs image processing to obtain a predetermined image. The HIS conversion module 41b converts the color information into hue, intensity, and saturation. Distance calculation module 41c
Then, the distance from the reference value is calculated based on the numerical value converted by the HIS conversion module 41b and the reference value called from the reference information storage unit 41f. The distance is compared and classified by the comparison module 41e.

The parameter calculation module 41d calculates the parameters and determines the acceptability based on the distance from the parameter reference value called from the reference parameter storage 41g.

With reference to FIG. 9, a second example of a color information acquiring apparatus that can be used in the present embodiment will be described. In the following description, various examples of color information acquisition including the second example will be described. The description of the distance is the same as that of the above-described embodiment, and thus will be omitted. In the figure, a monochrome CCD camera 24 is arranged as an image sensor instead of the color CCD camera 22 of the first example. Analyzer 21 and monochrome C
The color filter 23 is disposed between the color filter 23 and the CD camera 24. However, the position where the color filter 23 is arranged is not limited to the position between the analyzer 21 and the monochrome CCD camera 24, but may be any other position between the surface of the film package 1 to be inspected and the imaging device 24. May be. Here, it is shown that the color filter 23 is inserted, but instead, the image sensor 24 may be an image sensor that detects only light of a specific wavelength.

According to the second example, the film package 1
By selecting the color of the color filter 23 in correspondence with the color in each region, the region can be extracted and observed. For example, referring to FIGS. 1 and 4, the area where the film overlaps five sheets becomes pink, but if the color filter 23 is a filter that transmits this color, only the area where the film overlaps five sheets becomes The monochrome CCD camera captures an image in white, and the other parts are black. Therefore, it is possible to particularly extract and observe the five-overlap portion.

In this example, since a monochrome camera is used, the hue and saturation are meaningless, and a one-dimensional distance of only intensity is used.

An example of the third color information acquiring device will be described with reference to FIG. Here, a monochromatic light source 13 is arranged in place of the white light source in the second example (a). In this embodiment, in the reflected light from the film package,
Only light from a specific area that reflects light that matches the polarization direction of the analyzer 21 is transmitted through the analyzer 21 and the monochrome CCD
Observed by the camera 24. The specific area changes according to the color of the monochromatic light source 13. Also in this case, a one-dimensional distance is used.

In the above embodiment, when the polarizer and the analyzer are in a crossed Nicols relationship, the effect of removing the shining is obtained, and the light reflected inside the film or on the base can be observed. it can. Although the illuminator tends to be an isolated point, it is not necessary to remove the illuminated point by software.

The polarizer 12 and the analyzer 21 having the orthogonal Nicols relationship mean that the polarizers 12 and the analyzer 21 are coupled in an optical path such that the vibration planes of the electric vector of the linearly polarized light passing through the polarizer 12 and the analyzer 21 are perpendicular to each other. Refers to the state of being placed inside. In the above embodiment, the light transmitted through the polarizer 12 is reflected by the transparent film without any special change in the polarization state, and the vibration surface of the light is reflected by the transparent film. It means that the surfaces are orthogonal to each other.

Referring back to FIG. 2, and referring to FIG. 12, a fourth example of the color information acquiring apparatus used in the embodiment of the present invention will be described. In this example, as shown in FIG. 2, the polarizer 12 and the analyzer 21 are configured to be rotatable around respective optical axes AX1 and AX2. The polarizer 12 is rotated by an optical axis A by a rotation mechanism 31.
The analyzer 21 is rotated around X1, and the analyzer 21 is rotated by the rotation mechanism 32. The rotation mechanism 31 and the rotation mechanism 32 are configured to be rotatable by the controller 33 while maintaining the relative relationship between the polarizer 12 and the analyzer 21. In particular, the polarizer 12 and the analyzer 21 can be rotated while maintaining the orthogonal Nicol relationship.
With such a configuration, polarized light having a specific direction can be selected from the polarized light transmitted through the polarizer 12 and the transparent film by the analyzer 21 and observed by the imaging devices 22 and 24.

An image processing device 41 is connected to the CCD camera 22. The image processing device 41 can form an image of the surface of the film package 1 based on a signal from the CCD camera 22.

In particular, FIG. 10 shows a case where the structure is made rotatable while maintaining the relationship of orthogonal Nicols. (A)
The color classification of the surface region of the film package 1 when the polarization directions of the polarizer 12 and the analyzer 21 are inclined in the drawing is shown. (B) shows the color coding of the regions when the polarization directions of the polarizer 12 and the analyzer 21 are directed in the horizontal direction and the vertical direction in the figure. As described above, in the cases of (a) and (b), the color of each area is different. However, the characters and patterns on the base (not shown) are imaged by the CCD camera with the same brightness regardless of the polarization directions of the polarizer and the analyzer. That is, the polarizer 12 and the analyzer 21 are rotated around the optical axes AX1 and AX2 while maintaining the crossed Nicols relationship to obtain 2
When the image is picked up at two positions in the rotation direction, two different images as shown in FIGS. 10A and 10B are formed, and these include images of substantially the same background.

By subtracting the images as shown in FIGS. 10A and 10B to form a difference image between the two different images, FIG.
1 (a) is a character, picture, or pattern 3 of the background.
It becomes almost invisible as shown in FIG. Similarly, the flash 2 on the film surface is almost eliminated. Therefore,
The thickness unevenness of the transparent film and the folds of the package can be extracted and observed. Such a fold is a boundary of each area, and the area can be displayed as coordinates.
That is, each point can be classified into an area based on the color information of each point on the acquired image, and the shape parameter can be obtained. In this way, the material (paper,
Aluminum, etc.) and the problem that the fold state of the package cannot be observed depending on the state of the surface.

In the above description, two different images are used. However, three or more images may be used. If the number of images used is increased, finer observation is possible.

In the third example, the number of monochromatic light sources is one. However, a plurality of monochromatic light sources may be used, and a light source having a plurality of bright lines may be used. Accordingly, two or more regions can be extracted and observed.

Referring to FIG. 12, a description will be given of a color information acquiring apparatus according to a fourth example. In the above example, the polarizer is arranged in the illumination system and the analyzer is arranged in the imaging system. In addition to this, an element (birefringent plate) which provides an appropriate birefringence action is arranged in the illumination system and the imaging system. I do. However, it may be simply added to either the illumination system or the imaging system.

In the apparatus of this example, the film package 1 to be inspected and the analyzer 2 are utilized by utilizing the case of the first example.
1, a birefringent plate 51 is inserted. Here, when the illumination light source and the polarizer are one illumination system, and the analyzer and the imaging device are one imaging system, any part on the optical path of the illumination system and any part on the optical path of the imaging system , Typically, between the polarizer 12 and the inspection target 1, or between the inspection target 1 and the analyzer 21, the birefringent plate 51 is inserted and arranged. In any case, the birefringent plate 51 may be disposed between the light source 11 and the image sensor 22. Further, the birefringent plate 51 is configured to be rotatable around the optical axis by a rotation mechanism 52.

The operation of the device of the fourth example will be described. FIG.
As described in the principle with reference to, when birefringence is present, the amount of light detected in the orthogonal Nicol state changes, and the refractive index also has wavelength dependence. Therefore, changes in film thickness and the degree of film overlap and peeling,
The thickness unevenness of the sheet is observed as a difference in color.

Here, as shown in FIG. 12, the case where the birefringent plate is not in the illumination system but in the imaging system will be described.
The light having the wavelength that has passed through the analyzer 21 enters the imaging system due to the birefringence wavelength dependency of the inspection object 1 and is colored as a difference in the intensity of the three primary colors of R, G, and B.

Here, by further inserting the birefringent plate 51 and rotating it appropriately using the rotating mechanism 52, the intensity ratio of the three primary colors can be freely controlled to some extent. In this way, it is possible to emphasize different colors or adjust the degree of coloring. RGB color information can be converted to HIS.

Therefore, the inspection of the film thickness, the peeling of the film, and the like, which could not be inspected by the conventional illumination system, can be observed with a certain degree of free color, and the polarizer 12 and the analyzer can be observed. With only 21, depending on the material of the film, even in an area that is desired to be identified and is difficult to distinguish due to a similar color condition, the color difference can be emphasized, so that the distance can be calculated as an enlarged one. This makes it easier to make a decision.

Referring to the front view of FIG. 13, a description will be given of a fifth example of a color information acquiring apparatus which can be used in the embodiment of the present invention. In the above apparatus example, one light source is used for the illumination system, but in this example, two light sources are used. (B)
The configuration of (c) is a modified example of the device of (a).

In FIG. 9A, a white light source 11 a is disposed obliquely above the surface of the film package 1, and a polarizer 12 a as an illumination-side polarization filter is provided between the white light source 11 a and the surface of the film package 1. Are located. The optical axis AX1 of the polarizer 12a obliquely intersects the surface of the film package 1.

An analyzer 21a, which is a light-receiving-side polarization filter, is arranged in a direction in which light from the light source 11a via the polarizer 12a is regularly reflected on the surface of the film package 1. The optical axis AX2 of the analyzer 21a is in the regular reflection direction of the optical axis AX1 with respect to the surface of the film package 1. Analyzer 2
A color CCD camera 22a as an image sensor is arranged at a position where regular reflection light from the light source 11a is received via 1a. Thus, the light source 11a, the polarizer 12
a, film package 1, analyzer 21a, color CC
The D camera 22a is arranged in the same relationship as in FIG.

In this example, a beam splitter 26a is further provided in the optical axis AX1 between the polarizer 12a and the film package 1, and a beam splitter 26b is provided in the optical axis AX2 between the film package 1 and the analyzer 21a. Each is inserted and arranged.

Further, as a result of being reflected by the beam splitter 26a, the white light source 11b and the polarizer 12 move in the direction in which light is incident on the surface of the film package 1 so as to coincide with the optical axis AX1.
b is arranged. The polarizer 12b is disposed between the white light source 11b and the beam splitter 26a.

The optical axis A from the film package 1
The analyzer 21b and the color CCD camera 22b are arranged in this order in a direction in which light along X2 is reflected by the beam splitter 26b.

The color CCD cameras 22a and 22b have:
The image processing device 41 is connected.

Here, the polarizer 12a and the analyzer 21a are arranged such that their respective polarization directions have a crossed Nicols relationship, as described with reference to FIG.
Similarly, the polarizer 12b and the analyzer 21b are arranged such that their polarization directions are in a crossed Nicols relationship. However, they are arranged so that the polarization direction of the polarizer 12a and the polarization direction of the polarizer 12b are different. In particular, it is preferable to dispose it by 90 degrees.

In the configuration described above, the light source 1
1a is turned on, the light source 11b is turned off, light is detected by the camera 22a, and an image is formed on the image processing device 41. Next, the light source 11a is turned off, the light source 11b is turned on, light is detected by the camera 22b, and an image is formed on the image processing device 41. In this way, two different images are formed in the image processing device 41. These images are processed as described above.

The lighting time of the light sources 11a and 11b may be a time during which an image can be formed, and lighting and turning off can be performed in a very short time. In particular, compared to a case where two images are formed by rotating the polarizer and the analyzer mechanically, two different images can be formed in a short time almost at the same time.

In the configuration shown in (b), a polarization beam splitter 27b is inserted and arranged instead of the beam splitter 26b in the configuration of (a). In addition, the analyzer 21a,
21b is unnecessary and has been removed.

The polarization direction of the transmitted light of the polarization beam splitter 27b is set to be orthogonal to the polarization direction of the polarizer 12a, and the polarization direction of the reflected light of the polarization beam splitter 27b is the polarization direction of the polarizer 12b. And orthogonal Nicols. Also at this time, two different images can be formed almost simultaneously as in (a).

In the configuration shown in (c), a polarization beam splitter 27a is inserted and arranged instead of the beam splitter 26a of the configuration shown in (b). The polarization direction of the transmitted light of the polarization beam splitter 27b is determined by the polarization beam splitter 2
The polarization direction of the reflected light from the polarization beam splitter 27a is set to a relationship between the polarization direction of the reflected light of the polarization beam splitter 27a and the orthogonal Nicols. . Again,
As in (a) and (b), two different images can be formed almost simultaneously.

As described above, the polarization directions are arranged in a crossed Nicols relation by the polarizer 12a, the analyzer 21a, and the polarizer 1a.
2b and the analyzer 21b, but conversely, as the polarizer 12a and the analyzer 21b, and as the polarizer 12b and the analyzer 21a, the camera 22b forms an image when the light source 11a is turned on, and the camera 22a when the light source 11b is turned on. May be used to form an image.

The same applies to (b) and (c), where the light reflected by the polarizer 12a and the light reflected by the polarization beam splitter 27b are respectively expressed by
Further, the light transmitted from the polarizing beam splitter 27a and the reflected light from the polarizing beam splitter 27b are placed in a relationship of orthogonal Nicols, and the camera 22 is turned on when the light source 11a is turned on.
b when the image is formed and the light source 11b is turned on.
a may be used to form an image.

Referring to the partial sectional front view of FIG.
An example of the color information acquisition device will be described. A color C having an optical axis AX in a direction normal to the surface of the film package 1
A CD camera 22 is provided, and an analyzer 21 is inserted between the camera 22 and the film package 1. Further, an annular light source 11 c having a rotationally symmetric shape with respect to the optical axis AX, and an annular polarizer 12 c disposed between the annular light source 11 c and the film package 1 are provided.

A rotation mechanism 31a and a rotation mechanism 31b for rotating the polarizer 12c and the analyzer 21 around the optical axis AX are respectively connected to the polarizer 12c and the analyzer 21.
The rotation mechanism 31b and the controller 33 can be rotated by the controller 33 while maintaining a predetermined relationship between the polarization directions of the polarizer 12c and the analyzer 21. The camera 22 is connected to the controller 33. In the camera 22, the polarization directions of the polarizer 12c and the analyzer 21 are directed to a plurality of (for example, two) specific directions in response to a photographing instruction signal from the controller 33. A plurality of different images at that time can be captured.

In the color information acquisition devices of FIGS. 13 and 14, the birefringent plate 51 may be inserted as in the case of FIG. In particular, it is preferable to arrange between the film package 1 and the beam splitters 26a and 26b or the polarization beam splitters 27a and 27b.

In the above-described color information acquisition device, the irradiation device for irradiating light polarized in a predetermined polarization direction has been described using a combination of a light source and a polarizer, but is not limited to this. For example,
The light source may be a polarizing film attached to the light source.
In this case, rotating the polarized light around the optical axis does not rotate the polarizer but rotates the light source itself.

As described above, the folds and the like can be observed by extracting a transparent film or a film package portion by eliminating the influence of the underlayer, so that the quality of the film or the film package can be reliably determined. It becomes possible.

The inspection of the film package includes, for example, the inspection of the adhesive surface of the sealing in packaging of confectionery and the like. For example, it is possible to inspect the leakage of the contents from the adhesive surface. .

As described above, the test object having birefringence is described as a transparent film or a film package wrapped with a transparent film. However, the present invention is not limited to this, and a birefringent thin film is formed on the surface. An optical disk substrate such as a CD, a transparent film to be attached to a liquid crystal display device, and a liquid crystal display device to which the transparent film is attached can also be used.

[0112]

As described above, according to the present invention, an irradiation apparatus for irradiating a test object having birefringence with light polarized in a predetermined polarization direction, and detecting light from the test object to perform inspection. Since an image sensor for acquiring color information of each of a plurality of positions of the object and an analyzer arranged between the inspection object and the image sensor are provided, color information can be obtained from the inspection object. A classification mechanism for classifying the plurality of locations into a plurality of regions based on the distance from the reference information of each piece of color information obtained in the step, so that the plurality of locations can be clearly classified into each region. It is possible to provide an inspection device for an inspection object having refraction.

[Brief description of the drawings]

FIG. 1 is a plan view showing a stretching direction of a transparent film, a stretching direction of a package surface by the transparent film, and an overlapping state.

FIG. 2 is a conceptual perspective view showing a first example of a color information acquisition device that can be used in the embodiment of the present invention.

FIG. 3 is a diagram illustrating a wavelength range of white light.

FIG. 4 is a plan view showing an example of an image of the surface of a film package captured by a color CCD camera using the apparatus of FIG. 2;

FIG. 5A is a partial plan view showing a reference value of color information of two areas on an inspection object and color information of an arbitrary point, and a three-dimensional coordinate system showing a concept of a distance between color information. FIG.

FIG. 6 is a flowchart illustrating a method for acquiring reference information and reference parameters from a non-defective package.

FIG. 7 is a flowchart illustrating a method for determining the quality of an inspection object.

FIG. 8 is a block diagram illustrating a configuration example of an image processing apparatus.

FIG. 9 is a conceptual perspective view showing a second example (a) and a third example (b) of a color information acquisition device that can be used in the embodiment of the present invention.

FIG. 10 is a plane showing an example of an image obtained by rotating a polarizer and an analyzer and using a color CCD camera to image the surface of a film package at two different rotation positions using the color information acquisition apparatus of the first example. FIG.

11 is a plan view showing a state in which a difference image between two different images shown in FIG. 10 has been obtained, and characters on the base and surface luminosity have been removed.

FIG. 12 is a conceptual perspective view of a color information acquisition device that is a fourth example.

FIG. 13 is a conceptual perspective view of a color information acquisition device that is a fifth example.

FIG. 14 is a conceptual partial cross-sectional front view of a color information acquiring apparatus according to a sixth example.

FIG. 15 is a diagram illustrating the principle of acquiring color information.

FIG. 16 is a conceptual perspective view illustrating a conventional inspection method.

[Explanation of symbols]

 Reference Signs List 1 film package 11 white light source 12 polarizer 13 monochromatic light source 21 analyzer 22 CCD camera 23 color filter 24 CCD camera 26a, 26b beam splitter 27a, 27b polarization beam splitter 31, 32 rotation mechanism 33 controller 41 image processing device 41A control unit 41B Storage unit 41a Image processing module 41b HIS conversion module 41c Distance calculation module 41d Parameter calculation module 41e Comparison module 41f Reference information storage unit 41g Reference parameter storage unit 51 Birefringent plate AX, AX1, AX2 Optical axis

Claims (5)

[Claims] An irradiation device for irradiating a test object having birefringence with light polarized in a predetermined polarization direction; detecting light from the test object and detecting a color of each of a plurality of portions of the test object. An image sensor for acquiring information; an analyzer disposed between the object to be inspected and the image sensor; and the plurality of color information acquired by the image sensor based on distances from reference information. A classification mechanism for classifying the location into a plurality of regions; a device for inspecting an inspection object having birefringence. 2. The birefringent device according to claim 1, further comprising: a calculating mechanism that expresses the shape of the region by coordinates based on the coordinates of the classified plurality of locations and calculates a parameter that specifies the shape. Inspection device for inspection object. 3. An HIS conversion mechanism for converting the color information of the plurality of locations into hue, intensity, and saturation, respectively; wherein the classification mechanism calculates an average value of hue for each of a plurality of predetermined regions of the inspection object. Standard deviation, the average value and standard deviation of the intensity, has a reference information storage unit to determine and save the average value and the standard deviation of the saturation, hue, intensity, saturation of the plurality of locations of the inspection object, Based on the corresponding hue, intensity, saturation average value and standard deviation stored in the reference information storage unit, the distance is obtained, and the plurality of locations are configured to be classified into an area that gives the shortest distance. The inspection device for an inspection object having birefringence according to claim 1 or 2. 4. An irradiation step of irradiating a test object having birefringence with light polarized in a predetermined polarization direction; detecting light from the test object and detecting a color of each of a plurality of portions of the test object. An imaging step of acquiring information; an analysis step of analyzing light from the inspection object at the time of the imaging step; based on a distance of each color information acquired in the imaging step from reference information. And a classifying step of classifying the plurality of places into a plurality of regions; 5. A first irradiation step of irradiating a non-defective inspection object having birefringence with light polarized in a predetermined polarization direction; and detecting light from the non-defective inspection object to detect the non-defective item. A first imaging step of acquiring color information for each of a plurality of locations in each of a plurality of predetermined areas of the inspection target; and transmitting light from the non-defective inspection target to the first imaging step. A first light detection step for detecting light; a reference information obtaining step for obtaining an average value and a standard deviation of each color information obtained in the first image pickup step for each predetermined area; A second irradiation step of irradiating the inspection object to be determined having refractivity with light polarized in a predetermined polarization direction; and detecting the light from the inspection object to be determined and the inspection object to be determined. A second imaging step of acquiring color information of each of a plurality of locations; A second light detection step of detecting light from the inspection object to be performed at the time of the second image pickup step; and the respective predetermined regions of the respective pieces of color information acquired in the second image pickup step. A classifying step of classifying the plurality of places into the predetermined area based on a distance from each piece of reference information;