JP5258349B2 - Defect detection apparatus and method - Google Patents

Description

  The present invention relates to a defect detection apparatus and method for detecting a signal of a defective portion of a film as a defect signal based on a signal obtained by imaging a film.

  An optical compensation film (hereinafter referred to as “retardation film”) capable of improving the viewing angle of a liquid crystal display device by forming a liquid crystal layer having optical anisotropy on a transparent film is known. This retardation film is manufactured through a step of forming an alignment film on a long transparent film and a step of forming a liquid crystal layer by applying a liquid crystal thereon and drying (see, for example, Patent Document 1). Each manufacturing process is under strict quality control, but it completely eliminates defects such as molecular orientation unevenness due to foreign substance contamination and adhesion, transparent film thickness unevenness, and liquid crystal layer coating unevenness. It is not easy.

Until now, in order to detect such defects, the film to be inspected is irradiated with light from the projector, the light from the inspection object is received by the light receiver, and the light signal received by the light receiver is received. By analyzing, the position, size, and strength of the defect were grasped online. Furthermore, the first polarizing plate is disposed between the projector and the inspection target, and the second polarizing plate is disposed between the inspection target and the light receiver so that the polarization directions of the respective polarizing plates are orthogonal (crossed Nicols). Thus, an inspection method is known in which only light scattered and diffused by a defect on the inspection object enters the light receiver (see Patent Document 2). In addition, the image pickup signal obtained by the light receiver is subjected to differential processing to make the slope of the signal of the defective portion on the film (hereinafter referred to as “defect signal”) stand out, and the signal of the portion other than the defect (hereinafter referred to as “defect signal”) A method of making the slope of “noise signal” gentle is generally known.
JP-A-9-73081 Japanese Unexamined Patent Publication No. 6-148095

  As described above, in the defect inspection in which a pair of polarizing plates are arranged in crossed Nicols, the imaging signal in the width direction of the film detected by the light receiver is determined by the brightness of the central portion from the optical characteristics of the film and the sensitivity characteristics of the light receiver. And the brightness gradually decreases from the center to both ends. When differential processing is performed on such an image pickup signal, the defect signal is affected by the slope of the surrounding noise signal.

  For example, when there are multiple defects of the same degree on the film and the brightness of the light scattered and confused by each defect is the same, if differential processing is performed, the defect signal where the slope of the noise signal is negative is The inclination becomes smaller than the defect signal in the positive region (see FIG. 7B). Therefore, when the threshold value for defect detection is set to a constant luminance value Th (see FIG. 7B), the defect signal having a positive slope of the noise signal is detected because it is equal to or greater than Th. On the other hand, the defect signal in the region where the slope of the noise signal is negative is less than Th and is not detected. Therefore, there is a problem that the signal of the defective portion is not detected even though the scattered and confused light has the same luminance.

  An object of the present invention is to provide a defect detection apparatus and method that can detect a signal of a defective portion without omission.

In order to achieve the above object, the present invention is a defect detection apparatus for detecting a defect in a film in a state where the film is sandwiched between first and second polarizing plates arranged in crossed Nicols. Of the light that passes through the first polarizing plate, the first polarizing plate, the film, and the second polarizing plate, the removal optical system that removes light of 700 nm or more, and the light from the removal optical system Among the image processing means for receiving the light, the differential processing means for performing differential processing on the imaging signal obtained by the imaging means, and the differential processing signal obtained by the differential processing means, the luminance value is maximum within a predetermined pixel range. A signal detecting means for detecting a local maximum signal and a local minimum signal having a minimum luminance value within a predetermined pixel range, and a difference between the luminance value of the reference signal and the luminance value of the local maximum signal sequentially updated as a first difference As a value The difference value calculation means for obtaining the difference between the luminance value of the reference signal and the luminance value of the minimal signal as the second difference value, the addition means for obtaining the addition value obtained by adding the first difference value and the second difference value, and the addition value is constant When the value is equal to or greater than the value, a defect signal specifying means for specifying a signal within a predetermined pixel range as a defect signal is provided.

  The film is a retardation film, and the polarization directions of the first and second polarizing plates are preferably oriented in the direction of 45 ° with respect to the slow axis of the retardation film.

The present invention is directed to a defect detection method for detecting a film defect in a state in which a film is sandwiched between first and second polarizing plates arranged in crossed Nicols, and directs light having only one bright line to the first polarizing plate. The light that has been irradiated from the projector and passed through the first polarizing plate, the film, and the second polarizing plate is removed by the removal optical system, and the light from the removal optical system is received by the imaging means. A differential process is performed on the imaging signal obtained by the imaging means, and among the differential processing signals obtained by the differentiation process, a maximum signal having a maximum luminance value within a predetermined pixel range and a luminance within the predetermined pixel range The minimum signal having the minimum value is detected, and the difference between the luminance value of the reference signal and the luminance value of the maximum signal that are sequentially updated is obtained as the first difference value, and the luminance value of the reference signal and the luminance value of the minimum signal are calculated. Find the difference as the second difference value, It obtains a first difference value and the addition value obtained by adding the second difference value, if the sum value is equal to or greater than a predetermined value, and identifies a signal in a predetermined pixel range as a defect signal.

  According to the present invention, when a signal of a defective portion of a film is detected as a defect signal, a maximum signal having a maximum luminance value and a minimum signal having a minimum value within a predetermined pixel range are obtained, and the luminance value of the reference signal is determined. An addition value obtained by adding the second difference value obtained from the difference between the luminance value of the reference signal and the luminance value of the reference signal to the first difference value obtained from the difference between the luminance values of the maximum signal is obtained, and the added value is equal to or larger than a certain value. In this case, the defect signal can be reliably detected without omission by specifying the signal within the predetermined pixel range as the defect signal.

  As shown in FIG. 1, the retardation film production line 10 includes an alignment film forming device 11, a liquid crystal layer forming device 12, a defect detecting device 13, and a winding device 14.

  The alignment film forming apparatus 11 applies a coating solution containing an alignment film forming resin to the surface of the transparent resin film 15 fed from the film roll 18 and heat-drys it. Thereby, an alignment film forming resin layer is formed on the surface of the transparent resin film 15. The alignment film forming apparatus 11 performs a rubbing process on the alignment film forming resin layer to form an alignment film.

  The liquid crystal layer forming apparatus 12 applies a coating liquid containing a liquid crystal compound on the alignment film of the transparent resin film 15, and heat-drys after application to form a liquid crystal layer. Then, the liquid crystal layer is irradiated with ultraviolet rays to crosslink the liquid crystal layer. Thereby, a retardation film 16 (hereinafter referred to as “film”) in which an alignment film and a liquid crystal layer are formed on the transparent resin film 15 is manufactured. The film 16 is wound up by the winding device 14 after passing through the defect detection device 13 described in detail below. Here, let the conveyance direction of the film 16 until it leaves the film roll 18 and is wound up by the winding device 14 be an X direction. Also, the X direction coincides with the slow axis direction of the film 16.

  The defect detection device 13 detects defects generated on the film 16. Examples of defects include scratches, thickness unevenness, coating unevenness, and molecular orientation unevenness. The film to be inspected is not limited to the retardation film, and may be any member that transmits light, such as a transparent body or a translucent body, such as an antireflection film.

  The defect detection device 13 includes guide rollers 20 and 21, a halogen lamp 22, a light amount adjustment unit 23, a light receiver 24, first and second polarizing plates 25 and 26, a removal optical system 27, and a controller 28. The guide rollers 20 and 21 are arranged in the conveyance path of the film 16 so as to be spaced apart at a constant interval. These guide rollers 20 and 21 are rotatable, and rotate following the conveyance of the film 16. Further, the film 16 is held in a planar shape by being passed over the guide rollers 20 and 21. An encoder 30 is connected to the guide roller 21. The encoder 30 generates an encoder pulse signal every time the film 16 is conveyed for a certain length. The encoder pulse signal is transmitted to the controller 28 and used when specifying the defect position.

  The halogen lamp 22 is installed below the conveyance path of the film 16. As shown in FIG. 2, the halogen lamp 22 has only one bright line between wavelengths of 600 nm and 800 nm. Therefore, interference fringes hardly occur on the film 16. Even if an interference fringe is generated, the brightness of the bright part of the interference fringe where the light is intense is only slightly larger than the brightness of the light emitted from the halogen lamp 22. Therefore, interference fringes do not affect the accuracy of defect detection. Note that the projector need not be limited to the halogen lamp as long as the projector has no bright line or only one projector.

  The light amount adjusting unit 23 controls the halogen lamp 22 so that the light amount becomes constant based on a light amount detection signal detected by a sensor (not shown) installed in the vicinity of the halogen lamp 22. Thereby, since the film 16 can be irradiated with light having a uniform amount of light, defect detection can always be performed with the same sensitivity.

  The light receiver 24 is composed of a CCD camera, and is installed above the transport path of the film 16. The light receiver 24 includes a large number of imaging elements arranged in a line in the width direction of the film 16, and images the film 16 line by line in the width direction each time the film 16 is conveyed for a certain length. Signals obtained by imaging include signals of light scattered and confused due to defects on the film 16 (hereinafter referred to as “defect signals”), and signals such as light slightly transmitted from the first and second polarizing plates 25 and 26. (Hereinafter referred to as “noise signal”). These signals are transmitted to the controller 28. The number of light receivers is not limited to one and may be two or more.

  The first and second polarizing plates 25 and 26 are composed of iodine-based polarizing plates, and the first polarizing plate 25 is disposed between the halogen lamp 22 and the film 16 as shown in FIG. Is installed between the film 16 and the light receiver 24. The first and second polarizing plates 25 and 26 are arranged so that their polarization directions 25a and 26a are orthogonal (crossed Nicols). In addition, although the iodine type polarizing plate is used from the surface of performance and a price, you may use not only this but the polarizing plate which consists of a dye type polarizing plate, a metal polarizer, calcite.

  By disposing the first and second polarizing plates 25 and 26 in crossed Nicols, when the film 16 having no defect is positioned, the light polarized by the first polarizing plate 25 to the specific polarization plane is the other second polarized light. Since the light is blocked by the plate 26, the light hardly enters the light receiver 24. That is, the light receiver 24 is in a dark field state. On the other hand, when the film 16 having a defect is located, the light polarized on the specific polarization plane by the first polarizing plate 25 is scattered and confused by the defect, and the polarization plane changes. When the polarization plane changes in this way, light comes out from the other second polarizing plate 26. That is, the light receiver 24 is in a light receiving state.

  The polarization direction 25a of the first polarizing plate is set so that the angle θ with respect to the X direction (the slow axis direction of the film 16) is 45 °, and the polarization direction 26a of the second polarizing plate is an angle with respect to the X direction. (90 ° −θ) is set to be 45 °.

  FIG. 4 shows the ratio of the amount of incident light incident on the first polarizing plate 25 and the amount of transmitted light transmitted through the second polarizing plate 26 with respect to the angle θ (transmitted light amount ratio (%)). It is a graph which shows whether it changes. FIG. 5 is a graph showing how the transmitted light amount ratio changes with respect to the phase difference of the film 16, and “◯” indicates the transmitted light amount ratio when the angle θ is 0 °, and “□”. Indicates the transmitted light amount ratio when the angle θ is 15 °, “Δ” indicates the transmitted light amount ratio when the angle θ is 30 °, and “◇” indicates the transmitted light amount ratio when the angle θ is 45 °. .

  As shown in FIG. 4, when the angle θ is 45 °, the transmitted light amount ratio is the highest. Therefore, by setting the angle θ to 45 °, the amount of light scattered and confused by defects such as scratches increases, so that the S / N ratio can be improved. Further, as shown in FIG. 5, the phase difference dependency of the transmitted light amount ratio gradually increases as the angle θ increases from 0 ° to 45 °, and the phase difference dependency increases most when the angle θ is 45 °. Phase difference defects such as coating unevenness may not be detected because the amount of light scattered and confused by the phase difference defect is small, but by setting the angle θ to 45 °, the transmitted light amount ratio of the phase difference defect part Therefore, not only defects such as scratches but also phase difference defects can be reliably detected.

  The removal optical system 27 is composed of an infrared cut filter and is installed immediately before the light receiver 24. The removal optical system 27 removes light having a wavelength of 700 nm or more from the light from the second polarizing plate 26. As a result, only light having a wavelength of less than 700 nm is incident on the light receiver 24, so that the defect can be detected with high accuracy at a high S / N ratio for the following reason.

  FIG. 6 shows the state of the light of the first and second polarizing plates 25 and 26 when the angle θ of the polarization direction 25a of the first polarizing plate is 45 ° with the removal optical system 27 removed from the defect detection device 13. The first and second polarizing plates 25 and 26 substantially block light with a wavelength up to 700 nm, but transmit light with a high transmittance with respect to light with a wavelength exceeding 700 nm. On the other hand, the light receiver 24 is sensitive to light having a wavelength exceeding 700 nm. Therefore, light having a wavelength exceeding 700 nm is included in the noise signal of the light receiver 24.

  Therefore, light having a wavelength of 700 nm or more (light in the hatching area 35) is removed by the removal optical system 27 so that only light having a wavelength of less than 700 nm enters the light receiver 24. For this reason, light having a wavelength of 700 nm or more is not included in the noise signal of the light receiver 24. Thereby, since a noise signal can be suppressed to the minimum, a sufficiently large S / N ratio can be obtained. Note that the S / N ratio when a coating stripe that is a defect is detected is 1.5 when light having a wavelength of 700 nm or more is not removed, whereas it is 1.5 when light having a wavelength of 700 nm or more is removed. Will improve to 2.1.

  Although an infrared cut filter is used as the removal optical system 27, a band pass filter using a dielectric multilayer film, a monochromator, a wavelength cut filter, a colored glass filter, a diffraction grating, or the like may be used. Further, the wavelength range of the light removed by the removal optical system 27 is not necessarily limited to 700 nm or more, and may be appropriately changed according to the type of polarizing plate used for defect detection. Further, a removal optical system that removes light having a wavelength of 600 nm or more may be installed between the halogen lamp 22 and the first polarizing plate 25 to remove bright lines from the light from the halogen lamp 22.

  In addition to the position immediately before the light receiver 24 (hereinafter referred to as “installation position of the present invention”), the following six installation positions are conceivable for the removal optical system 27. The first installation position is between the halogen lamp 22 and the first polarizing plate 25, the second installation position is between the light receiver 24 and the second polarizing plate 26, and the third installation position is the second polarizing plate 26. Between the film 16, the fourth installation position is between the film 16 and the first polarizing plate 25, the fifth installation position is inside the halogen lamp 22, and the sixth installation position is integrated with the halogen lamp 22. is there.

  Which of the first to sixth installation positions is optimal will be examined below. Generally, since the polarizing plate is weak against the light and heat of the halogen lamp 22, the second to fourth installation positions are not preferable. In contrast, at the first installation position, the light from the halogen lamp 22 indirectly strikes the polarizing plate via the removal optical system 27, so that the performance of the polarizing plate is not deteriorated by light or heat. Therefore, the first installation position is preferable.

  Since the focus of the light receiver 24 is adjusted to the film, the distance between the removal optical system 27 and the film 16 should be as far as possible. Therefore, the third and fourth installation positions are not preferable. Further, the sixth installation position is not preferable because the defect detection accuracy changes every time the halogen lamp 22 is replaced. On the other hand, if the removal optical system 27 is disposed in the halogen lamp 22, the removal optical system 27 can be small, and hence the cost is low. Therefore, the fifth installation position is preferable.

  From the above, it is preferable to install the removal optical system 27 at any of the installation position, the first installation position, and the fifth installation position of the present invention, but the removal optical system 27 is installed at the installation position of the present invention. Sometimes the S / N ratio is highest. Therefore, it is most preferable to install the removal optical system 27 at the installation position of the present invention.

  The controller 28 specifies the position of the defect on the film 16 based on the defect signal detection unit 28 a that detects the defect signal based on the imaging signal obtained by the light receiver 24, and the defect signal and the encoder pulse signal from the encoder 30. And a defect position specifying unit 28a. In the defect signal detection unit 28a, first, differentiation processing is performed on the imaging signal 40 shown in FIG. As a result, as shown in FIG. 7B, a signal 42 (hereinafter referred to as “differential processing signal”) in which a portion with a large change in the luminance value is highlighted is obtained. Note that the signals in FIG. 7 and FIG. 8 shown below are examples, and the present invention is not limited to this.

  Next, as shown in FIG. 8, from the differential processing signal 42, maximum signals 50a, 51a, and 52a in which the luminance value is maximum within a certain pixel range, and the luminance value is minimum within the certain pixel range. The minimum signals 50b, 51b, and 52b are specified. Then, differences LA1, LB1, and LC1 (hereinafter referred to as “first difference values”) between the preset luminance value of the reference signal 55 and the luminance values of the local maximum signals 50a, 51a, and 52a are obtained, and the luminance value of the reference signal 45 is also obtained. And differences LA2, LB2, and LC2 (hereinafter referred to as “second difference values”) between the luminance values of the minimum signals 50b, 51b, and 52b. Here, the reference signal 55 is obtained by averaging the noise signals detected by the light receiver 24 several seconds ago, and is sequentially updated.

  Next, addition values LA, LB, and LC obtained by adding the first difference values LA1, LB1, and LC1 and the second difference values LA2, LB2, and LC2 are obtained. Then, it is determined whether or not the added values LA, LB, and LC are equal to or greater than a certain value. If the addition values LA, LB, and LC are greater than or equal to a certain value as a result of the determination, it is determined that the signals 50, 51, and 52 within the pixel range for which the addition value has been obtained are defective signals. In FIG. 8, since the addition values LA, LB, and LC are all the same value, the signals 50, 51, and 52 within the pixel range for which the addition values LA, LB, and LC are obtained are specified as defect signals.

  Conventionally, as shown in FIG. 7B, when a certain threshold value Th is set for the differential processing signal 42 and a signal exceeding the threshold value Th is determined as a defect signal, a change in luminance occurs. Even a large signal was not determined as a defect signal when the signal was equal to or less than the threshold Th. On the other hand, in the present invention, instead of the threshold value Th, as described above, the defect signal is detected based on the added value indicating the amount of change in the luminance value, so that the luminance change even when the threshold value Th is not more than the threshold value Th. If the signal is large, it can be reliably detected as a defect signal without omission.

  As described above, by setting the angle θ of the polarization direction of the first polarizing plate to 45 ° with respect to the slow axis of the film 16, the transmitted light amount ratio increases according to the phase difference of the film 16. Is possible. However, since the film 16 has a phase difference distribution in which the phase difference is high in the central portion in the width direction and gradually decreases from the central portion to the end portion, the defect determination is performed with the constant threshold Th. If performed, defect detection may not be performed reliably. On the other hand, in the present invention, defect detection can be performed reliably regardless of the phase difference distribution of the film 16 by performing defect detection based on the amount of change in luminance value instead of the constant threshold Th. it can.

  Next, the operation of the defect detection apparatus of the present invention will be described. A film 16 manufactured by the alignment film forming device 11 and the liquid crystal layer forming device 12 is fed into the defect detection device 13. The film 16 travels between the first polarizing plate 25 and the second polarizing plate 26. Light is emitted from the halogen lamp 22 to the first and second polarizing plates 25 and 26.

  Light from the halogen lamp 22 is polarized to a specific polarization plane by the first polarizing plate 25. Here, when there is no defect on the film 16, the polarized light is blocked by the second polarizing plate 26. On the other hand, when there is a defect on the film 16, the polarized light is scattered and confused by the defect. The scattered / confused light comes out of the second polarizing plate 26. The light emitted from the second polarizing plate 26 is removed by the removal optical system 27 with a wavelength of 700 nm or more. The light that has passed through the removal optical system 27 is received by the light receiver 24.

  The light receiver 24 performs imaging for one line every time the film 16 is fed for a certain length. In the defect signal detection unit 28a, differential processing is performed on the imaging signal obtained by the light receiver 24 to generate a differential processing signal. Then, a maximum signal and a minimum signal are detected from the differential processing signal. Then, the difference value between the luminance value of the reference signal and the luminance value of the maximum signal is obtained as the first difference value, and the difference value between the luminance value of the reference signal and the luminance value of the minimum signal is obtained as the second difference value. And when the addition value which added the 1st difference value and the 2nd difference value is more than a fixed value, the signal in the pixel range which calculated | required the addition value is specified as a defect signal. And the defect position specific | specification part 28b specifies the defect position on the film 16 based on a defect signal and an encoder pulse signal. The result of the controller 28 is displayed on a display (not shown).

It is the schematic of a phase difference film manufacturing line. It is a graph which shows the spectral radiant intensity of a halogen lamp. It is a perspective view of a defect detection apparatus. It is a graph which shows the relationship between angle (theta) and transmitted light amount ratio. It is a graph which shows the relationship between the phase difference of a film, and transmitted light amount ratio. It is a graph which shows the spectral transmission factor of a pair of polarizing plate arrange | positioned at cross Nicol. (A) is a graph which shows an imaging signal, (B) is a graph which shows a differentiation process signal. It is explanatory drawing explaining the method of calculating | requiring the addition value which shows the change of a luminance value.

Explanation of symbols

DESCRIPTION OF SYMBOLS 13 Defect detection apparatus 16 Film 24 Light receiver 28 Controller 28a Defect signal detection part 28b Defect position specific | specification part 40 Imaging signal 42 Differentiated signal 50a, 51a, 52a Maximum signal 50b, 51b, 52b Minimum signal 55 Reference signal LA1, LB1, LC1 1st difference value LA2, LB2, LC2 2nd difference value LA, LB, LC addition value

Claims (3)

In the defect detection device for detecting defects of the film in a state where the film is sandwiched between the first and second polarizing plates arranged in crossed Nicols,
A projector for irradiating light having only one emission line toward the first polarizing plate;
A removal optical system that removes light of 700 nm or more out of light that has passed through the first polarizing plate, the film, and the second polarizing plate;
Imaging means for receiving light from the removal optical system;
Differential processing means for performing differential processing on the imaging signal obtained by the imaging means;
Signal detection for detecting a maximum signal having a maximum luminance value within a predetermined pixel range and a minimum signal having a minimum luminance value within the predetermined pixel range among the differential processing signals obtained by the differential processing means. Means,
The difference between the luminance value of the reference signal and the luminance value of the maximum signal that are sequentially updated is determined as a first difference value, and the difference between the luminance value of the reference signal and the luminance value of the minimum signal is determined as a second difference value. A value calculating means;
Adding means for obtaining an added value obtained by adding the first difference value and the second difference value;
A defect detection apparatus comprising defect signal specifying means for specifying a signal within the predetermined pixel range as a defect signal when the added value is equal to or greater than a certain value.   The said film is a phase difference film, The polarization direction of the said 1st and 2nd polarizing plate is orient | assigned to the direction of 45 degrees with respect to the slow axis of the said phase difference film. The defect detection apparatus described. In a defect detection method for detecting defects in the film, with the film sandwiched between first and second polarizing plates arranged in crossed Nicols,
Irradiating light having only one bright line from the projector toward the first polarizing plate,
Of the light that has passed through the first polarizing plate, the film, and the second polarizing plate, light having a wavelength of 700 nm or more is removed by a removal optical system,
The image pickup means receives light from the removal optical system,
A differential process is performed on the imaging signal obtained by the imaging means,
Among the differential processing signals obtained by the differential processing, a maximum signal having a maximum luminance value within a predetermined pixel range and a minimum signal having a minimum luminance value within the predetermined pixel range are detected,
The difference between the luminance value of the reference signal and the luminance value of the maximum signal that are sequentially updated is determined as a first difference value, and the difference between the luminance value of the reference signal and the luminance value of the minimum signal is determined as a second difference value.
Obtain an added value obtained by adding the first difference value and the second difference value;
A defect detection method characterized by identifying a signal within the predetermined pixel range as a defect signal when the addition value is a certain value or more.

Images