JP2012107893A - Optical characteristics measuring device and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To promptly and accurately measure the optical characteristics of an optical film of a wide area.SOLUTION: An optical film 12 which is a measuring object is placed on a surface illuminating portion 14. The surface illuminating portion 14 projects circular polarization toward the optical film 12. When the optical film 12 moves in an X-direction, measurement pixels E on the optical film 12 successively pass through a first to a forth imaging areas of a CCD camera in an imaging portion 15. The CCD camera images the measurement pixels E several times when the measurement pixels E pass through each imaging area. A value calculated by adding output values obtained by several imaging is made to be a measurement value in the imaging area. Stokes parameters of the measurement pixels E are calculated by the measurement values in the four imaging areas. The stokes parameters are obtained about all measurement pixels E on the optical film 12.

Description

  The present invention relates to an optical property measuring apparatus and method for measuring the polarization property of an optical film.

  In the liquid crystal display device, a functional plastic resin film (hereinafter referred to as “optical film”) having various optical characteristics such as a polarizing plate, a viewing angle correction film, and an antireflection film is used. Since the liquid crystal display device obtains contrast using the birefringence characteristic of the liquid crystal, the optical film to be used must have a predetermined birefringence characteristic. If the birefringence characteristic of the optical film is not uniform over the entire surface, unevenness in image display in the liquid crystal display device occurs.

  Therefore, before the optical film is incorporated into the liquid crystal display device, it is necessary to measure whether the film has a desired birefringence characteristic. Birefringence characteristics are measured using a light source that emits measurement light to the optical film to be measured, a light receiver that receives light emitted from the optical film, and a phase difference for measuring the polarization characteristics of the optical film. This is performed using various optical members such as a plate and a polarizing plate.

  For example, in Patent Document 1, various polarization states are created by rotating a retardation plate between a light source and a CCD camera as a light receiver around an optical axis. Then, images of different polarization states are picked up by a CCD camera, and birefringence characteristics are calculated for each pixel from the luminance value change of each pixel of the image group obtained by the image pickup. Patent Document 2 discloses a method for measuring the birefringence characteristics of an optical film being conveyed in a predetermined direction online. In Patent Document 3, when performing measurement while moving a film, imaging is repeated in consideration of the field of view size of the CCD camera and the moving speed of the optical film, so that the birefringence distribution of a wide area optical film is obtained. An apparatus for measuring is disclosed.

JP 2009-229279 A JP-A-5-346397 JP 2007-263593 A

  In recent years, since liquid crystal display devices have become larger, an optical film having a large area has been used as an optical film incorporated therein. Accordingly, there is a need for an apparatus and method that can measure the birefringence characteristics of an optical film having a large area. For example, an inspection of an optical film having a size of about A3 is necessary for a liquid crystal display device of about 20 inches.

  In this regard, the conventional techniques as shown in Patent Documents 1 to 3 have a problem that the birefringence characteristics of a wide area optical film cannot be measured quickly and accurately for the following reasons. In the case of Patent Document 1, it is desirable to use a telecentric lens as the imaging lens. However, since the field of view of the lens is at most about 5 cm on a side, the A3 size cannot be inspected with one field of view.

  Accordingly, it is necessary to divide the optical film into a plurality of measurement areas in accordance with the field of view of the CCD camera, and to perform the entire inspection by performing birefringence measurement for each measurement area. At this time, in each measurement area, in order to measure the polarization state of the optical film, it is necessary to take an image while rotating the angle of the phase difference plate while the CCD camera is stationary. Therefore, imaging (stationary) in a given measurement area → moving the CCD camera to another measurement area → imaging (stationary) in another measurement area → ... repeats the measurement, so it takes time without progressing. There was a problem.

  Moreover, in patent document 2, one point on an optical film is measured along a conveyance direction. Since there is no special retardation plate rotation, measurement is possible without stopping the camera or film. Therefore, in order to extend the single-point measurement to the surface measurement, it is conceivable to arrange this measuring device in the width direction of the optical film. However, if the measurement spatial resolution of the measurement apparatus disclosed in Patent Document 2 is 1 mm square, for example, a total of 294 apparatuses are required to arrange this measurement apparatus in the width direction of the A3 size optical film. Therefore, it can be said that there is no feasibility. The term “measurement spatial resolution” used here means the size of one measurement point on the measurement object, and when the measurement result distribution is finally imaged, it becomes the pixel size of the image.

  Moreover, since the CCD camera used in Patent Document 3 is provided with a polarizer necessary for measuring the polarization state of the optical film in each light receiving element, a retardation plate is provided for each imaging as in Patent Document 1. There is no need to rotate. However, in this patent document 3, since the noise problem which a CCD camera has cannot be solved, a highly accurate measurement cannot be performed.

  In Patent Document 3, one visual field size is measured based on one image obtained by one imaging with a CCD camera. In a CCD camera, even when continuous imaging is performed under the same conditions, the output value obtained each time varies due to variations in luminance due to CCD noise. That is, in Patent Document 3, it cannot be said that measurement reproduction is insufficient.

When measuring polarization characteristics of an optical film having an area larger than the imaging field of view using a two-dimensional image sensor, the present invention requires the image sensor to be stationary on the optical film for imaging in a plurality of polarization states. In addition, the image sensor can be taken multiple times without stopping the image sensor in order to improve measurement accuracy. Therefore, it is an object of the present invention to provide a polarization characteristic measuring apparatus and method that have a rapid and high measurement accuracy.

  The optical characteristic measuring apparatus according to the present invention includes at least four types of polarization characteristics in the first direction by arranging a light projecting unit that irradiates a measurement object with specific polarized illumination and a wave plate aligned in the first direction. And an imaging means including an image sensor partitioned into an imaging area for individually imaging light transmitted through the wave plate, and the imaging means relative to the measurement target in the first direction. A scanning means for moving, and with the relative movement of the imaging means by the scanning means, an output value obtained by imaging each measurement pixel to be measured a plurality of times in each imaging area is added for each measurement pixel and imaging is performed. In addition to creating a measurement value in the area, a Stokes parameter of light for each measurement pixel is calculated from the measurement values collected from each imaging area. It is preferable that the image sensor captures an image with an image sensor including a large number of pixels, and outputs one output value for each combined cell obtained by combining a plurality of adjacent pixels. Before the measurement, a polarization transfer matrix representing the relationship between the light emitted from the measurement object and the output value output from each imaging area of the image sensor is obtained for each coupled cell, and the polarization transfer matrix and each imaging It is preferable to obtain the Stokes parameter to be measured using the output value output from the area.

  As shown in FIG. 1, the measurement spatial resolution is determined as a condition for the measurement of the optical film 12 as a measurement target. Therefore, the final measurement result is a microscopic image obtained by subdividing the optical film 12 by the size of the measurement spatial resolution. It is a set of measurement results for each area, that is, surface distribution information of optical characteristics. On the optical film 12 to be measured, it is assumed that there are virtually subdivided areas with the measurement spatial resolution, and each of these areas is collectively referred to as “measurement pixel E” hereinafter.

  The image sensor preferably outputs an output value obtained by imaging in units of combined cells. The “combined cell” referred to here is a large cell (combined cell) in which a predetermined number of imaging cells of adjacent image sensors are vertically and horizontally arranged, and the output value of the cell is all the imaging cells included in the combined cell. The output value is averaged. Each of the combined cells is collectively referred to as “combined cell CP” hereinafter.

  The reason why imaging is performed in units of coupled cells CP in this way is as follows. The graph of FIG. 2 shows variations in output values when a 12-bit output CCD camera is used as an image sensor. In this graph, relatively bright light (light whose output value is near 3740) is input to a combined cell in which a predetermined number of pixels are combined, and the measurement is simply performed 256 times, from the maximum value of all output values. This is a plot of the variation width minus the minimum value. In this graph, the vertical axis represents the variation in the output value of the CCD camera, and the horizontal axis represents the number of combined cells indicating the number of combined pixels. Here, for example, a combined cell with 4 combined cells has 2 vertical pixels and 2 horizontal pixels. In FIG. 2, the black circles represent the variation width (measurement result) of the output values of the CCD camera, and the dotted line represents the approximate curve obtained from the black circles. From this result, it can be seen that the noise on the output of the CCD camera is proportional to the minus half power of the number of combined cells and has the property of random noise.

  As shown in this graph, as the cell size (number of combined cells) is increased to 1 cell, 4 cells, 9 cells, 16 cells,. Therefore, unless the number of coupled cells is increased to some extent, the output value of the CCD camera varies greatly, so that accurate measurement cannot be performed unless the number of captured images is increased and averaging is performed. The CCD used to obtain this graph is 1 / 1.8 inches, 2 million pixels, and the imaging cell size is 4.4 μm.

The number of pixels constituting the combined cell CP is preferably a number of N ² (N is a natural number) starting from 1, such as 1, 4, 9,..., And the maximum value is formed on the image sensor. It is assumed that the measurement spatial resolution of the measurement target to be measured (that is, the size of the measurement pixel E) is the number that becomes the size of the combined cell.

  FIG. 3 shows the imaging unit 15. Here, an example using a CCD as an image sensor is shown, but a CMOS may be used as the image sensor. In this example, four wave plates are used to realize one kind of polarization characteristic per wave plate. The number of wave plates is not limited to four as will be described later. The structure of the imaging unit 15 includes a camera case 40, a CCD camera 41, a telecentric lens 42, a CCD camera rotation mechanism 43, first to fourth wavelength plates 45 to 48, and a polarizing plate 49. The camera case 40 has a substantially rectangular parallelepiped shape, and one opening 40a (see FIG. 4) for attaching the first to fourth wave plates 45 to 48 and the polarizing plate 49 is formed. In this camera case 40, a CCD camera 41, a telecentric lens 42, and a CCD camera rotation mechanism 43 are provided. The CCD camera rotation mechanism 43 is provided for the purpose of adjusting the one direction of the two-dimensional array of coupled cells to coincide with the scanning direction of the scanning means. Note that an arrow X in FIG. 3 indicates the relative movement direction of the measurement target (either the measurement target or the imaging unit may move).

  As the telecentric lens 42, a double-sided telecentric lens or an object-side telecentric lens is used. The telecentric lens 42 forms an image of the measurement object on the CCD with a size obtained by multiplying the lens magnification. (A magnification of 1 to 1/3 is used.) The light transmitted through each wave plate is mixed because of the deep focal depth of the telecentric lens and the ability to capture the light beam parallel to the optical axis. It reaches the image sensor without matching and forms an individual area corresponding to each wave plate. Hereinafter, each area on the CCD substantially divided by the wave plate that is formed by the light beam that has passed through each wave plate is collectively referred to as an “imaging area”.

How one measurement pixel E on the optical film 12 is measured by the imaging unit 15 through the first to fourth wavelength plates 45 to 48 will be described with reference to FIG.
In FIG. 4, the reduction or enlargement effect in the telecentric lens 42 and the inverted imaging effect are not included in the figure for the sake of easy understanding, and one point on the optical film 12 is the same magnification on the CCD 55. It is drawn to form a vertical image.

  First, when a certain measurement pixel E on the optical film 12 enters the field of view of the imaging unit 15 by the scanning means, the light emitted from the measurement pixel E first enters the section of the first wavelength plate 45 for a while. Cross this section. The first wave plate 45 is imaged in the imaging area 50 corresponding to the CCD 55 by the action of the telecentric lens 42. Therefore, while the measurement pixel E crosses the section of the first wave plate 45, the image of the measurement pixel E is the imaging area. Crossing 50 sections, imaging is performed a plurality of times in accordance with the movement of the measurement pixel E by a certain distance. Similarly, the second to fourth wavelength plates 46 to 48 are imaged in the imaging areas 51 to 53, and the measurement pixel E is also imaged a plurality of times.

  A polarization transfer matrix representing the relationship between the Stokes parameter of light incident on the imaging means and the output value output from the image sensor is obtained in advance for each CCD combined cell CR.

  The CCD output value for the measurement pixel E obtained by imaging is the matrix product of the Stokes parameter of the light emitted from the measurement pixel E and the polarization transfer matrix of the combined cell that has imaged it.

  The output values of the CCD relating to the measurement pixel E obtained by imaging are added for each imaging area, and are taken out as measurement values in the imaging area. At the same time, the matrix sum of the polarization transfer matrix of the combined cell CR used when imaging the measurement pixel E is also performed for each imaging area. Thus, the relational expression of the measurement value E, the Stokes parameter of light, and the polarization transfer matrix for the measurement pixel E is obtained by the number of imaging areas (or the number of types of polarization states embodied). Since there are four types of measurement values, the number of imaging areas, Stokes parameters can be calculated from this. By performing this process for each measurement pixel E, the surface distribution information of the Stokes parameter to be measured is calculated.

  In the image sensor, it is preferable that a predetermined measurement pixel is imaged across two adjacent combined cells. The output value of the predetermined measurement pixel is obtained by multiplying the ratio of the pixel that has captured the predetermined measurement pixel in one combined cell by the output average value of the pixel, and the predetermined measurement pixel in the other combined cell. It is preferable that the ratio of a pixel occupied by imaging is multiplied by an average output value of the pixel.

  The moving unit moves at least one of the measurement target and the imaging unit in the first direction, and each time imaging of the measurement target in the first direction by the imaging unit is completed, the measurement unit or the imaging unit It is preferable that at least one of them is moved in a second direction perpendicular to the first direction and parallel to the measurement object.

  When the moving means moves at least one of the measuring object and the imaging means in a second direction perpendicular to the first direction and parallel to the measuring object, the second means of the imaging means and the light projecting means It is preferable to move so that the positional relationship in the direction does not change so that the polarized light irradiation width of the light projecting means is narrow enough to irradiate the visual field width in the second direction of the imaging means.

  The number of types of polarized light embodied by the wave plate is preferably 4 to 40. The wave plate preferably has a retardation effect similar to that of a wave plate having a retardation amount of 70 ° to 170 ° or 190 ° to 290 °.

  In the optical characteristic measuring method of the present invention, the wave plates are aligned and arranged in the first direction to realize at least four kinds of polarization characteristics in the first direction, and the light transmitted through the wave plate is individually provided. The image pickup means having an image sensor divided into the image pickup areas to be picked up is used to move the image pickup means relative to the measurement target in the first direction so that each measurement pixel of the measurement target is in each image pickup area. The output values obtained by imaging multiple times are added to create a measurement value in the imaging area for each measurement pixel, and the measurement values collected in the same manner from each imaging area are output from the measurement target for each measurement pixel. The Stokes parameter of the obtained light is calculated.

  According to the present invention, at least four types of wave plates having different polarization characteristics are arranged in alignment in the first direction, and from the four imaging areas of the wave plate for individually imaging the light transmitted through each of the wave plates. Since the configured image pickup means is used, each measurement pixel on the measurement object is continuously subjected to four or more types of polarization measurement image pickup during the relative movement as the measurement object moves in the first direction with respect to the image pickup means. Done. Furthermore, within one imaging area, each measurement pixel is imaged multiple times with different measurement pixels, and the polarization transfer matrix of each measurement pixel is pre-measured, so the real average processing is performed as multiple measurements of the same measurement. It is possible to improve the S / N ratio. In this way, it is possible to simultaneously achieve four or more types of polarization state measurements and a plurality of times of imaging while moving the imaging means relative to the measurement object without stopping. After completion of movement of the measurement object in the first direction (ie, completion of measurement), the measurement surface can be widened by moving the imaging object in the second direction by the visual field width, and movement in the first direction is performed. The entire surface of the measurement object can be measured by repeating the imaging and the movement in the second direction. Since four or more polarization states are measured, the measurement Stokes parameter can be determined. By comparing the Stokes parameter of the light source with the measured Stokes parameter, the polarization characteristic of the measurement object can be calculated.

  As described above, according to the present invention, the polarization characteristics of a wide area optical film can be measured quickly and accurately. For example, in the present invention, the measurement can be performed in about two and a half minutes while the conventional method requires about 10 minutes under conditions of a spatial resolution of 1 mm □ and an axial direction measurement accuracy of 0.1 °. That is, according to the present invention, a speed increase of about 4 times can be achieved as compared with the conventional method.

It is explanatory drawing explaining the measurement pixel of the optical film which is a sample. It is a graph showing the relationship between the number of combined cells and the variation in the output value of the CDD camera (data near high brightness 3740/4096 in a camera with an output value of 12 bits). It is the schematic of an imaging part. It is explanatory drawing for demonstrating imaging the measurement pixel E in the 1st-4th imaging area of a CCD camera. It is the schematic of the optical characteristic measuring apparatus of this invention. It is a flowchart showing the effect | action of this invention. It is explanatory drawing explaining the light projection pixel of a surface illumination part. It is the schematic which shows the light used in order to obtain | require a polarization transfer matrix, and an imaging part. It is the schematic which shows the two-dimensional array structure of 1 element in the storage area with the XY address used in a calibration measurement and a measurement. It is explanatory drawing for demonstrating imaging of the measurement pixel E1 by each combined cell in the 1st imaging area of a CCD camera. It is explanatory drawing for demonstrating imaging the measurement pixel E1 by each combined cell in the 2nd-4th imaging area of a CCD camera. It is explanatory drawing for demonstrating memorize | stored in the 1st-4th memory | storage part E11-En4 the output value obtained by imaging of the measurement pixels E1-En. It is explanatory drawing for demonstrating the case where combined cell CP11 and CP12 image the measurement pixel E1 in the ratio of 7: 3. It is explanatory drawing for demonstrating that the measurement pixel E1 is imaged 11 times by combined cell CP11-CP15. It is explanatory drawing for demonstrating the case where combined cell CP11 and CP12 image the measurement pixel E1 in the ratio of 5: 5. It is a graph which shows the relationship between the amount of retardation of the wavelength plate to be used, and the amount of calculation errors. It is a table | surface which shows the relationship between the amount of retardation of the wavelength plate to be used, and the amount of calculation errors. It is the schematic which shows the imaging part in which the 1st-4th wavelength plate was provided just before the CCD camera. It is the schematic of the measuring apparatus manufactured by making an illumination part thin. It is the schematic which shows a measuring apparatus provided with two CCD cameras.

  As shown in FIG. 5, the optical property measuring apparatus 10 of the present invention measures an optical film 12 having a predetermined birefringence property as a measurement object. In the optical characteristic measuring apparatus 10, the optical film 12 to be measured is placed on the surface illumination unit 14 attached to the sample stage 13. The optical film 12 is illuminated with circularly polarized illumination light emitted from the surface illumination unit 14, and the light emitted from the optical film 12 is imaged while the sample stage 13 is moved in the X direction by the imaging unit 15. Then, the computer 16 obtains the optical characteristics of the optical film 12 by performing various analyzes based on the output values obtained by the imaging unit 15. Note that elliptically polarized light may be used as the polarized illumination.

  The sample stage 13 can be moved in the X direction along two rails 22 a and 22 b on the base 22 by the X direction moving mechanism 20. The X-direction moving mechanism 20 includes a servo motor that is driven based on a driving pulse output from the X motor driver 24.

  Similarly, the imaging unit 15 is attached to an arm 31 provided on the support base 30. The arm 31 can be moved in the Y direction orthogonal to the X direction by the Y direction moving mechanism 33, and can be moved in the Z direction orthogonal to the X direction or Y direction by the Z direction moving mechanism 34. Yes. As described above, when the arm 31 moves in the Y direction or the Z direction, the imaging unit 15 can also move in the Y direction or the Z direction. The purpose of moving in the Z direction is to adjust the focus of the imaging unit 15. The Y-direction moving mechanism 33 is driven based on a drive pulse output from a Y motor driver (not shown).

  Drive pulses from the X motor driver 24 and the Y motor driver are also transmitted to an X pulse counter 26 and a Y pulse counter (not shown), respectively. Each pulse counter counts the received drive pulses. The value counted by the pulse counter is sent to the computer 16. Since the computer 16 stores the amount of movement of the sample stage 13 per pulse and the amount of movement of the imaging unit 15, the visual field of the imaging unit 15 is located on the sample stage 13 from the count values of both pulse counters. I can understand.

  Next, the operation of the present invention will be described with reference to the flowchart of FIG. First, measurement preparation is performed first, and here, the polarization transfer matrix of the imaging unit 15 is specified for each coupled cell CP of the CCD camera 41. This operation is performed only once as an initial setting. The obtained polarization transfer matrix is stored in the computer 16, and after the initial setting, the obtained polarization transfer matrix is used.

  Calibration measurement is performed after measurement preparation. Here, the Stokes parameter (hereinafter referred to as S parameter) of the light emitted from the surface illumination unit 14 is measured in units of measurement resolution over the entire surface of the surface illumination unit 14.

  As shown in FIG. 7, the surface illumination unit 14 serving as a light projecting unit is regarded as having virtually subdivided areas in units of measurement resolution, and the individual areas are collectively referred to hereinafter. , Called “projection pixel L”. Therefore, the calibration measurement is a step of obtaining the S parameter in the light projection pixel unit L.

  The calibration measurement need not be performed unless there is a light source variation, but it is desirable to perform the calibration measurement at the first measurement day.

  The actual measurement is performed next to the calibration measurement. Here, the optical film 12 is placed on the surface illumination unit 14 and the S parameter of the light transmitted through the optical film 12 is spread over the entire surface of the optical film 12. Measurement is performed in units of measurement resolution (ie, measurement target E).

  Finally, the birefringence characteristic of the measurement object is calculated by comparing the S parameter obtained by actual measurement with the S parameter of the illumination unit. What is important here is to match the position of each measurement pixel E in actual measurement with the position of each projection pixel L in calibration measurement, and a counter of the measurement target stage is used for this purpose. The current value of the counter tells where each combined cell of the camera captures the stage to be measured. Therefore, the output value of each combined cell obtained at the time of measurement is accurately distributed to each projection pixel L or each measurement pixel E. be able to.

  The measurement preparation process will be described in detail below. The measurement preparation step is a step of specifying the polarization transfer matrix of each coupled cell of the CCD camera used in the imaging unit 15 of the present apparatus prior to actual measurement. The measurement mechanism itself used for this specification may be incorporated in the apparatus, or may be separated from the apparatus, and the polarization transfer matrix measurement may be performed outside, and only the data may be taken into the computer 16 using a means such as a USB memory.

  The reason why the polarization transfer matrix of the imaging unit 15 is obtained in the unit cell unit of the CCD camera 41 is that the measurement values obtained from a plurality of measurement pixels E in the same wavelength plate section are averaged and one wavelength plate is used. This is to obtain a representative measurement value with high reliability. Even light that has passed through the same wavelength plate section cannot be treated in the same way because it is a measurement value using different parts of the wavelength plate, polarizing plate, telecentric lens, and measurement pixels of different CCDs. This is because there is local variation in local polarization transmission characteristics (locality). However, if a polarization transfer matrix is obtained in advance for each combined cell unit, the influence of locality can be corrected from a plurality of measured values and combined into one highly reliable representative measured value.

このＭ行列の第１行だけを取り出して、Ｍ₁₁ 要素で規格化すると［数２］となる。この行列をその結合セルでの偏光伝達行列と定義する。

また、あらためて、M₁₂／Ｍ₁₁ 、M₁₃／Ｍ₁₁ 、M₁₄／Ｍ₁₁ を、M行列になじみの深い記号 Ｍ₁₂、M₁₃、M₁₄、で置き換えなおし、Ｍ₁₁を係数 Ｋ で置き換えて、［数３］を得る。

［数３］に示された形式の行列を、結合セルにおける偏光伝達行列を表現する一般的な記号とする。ここで、Ｋ は偏光伝達行列の比例係数であるが、この値には、ＣＣＤカメラのシェーディング効果（ＣＣＤカメラ４１の各結合セルの量子効率やゲイン係数のばらつき）も含まれている。 The polarization transfer matrix is a matrix that represents the relationship between the S parameter of the light incident on the imaging unit 15 and the output value output from the CCD camera 41. This polarization transfer matrix can be determined from the product of a Mueller matrix (hereinafter referred to as M matrix) such as an optical member forming the imaging unit 15. The M matrix associated with a certain coupled cell on the CCD is the M matrix of the light beam passing portion of any of the wave plates 45 to 48 through which the light beam incident on the coupled cell has passed, and the M matrix of the light beam passing portion of the polarizing plate 49. Is multiplied by the M matrix of the beam passing portion of the telecentric lens and the M matrix of the coupled cell of the CCD. [Formula 1] represents a general form of the M matrix. The x mark is an element that does not need to be specified because it does not relate to subsequent calculations.

When only the first row of this M matrix is taken out and normalized by the M ₁₁ element, [Equation 2] is obtained. This matrix is defined as the polarization transfer matrix in the combined cell.

In addition, M ₁₂ / M ₁₁ , M ₁₃ / M ₁₁ , and M ₁₄ / M ₁₁ are replaced with symbols M ₁₂ , M ₁₃ , and M ₁₄ that are familiar to the M matrix, and M ₁₁ is replaced with the coefficient K. Thus, [Equation 3] is obtained.

Let the matrix of the form shown in [Equation 3] be a general symbol representing the polarization transfer matrix in the combined cell. Here, K is a proportional coefficient of the polarization transfer matrix, and this value also includes the shading effect of the CCD camera (quantity efficiency of each coupled cell of the CCD camera 41 and variations in the gain coefficient).

  In the measurement preparation process, as shown in FIG. 8, the polarization transfer matrix is obtained individually for every combined cell of the imaging unit 15 using the light 70 whose S parameter is known. Here, the light 70 whose S parameter is known is obtained by allowing the parallel light from the parallel monochromatic light source 72 to pass through the polarizing plate PL1 and the quarter-wave plate QWP1 in the reference projector 71. The polarizing plate PL1 is fixed in orientation and the transmission axis is arranged at 0 ° with the reference direction of the orientation in the imaging unit 15 as 0 °. The quarter-wave plate QWP1 has a motor-driven continuous rotation mechanism (not shown), and is used by continuously rotating this in the measurement of the polarization transfer matrix. The actual retardation amount and axial direction of the wave plate QWP1 are known, and the fast axis direction is defined with the reference direction in the imaging unit 15 being 0 °.

  The known light 70 is a light beam around the center of the optical axis of the reference projector 71. The polarization transfer matrix of one coupled cell is measured in one measurement. The reference projector 71 and the imaging unit 15 are made to relative to each other using the XY moving mechanism 71a of the reference projector 71 so that the optical axis center of the light beam of the reference projector 71 passes through the center of the coupled cell that is the measurement target of the imaging unit 15. . When the polarization transfer matrix measurement of one coupled cell is completed, the polarization transfer matrix measurement of the adjacent measurement pixel is performed using the XY moving mechanism 71a. In this way, the polarization transfer matrix of all the coupled cells of the imaging unit 15 is measured.

Here, a method for measuring the polarization transfer matrix of a certain coupled cell will be described, assuming that the known S parameter of the light 70 is | P ₀ P ₁ P ₂ P ₃ | ^T.
The signal output value of the combined cell and the S parameter of the light 70 used for measurement have the relationship of [Expression 4] when the polarization transfer matrix of this combined cell is described by [Expression 3].

ここでは、ＱＷＰ１の方位をγ、位相差をε、Ｃ＝cos２γ、Ｓ＝sin２γとしている。
Ｋ´は実際のＣＣＤカメラ４１の出力値との整合をとるための係数でこの測定中に決まる実数である。 When each element of the S parameter of the light 70 is described in detail using the constant of QWP1, it is expressed by [Equation 5].

Here, the orientation of QWP1 is γ, the phase difference is ε, C = cos2γ, and S = sin2γ.
K ′ is a coefficient for matching with the actual output value of the CCD camera 41 and is a real number determined during this measurement.

［数６］を［数４］へ代入すると［数７］が得られる。

From the above, P ₀ , P ₁ , P ₂ , and P ₃ are expressed by the following [Equation 6].

[Expression 7] is obtained by substituting [Expression 6] into [Expression 4].

When DFT (Discrete Fourier Transformation) is performed on the [Equation 7] in the azimuth γ of QWP1, as shown in [Equation 8], four relational expressions indicating the output values of the DC component and the following frequency component are obtained. Here, Fdc represents the DC component, Fcos4 represents the cos4γ component, Fsin4 represents the sin4γ component, and Fsin2 represents the measured amplitude of the sin2γ component.

［数９］に示す４本の式において、未知数はＫ・Ｋ´、Ｍ_１２、Ｍ_１３、Ｍ_１４の4個であるので値が求まる。例えば、（１）÷（２）からM_１２が特定でき、続いてK・K´が特定でき、その後M_１３、M_１４が特定できる。 Here, it should be noted that the Fdc is raised by the amount of the dark current of the CCD (the CCD that outputs zero light outputs a certain value). When this numerical value is BG, a value obtained by subtracting BG from Fdc in [Equation 8] is a direct current component to be used in the subsequent calculation, and Equation 4 of [Equation 8] is corrected to [Equation 9]. The BG can be specified by completely blocking the light from the CCD camera, and this value is obtained in advance.

In the four equations shown in [Equation 9], the number of unknowns is K · K ′, M ₁₂ , M ₁₃ , and M ₁₄ , so the value is obtained. For example, M ₁₂ can be specified from (1) ÷ (2), then K · K ′ can be specified, and then M ₁₃ and M ₁₄ can be specified.

  In this way, all elements of the polarization transfer matrix of one combined cell and the values of K · K ′ can be specified. By repeating this in all the coupled cells CP, the polarization transfer matrix and the values of K · K ′ in all the coupled cells CP of the imaging unit 15 can be specified. Among the values of K · K ′, K is a different value for each combined cell, but K ′ is a value related to the light intensity used in the preparation for the measurement, and can be regarded as being constant during this measurement. Therefore, the value of K · K ′ specified here can be used as a relative signal strength ratio between the combined cells in the subsequent measurement (calibration measurement, actual measurement). Further, in the subsequent measurement, even if the light intensity may be different from the measurement of the current measurement preparation, this value can be used as the relative signal intensity ratio between the combined cells.

Here, each element of the polarization transfer matrix in each identified combined cell CP is stored in the computer 16. At this time, the value of K · K ′ specified here is referred to again by the symbol K in the following description.

  As described above, the wavelength plate, the polarizing plate, the lens, and the CCD are collectively measured as a polarization transfer matrix after being assembled, so that it is not necessary to specify each one. This creates a significant benefit in reducing the measurement load.

  Next, the details of calibration measurement and actual measurement will be described (see the flowchart in FIG. 6). Both operations are exactly the same except that the measurement object is the surface illumination unit 14 in the calibration measurement and the optical film 12 in the actual measurement. In the following, an explanation will be given using actual measurement as an example.

  For both calibration measurement and actual measurement, a temporary output value storage area is provided in the computer 16 to store the CCD output value. This output value storage area is an array distinguished by two-dimensional XY addresses on the measurement stage of the measurement pixel E, and one array element is the number of wave plates as shown in FIG. 9 and the number of measurements on one wave plate. The two-dimensional structures are distinguished from each other, and the whole is a four-dimensional structure.

  In the calibration measurement and the actual measurement, the sample stage 13 is first moved from one end to the other end in the X direction while the imaging unit 15 is stationary at a predetermined position in the Y direction. When the sample stage 13 moves, an imaging trigger is issued and imaging is automatically performed. On the sample stage 13, there are a surface illumination unit 14 and an optical film 12 (hereinafter simply referred to as “optical film 12”). When the imaging unit 15 captures an image up to the other end of the optical film 12, measurement is performed for the visual field width of the camera. Therefore, in order to change the camera field of view, the imaging unit 15 is now moved in the Y direction by the field of view width. Thereafter, the optical film 12 is moved again from the other end to the other end, and an unimaged portion such as the optical film 12 is imaged. The above procedure is repeated to image the entire optical film 12 and the like.

  The process from imaging until S parameter is calculated is shown below. In addition, since all the pixels are imaged simultaneously at the imaging timing of the CCD, data is acquired from all coupled cells by one imaging, but in the following description, a description will be given focusing on one measurement pixel E as a representative example. . For the sake of simplicity, it is assumed that five coupled cells CP are arranged in the X direction in each of the imaging areas 50 to 53 of the CCD camera 41, and an image of the target measurement pixel E passes therethrough. A description will be given with one cross section in the X direction. Here, the measurement pixel E is set to be the same as the size of the combined cell CP on the CCD. This setting can be adjusted by the magnification setting of the telecentric lens or the number of combined cells.

  As shown in FIG. 10A, the five first combined cells provided in the first imaging area 50 are CP11 to CP15, and the five second combined cells provided in the second imaging area 51 are CP21 to CP25. The five third combined cells provided in the third imaging area 52 are CP31 to CP35, and the five fourth combined cells provided in the fourth imaging area 53 are CP41 to CP45. The measurement pixels of the optical film 12 are E1 to En (n is a natural number of 2 or more). The interval between imaging triggers generated when the optical film 12 is moved in the X direction is the length L in the X direction of the coupled cell on the CCD of the measurement pixel E (= the length in the X direction of one measurement pixel). Set the distance to advance. Since five coupled cells are arranged in the X direction in each imaging area of the CCD, when one measurement pixel E passes through each imaging area, imaging is performed five times in each imaging area.

First, when the optical film moves in the X direction, the image of the measurement pixel E1 reaches the first combined cell CP11 in the first imaging area at a certain timing. As shown in FIG. 10B, when the image of the measurement pixel E1 is positioned on the first combination cell CP11, the first combination cell CP11 images the measurement pixel E1. The output value obtained by this imaging is stored in the output value storage area EM1 for the measurement pixel E1 in the computer 16. EM1 is a two-dimensional array as shown in FIG. 9, where the row direction is the number of imaging areas and the column direction is the number of measurement values. This output value is stored in the row EM11 for the first imaging area in the output value storage area EM1.

Next, when the optical film 12 moves in the X direction by the movement amount of one imaging trigger, the image of the measurement pixel E1 of the optical film 12 is positioned on the first combined cell area CP12. Then, the first combined cell area CP12 images the measurement pixel E1. The output value obtained by this imaging is
It is stored in another area of the row EM11 for the first imaging area in the output value storage area EM1 for the measurement pixel E1. Similarly, the measurement pixel E1 is imaged in the first combined cell areas CP13 to CP15, and the output value obtained by this imaging is stored in the EM11. Therefore, the measurement pixel E1 passes through the first imaging area, and thus imaging is performed five times in total.

  Next, as shown in FIG. 11A, when the measurement pixel E1 reaches the combined cell CP21 in the second imaging area 51, the combined cell CP21 images the measurement pixel E1. The output value obtained by this imaging is stored in the second imaging area row EM12 of the output value storage area EM1 in the computer 16. Similarly, imaging is performed every time the measurement pixel E1 passes through the second combined cells CP22 to CP25, and output values obtained by the imaging are sequentially stored in the EM12.

  As shown in FIG. 11B, when the measurement pixel E1 passes through the third combined cells CP31 to CP35 in the third imaging area 52, imaging is performed in the same manner. The output value when the measurement pixel E1 is imaged by the third combined cells CP31 to CP35 is stored in the third imaging area row EM13 of the output value storage area EM1. As shown in FIG. 11C, when the measurement pixel E1 passes through the fourth combined cells CP41 to CP45 in the fourth imaging area 53, imaging is performed in the same manner, and the output value is an output value storage area. It is stored in the fourth imaging area row EM14 of EM1.

  Then, when the measurement pixel E1 passes through the fourth combined cell CP45 in the fourth imaging area 53, the measurement for the measurement pixel E1 is completed. The measurement pixel E2 located one backward with respect to the traveling direction of the measurement pixel E1 starts and ends with a delay of one imaging trigger with respect to the measurement pixel E1, but the same contents are performed. Further, the measurement pixels E3, E4, E5,... En positioned further rearward are each imaged while being shifted by one imaging trigger, and the process ends. These output values are stored in output value storage areas EM21 to EMn4 for the measurement pixels E2 to En corresponding to the output value storage areas EM11 to EM14 shown in FIGS. The correspondence is shown in FIG.

  Next, a method for obtaining the S parameter in the measurement pixel E1 from the output value of the measurement pixel E1 stored in the output value storage area EM1 and the polarization transfer matrix of the combined cell specified in the measurement preparation step will be described.

When obtaining the S parameter in the measurement pixel E1, first, as shown in [Equation 10] below, the total of output values obtained by imaging in the first to fourth imaging areas, that is, stored in each of EM11 to EM14. The sum of the five output values S11- _{Σ to} S14- _Σ is obtained. In the present invention, the total of such output values is referred to as a measured value, and S is represented by adding a subscript “_Σ”. The first number 1 attached to S, S11 to S14, is a number for identifying a measurement pixel and is originally a number group corresponding to a two-dimensional address, but corresponds to 1 of E1 in this example. The second numbers 1 and 4 correspond to the imaging areas 1 to 4 and are wave plate numbers used for measurement. The subscripts _A1 to _A5 indicate the order of measurement in one imaging area, that is, the first measurement to the fifth measurement result.

ここで、添字_CP11等の番号は、結合セルを区別する番号である。 On the other hand, each of the coupled cells CP11 to CP45 has a unique polarization transfer matrix, and this value is specified in the measurement preparation process. When the polarization transfer matrices of CP11 to CP45 are written out in order according to the definition, [Equation 11] is obtained.

Here, the numbers such as subscript_CP11 are numbers for distinguishing combined cells.

In this example, the output value S11 _{_CP11} ~S14 _{_CP45} is because it was separately measured by binding cell CP11～CP45, as shown in the following [Expression 12], the measurement pixel polarization transfer matrix of the binding cells It should be defined by the matrix product of the S parameter of E1 with | _{S0_E1 S1_E1 S2_E1 S3_E1} | ^T. Here, the suffix _E1 indicates the S parameter of the measurement pixel E1.

Here, the measured value S11_ _sigma [Expression 10], and rearranging the S parameter by substituting the definition of the output value of Equation 12] [Expression 13] is obtained.

既知量の添字の最初の数は、係数となるSパラメータの要素番号を示し、２番目の数字は波長板を区別する番号である。＿Nは、結合セルのY方向での位置を区別する添字であるが、この例ではX方向の１つの断面を扱っているので、ある断面位置を示す値になる。 This measurement S11_ _sigma in Equation 13], each S-parameter element, and has a shape which took some coefficients, respectively. These coefficients are obtained by multiplying and summing known values already identified in the measurement preparation process. Here, as shown in [Equation 14], this coefficient group is defined by the name of a known quantity. The known amount can be calculated in advance and can be treated as one numerical value thereafter.

The first number of known subscripts indicates the element number of the S parameter that is a coefficient, and the second number is a number that distinguishes the waveplate. _N is a subscript for distinguishing the position of the combined cell in the Y direction. In this example, since one cross section in the X direction is handled, the value indicates a certain cross section position.

By substituting [Equation 14] into [Equation 13], the following [Equation 15] is obtained.

Number 10] can be applied the same processing on the measurements _{S12_} Σ ~ measurements S14_ _Σ, it is obtained [number 16].

  The four equations combining [Equation 15] and [Equation 16] can be solved because the unknown is four S-parameters and the number of equations is four. Thereby, the S parameter in the measurement pixel E1 is obtained.

  As described above, it has been described that one measurement pixel E1 is sequentially imaged by the coupled cells arranged in the X direction on the CCD on which the image is formed, and finally the S parameter is calculated. Since all the pixels are imaged simultaneously in the CCD, all the coupled cells capture some measurement pixel E on the optical film 12 and the same processing is performed at the same time while the above processing is performed. . If you look at the CCD as a whole, it has the processing capacity of the combined cells in the field of view.

In this way, the S parameter for all the measurement pixels E can be obtained in the same manner as the method for obtaining the S parameter for the measurement pixel E1. The obtained S parameter is stored in the actual measurement S parameter storage area in the computer 16 in accordance with the XY address of the measurement pixel E.

  When the measurement is calibration measurement, the S parameter of all the light projection pixels L of the surface illumination unit 14 is measured instead of the measurement pixel E. The S parameters of all the projection pixels L are stored in an S parameter storage area for calibration measurement in the computer 16.

In addition, the XY address of the measurement pixel E and the XY address of the projection pixel L are determined by the position of the sample stage when the imaging trigger is issued. , The positional relationship is overlapped vertically. Further, the light that illuminates the measurement pixel E can be regarded as being emitted from the projection pixel L having the same XY address by the function of selecting and supplementing the deep focal depth of the telecentric lens and the light component parallel to the optical axis.

  Furthermore, since the position at which the imaging trigger is issued on the sample stage 13 can be designated in advance by the computer 16, a numerical value called a known amount can be calculated in advance. This has an effect that it is very advantageous for speeding up the processing in an application that needs to process a large amount of measurement data such as two-dimensional measurement.

ここで、S＝ｓｉｎ２α、 Ｃ＝cos２α である。この関係式を用い、光学フイルム１２のすべての測定画素Eの複屈折の主軸方位αと遅相量δを計算することで、光学フイルム１２の偏光特性分布を試料解像度で測定することができる。 Finally, a method for calculating the polarization characteristics of the optical film 12 will be described. The computer 16 calculates the principal axis direction and retardation of the entire surface of the optical film based on the S parameter of the entire surface illumination unit 14 obtained by the calibration measurement and the S parameter of the entire surface of the optical film 12 obtained by the actual measurement. The S parameter of the light transmitted from the measurement pixels E with the optical film 12 | is ^T, light projecting pixel L of emitting light of S parameters directly under the E | | 1 Φ Ψ ξ measurements when the ^T | 1 XYZ The principal axis direction α and the slow phase amount δ of the birefringence of the optical film 12 corresponding to the pixel E can be expressed by [Equation 17].

Here, S = sin 2α and C = cos 2α. By calculating the birefringence principal axis direction α and the retardation amount δ of all the measurement pixels E of the optical film 12 using this relational expression, the polarization characteristic distribution of the optical film 12 can be measured at the sample resolution.

  In the description of the present embodiment, a total of five times of imaging is performed in each imaging area. However, in order to increase measurement accuracy, five or more times, for example, ten times of imaging may be performed. In this case, as shown below, it is performed by overlapping imaging (one imaging field of view and the next imaging field of view partially overlap).

The overlap imaging is performed by setting the amount of movement of the sample stage between one imaging trigger to be equal to or less than the length L in the X direction of the measurement pixel. When such a movement amount is set, since one measurement pixel is imaged across two adjacent combined cells, it is necessary to distribute output values. For example, FIG. 13 shows a state of the second imaging in the example of FIG. 10 when the movement amount between one imaging trigger is set to 3/10 of L. At this imaging timing, the measurement pixel E1 is positioned on the 7/10 region of the combined cell CP11 and is positioned on the 3/10 region of the combined cell CP12. In such a case, a SUB combined cell is defined. The SUB merged cell is an integer multiple of 1/10 of the merged cell CP in the X direction and the same size as the merged cell in the Y direction. The output average of the CCD imaging cell contained in the SUB merged cell is the SUB merged cell. The output value of the combined cell. In FIG. 13, a SUB combined cell of CP11 is formed in area 7 (indicated by hatching), and a SUB output value _{S11_CP11} is formed as an output value. Then, SUB output value _{S11_CP12} is created as the output value, and each is stored in EM11.
Further, at this time, the output value is treated as having the contents shown in [Equation 18].

´記号と´´記号は、測定のタイミングの違いを明示するためにつけた記号で他意はない。この結果における既知量を計算すると［数２０］となる。これは、［数１４］の2倍であり、撮像が2倍に増加した結果と解釈できる。

FIG. 14 shows a state in the imaging area 50 when the movement amount between one imaging trigger is set to 5/10 of L, and FIG. 15 shows a state of the SUB coupled cell at the time of the third imaging. The measurement pixel E1 is imaged a total of 11 times with CP11 to CP15. The definition of the 11 output values is [Equation 19].

The “symbols” and “symbols” are symbols used to clearly indicate the difference in measurement timing. The known amount in this result is calculated as [Equation 20]. This is twice the [Equation 14], and can be interpreted as a result of an increase in imaging.

  The known amount is corrected by a distribution relational expression such as [Equation 18] and [Equation 19]. Since the imaging timing is determined by the position of the sample stage, it is possible to design in advance in which coupling cell CP or SUB coupling cell the measurement pixel E is imaged at all imaging timings. The size and known amount of the combined cell can be calculated in advance at all imaging timings.

  As described above, there is a case where data from two SUB coupled pixels is allocated to each measurement pixel E for one imaging, but a measurement value of one imaging area is still obtained in the imaging area. It can be handled by adding all output values. As a result, the formulas of the measurement values are collected only for the number of types of wave plates, and thereafter, the birefringence distribution of the measurement object is obtained by the same method as described above.

  In the present embodiment, the first to fourth wave plates 45 to 48 are wave plates having a retardation amount of about 135 °, and the main axis direction (fast axis direction) is the same as that of the first wave plate 45 in FIG. When the direction is set to 0 °, the second wave plate 46 is arranged so as to have an axial direction of about 20 °. The second wave plate 46 is arranged in an direction in which the axial direction is added to the first wave plate 45 by about 36 °. 47 is arranged in an orientation in which the axial direction is further added by about 36 ° with respect to the second wavelength plate 46, and the fourth wavelength plate 48 is arranged in an direction in which the axial direction is further added by about 36 ° with respect to the third wavelength plate 47. The polarizing plate 49 was arranged so that the transmission axis was oriented at 0 °. The reason why the word “abbreviated” is used here is that, in the measurement preparation process shown in the flowchart of FIG. 6, the imaging unit specifies the polarization transfer matrix in units of coupled cells CP, and thereafter the polarization transfer matrix is used. This is because strictness is not necessary here. Generally, it should be in the range of ± 0.5 °.

  In this example, the number of wave plates is four, but the number of wave plates may be four or more. For example, when N types of wave plates (N is a natural number of 5 or more) are used, it is preferable that the N wave plates are arranged in such a manner that the main axis orientation is uniformly distributed at 180 °. Further, the amount of retardation of each wave plate is preferably about 135 °. With such a configuration, the principal axis orientation of the first wave plate may be an arbitrary direction.

  The reason for this arrangement is that the error in solving [Equation 15] and [Equation 16] with simultaneous equations is minimized. When the S parameter of light is obtained from [Equation 15] and [Equation 16], the measured value (left side) of each equation includes noise included in the output value of the CCD. Although these are reduced by the average effect due to multiplex imaging but are not zero, they are included as errors in the resulting S parameter. The amount of noise on the left side of [Equation 15] and [Equation 16] reflected as an error in the S parameter in the calculation process by the computer 16 is the variable (S parameter) of [Equation 15] and [Equation 16]. It is determined by the coefficient for. In other words, if the system is assembled so as to have an appropriate coefficient, this CCD calculation error (effect of CCD noise) can be minimized. The selection of this coefficient is nothing but the specifications of the wave plate (selection of the amount of slow phase and principal axis orientation).

  As a result of repeated simulations by the authors, the following was found. First, when implementing a change in polarization state by changing the main axis orientation of the installation using wave plates with the same amount of slow phase, even if the number of wave plates is increased, the calculation accuracy may not be significantly improved. all right. This is because even if the number of wave plates is increased, the area of one wave plate is reduced, so that the reliability of the measured value with one wave plate is lowered. Furthermore, if the number of types of wave plates is increased, the coupled cell that is the boundary between the wave plates cannot be used, so that the light receiving area of the CCD is substantially reduced. A decrease in the light receiving area means a decrease in the S / N ratio. On the other hand, when the number of wave plates is large, there is an effect that the cutoff frequency of noise contained in the signal can be increased. Considering both, it can be said that the minimum value of the number of wave plates is four kinds necessary for determining the S parameter, and the maximum value is up to about 40.

  For the types of wave plates, it is preferable to use wave plates with the same amount of retardation, and to make a difference in the orientation of the main axis, and to arrange the directions different from each other by an angle obtained by dividing 180 ° by the number of wave plates. But the calculation error was the smallest. Such an installation method of the installation wave plate is hereinafter referred to as equal division, and the number of wave plates is referred to as the division number.

  On the other hand, when the difference in angle between the wave plates is 45 °, the calculation error becomes large regardless of the number of types of wave plates. In the case of four types of wave plates, the equally divided angle is 45 °, so the equally divided angle that should have the smallest calculation error cannot be used. For this reason, the arrangement of this example was selected.

  On the other hand, as shown in FIGS. 16A and 16B, it was found that the retardation amount of the wave plate has a region where the calculation error is minimized around 135 ° in the number of all wave plates. The angle division of the direction of the wave plate is set to be equal division (however, only four divisions are 36 ° difference). This tendency was confirmed in all of the 4 to 40 divisions.

  Regarding the position of the wave plate, in the present embodiment, the first to fourth wave plates 45 to 48 are provided on the object side of the telecentric lens. Instead, as shown in FIG. Wave plates 45 to 48 may be provided immediately before the CCD camera. Here, a polarizing plate 49 is provided between the first to fourth wave plates 45 to 48 and the CCD camera 41.

  In this case, there is an advantage that the wavelength plate can be made small and the cost is reduced. On the other hand, since the telecentric lens comes out of the polarizing plate, there is a demerit that an error in obtaining the birefringence transfer function becomes large.

  In the present embodiment, the surface illumination unit 14 having a size for illuminating the entire surface of the optical film 12 is used and the imaging unit 15 is moved in the X direction and the Y direction. However, as shown in FIG. Instead of the illuminating unit 14, the width is narrowed to illuminate the visual field width of the imaging unit 15 in the Y direction, and the illuminating unit 14 is also moved in the Y direction when the imaging unit 15 moves Y. 101 may be used. In the case of such a configuration, an effect of reducing the cost of the surface illumination unit 14 and shortening the acquisition time of the Stokes parameters can be obtained. When the illumination unit 101 is used, a sample stage 102 in which an opening is formed in a portion where the optical film 12 is placed is used instead of the sample stage 13.

  In the present embodiment, the optical characteristics of the entire optical film 12 are measured by moving one CCD camera 41 in the X and Y directions. However, in order to further shorten the measurement time, a plurality of CCD cameras 41 are measured. Measurement may be performed with a single CCD camera. At that time, as shown in FIG. 19, one dedicated camera CPU is provided for each camera, and a main CPU that integrates a plurality of camera CPUs is provided at the upper level. The polarization transfer matrix of each combined cell of the camera is placed on the main CPU side, and each known amount is calculated in advance. The S parameter storage area for calibration measurement and actual measurement is also placed in the main CPU. An output value storage area for storing the measurement results of each camera is also placed on the main CPU side, but a copy (replica) of the output value storage area is prepared on the camera CPU side. As the measurement stage moves in the X direction, each CCD camera repeats imaging. The camera CPU calculates output values in units of combined cells (in some cases, SUB combined cells) and stores them in a replica (replica) of the output value storage area. When the imaging in the X direction reaches the scanning end, the Y movement corresponding to the visual field width of each camera and the movement of the sample stage to the scanning start end in the X direction are performed, but since the imaging is not performed during this movement, the camera The CPU has no imaging load. Using this load valley, the camera CPU performs data copying from the copy of the output value storage area to the output value storage area on the main CPU side. The main CPU detects a copy of data in the output value storage area, calculates and stores S parameters, and calculates the polarization characteristics of the optical film. Since the main CPU is immune from the imaging load even during the imaging of the CCD camera, most of the C`PU power can be allocated to this calculation. As described above, both the camera CPU and the main CPU can be operated efficiently, and the effect of speeding up the processing by increasing the number of cameras can be obtained.

  In the present embodiment, in the calibration measurement and the actual measurement, the start and end of imaging are not described in detail. However, since the entire field of view of the CCD camera is not effective at the start and end of imaging, the output obtained by imaging at this time The value should not be used in subsequent calculations. Although the boundary between the first to fourth wavelength plates was not described in detail, when the measurement pixel E was imaged on the boundary between the first to fourth wavelength plates, the output value obtained by this imaging. Should not be used in subsequent calculations. This is because the position detection means can always identify where each CCD-coupled cell captures the sample stage 13, so that information at unnecessary positions on the sample stage can be excluded. Furthermore, since the coupled cell at the position where the joint of the wave plate is imaged is known in advance, it is possible to avoid using this portion.

  In the present invention, imaging is performed in units of combined cells obtained by combining a plurality of adjacent imaging cells. However, if the imaging cell size is sufficiently large and there is no noise problem, imaging may be performed in units of imaging cells. Good.

DESCRIPTION OF SYMBOLS 10 Optical characteristic measuring apparatus 12 Optical film 13 Sample stage 14 Surface illumination part 15 Imaging part 16 Computer 20 X direction moving mechanism 33 Y direction moving mechanism 41 CCD camera 45-48 1st-4th wavelength plate 50-53 1st-1st 4 imaging area 55 CCD
CP1 to CP4 1st to 4th combined cells

Claims (10)

  Projecting means for irradiating a measurement object with a specific polarized illumination, and arranging a wave plate in a first direction so as to arrange at least four kinds of polarization characteristics in the first direction, and the wave plate An imaging unit including an image sensor that is divided into imaging areas that individually capture light transmitted through the scanning unit, and a scanning unit that moves the imaging unit relative to the measurement target in the first direction, Along with the relative movement of the imaging means by the scanning means, an output value obtained by imaging each measurement pixel to be measured a plurality of times in each imaging area is added for each measurement pixel to create a measurement value in that imaging area, An optical characteristic measuring apparatus, wherein a Stokes parameter of light for each measurement pixel is calculated from measurement values collected from each imaging area.   The optical sensor according to claim 1, wherein the image sensor captures an image with an image sensor including a plurality of pixels, and outputs one output value for each combined cell obtained by combining a plurality of adjacent pixels at the time of imaging. Characteristic measuring device.   Before the measurement, a polarization transfer matrix representing the relationship between the light emitted from the measurement object and the output value output from each imaging area of the image sensor is obtained for each coupled cell, and the polarization transfer matrix and each imaging 3. The optical characteristic measuring apparatus according to claim 1, wherein a Stokes parameter to be measured is obtained using an output value output from the area.   4. The optical characteristic measurement apparatus according to claim 2, wherein the image sensor captures an image of a predetermined measurement pixel across two adjacent combined cells. 5.   The output value of the predetermined measurement pixel is obtained by multiplying the ratio of the pixel that has captured the predetermined measurement pixel in one combined cell by the output average value of the pixel, and the predetermined measurement pixel in the other combined cell. 5. The optical characteristic measuring apparatus according to claim 4, wherein the optical characteristic measuring apparatus is composed of a ratio of a pixel that captures the image and an output average value of the pixel.   The moving unit moves at least one of the measurement target and the imaging unit in the first direction, and each time imaging of the measurement target in the first direction by the imaging unit is completed, the measurement unit or the imaging unit 6. The optical characteristic measuring apparatus according to claim 1, wherein at least one of the optical characteristics is moved in a second direction perpendicular to the first direction and parallel to the object to be measured.   When the moving means moves at least one of the measuring object and the imaging means in a second direction perpendicular to the first direction and parallel to the measuring object, the second means of the imaging means and the light projecting means 7. The positional relationship in the direction is moved so as not to change, and the polarized light irradiation width of the light projecting means is narrowed to such an extent that the visual field width in the second direction of the imaging means can be irradiated. Optical property measuring device.   The optical characteristic measuring apparatus according to any one of claims 1 to 6, wherein the number of types of polarized light embodied by the wave plate is 4 to 40.   9. The wave plate according to claim 1, wherein the wave plate has a retardation effect similar to that of a wave plate having a retardation amount of 70 ° to 170 ° or 190 ° to 290 °. The optical characteristic measuring apparatus as described.   The wave plates are aligned and arranged in the first direction to realize at least four types of polarization characteristics in the first direction, and are divided into imaging areas for individually imaging the light transmitted through the wave plate. Output obtained by moving the imaging means relative to the measurement object in the first direction using the imaging means provided with the image sensor, and imaging each measurement pixel of the measurement object multiple times in each imaging area A value is added to create a measurement value in the imaging area for each measurement pixel, and a Stokes parameter of the light emitted from the measurement target is calculated for each measurement pixel from the measurement values collected in the same manner from each imaging area. A method for measuring optical characteristics.

Images