JP2017037004A - Information processing device, information processing method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for easily calculating an anisotropy evaluation value representing an intensity of the anisotropy of BRDF (a bidirectional reflectance distribution function) in a measurement target object at a high speed.SOLUTION: An information processing device for evaluating the anisotropy of a reflection characteristic of an object includes: image data acquisition means 605 for using an imaging device and a light source arranged on an optical axis of the imaging device to image an object, and acquiring image data; calibration image data acquisition means 604 for acquiring calibration image data for calibrating the imaging device and the light source; reflectance map calculation means 606 for calculating a reflectance map of the object on the basis of the image data and the calibration image data; and anisotropy evaluation value calculation means 607 for using reflectance corresponding onto a circumference with the optical axis of the imaging device as a center in the reflectance map to calculate an anisotropy evaluation value about the object.SELECTED DRAWING: Figure 6

Description

  The present invention relates to a method for calculating an evaluation value of anisotropy in reflection characteristics of an object.

  In recent years, the use of computer graphics (CG) has expanded in various fields, and there is a strong demand for a technique for generating an image that faithfully reproduces the real texture. The texture of an object's appearance is called a reflection characteristic, and the reflection characteristic of an opaque material such as metal is represented by a bi-directional reflectance distribution function (BRDF). Therefore, BRDF measurement is indispensable for generating a CG faithful to the actual product. In addition, each object has a specific texture, and it is indispensable to measure BRDF for various objects in order to reproduce a real landscape with CG.

  However, since BRDF is a four-variable function in which the incident direction (two variables) and the exit direction (two variables) of the light beam are variables, when BRDF is measured, reflection is performed for a huge combination of the incident direction and the exit direction. It is necessary to measure the rate. Therefore, the number of times of measurement becomes enormous, and as a result, a very long measurement time is required for BRDF measurement.

  Conventionally, a technique using a turntable and an arch arm is known as a conventional technique for automatically executing BRDF measurement (Patent Document 1). A BRDF is automatically measured by installing a measurement target object on a turntable, attaching a light source and a light receiver to an arch arm, and installing them in various incident directions and outgoing directions.

  Further, an isotropic BRDF measurement technique for measuring an isotropic BRDF (Isotropic BRDF) of an object to be measured is also known as a conventional technique (Non-Patent Document 1). The isotropic BRDF measurement technique measures BRDF at high speed with a limited combination of incident and outgoing directions, assuming isotropic reflection of the BRDF of an object to be measured.

JP 2006-275955 A

Wojciech Matusik, Hanspetter Pfister, Matt Brand, and Leonard McMillan. “A Data-Driven Reflection Model”. ACM Transactions on Graphics. 22 (3) 2002.

  However, although the technology of Patent Document 1 can automate BRDF measurement, the number of times of measurement itself has not been reduced.

  In addition, the method of Non-Patent Document 1 does not target anisotropic BRDF (Anisotropic BRDF), and it is difficult to accurately measure BRDF for an object with significant anisotropic reflection. Therefore, when measuring the BRDF of an unknown object, it is necessary to determine in advance whether the BRDF has anisotropy.

  The presence or absence of anisotropy of BRDF can be determined by measuring the reflected light of a measurement target object placed on a turntable or by visual observation with a light receiver. It took time.

  Accordingly, an object of the present invention is to provide an information processing apparatus that can easily and rapidly calculate an anisotropy evaluation value representing the strength of anisotropy of BRDF in a measurement target object.

In order to achieve the above object, an information processing apparatus according to the present invention provides:
An information processing apparatus for evaluating the anisotropy of reflection characteristics of an object,
(605) an image data acquisition unit that images the object using an imaging device and a light source disposed on the optical axis of the imaging device and acquires image data;
Calibration image data acquisition means for acquiring calibration image data for calibrating the imaging device and the light source (604);
(606) a reflectance map calculating means for calculating a reflectance map of the object based on the image data and the calibration image data;
(607) anisotropy evaluation value calculation means for calculating the anisotropy evaluation value for the object using a reflectance corresponding to a circumference around the optical axis of the imaging device in the reflectance map;
It is characterized by having.

  According to the information processing apparatus of the present invention, the BRDF anisotropy evaluation value of the object is calculated, and the measurement method of the BRDF is selected based on the calculated value. Play.

The figure explaining the anisotropy of BRDF which concerns on Example 1. FIG. 5 is a diagram for explaining an optical system for calculating anisotropy evaluation values according to Example 1; FIG. 1 is a diagram illustrating an example of a system configuration according to the first embodiment. 1 is a diagram illustrating an appearance of an image data acquisition apparatus according to a first embodiment. The figure which shows the internal structure of the information arithmetic processing apparatus which concerns on Example 1. FIG. The figure showing the outline functional composition concerning Example 1. Flowchart showing a flow of a series of processes according to the first embodiment. FIG. 3 is a diagram illustrating an optical system for measuring a spherical measurement target object according to the first embodiment. FIG. 5 is a diagram illustrating an optical system when using the half mirror according to the first embodiment. FIG. 5 is a diagram illustrating an optical system when using a ring-shaped light source according to the first embodiment.

Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings.
[Example 1]
In order to generate a CG image that reflects the texture of an actual object (real object), it is necessary to perform rendering based on measurement of the reflection characteristics of the real object. Generally, the reflection characteristic of an object is represented by BRDF which is a ratio of incident light and reflected light in a certain direction.

  First, the anisotropy of BRDF, which is a target for calculating an evaluation value in the present invention, will be described with reference to FIG.

  A flat measurement target object 101 is set at the origin O, the normal vector of the measurement surface is n, the tangent vector is t, and the subnormal vector is bn. In addition, the direction vector (incident direction vector) of light incident on the origin O is denoted by wi, and the direction vector (outgoing direction vector) of light exiting from the point O by reflection is denoted by wo. At this time, BRDF fr of the measurement target object is a reflectance from an arbitrary incident direction wi to an outgoing direction wo, and is generally expressed as follows.

Note that Lr is the radiance in the wo direction at the point O, and Ei is the irradiance in the wi direction at the point O. Further, when the incident direction vector wi and the outgoing direction vector wo are displayed in polar coordinates in the tangent space composed of the tangent vector t, the normal vector bn, and the normal vector n, respectively,

It is expressed. Therefore, BRDF can be rewritten as follows using the polar coordinate display.

However, φd is determined by the following equation.

At this time, if the BRDF does not depend on φi, the BRDF of the object is said to be isotropic, and conversely if it depends on φi, it is said to be anisotropic.

  Note that φi is an angle formed by a vector obtained by projecting wi on a tangent plane composed of t and bn and the tangent vector t, and is generally expressed by the following equation.

  Therefore, the anisotropy of BRDF can be evaluated essentially by the tangent vector dependency, but in this embodiment, the anisotropy of BRDF is evaluated as the dependency of φi for simplicity.

  In this embodiment, paying attention to the dependence of BRDF on φi, the anisotropy is evaluated using the optical system shown in FIG. Below, the detail of the calculation method of a specific anisotropy evaluation value is demonstrated.

  In FIG. 2, the position of the light receiver 201 is a point E, the position of the light source 202 disposed on the optical axis 204 of the light receiver is a point L, and the intersection of the measurement target object 203 and the optical axis 204 is a point C0. In this embodiment, an example of a point light source will be described as the light source 202. In addition, in a circle C having a radius r centered on the point C0, two arbitrary points on the circumference are defined as a point P1 and a point P2, and the incident light vector and the emitted light vector at the respective points are denoted by wip1, wop1, wip2, Let it be wop2. These vectors are expressed by the following equations using a polar coordinate system.

  At this time, since the triangle EC0P1 and the triangle LC0P1 exist on the same plane, and the triangle EC0P2 and the triangle LC0P2 also exist on the same plane, the following equation holds.

  Further, since the point P1 and the point P2 are on the same circumference, the following equation is established.

  At this time, the reflectance fr, p1 obtained by observing the point P1 is expressed according to the equation (4).

The reflectance fr, p2 obtained by observing the point P2 is

It is. fr, p1 and fr, p2 are equal in the first to third variables (that is, θip, θop, 0). Therefore, if the BRDF of the measurement target object is isotropic (does not depend on φi), the reflectance observed at any point on the circle C is equal.

  Therefore, the reflectance at the point N on the circle C is expressed by the following equation.

  At this time, the anisotropy evaluation value β is expressed by the reflectance variance σ2, and is calculated by the following equation.

  However,

far.

  In this embodiment, the dispersion of the reflectance is used to calculate the anisotropy evaluation value. However, the present invention is not limited to this. For example, the absolute value of the difference between the minimum value and the maximum value of the reflectance may be used.

  In the present embodiment, the anisotropy evaluation value described above is calculated based on image data obtained by imaging the measurement target object. Furthermore, by measuring BRDF adaptively according to the calculated anisotropy evaluation value, the time required for measurement is shortened.

  Hereinafter, preferred embodiments of the present invention will be described in detail. FIG. 3 is a diagram illustrating a system configuration example according to the present embodiment. This system includes an image processing device 301 that calculates an anisotropy evaluation value and a reflection property measurement device 302 that measures the reflection property of a measurement target object. The image processing apparatus 301 includes an image data acquisition apparatus 303 that acquires image data, and an information arithmetic processing apparatus 304 that performs various processes described later including calculation of anisotropy evaluation value from the obtained image data. .

  In this embodiment, the measurement target object is photographed by the image data acquisition device 303, and the anisotropy evaluation value is calculated by the information calculation processing device 304 based on the obtained data. The reflection characteristic measuring apparatus 302 is an optical measuring instrument as described in Patent Document 1 and Non-Patent Document 1, for example. This measurement apparatus can switch the measurement method, and adaptively according to the anisotropy strength of the measurement target object by using the anisotropy evaluation value when measuring the BRDF of the measurement target object. It is possible to measure BRDF.

  For example, for an object whose anisotropy evaluation value is sufficiently small and the BRDF can be regarded as isotropic, the isotropic BRDF is measured at high speed using the method proposed in Non-Patent Document 1. . For an object having a large anisotropy evaluation value and a relatively large BRDF anisotropy, the anisotropic BRDF is measured using the method proposed in Patent Document 1. In general, it is known that measurement of isotropic BRDF takes less time than measurement of anisotropic BRDF. Therefore, for example, when there are a plurality of measurement target objects, the measurement target object that is determined to have a small BRDF anisotropy rather than all of the measurement target techniques as in Patent Document 1 is as in Non-Patent Document 1. Measurement time can be reduced by measuring with this method.

  FIG. 4 shows the appearance of the image data acquisition device 303 as an example of the optical system shown in FIG. The image data acquisition device 303 includes an imaging device 401 as a light receiver, a light source 402, a stage 403, and a support unit 404 for the light source 402. The light source 402 is installed on the optical axis 204 of the imaging device 401. The flat measurement target object 203 is installed on the stage 403 so that the normal line of the measurement surface is parallel to the optical axis 204.

  FIG. 5 is a diagram showing an internal configuration of the information processing unit 304. The information processing unit 304 includes a CPU 501, RAM 502, ROM 503, HDD interface (I / F) 504, hard disk drive (HDD) 505, input I / F 506, output I / F 507, and system bus 508.

  The CPU 501 that is an arithmetic device executes a program stored in the ROM 503 or the HDD 505 using the RAM 502 as a work memory, and comprehensively controls each component described later via the system bus 507. Thereby, various processes to be described later such as calculation of anisotropy evaluation value are executed. The HDD I / F 504 is an interface such as serial ATA, for example, and connects the HDD 505 as a secondary storage device and a serial bus 508. The CPU 501 can read data from the HDD 505 and write data to the HDD 505 via the HDD I / F 504.

  Further, the CPU 501 can read the data stored in the HDD 505 to the RAM 502 and similarly can instruct the HDD 505 to save the data expanded in the RAM 502. The CPU 501 can read the data developed in the RAM 502 and execute it as a program. The secondary storage device may be a storage device such as a flash memory or an optical disk drive in addition to the HDD. The input I / F 506 is an interface such as RS232C or USB, for example, and is connected to the image data acquisition device 303 via this I / F to receive image data.

  The output I / F 507 is an interface such as RS232C or USB, for example, and is connected to the reflection characteristic measurement device 302 via this I / F to transmit the calculated anisotropy evaluation value. There are no restrictions on the form of the input I / F 506 and the output I / F 507, and data may be transferred via a medium such as an SD memory card.

  FIG. 6 is a functional block diagram according to the present embodiment, which includes an image acquisition unit 601, a calculation unit 602, and a reflection characteristic measurement unit 603. The image acquisition unit 601 includes a calibration image data acquisition unit 604 and an image data acquisition unit 605. Further, the calculation unit 602 includes a reflectance map acquisition unit 606 and an anisotropy evaluation value calculation unit 607.

  The calibration image data acquisition unit 604 images an object whose BRDF is known using the image data acquisition device 303. The image data acquisition unit 605 images the measurement target object using the image data acquisition device 303. The reflectance map calculation unit 606 calculates a reflectance map using the imaging data and the calibration imaging data. The anisotropy evaluation value calculation unit 607 calculates the anisotropy evaluation value of the measurement target object using the reflectance map. The reflection characteristic measurement unit 603 measures BRDF using a measurement method suitable for the measurement target material based on the anisotropy evaluation value.

  FIG. 7 is a flowchart showing the flow of calculation of the anisotropy evaluation value and the BRDF measurement based on the calculation of the anisotropy evaluation value according to the present embodiment.

  In step S <b> 701, the calibration image data acquisition unit 604 captures a calibration object using the image data acquisition device 303 to acquire calibration image data, and transfers the calibration image data to the information processing unit 304.

  The calibration object to be measured in this step only needs to have BRDF, for example, a standard white plate is used. The luminance value at each pixel of the calibration image data obtained at this time is defined as Iw (w, h). Note that w and h are the horizontal and vertical pixel positions of the calibration image data, respectively.

  In step S <b> 702, the image data acquisition unit 605 captures the measurement target object, acquires image data, and transfers the image data to the information calculation processing device 304. The luminance value at each pixel of the image data obtained at this time is defined as I (w, h).

  In step S703, the reflectance map calculation unit 606 calculates a reflectance map from the calibration image data acquired in step S701 and the image data acquired in step S702. The reflectance map is a two-dimensional map that stores the reflectance of the object to be imaged corresponding to each pixel position of the image data. The reflectance fr (w, h) corresponding to the pixel position (w, h) is represented by the ratio between the luminance value of the calibration image data and the luminance value of the image data, and is generally calculated using the following equation.

  Fr and w are reflectances of the calibration object. In step S704, the anisotropy evaluation value calculation unit 607 calculates the anisotropy evaluation value of the measurement target object based on the reflectance map calculated in step S703.

  First, a set of reflectances on the same circumference centered on the optical axis 204 is extracted from the reflectance map. As an example, assuming that the pixel position of the optical axis 204 in the reflectance map is (wo, ho) and the radius is r0, N pixels (wc1, hc1), (wc2, hc2), ..., (wcN, When both hcN) satisfy the following equation, these pixels exist on the same circumference.

However, iε [1, N].

  Next, the reflectance fr (wci, hci) corresponding to these pixel positions is extracted from the reflectance map, and this is expressed as fr, i in equation (18) as follows.

  At this time, anisotropy evaluation is calculated using Expression (18) and transmitted to the reflection characteristic measurement unit 603.

  In step S705, the reflection characteristic measurement unit 608 measures the BRDF of the measurement target object by a measurement method according to the anisotropy evaluation value calculated in step S704.

  For example, when the anisotropy evaluation value is smaller than a predetermined threshold value, the method proposed in Non-Patent Document 1 capable of measuring BRDF at high speed is used. Further, when the anisotropy evaluation value is larger than a predetermined threshold value, measurement is performed using the method proposed in Patent Document 1 capable of automatically measuring BRDF.

  Note that the image data acquired by the image data acquisition unit 605 and the image data for calibration acquired by the calibration image data acquisition unit 604 are reflected by the support unit 404 of the light source 202, and the measurement target object and the calibration material are partially included. It is shielded and the anisotropy evaluation value may not be calculated accurately. In this case, by acquiring in advance information on the position and shape of the support unit 404 viewed from the imaging device 201, it becomes possible to detect a shielded portion in image data and calibration image data based on this information.

  With respect to this shielding portion, the calculation of the reflectance is canceled, and the reflectance is not stored in the shielding pixel of the reflectance map. Furthermore, in the anisotropy evaluation value calculation unit 607, when there are shielding pixels on the circumference from which the reflectance is extracted, the reflectance is not extracted from these pixels, but the reflectance extracted from other pixels is extracted. The anisotropy evaluation value may be calculated using

  In this embodiment, an example of a flat plate is described as the shape of the measurement target object, but the shape is not limited to this. Any rotating body centered on the optical axis 204 may be used, such as a sphere or a cone. FIG. 8 shows an example in which a sphere 801 is used as a measurement target object. The optical axis 204 is disposed so as to pass through the center Q of the sphere 801. In this case, the first to third variables of BRDF that satisfy the relationship between Expression (15) and Expression (16) and are observed at two arbitrary points on the same circumference are equal. For this reason, it is possible to calculate the anisotropy evaluation value by the information processing method described above.

  In step S704, the size of the radius r0 is not particularly limited and may be arbitrarily determined. In the present embodiment, the example of the point light source shown in FIG. 2 is described as the light source, but the present invention is not limited to this. If the shape of the light source is a rotating body centered on the optical axis, such as a sphere or a circle, observation is performed at two arbitrary points on the same circumference of the measurement surface, satisfying the relationship between Equation (15) and Equation (16). The first to third variables of BRDF to be equalized. For this reason, it is possible to calculate the anisotropy evaluation value by the information processing method described above.

  When the optical system shown in FIGS. 2 and 8 is used, the image data acquired by the image acquisition unit 601 and the image data for calibration include a shielding portion where the light source covers the measurement target object. In order to avoid this, a half mirror 901 may be used in the optical system as shown in FIG. As a result, the image data and the calibration image data can be acquired without being blocked by the light source.

  Further, for the purpose of eliminating the shielding portion where the light source covers the measurement target object, a ring light source 1001 shown in FIG. 10 may be used as an example of a rotating body centered on the optical axis. In this case, since the imaging apparatus 201 can image the measurement target object from the hollow portion of the ring light source, the image acquisition unit 601 can acquire image data and calibration image data without being shielded by the light source.

  In this embodiment, the calibration image data is acquired by measurement, but the present invention is not limited to this. For example, the image data for calibration previously recorded in the hard disk drive (HDD) 505 may be acquired by reading through the HDD interface (I / F) 504.

  If the information processing method described above is used, the anisotropy evaluation value of the measurement target object can be calculated. Further, by measuring BRDF by a method according to the calculated anisotropy evaluation value, it is possible to measure BRDF at high speed.

n normal vector, t tangent vector, bn binormal vector,
wi direction vector (incident direction vector), wo direction vector (exit direction vector)

Claims (10)

An information processing apparatus that evaluates the anisotropy of reflection characteristics of an object, and that captures the object using an imaging device and a light source disposed on the optical axis of the imaging device to obtain image data. Means (605),
Calibration image data acquisition means for acquiring calibration image data for calibrating the imaging device and the light source (604);
(606) a reflectance map calculating means for calculating a reflectance map of the object based on the image data and the calibration image data;
(607) anisotropy evaluation value calculation means for calculating the anisotropy evaluation value for the object using a reflectance corresponding to a circumference around the optical axis of the imaging device in the reflectance map;
An information processing apparatus comprising: The information processing apparatus according to claim 1, wherein the anisotropy evaluation value calculation unit calculates a tangent vector dependency of the reflection characteristic of the object as the anisotropy evaluation value. The information processing apparatus according to claim 1, wherein the anisotropy evaluation value acquisition unit calculates based on a variance of reflectance on the circumference. The information processing apparatus according to any one of claims 1 to 3, wherein a normal line of the measurement surface of the object is parallel to the optical axis. The information processing apparatus according to claim 1, wherein the shape of the measurement surface of the object is a rotating body centered on the optical axis. The information processing apparatus according to claim 1, wherein the light source is a point light source. The information processing apparatus according to claim 1, wherein the shape of the light source is a rotating body centered on the optical axis. 8. The calibration image data acquisition means captures a calibration object with a known reflectance using the imaging device and the light source, and acquires calibration image data. The information processing apparatus according to any one of claims. 8. The information processing apparatus according to claim 1, wherein the calibration image data acquisition unit acquires the calibration image data by reading data stored in the storage device. The said information processing apparatus has the reflection characteristic measurement means (603) which measures the said reflection characteristic of the said object based on the said anisotropy evaluation value, The one of Claim 1 thru | or 9 characterized by the above-mentioned. The information processing apparatus described.