JP2019153931A - Measuring device, method for setting parameter for color measurement conversion in measuring device, and industrial product inspected by measuring device - Google Patents

Abstract

To provide a measuring device that can photograph an object to be measured a plurality of times with a plurality of setting conditions and accurately convert photographic data into tristimulus values that are color information corresponding to the color sensitivity characteristics of a human for each of the setting conditions.SOLUTION: A measuring system 100 measuring an object to be measured comprises: one or more illumination units 11, 12 that can irradiate the object to be measured with light; one or more photographing units 21 that photograph the object to be measured that is irradiated with light; and an information processing apparatus that includes a conversion unit converting the photographed images into tristimulus values. The illumination units and photographing units photograph the object to be measured a plurality of times with a plurality of setting conditions in which the illumination angles of the illumination units and/or the photographing angles of the photographing units are changed, and in the conversion unit, a condition to convert into the tristimulus values is different for each of the setting conditions.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a measuring device for measuring appearance characteristics, a method for setting a colorimetric conversion parameter in the measuring device, and an industrial product inspected by the measuring device.

  The appearance of the industrial product, that is, the texture, is an important factor that affects the willingness to purchase. Therefore, it is important to evaluate the texture in order to manage and improve the quality of the texture, but there is a risk that if a person visually evaluates the texture, it will vary. Therefore, a measuring instrument that digitizes the texture is required.

  When a person looks at the appearance of a product, the color and gloss change depending on the viewing angle, so a person observes the product at various angles. Therefore, to quantify the texture, the product must be measured under a plurality of angle conditions. For this reason, colorimeters that can measure by changing the illumination angle and the light receiving angle are on the market. However, the colorimeter can measure only an average color within the measurement diameter of a minute point, and cannot measure in-plane texture information. The texture information here indicates a pattern on the sample surface, a rough appearance, or a bright appearance. Without texture information, the texture cannot be quantified. Therefore, it is necessary to photograph with a camera.

  If you try to measure with a combination of multiple lighting and color camera, RGB data is tristimulus value XYZ corresponding to human sensitivity or L * a * b * color system L * a * b * Need to be converted to However, no matter how much the RGB data is converted to XYZ or L * a * b * using a general conversion formula and the texture is quantified, RGB is not compatible with human sensitivity, so it is misaligned with human senses. It becomes a value.

  Therefore, as a method of converting RGB data into XYZ corresponding to human color sensitivity, Patent Document 1 discloses a multi-grid 3D-LUT (for color conversion) acquired in advance for the purpose of converting RGB data into XYZ data. A conversion method using Look Up Table) is disclosed.

  However, when using the method of Patent Document 1 to measure a sample having a strong angle dependency such as a glossy finish that changes color depending on the illumination angle or the light receiving angle, a color conversion table at a certain angle. If only this is adopted for other angles, the conversion error becomes large.

  Therefore, in view of the above circumstances, the present invention can capture a plurality of times under a plurality of setting conditions and accurately convert the captured data into tristimulus values that are color information corresponding to human color sensitivity characteristics for each setting condition. The purpose is to provide a measuring device.

In order to solve the above problems, in one embodiment of the present invention,
A measuring device for measuring a measurement object,
One or more illumination units capable of irradiating the measurement object with light; and
One or more imaging units for imaging the measurement object irradiated with the light;
A conversion unit that converts the captured image into tristimulus values,
The illuminating unit and the photographing unit photograph a plurality of times under a plurality of setting conditions in which the illumination angle of the illuminating unit or / and the photographing angle of the photographing unit is changed,
The conversion unit provides a measurement device characterized in that conditions for converting to the tristimulus values differ for each set condition.

  According to one aspect, the measurement device can capture a plurality of times under a plurality of setting conditions, and can accurately convert the captured data into tristimulus values that are color information corresponding to human color sensitivity characteristics for each setting condition. .

The external appearance characteristic measurement system of 1st Embodiment of this invention. The example of the hardware block diagram contained in the external appearance characteristic measurement system of FIG. 2 is an example of a functional block diagram of the information processing apparatus in FIG. 2 is an overall flow showing a color measurement process of the first embodiment of the present invention. The figure explaining the difference in reflection by the material of a measuring object. Explanatory drawing which shows the state which initializes the parameter for colorimetric conversion at the time of shipment. Initial setting flow of colorimetric conversion parameters. Update flow for colorimetric conversion parameters. The figure which shows the value which converted the RGB data of the shade of 12 color tiles used to XYZ with the color conversion formula of the parameter produced on the shade conditions. The figure which shows the value which converted the RGB data of the shade of 12 color tiles used to XYZ by the color conversion formula of the parameter produced on highlight conditions. 12 is a detailed flow for creating composite RGB data in the color measurement process in the second control example of the present invention. 1 is a schematic configuration example of an integrated appearance characteristic measuring device of the present invention. The external appearance characteristic measurement system of 2nd Embodiment of this invention. The external appearance characteristic measurement system of the modification of 2nd Embodiment of this invention. Schematic of the spectroscopic part of the spectroscopic camera included in the appearance characteristic measurement system of the third embodiment of the present invention.

  Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. In the following drawings, the same components are denoted by the same reference numerals, and redundant description may be omitted.

[First Embodiment]
The configuration of the appearance characteristic measuring apparatus according to the first embodiment of the present invention will be described with reference to FIGS.

  The appearance characteristic measuring apparatus of the present invention irradiates light on a surface of an object at a plurality of angles, images the surface of the object with an imaging unit, and captures the captured image data for each illumination angle or each light reception angle in advance. Using the acquired color conversion formula, it is converted into XYZ data (colorimetric values) and a value indicating the texture and output. A configuration for realizing this function will be described.

  FIG. 1 is an overall schematic diagram of an appearance characteristic measuring system 100 according to the first embodiment of the present invention. FIG. 2 is a hardware block diagram of the appearance characteristic measurement system 100 according to the first embodiment.

  As shown in FIGS. 1 and 2, an appearance characteristic measuring system 100 that is a measuring device according to the present embodiment includes a light source device 1, a photographing device 2, an examination table 3, an information processing device 4, and a monitor device 5. With.

  In the present embodiment, the light source device 1 includes two illumination units so as to irradiate the sample S, which is a measurement object, placed on the examination table 3 with at least two illumination angles. 11 and 12 are provided.

  In the present embodiment, a high color rendering type surface-mounted white LED (light emitting diode) is used for the illumination units 11 and 12. The color rendering index Ra is 95 or more. Since general LEDs have a special spectrum shape, the color rendering is low, and the colors are different from those seen under sunlight (natural light), so accurate colors cannot be captured. Therefore, in the present invention, color conversion accuracy can be improved by using high color rendering LEDs.

  The 1st illumination part 11 is arrange | positioned in the position rotated 15 degrees from the angle of the regular reflection direction with respect to the imaging | photography part 21. FIG. The 2nd illumination part 12 is arrange | positioned in the position rotated 45 degrees from the angle of the said regular reflection direction.

  Therefore, the 1st illumination part 11 is irradiating the sample S with the light from the regular reflection vicinity with respect to the imaging | photography part 21 (highlight conditions), and the 2nd illumination part 12 is from a diffusion direction. Light is being irradiated (shade condition).

  In this example, the highlight condition and the shade condition will be described. However, the first setting condition for irradiating the sample S with illumination light from the first angle with respect to the angle of the imaging unit, and the imaging unit One or more illumination units so that photographing can be performed under two kinds of setting conditions, a second setting condition for irradiating the sample S with light of illumination from a second angle different from the first angle with respect to the angle And at least one photographing unit may be disposed.

  The imaging device (imaging device) 2 includes an imaging unit (camera) 21 and acquires image data (RGB: Raw data) of the sample S placed on the examination table 3 by imaging.

  In the present embodiment, the illumination units 11 and 12 and the camera 21 are supported by an arc-shaped base plate 8.

  In the present embodiment, the camera of the photographing unit 21 is a Bayer array RGB type camera. Due to the Bayer arrangement, the photodiodes are arranged so that the rows with R-G (red-green) filters and the rows with G-B (green-blue) filters alternate.

  In addition, the imaging unit 21 is configured so that, for example, the surface of the sample S of about several tens mm × several tens mm (for example, 50 mm × 50 mm) can be imaged at a time as a measurement size.

  Furthermore, the photographing unit 21 is configured by a camera that can acquire RGB in 10 bits each. For example, the photographing unit 21 adjusts the focus and working distance of the camera so that the resolution of the photographed image data is 20 μm per pixel.

  The information processing apparatus 4 has functions of colorimetric value conversion and texture calculation that calculate colorimetric values for each setting condition based on image data (RGB: Raw data).

  1 shows a configuration in which the information processing device 4 is provided away from the illumination unit and the photographing unit, the functions of the color calculation unit performed by the information processing device 4 are the same as those of the illumination unit and the photographing unit. It may be provided as an apparatus by being provided in an integrated housing. The integrated configuration will be described later with reference to FIG.

  Alternatively, the function of the colorimetric processing unit (information processing device) that operates as the color calculation unit may be executed by a calculation device (information processing device) such as a separate computer that is completely independent of the illumination unit and the imaging unit. .

  The monitor device 5 displays information on the captured image, tristimulus values, and texture.

  In the present embodiment, by providing a plurality of illuminations (illumination units 11 and 12), the sample S can be irradiated with light from at least two illumination angles. In this configuration, although irradiation can be performed from two illumination angles, light is irradiated from one direction at a single illumination angle for each photographing, not from two directions at a time.

  In this example, the sample S is shown as being placed on the inspection table 3. However, for example, the inspection table 3 may be a conveyor belt. In this case, for example, the sample S, which is an industrial product conveyed in the front or depth direction of FIG. 1 by the conveying belt, is temporarily stopped and photographed in a plurality of irradiation directions or a plurality of photographing directions by the appearance characteristic measurement system shown in FIG. This makes it possible to inspect the color (colorimetric value) and texture of the industrial product during manufacturing.

  An industrial product is a product processed from a metal material, a non-metal material, a material combining both, and the like, and particularly refers to a product made by surface treatment. As an example of industrial products, automobiles including two and four wheels, vehicles such as railways, sheet metal used in vehicles, interiors such as seat seats and dashboards in automobiles, aircraft, ships, building materials and construction using them Products, imaging devices, information processing devices such as personal computers, mobile terminals such as smartphones and tablets, home appliances such as watches, TVs, refrigerators and air conditioners, and cooking equipment such as dishes and pans. Any industrial product that can be used can be measured.

  Referring to FIG. 2, the light source device 1 includes a first illumination unit 11 that is a plurality of illumination units, a second illumination unit 12, and a lighting control unit 13 that drives each of the illumination units 11 and 12 to be lit. Have. In addition, in FIG. 2, although the lighting control part 13 is common and 1 case is shown, you may provide the lighting illumination part in each of the illumination parts 11 and 12. FIG.

  The imaging device 2 includes one imaging unit (camera) 21 and an image processing unit 22, and corresponds to two irradiation angles (illumination angles) of the illumination units 11 and 12 of the light source device 1 fixed at each angle. Then, each is acquired by one shooting operation (one shot).

  Further, as the information processing apparatus 4, a general computer apparatus can be used. Specifically, the computer device may be dedicated within the appearance characteristic measuring device of the present application, or an external computer may be used for colorimetric value conversion by reading a colorimetric value conversion program.

  Referring to FIG. 2, the information processing apparatus 4 includes a CPU (Central Processing Unit) 41, a ROM (Read Only Memory) 42, a RAM (Random Access Memory) 43, and an HDD (Hard Disk Drive) 44. Further, the information processing apparatus 4 includes various interfaces (I / F) 45, an I / O controller, and an input / output interface (I / O) 46. The CPUs 41 to I / O 46 are connected to each other via a bus line 48.

  The HDD 44 measures the surface of the measurement object such as a sample or a patch using the acquired RGB / Raw data together with a program for shooting control of the shooting device 2 and light source lighting control of the light source device 1. A colorimetric value conversion program for performing color value conversion and the like, and a texture calculation program are stored.

  As the monitor device 5, for example, a liquid crystal monitor device can be used. The monitor device 5 can display a setting menu, an operation menu, and the like in addition to shooting data and calculation results. In addition, the monitor device 5 has an RGB image, an RGB numerical value, a calculated tristimulus value XYZ numerical value, an L * a * b * numerical value, an L * a under each of the photographed highlight condition and shade condition. It is possible to display various graphs and images for reference that can be created from the variance value of * b * and each numerical value.

  For example, as a graph or image for reference, a chromaticity diagram corresponding to tristimulus values XYZ, a coordinate position in L * a * b * color space based on L * a * b * values, L * a * b * You may display a simulation example of how the color appears under each specified light source based on the color value.

<Information processing device>
FIG. 3 shows a functional block diagram of the information processing apparatus of the appearance characteristic measuring apparatus according to the embodiment of the present invention. Note that functions used in the present embodiment are indicated by solid lines, and those related to control examples described later are indicated by dotted lines.

  FIG. 3 shows a functional block diagram of each function realized by the CPU 41 operating according to the colorimetric value conversion program. FIG. 3 shows the functions of the part of the information processing apparatus 4 relating to the colorimetric processing unit of the present invention.

  The colorimetric value conversion program may be provided by being recorded in a computer-readable recording medium such as a CD-ROM or a flexible disk (FD) in an installable or executable file format. . Further, the program may be provided by being recorded on a computer-readable recording medium such as a CD-R, a DVD (Digital Versatile Disk), a Blu-ray Disc (registered trademark), or a semiconductor memory. The colorimetric value conversion program may be provided by being installed via a network such as the Internet. The colorimetric value conversion program may be provided by being incorporated in advance in a ROM or the like in the device.

  The information processing apparatus 4 includes a data input unit 80A, an illumination control unit 81, a photographing control unit 82, an image data storage unit 83, a calculation data storage unit 84, a colorimetric conversion parameter update unit 85, and a measurement unit. A color value calculation unit 86, a texture calculation unit 87, a measurement data storage unit 88, a communication unit 89, and a monitor output unit 80B can be realized.

  The CPU 41 shown in FIG. 2 has the functions of the illumination control unit 81, the shooting control unit 82, the colorimetric conversion parameter update unit 85, the colorimetric value calculation unit 86, and the texture calculation unit 87 of FIG. Realized in software. The description will proceed as being realized as software. Among the illumination control unit 81, the imaging control unit 82, the colorimetric conversion parameter update unit 85, the colorimetric value calculation unit 86, and the texture calculation unit 87, You may implement | achieve part or all by hardware.

  Further, the image data storage unit 83, the calculation data storage unit 84, and the measurement data storage unit 88 are realized by any of the HDD 44, the ROM 42, the RAM 43, and the EEPROM shown in FIG.

  The data input unit 80A, the communication unit 89, and the monitor output unit 80B are realized by any one of various interfaces (I / F) 45, an I / O controller, an input / output interface (I / O) 46, and the like.

  With reference to FIG. 3, the illumination control unit 81 has a function of controlling lighting / extinguishing by alternatively selecting an illumination unit.

  The imaging control unit 82 has a function that allows the imaging unit to perform imaging at a predetermined timing after the illumination is turned on.

  The image data storage unit 83 has at least functions of a highlight image storage unit 831 and a shade image storage unit 832 that store captured images for colorimetric value calculation.

  The calculation data storage unit 84 stores data referred to by the colorimetric value calculation unit and the texture calculation unit.

  Specifically, the calculation data storage unit 84 includes a noise processing data storage unit 841, an RGB composition data storage unit 842, a highlight calculation conversion formula storage unit 843, a shade calculation conversion formula storage unit 844, An L * a * b * calculation data storage unit 845, a conversion source data storage unit 846, and a fixed true value storage unit 847 are provided.

  The colorimetric conversion parameter update unit 85 updates the parameters substituted in the conversion formula (or conversion table) stored in the highlight calculation conversion formula storage unit 843 and the shade calculation conversion formula storage unit 844. The parameter update will be described later with reference to FIGS.

  The calorimetric value calculation unit (conversion unit) 86 includes at least a calibration processing unit 861, a demosaicing processing unit 862, and a tristimulus value XYZ calculation unit 864.

  The calibration processing unit 861 performs calibration for correcting image distortion caused by the lens of the photographing unit 21.

  The demosaicing processing unit 862 performs demosaicing to convert a Bayer array raw image into a general RGB array.

  In the case of a highlight image, the tristimulus value XYZ calculator 864 uses the highlight conversion formula (color conversion formula, colorimetric value conversion formula) stored in the highlight calculation conversion formula storage section 843, Convert RGB data to tristimulus values XYZ.

  In the case of a shade image, the tristimulus value XYZ calculation unit 864 converts each pixel into a tristimulus value XYZ using the shade conversion formula stored in the shade calculation conversion formula storage unit 844.

  The colorimetric value calculation unit 86 may be provided with a composite RGB data creation unit (RGB data synthesis unit) 863. The creation of the combined RGB data will be described later with reference to FIG.

  The texture calculation unit 87 includes an L * a * b * calculation unit 871 and an L * a * b * variance value calculation unit 872.

  Based on the tristimulus value XYZ calculated by the tristimulus value XYZ calculation unit 864, the L * a * b * calculation unit 871 digitizes the color for each pixel in the highlight condition and the shade condition. The L * a * b * color value that is the value is calculated.

  The L * a * b * variance value calculation unit 872 calculates the variance value of the L * a * b * color value and uses it as texture data.

  The measurement data storage unit 88 stores the tristimulus values XYZ for each pixel calculated by the colorimetric value calculation unit 86, the L * a * b * color value and the variance value for each pixel calculated by the texture calculation unit 87. Accumulate and remember.

  The communication unit 89 transmits the measurement data to another device (for example, another information processing device) connected by wire or wireless. Moreover, it communicates with the host system before shipment shown in FIG.

  The monitor output unit 80B outputs colorimetric data such as a photographed image, tristimulus values XYZ, L * a * b * color values, and L * a * b * variance values according to the display format on the monitor device 5. To do.

<First colorimetric value calculation process>
FIG. 4 is an overall flow showing the colorimetric value calculation process of the first embodiment of the present invention.

  In S1, the sample S is set on the inspection table 3.

  In S2, the first illumination unit 11 is turned on.

  In S <b> 3, the sample S is photographed by the photographing unit 21. With this photographing, the photographing unit 21 obtains an RGB image (Raw data) of the sample S at the illumination angle under the highlight condition. The photographed data is stored in the highlight image storage unit 831 as raw data in which luminance information is described.

  In S4, the first illumination unit 11 is turned off and the second illumination unit 12 is turned on.

  In S <b> 5, the sample S is photographed by the photographing unit 21. With this photographing, the photographing unit 21 obtains an RGB image (Raw data) of the sample S at the illumination angle of the shade condition. The captured data is stored in the shade image storage unit 832 as raw data in which luminance information is described.

  In step S6, the calibration processing unit 861 performs data noise reduction processing and smoothing processing, and then performs image calibration using measurement data of the white reference plate.

  In S7, the demosaicing processing unit 862 performs a demosaicing process on the acquired data using, for example, an adaptive color brain interpolation method. Note that the order of the correction processes in S6 and S7 may be reversed.

  Two types of highlight RGB luminance data and shade RGB luminance data are acquired by the above procedure.

  In S8, the tristimulus value XYZ calculation unit 864 converts the highlight RGB luminance data into tristimulus values using a color conversion formula set for the highlight condition, and creates highlight XYZ data. Further, the tristimulus value XYZ calculation unit 864 converts the shade RGB luminance data into tristimulus values using a color conversion formula set for the shade condition, and creates shade XYZ data. S8 is the main process, and the colorimetric value calculation process including the correction processes of S6 and S7 is used.

  Next, in S9 and S10, the texture of the sample is digitized based on the above XYZ data (texture digitization process).

  In S9, the highlight XYZ data and shade XYZ data are converted to L * a * b * by the conversion formula formulated by the International Commissioner for Illumination (CIE).

  Although XYZ data may be used to quantify the texture, XYZ is not linear in human color sensation, so in this example it is converted to the L * a * b * color system and then the texture Perform arithmetic processing. In S9, the conversion from the XYZ image to the L * a * b * image is performed using the following conversion formula (Formula 1) defined by the CIE.

In (Expression 1), X ₀ , Y ₀ , and Z ₀ indicate tristimulus values on the reference white surface.

  Here, as a kind of texture, there is a “glitter” peculiar to a painted surface containing a glittering material. An example of digitizing it will be described. A strong feeling also means that the colors in the image are fluctuating.

  Therefore, in S10, the variance value of L * a * b * in each of the highlight L * a * b * data and shade L * a * b * data is calculated. The glitter feeling is quantified by the dispersion of L *, a *, and b * of all the pixels. In the present embodiment, the four arithmetic operations of the variance value of L * a * b *, for example, the multiplication value, are made to sparkle.

  If the object to be measured is solid, the chromaticity does not change, so the multiplication value is small. On the other hand, if it is sparkling, the chromaticity fluctuation in the image is large, so the multiplication value is large. In this way, it is possible to evaluate the feeling of glitter with the amount of change in luminance.

  In step S11, the variance value of L * a * b * is output as a texture.

  In this flow, in S3 and S5, a measurement object is photographed for each illumination angle, and in S8, luminance data for each illumination angle is acquired in advance for each illumination angle condition (highlight condition, shade condition). It is converted into XYZ data (colorimetric values) and values indicating the texture. Therefore, it is possible to accurately convert the photographing data into tristimulus values XYZ that are color information corresponding to human color sensitivity characteristics.

  Further, by converting the tristimulus values XYZ into color values of the L * a * b * color system in S9, it is possible to digitize the accurate texture.

  Furthermore, by calculating the dispersion values of L *, a *, and b * in S10, it is possible to quantify the “glitter” that is a characteristic of the particle including the color of the particle as a texture.

<Texture of measurement object>
The glittering feeling of the measurement object will be described below with reference to FIG. FIG. 5 is a diagram for explaining the difference in reflection depending on the material of the measurement object.

  FIG. 5A shows light reflection in a general printed matter. General printed matter has low angle dependency. That is, even if the printed material is observed at various angles, the color does not change greatly. This is because the printed material has irregularities on the printing surface, and light diffuses on the surface.

FIG. 5B shows the reflection of light in metallic or pearl. The painted surface such as metallic or pearl having a high gloss shown in FIG. 5B has a large angle dependency for the following two reasons.
(1): The surface is smooth.
(2): Bright material is included.

  (1) will be described. When the surface of the object to be measured is smooth, light is reflected most strongly under the conditions of specular reflection (when the incident angle and the light receiving angle are the same), whereas the diffuse reflection direction (direction other than regular reflection) is light. The amount of reflection becomes smaller. For this reason, for example, in photographing a measurement object, highlights (a condition in which light is strongly applied) become extremely bright, but in shades (conditions in which light is weakly applied), the color rapidly becomes dark.

  (2) will be described. Metallic or pearl painted surfaces (for example, car painted surfaces) applied to the surface of the object to be measured contain aluminum flakes called luster or mica flakes, so the brightness and color vary greatly depending on the angle. . The reflection of aluminum flakes and mica flakes will be described below.

  FIG.5 (c) shows the example of the reflection of the light in an aluminum flake. The coating containing aluminum flakes on the surface of the object to be measured increases the angle dependency of light entering the paint layer. This is because light is strongly reflected by the aluminum metal in the paint layer.

  FIG.5 (d) shows the example of the reflection of the light in a mica flake. The color of the surface of the object to be measured containing mica flakes changes with respect to the reflection angle. This is because light interferes with mica flakes.

  In this embodiment, since the camera of the imaging unit 21 is color, color information can be obtained. For this reason, it is possible to acquire, with RGB data, characteristics of glittering colors such as a pearl painted surface containing mica flakes.

  However, since the metallic / pearl coating surface has a large angle dependency, the color changes greatly depending on the illumination angle. Therefore, if only the RGB → XYZ color conversion formula created at one angle and color conversion at another angle is performed, the conversion error becomes large.

  Therefore, in the present invention, a color conversion formula is prepared for each illumination angle or light receiving angle, and a separate color conversion formula is used for each corresponding angle, thereby reducing the conversion error.

The following two points are indispensable for quantifying the texture such as the above-mentioned “glitter” from the image of the measurement object.
(A): It is necessary to consider the color. The color must be a device-independent color corresponding to human color sensitivity characteristics, for example, tristimulus values XYZ or L *, a *, b *.
(B): The color value in (A) must be an exact XYZ value. This is because when the error is large, the numerical value of the texture also has an error. For this reason, when shooting using an RGB camera, it is necessary to minimize the conversion error.

  Therefore, in step S8 in the flow shown in FIG. 4, a color conversion equation is set in advance before shipment so that the conversion from the RGB luminance data to the tristimulus value XYZ value can be performed with high accuracy. Specifically, parameters to be substituted in a conversion formula or conversion table corresponding to “conditions for converting to tristimulus values” for converting RGB data into tristimulus values XYZ that are colorimetric values are set in advance. . The setting of the parameter (colorimetric conversion parameter) will be described below.

<Setting of colorimetric conversion parameters>
FIG. 6 is an explanatory diagram showing a state in which the colorimetric conversion parameters are initially set at the time of shipment. As a preparation for setting the colorimetric conversion parameters after the appearance characteristic measurement system 100 is manufactured and before shipment, the spectrophotometer 6 is used in advance to satisfy the illumination angle condition of FIG. Acquire corresponding tristimulus values XYZ of 15 ° (highlight) and 45 ° (shade).

  Then, the tristimulus values XYZ acquired using the spectrocolorimeter are set as fixed true values, and the parameters in the color conversion equation are set so as to match the values as much as possible.

  Specifically, the tristimulus values XYZ acquired using the spectrocolorimeter 6 are stored in the host system 9 in advance. And in each appearance characteristic measurement system 100 (A, B), after assembling manufacture of the imaging | photography part 21 and the illumination parts 11, 12, it is common in several appearance characteristic measurement system A, B ... as a setting before shipment, The tristimulus values XYZ acquired using the spectrocolorimeter 6 are stored in the information processing apparatus 4 (or the control unit 74 or the storage unit 75 in FIG. 12) as a fixed true value.

  For the setting of the colorimetric conversion parameters, both the measurement with the spectrocolorimeter 6 and the measurement with the appearance characteristic measurement system 100 (A, B) of the present invention are measured using a plurality of color patches. In this example, an example in which a CCS II color tile set including 12 ceramic tiles is used as an example of a solid color patch will be described.

  Here, the solid color patches may include 12 color tiles having a reflectance of less than 100%, but the plurality of colors include color tiles having a reflectance of 100% or more or a brightness of 100 or more. And more preferable. This is because the reflection brightness of the coated surface including the glittering material is very high, so that color conversion accuracy when a sample having a high reflectance is measured is improved by performing color calibration including a patch having a high reflectance.

  In order to optimize the parameters of the color conversion equation, it is preferable to measure eight or more colors as a solid color patch.

  In addition, as an example of the spectrocolorimeter 6, BYK maci of BYK Gardner is used, and the XYZ values are obtained from the spectral reflectance measured with the spectrocolorimeter under the conditions of the color matching function of the D65 light source and the 10 degree field of view. Calculated.

  FIG. 6 shows an example in which the colorimetric values of the spectrocolorimeter 6 are stored in the plurality of appearance characteristic measuring devices A and B via the host system 9. The colorimeter 6 and the appearance characteristic measuring apparatuses A and B may be connected to each other, and the colorimetric values acquired by the spectrocolorimeter 6 may be directly transmitted to the appearance characteristic measuring apparatuses A and B and stored.

  FIG. 7 shows an initial setting flow of the colorimetric conversion parameters.

  FIG. 7 is a process for setting in advance the parameters in the calorimetric value calculation of S8 in FIG. 4, and this setting is performed after manufacture and before shipment.

  In S801, the original data of the conversion formula is stored in advance in the storage unit (conversion formula original data storage unit 846).

  (Formula 2), which is an example of conversion formula original data used for conversion, is shown below. a1 to a7 are parameters. (Expression 2) is an expression for obtaining XYZ using RGB values.

In this embodiment, an example using (Expression 2) will be described, but a primary or tertiary conversion expression may be used. Further, when the RGB values are all 0, the constant term (a7 in the case of (Expression 2)) may be set to 0 so that the XYZ value becomes 0.

  In step S <b> 802, the spectrocolorimeter 6 measures the tristimulus values XYZ of the 12 color patches under the same angle conditions as the highlight conditions.

  In step S803, the spectrocolorimeter 6 measures the tristimulus values XYZ of the 12 color patches under the same angle conditions as the shade conditions.

  In S804, the tristimulus values XYZ measured in S802 and S803 are stored as fixed true values in the fixed true value storage unit 847 of the appearance characteristic measurement system 100.

  In step S <b> 805, an RGB image (raw data) of 12 color patches is acquired by the photographing unit 21 of the appearance characteristic measurement system 100 at the illumination angle under the highlight condition.

  In step S806, highlight RGB luminance data is created based on the raw data acquired in step S805.

  At this time, an image area at the center of the RGB data (Raw data) is extracted, and average R, G, and B values in the area are obtained and used as highlight RGB luminance data. Since the measurement range of the image captured by the photographing unit (camera) 21 of the appearance characteristic measurement system 100 is wider than the range of the minute diameter measured by the spectrocolorimeter 6, strictly speaking, each other than the central portion of the image, This is to cause a deviation from the illumination angle condition. In this example, 128 × 128 pixels at the center are extracted and averaged. Note that, before performing the above average value extraction, as shown in FIG. 4, noise processing, calibration processing, and demosaicing processing may be performed.

  In step S <b> 807, an RGB image (raw data) of 12 color patches is acquired by the imaging unit 21 of the appearance characteristic measurement system 100 at the illumination angle of the shade condition.

  In step S808, shade RGB luminance data is created based on the raw data acquired in step S807. At this time, similarly to S806, an image area at the center of the RGB data (Raw data) is extracted, and average R, G, and B values in the area are obtained to obtain shade RGB luminance data.

  In step S809, the colorimetric conversion parameter update unit 85 reads the conversion formula (formula 2) stored in step S801 and the fixed true value stored in step S804.

  In S810, the colorimetric conversion parameter update unit 85 defines the tristimulus values XYZ converted from the highlight RGB data created in S806 using a conversion formula in S804 for the highlight condition. Set the parameter so that it approaches the fixed true value of the tristimulus values XYZ.

  Specifically, the parameters a1 to a7 in the color conversion equation (Formula 2) are set by calculation based on the 12-color highlight RGB luminance data and the 12-color highlight XYZ data measured by the colorimeter. .

  For example, the X, Y, and Z values under 12 highlight conditions measured by the spectrocolorimeter 6 are objective variables, and the R, G, and B values under 12 highlight conditions measured by the photographing unit 21 are described. Parameters a1 to a7 are calculated as variables by the least square method.

  The conversion formula into which the set parameters a1 to a7 are substituted is stored in the highlight calculation conversion formula storage unit 843 as a highlight calculation conversion formula.

  In step S811, the colorimetric conversion parameter update unit 85 converts the tristimulus values XYZ converted from the shade RGB data created in step S808 using a conversion formula into the tristimulus defined in step S804 for the shade condition. Set the parameter so that it approaches the fixed true value of value XYZ.

  Specifically, the parameters a1 to a7 in the color conversion equation (Equation 2) are set by calculation based on the shaded RGB luminance data of 12 colors and the shaded XYZ data of 12 colors measured by the colorimeter. Similarly to S810, for example, X, Y, and Z values under 12 shade conditions measured by the spectrocolorimeter 6 are objective variables, and R, G, and B values under 12 shade conditions measured by the photographing unit 21 are used. Are the explanatory variables, and parameters a1 to a7 are calculated by the least square method.

  The conversion formula into which the set parameters a1 to a7 are substituted is stored in the shade calculation conversion formula storage unit 844 as a shade calculation conversion formula.

  As described above, using the obtained XYZ data of the spectrocolorimeter and the RGB data measured by the imaging unit (camera) of the appearance characteristic measuring device of the present invention, each illumination angle (highlight condition, shade condition) By determining and storing the color conversion formulas in, the RGB data captured are converted into XYZ data using these conversion formulas at the time of measurement by the appearance characteristic measuring apparatus.

  Therefore, in the appearance measuring apparatus of the present invention, accurate XYZ data close to the value measured by the spectrocolorimeter can be measured for each pixel with respect to an image having a wider measurement range than the spectrocolorimeter. .

  In this embodiment, as an example of a solid color patch, a color conversion formula is created from 12 color tiles. For example, a conversion table from RGB to XYZ is created from 100 or more color patches. The conversion method may be used.

  FIG. 8 shows a flow for updating the colorimetric conversion parameters.

  Even if the parameters of the colorimetric conversion equation are approximated so that they are close to the values of the spectrophotometer in the initial setting, when the illumination deteriorates due to long-term use, or when the illumination unit or the imaging unit lens is replaced In addition, since it is assumed that the actual value deviates from a fixed true value, when such an event occurs, it is desirable to update it appropriately.

  A flow for updating is shown in FIG. As described above, START in FIG. 8 occurs when the conversion formula needs to be updated due to deterioration of illumination due to long-term use or replacement of the LED of the illumination unit or the lens of the photographing unit. Compared to FIG. 7, since the fixed true value has already been measured at the time of shipment, only the parameter update using the solid color patches in the target appearance characteristic measurement device is performed at the time of update. .

  S901 to S907 in FIG. 8 are substantially the same as S805 to S811 in FIG.

  In S906, the tristimulus value XYZ converted using the conversion formula from the highlight RGB data measured this time is fixed true of the tristimulus value XYZ previously defined in S804 at the time of shipment with respect to the highlight condition. Update the parameter to approach the value. The conversion formula into which the updated parameters a1 to a7 are substituted is updated and stored in the highlight calculation conversion formula storage unit 843 as a highlight calculation conversion formula.

  In S907, the tristimulus value XYZ converted using the conversion formula from the shade RGB data measured this time is fixed true of the tristimulus value XYZ defined in S804 at the time of shipment in advance for the shade condition. Update the parameter to approach the value. The conversion formula into which the updated parameters a1 to a7 are substituted is updated and stored in the shade calculation conversion formula storage unit 844 as a shade calculation conversion formula.

  In this way, by updating the colorimetric conversion equation or conversion table parameter for each set condition, the calculated colorimetric conversion value becomes the true value measured by the spectrocolorimeter even after many years of use. It is possible to maintain a state of outputting close high-precision colorimetric values.

<Verification of colorimetric result>
Next, the results of verifying the effects of the present invention are shown below.

  In order to verify the effect of the present invention, (i) a case where an RGB value photographed under a shade condition is converted into XYZ using a color conversion formula created under the shade condition using the colorimetric value conversion method of the present invention, and (ii) ) Assuming that no color conversion formula is set for each angle condition, compare the error when RGB values shot under shade conditions are converted to XYZ with the color conversion formula created under highlight conditions. did.

  The table shown in FIG. 9 is the result when (i) the RGB data photographed under the shade condition is converted into XYZ by the colorimetric conversion formula created under the shade condition by the colorimetric value conversion method of the present invention.

  | ΔX |, | ΔY |, and | ΔZ | in FIG. 9 and FIG. 10 are the XYZ values of a commercially available spectrophotometer and the XYZ values when the RGB data photographed by the photographing unit 21 are converted by a conversion formula The absolute value of the difference, that is, the error is shown. A commercially available spectrophotometer was BYK maci manufactured by BYK Gardner.

  Note that the image data captured by the imaging unit 21 used for the calculation is an average XYZ value of the center 128 × 128 pixels of the image. The average error in FIG. 9 is as small as 0.7 for X, 0.46 for Y, and 0.49 for Z.

  Next, the table of FIG. 10 shows the result when (ii) RGB data photographed under the shade condition is converted into XYZ by the colorimetric conversion formula created under the highlight condition. The average error of X in FIG. 10 is 12.63, Y is 9.72, and Z is 20.93.

  From the results shown in FIG. 9 and FIG. 10, it can be seen that the conversion accuracy is remarkably improved by creating conversion equations for each illumination angle (for each set condition) as in the present invention.

  In other words, if RGB data measured at other illumination angles is converted into XYZ data using a color conversion formula or color conversion table (Look Up Table (LUT)) created at a specific illumination angle, the error will increase. Can be prevented.

  Also, in this example, the color conversion formula was created using the captured RGB data as it was, but first the white reference plate was measured, and at the time of sample measurement, the captured data was normalized for each pixel with the white data. RGB data after conversion may be used. By performing normalization, it is possible to correct uneven illumination over time.

<Second colorimetric value calculation process>
In the calorimetric value calculation process of FIG. 4, one-shot data is used at each angle. Actually, industrial products that exist in the world exist from very bright samples to dark samples, and with a constant exposure time. When a bright sample is measured, it may saturate beyond the camera's dynamic range. In addition, when a very dark sample is measured, there is a risk that the data will be corrupted.

  Therefore, in the second colorimetric value calculation process, multi-exposure is performed in which exposure is performed for a plurality of exposure times, data obtained by measuring samples is synthesized, and color conversion is performed using the synthesized RGB data. By this method, it is possible to measure from a dark sample to a bright sample.

  Here, a data synthesis method in the colorimetric value calculation process will be described. Here, for ease of explanation, a method under highlight conditions will be described. Data synthesis is performed in the same manner in the shade.

  Three types of exposure time are set in advance. For example, when a reference exposure time is set, a value that is 1/2 times or twice that of the reference exposure time is set. Here, assuming that the reference exposure time is T seconds, the exposure time has the following three types. "1 / 2T seconds, T seconds, 2T seconds"

  FIG. 11 shows a detailed flow for creating composite RGB data in the second colorimetric value calculation process of the present invention. In FIG. 11, an example in which the sample S is photographed and the colorimetric value is measured at the time of measurement will be described. However, a similar flow is performed when measuring a specific color patch when updating the parameters. I do.

  In S101, it is confirmed that the sample S is placed on the inspection table 3.

  In S102, the first illumination unit 11 is turned on.

  In S103, the RGB brightness data of the sample S is acquired with the exposure time 1 / 2T, T, and 2T by the camera at the illumination angle of the highlight condition, and stored in the highlight image storage unit 831.

  Specifically, one sample S is photographed three times of 1/2 T seconds, T seconds, and 2 T seconds with the camera that is the photographing unit 21. The photographed image data is darkest when the exposure time is short for 1 / 2T seconds, and brightest when the exposure time is 2T seconds. Then, the central image area is extracted from the captured image data in the same manner as in S806 in FIG. 7, and the average R, G, and B values in the area are obtained, and 1/2 T seconds, T seconds, and 2 T seconds are obtained. The three types of highlight RGB brightness data are created.

  Next, in S104 to S109, RGB data is synthesized. Data measured in 2T seconds is first data, data measured in T seconds is second data, and data measured in 1 / 2T seconds is third data.

  In S104, the composite RGB data creation unit 863 extracts a pixel position of data that is saturated or close to a saturation value out of the R luminance of all pixel data (first data) with an exposure time of 2T. In this embodiment, since a 10-bit camera is used for the photographing unit 21, the saturation value is 2 to the 10th power of -1, that is, 1023. For example, a case where the luminance is a value of 1000 or more is extracted as a saturated pixel region.

  In S105, the composite RGB data creation unit 863 replaces the pixel position extracted in S104 with the luminance of the pixel data (second data) of the exposure time T. That is, a process of replacing a saturated pixel area with a pixel value that is not saturated is performed.

  In S106, the composite RGB data creation unit 863 multiplies the luminance value of the pixel area of the image data (first data) not extracted in S104 by 1/2. That is, after the replacement, the pixel value of the pixel area that has not been replaced is multiplied by 1/2. This is because the exposure time is normalized to T seconds.

  In S107, the composite RGB data creation unit 863 extracts a pixel position “saturated or close to the saturation value” from the R luminance of the pixel data (second data) of the exposure time T replaced in S105. . For example, a pixel position having a value of 1000 or more in luminance is extracted in the pixel area where the replacement is further performed.

  In S108, the composite RGB data creation unit 863 replaces the pixel position extracted in S107 with the luminance of the pixel data (third data) having the exposure time 1 / 2T.

  In S109, the composite RGB data creation unit 863 doubles the luminance value of the pixel area of the image data (third data) replaced in S108. This normalizes the exposure time.

  Using the composite RGB data created in FIG. 11, the process proceeds to S9 in FIG. 4, where XYZ data is acquired using a conversion formula, and the texture is digitized.

  In this flow, an example is shown in which the luminance value after replacement is doubled or halved for normalization. However, before the replacement, data for exposure times 2T and 1 / 2T are standardized in advance. They may be used to make substitutions.

  In this embodiment, when only R is saturated in the RGB data, processing is performed to replace only R. For example, G and B at the same pixel position are also replaced even if they are not saturated. It is also possible to use a processing method in which The processing method may be determined by the ease of creating a program (algorithm).

  As a pre-preparation for carrying out the above RGB data composition process, even when measuring a plurality of solid color patches and setting / updating the parameters of the conversion equation, it is necessary to use multi-exposure in advance as in FIG. Create RGB data.

  For this reason, when the colorimetric value is obtained by the spectrocolorimeter 6 before shipment, measurement can be performed using a plurality of exposure conditions under the same conditions as the highlight condition and the shade condition so that this process can be performed. The color value is acquired and stored.

  Using the combined RGB data for the solid color patches obtained as described above, as shown in FIGS. 7 and 8, parameters to be substituted into the colorimetric value conversion equation are obtained, and the conversion equation is obtained. create. The same processing is performed under the shade condition, and a color conversion formula under the shade condition is created.

  In this embodiment, by synthesizing data measured by multi-exposure, it is possible to increase the dynamic range in a pseudo manner and create HDR (high dynamic range) data. For this reason, data from a very high gloss and bright sample to a low gloss and low brightness sample can be made in a state where there is no blackening or saturation in the data.

  Therefore, as in the present embodiment, a color conversion formula or color conversion table is created using HDR data for a plurality of color patches, and measurement is performed using the HDR data for a sample. It is possible to improve the accuracy of colorimetric value conversion and the accuracy of digitizing the texture in the external appearance characteristic measuring apparatus.

<Integrated measuring device>
FIG. 12 is a schematic configuration example of the integrated appearance characteristic measuring apparatus of the present invention.

  Although the example which provides the information processing apparatus 4 in the external appearance characteristic measurement system 100 separately from the illumination parts 11 and 12 and the imaging | photography part 21 was demonstrated in FIG. 1, the control part and memory | storage part which perform the information processing function of this structure May be integrated with the illumination unit and the imaging unit.

  The integrated appearance characteristic measuring apparatus 7 shown in FIG. 12 is entirely surrounded by a housing 70, and includes an illumination unit 71, an illumination unit 72, a photographing unit (camera) 73, a control unit 74, a storage unit 75, a display unit 76, and an operation. A button 77 is provided. The integrated appearance characteristic measuring device 7 is an example of a measuring device according to an embodiment of the present invention.

  The integrated appearance characteristic measuring device 7 performs photographing in a state where the casing 70 is in contact with the surface of the sample S.

  In the housing 70, since the holes irradiated with the light from the illumination units 71 and 72 have holes, it is possible to take an image reflected from the sample S outside the holes with the photographing unit 73. It is.

  Also in this configuration, the first illumination unit 71 irradiates the sample S with light from the vicinity of regular reflection (highlight conditions) with respect to the imaging unit 73, and the second illumination unit 72 has a diffusion direction (shade conditions). Irradiate light from.

  The control unit 74 and the storage unit 75 can execute a function related to color measurement of the information processing apparatus 4 illustrated in FIG. 3.

  The display unit 76 is a color display such as an LCD (Liquid Crystal Display) or an organic EL (Electro Luminescence) or a monochrome display. The display unit 76 has the function of the monitor device 5 and can display a setting menu, an operation menu, and the like in addition to shooting data and calculation results.

  The operation button 77 is an input unit for inputting an instruction from the operator.

  In FIG. 12, an illumination unit and a photographing unit are provided on the upper side, and a configuration in which the sample S is placed on the lower side is illustrated. However, if a part of the housing can be brought into contact with the sample, for example, The wall surface may be used as a sample, and the integrated appearance characteristic measuring device 7 may be used in a landscape orientation.

  Also in this configuration, as described above, by creating a conversion equation or a conversion table for each illumination angle condition, it is possible to improve the accuracy in converting captured RGB data into XYZ data.

  As an integrated appearance characteristic measuring apparatus, FIG. 12 shows an example with two illumination sections and one photographing section. However, as shown in FIG. 13 described later, the integrated appearance characteristic measurement apparatus also has two cameras. A configuration may be provided in which one illumination is provided.

Second Embodiment
Next, the appearance characteristic measuring system according to the second embodiment of the present invention will be described with reference to FIG. FIG. 13 is an overall schematic diagram of an appearance characteristic measurement system 100A according to the second embodiment of the present invention.

  In the first embodiment, a plurality of illumination units are provided to measure at a plurality of illumination angles. However, as shown in FIG. 13, a plurality of imaging units (cameras 23 and 24) are provided to provide a plurality of illumination units. You may measure from an imaging angle.

  In the configuration shown in FIG. 13, the first imaging unit (camera) 23 and the second imaging unit (camera) 24 are provided for one illumination unit 14 with respect to the examination table 3 on which the sample S is placed. Are arranged at different shooting angles. Therefore, the imaging device 2A can image the sample on the examination table 3 at two imaging angles. In addition, you may increase the setting number of imaging | photography angle further by increasing the number of cameras to install.

  In the present embodiment, selection of the photographing unit (camera 23 or camera 24) used for photographing is instructed instead of instructing selection of illumination used in photographing. Therefore, in the information processing apparatus 4 of the present embodiment, the shooting control unit 82 selects the shooting units 23 and 24 instead of the lighting control unit 81 of FIG. 3 selecting the lighting unit as in the first embodiment. It is assumed that the function is executable.

  In the present embodiment, by measuring at a plurality of shooting angles, it is possible to obtain a measurement value that conforms to the visual impression when the sample is observed.

  Further, in this configuration, the first imaging unit 24 images the illumination unit 14 irradiated with light from the vicinity of regular reflection (highlight conditions) on the sample S, and the second imaging unit 24 A sample of the illumination unit 14 irradiated with light from the diffusion direction (shade condition) is photographed.

  In this configuration, instead of the illumination angle condition, by creating a conversion formula or a conversion table for each shooting angle condition, it is possible to improve the accuracy when the captured RGB data is converted into XYZ data. .

<Modification>
In FIG. 13, images with a plurality of shooting angles are acquired using a plurality of shooting units. However, as shown in FIG. 14, a single image sensor having a plurality of shooting angles includes a single line sensor 20. An imaging apparatus including an imaging unit may be used.

  By using the line sensor 20 that is a line scan type imaging apparatus, an image is acquired while continuously changing the illumination angle or the imaging angle, so that one measurement image can be obtained at a plurality of imaging angles. The measured image data can be stored.

  It is also possible to acquire image data at a plurality of illumination angles and / or shooting angles at a time by shooting the sample S with the above-described line sensor.

<Third Embodiment>
Next, a third embodiment in which a spectroscopic camera is used for the photographing unit will be described with reference to FIG.

  The camera of the imaging unit of the present embodiment uses a spectroscopic camera 25 that can acquire two-dimensional spectroscopic information in a wavelength band corresponding to the visible light region, a multispectral camera that can obtain spectroscopic information in a plurality of bands, and wavelength resolution. A hyperspectral camera that can acquire high spectral information can be used.

  Here, the outline of the spectral imaging of the spectral camera used as the imaging unit will be described with reference to FIG. FIG. 15 is a main part structural diagram of an imaging unit (spectral camera) 25 used in the appearance characteristic measuring apparatus according to the third embodiment. The spectroscopic camera 25 of FIG. 15 shows an example of a multispectral camera configured by a set of filters and a diffraction grating as an example, but the spectroscopic camera 25 configuring the imaging unit includes one or more sets of filters and A hyperspectral camera device including a diffraction grating (or prism) may be used.

  In the spectroscopic camera 25, as a spectroscopic information acquisition unit that acquires two-dimensional spectroscopic information, a spectroscopic filter group inserted in the main lens and a microlens array inserted between the main lens and the light receiving element are used for each microlens. Spectral information corresponding to the number of spectral filters is acquired.

  As shown in FIG. 15, a microlens array (MLA) 53 including a plurality of microlenses (small lenses) is disposed in the vicinity of the condensing position of the main lens 54. On the image plane, a light receiving element array 55 including a plurality of light receiving elements (sensors) that convert optical information collected by the main lens 54 into electronic information (electrical signals) is arranged.

  In FIG. 15, for ease of understanding, the main lens 54 as an optical system is shown as a single lens, and the aperture position S of the main lens 54 is shown as the center of the single lens, but actually, as shown in FIG. The color filter 56 is not positioned in the lens. The color filter 56 is disposed near the stop of the main lens 54. “Near the aperture” means a portion including the aperture position through which light rays with various angles of view can pass. In other words, it means an allowable range in design of the color filter 56 with respect to the main lens 54.

  In FIG. 15, a color filter 56 as an optical bandpass filter is disposed at the center of the main lens 54. The color filter 56 is a filter corresponding to a tristimulus value of a color having a spectral transmittance based on the color matching function of the XYZ color system. That is, the color filter 56 includes a plurality (three in this case) of color filters 56a, 56b, and 56c having different spectral transmittances based on the color matching functions of the XYZ color system.

  Such an optical bandpass filter may be configured by combining a plurality of filters having different spectral transmittances, or may be configured to vary the spectral transmittance for each region on one filter.

  With the above configuration, for example, when 31 types of optical bandpass filters having transmission wavelength peaks in increments of 10 nm are used in the wavelength range from 380 nm to 780 nm, the spectral information in the wavelength range from 380 nm to 780 nm is 10 nm. It is possible to obtain in increments.

  In this way, the spectral range of the spectrum that can be measured (captured) by the camera covers from 380 nm to 780 nm, thereby covering the human visible light region, and spectral information in accordance with the human appearance. Can be obtained.

  Similarly, in the multispectral camera according to the present embodiment, conversion accuracy can be improved by preparing a conversion formula or conversion table for each angle. Since the amount of information is larger for 4 channels or more, the conversion accuracy is higher than for 3 channels.

  Although the embodiments of the RGB camera and the multispectral camera have been described above, the same processing may be performed even with an XYZ camera. This is because even if a filter or the like close to the characteristic of the color matching function is employed in the camera, it is difficult to make the color matching function and the color characteristic completely coincide with each other, resulting in a conversion error. In that case, the color conversion equation may be obtained in the same manner using RGB of this embodiment as pseudo XYZ data of the XYZ camera. That is, the XYZ camera can also improve the colorimetric accuracy by performing “pseudo XYZ (camera imaged value) → XYZ (true value) conversion”.

  Further, the camera used for the photographing unit is not limited to a single plate type, and may be a three plate type camera.

  Furthermore, in the above example, the illumination unit is a white camera and the photographing unit is a color camera. However, the illumination unit may be an LED that emits light in three colors of RGB, and the photographing unit may be a monochrome camera. In this case, the LED illumination light is sequentially turned on in the order of R, G, and B, and shooting is performed three times with a monochrome camera. Thereby, RGB data can be acquired. In the case of this configuration, there is an advantage that the demosaicing process is unnecessary. However, since there is a demerit that it is necessary to shoot three times for one angle and it takes a measurement time, the configuration is appropriately selected according to the application.

  The preferred embodiment of the present invention has been described in detail above, but the present invention is not limited to the specific embodiment, and within the scope of the gist of the embodiment of the present invention described in the claims, Various modifications and changes are possible.

100, 100A, 100B Appearance characteristic measurement system (measuring device)
DESCRIPTION OF SYMBOLS 1 Light source device 2 Imaging device 3 Inspection table 4, 4A, 4B Information processing device 5 Monitor device 6 Spectral colorimeter (colorimeter)
7 Integrated appearance characteristic measuring device (measuring device)
9 Host system 11 Illumination unit (first illumination unit, LED)
12 Illumination unit (second illumination unit, LED)
14 Illumination unit 2, 2A, 2C Imaging device 20 Multi-spectral camera 21 Imaging unit (camera)
23 First photographing unit (camera)
24 Second photographing unit (camera)
25 spectroscopic camera 71 first illumination unit 72 second illumination unit 73 imaging unit 74 control unit 75 storage unit 76 display unit 77 operation buttons 80A input unit 80B monitor output unit 83 image data storage unit 84 calculation data storage unit (storage) Part)
85 Parameter update unit for colorimetric conversion 86 Colorimetric value calculation unit (conversion unit)
863 Composite RGB data creation part (RGB data composition part)
864 Tristimulus XYZ calculation unit 87 Texture calculation unit 88 Measurement data storage unit S Sample (measurement object)
P color tile (multiple color patch)

JP 2009-239419 A

Claims (15)

A measuring device for measuring a measurement object,
One or more illumination units capable of irradiating the measurement object with light; and
One or more imaging units for imaging the measurement object irradiated with the light;
A conversion unit that converts the captured image into tristimulus values,
The illuminating unit and the photographing unit photograph a plurality of times under a plurality of setting conditions in which the illumination angle of the illuminating unit or / and the photographing angle of the photographing unit is changed,
In the conversion unit, a condition for converting to the tristimulus value is different for each setting condition. The one or more illumination units and the one or more imaging units include a first setting condition for irradiating the measurement object with illumination light from a first angle with respect to the angle of the imaging unit; A second setting condition for irradiating the object to be measured with illumination light from a second angle different from the first angle with respect to the angle; and
With respect to the first setting condition and the second setting condition, a condition for converting the data for each pixel of the captured image into the tristimulus value for converting to the tristimulus value is set for each setting condition. The measuring apparatus according to claim 1, further comprising a storage unit that stores the storage unit. In the storage unit,
The true value of the reference color defined in advance for a solid patch of multiple colors;
A conversion formula or conversion table which is a condition for converting to the tristimulus values for calculation used in the conversion unit;
An updatable parameter to be substituted in the conversion formula or conversion table is stored;
Tristimulus values converted using the conversion conditions from the captured images obtained by capturing the solid color patches of the plurality of colors under the first setting condition and the second setting condition, respectively, are converted into the true values. The measurement apparatus according to claim 2, further comprising: a parameter setting unit that sets or updates a parameter to be substituted under the condition for conversion so as to be close to each other. The first setting condition is a highlight condition for irradiating the measurement object with illumination light from a regular reflection or an angle near the regular reflection with respect to the angle of the imaging unit,
The second setting condition is a highlight condition for irradiating the measurement object with illumination light from a diffuse reflection angle with respect to the angle of the imaging unit. 4. The measuring device according to one item. The measuring apparatus according to claim 1, wherein the one or more illumination units are LEDs having an average color rendering index Ra of 95 or more, or color LEDs that emit at least three colors. The measurement apparatus according to any one of claims 1 to 5, wherein the data after being photographed by the one or more photographing units is 3ch color data. The one or more photographing units obtain RGB data by photographing with a plurality of exposure times for each of the plurality of setting conditions.
The conversion unit includes an RGB data synthesis unit that synthesizes RGB data captured at the plurality of exposure times.
In the RGB data synthesis unit,
In the R value of the RGB data with the longest exposure time, if there is a “saturated or nearly saturated pixel value”, that region is extracted as the first saturated pixel region, and the first saturated pixel Processing to replace the area with RGB data with the next longest exposure time;
When the longest exposure time is longer than the reference exposure time, a process of standardizing the luminance value to be reduced in accordance with the reference exposure time;
In the first saturated pixel area replaced with the RGB data having the next longest exposure time, if there is a “saturated or nearly saturated pixel value”, that area is designated as the second saturated pixel area. And replacing the second saturated pixel region with RGB data having the third longest exposure time;
When the third longest exposure time is shorter than the reference exposure time, the process of normalizing the luminance value so as to increase in accordance with the reference exposure time is repeated,
Get R composite data,
The measurement apparatus according to claim 6, wherein combined data is acquired for G and B by the same processing as R. 8. A measurement apparatus according to claim 1, further comprising a texture calculation unit that digitizes the appearance characteristic of the measurement object as a texture using the converted tristimulus values. . The texture calculation unit
An L * a * b * calculation unit that converts the XYZ that is the tristimulus value for each pixel into a value of the L * a * b * color system for each pixel;
The measurement apparatus according to claim 8, further comprising: a texture evaluation unit that evaluates a texture based on a variation amount corresponding to a dispersion value in the image of the L *, a *, and b * values. The data photographed by the photographing unit is pseudo XYZ data, or data of three or more channels photographed by a multispectral camera. The measuring device described in 1. A method for setting a colorimetric conversion parameter in a measuring device,
The measurement apparatus includes a storage unit, a first setting condition for irradiating light of a plurality of colors on a solid color patch from a first angle with respect to an angle of the photographing unit, and an angle of the photographing unit. One or more illumination units so as to satisfy the second setting condition of irradiating the light of illumination from a second angle different from the first angle to the plain patches of the plurality of colors, and One or more shooting units,
The setting method is as follows:
In the storage unit, acquiring in advance a conversion formula or conversion table for colorimetric conversion including updatable parameters;
In the storage unit, a fixed true value of tristimulus values for color reference defined for each of the first setting condition and the second setting condition is previously stored for a plurality of solid color patches. Steps to get,
Using the measuring device to capture a solid color patch of the plurality of colors under the first setting condition to obtain a captured image;
Using the measuring device to capture a solid color patch of the plurality of colors under the second setting condition to obtain a captured image;
Tristimulus values that are converted using the conversion formula or conversion table from the captured images obtained by capturing the solid color patches of the plurality of colors under the first setting conditions are defined for the first setting conditions. Setting or updating a parameter to be substituted in the conversion formula or conversion table so as to approach a fixed true value of the tristimulus value for color reference that has been performed,
Tristimulus values that are converted using the conversion formula or conversion table from the captured images obtained by capturing the solid color patches of the plurality of colors under the second setting conditions are defined for the second setting conditions. Setting or updating a parameter substituted in the conversion formula or conversion table so as to approach a fixed true value of the tristimulus value for color reference that has been performed. Color conversion parameter setting method. In the storage unit, a fixed true value of tristimulus values for color reference defined for each of the first setting condition and the second setting condition is previously stored for a plurality of solid color patches. In the step to get
Using a colorimeter, the tristimulus values of the solid color patches are measured at the same angle as the first setting condition,
Using the colorimeter, under the same angle condition as the second setting condition, measure the tristimulus values of the plain patches of the plurality of colors as true values,
In the storage unit, tristimulus values of the solid color patches of each of the first setting condition and the second setting condition measured by the colorimeter are defined. The method for setting a colorimetric conversion parameter in the measuring apparatus according to claim 11, wherein the setting is stored as a fixed true value of tristimulus values for color reference. The method for setting a colorimetric conversion parameter in a measuring apparatus according to claim 11 or 12, wherein the solid color patches include a patch having a reflectance of 100% or more or a lightness of 100 or more. The method for setting a colorimetric conversion parameter in a measuring apparatus according to any one of claims 11 to 13, wherein the plurality of solid color patches are at least eight colors or more. An industrial product that is inspected and produced using the measuring device according to claim 1.