Patent Application
20060146326
This application is based on patent application No. 2005-000583 filed in Japan, the contents of which are hereby incorporated by references.
The present invention relates to a light measuring apparatus for obtaining characteristics of light from a light source or light to be measured such as light reflected by or transmitted through an illuminated sample and particularly to a spectral color meter for measuring color luminance values and the like of visible light (monochromatic light) and a method for measuring monochromatic light.
A color meter, one mode of a light measuring apparatus, for measuring color values of an incident light needs to have spectral responsitivities approximate to color matching functions of a standard observer recommended by the CIE (International Commission on Illumination). In recent years, with an increase of light sources for emitting a monochromatic light or a combination of monochromatic lights such as liquid crystal displays, plasma displays or LEDs (semiconductor light emitting devices), there has been an increasing demand for a further improvement in approximation precision to the color matching functions in order to improve the measurement precision of color values of these light sources.
Systems for realizing the spectral responsitivities approximate to the color matching functions include (a) a system by a color luminance meter directly reading stimulus values and (b) a system by a spectral luminance meter.
(a) The color luminance meter directly reading stimulus values builds filter characteristics approximate to the color matching function by combining a plurality of optical filters. This system has an advantage of compacting the light measuring apparatus since an optical system is simple and it is sufficient to provide a small number of light receiving sensors. However, there is a disadvantage of being difficult to meet a required approximation precision due to the restriction of usable filters and variations in the characteristics of the usable filters.
(b) The spectral luminance meter is for measuring spectral data at minute wavelength intervals using a narrow half power band width, for example, for measuring the luminance or chromaticity of a light to be measured (measurement light) from a spectral data obtained at wavelength pitches of 1 nm using a half power band width of about 5 nm (i.e., a visible wavelength region is covered by about 400 light receiving sensors). With such a spectral luminance meter, the spectral responsitivities can be caused to precisely coincide with the color matching functions, enabling highly precise luminance and chromaticity measurements, since the spectral data is obtained at the minute wavelength intervals. However, a bright optical system having a high resolving power is necessary in order to obtain spectral data as above, thereby presenting a problem of necessitating the use of a light measuring apparatus having a relatively large size.
In view of the above problems, there has been known a light measuring apparatus adopting a so-called spectral fitting method for conducting a measurement after obtaining a composite spectral responsitivity approximate to a specific spectral responsitivity by multiplying spectral responsitivities obtained at relatively long wavelength intervals (about several tens nm) using a wide half power band width by specified weights set beforehand. Such a light measuring apparatus is disclosed, for example, in Japanese Unexamined Patent Publication No. 2002-13981. Such a light measuring apparatus adopting the spectral fitting method has advantages that:
(I) The approximation precision of the spectral responsitivities is higher than the color luminance meter directly reading the stimulus values (a),
(II) A light amount is larger and a signal-to-noise (S/N) ratio is higher since each light receiving channel lies in a wide wavelength region, and
(III) A silicon photodiode array having better linearity, S/N ratio and dynamic range due to a smaller number of light receiving channel can be used, and it is not necessary to use a CCD (charged-coupled device) sensor array or a like electric charge transferring type sensor array inferior in these points.
Although the light measuring apparatus adopting the spectral fitting method has various merits as described above, the inventors of the present invention found out that the approximation precision of the composite spectral responsitivities might decrease unless the half power band width and the wavelength pitch (interval between center wavelengths of adjacent light receiving sensors) of the respective light receiving sensors used were optimized. Particularly, the influence of the decreased approximation precision is large in the measurement of color luminance values of monochromatic lights emitted from the aforementioned various displays and LEDs, with the result that there have been cases where the characteristics of the monochromatic lights could not be obtained with high precision.
It is an object of the present invention to provide a light measuring apparatus and a light measuring method which are free from the problems residing in the prior art.
According to an aspect of the present invention, a light measuring apparatus is provided with a spectral device for dispersing measurement light at each of specified wavelengths, a predetermined number of photoelectrical converters for receiving the dispersed measurement light and outputting light reception signals corresponding to the predetermined number of wavelengths, a storage device for storing the predetermined number of weight coefficients set beforehand to obtain a specified spectral responsitivity in accordance with the light reception signals outputted from the respective photoelectrical converters, and a calculator for calculating characteristics of the measurement light based on the specified spectral responsitivity using the respective light reception signals and the respective weight coefficients.
The spectral responsitivities and wavelength pitch of the respective photoelectrical converters are so determined as to satisfy the following conditions (1), (2):
(1) B≧5 nm,
(2) A/B=1.5 to 4.0.
wherein A denotes a half power band width of the spectral responsitivities of the respective photoelectrical converters and B denotes a center wavelength interval of the spectral responsitivities of the respective photoelectrical converters.
A monochromatic light is measured using the above-mentioned light measuring apparatus by calculating a composite intensity by adding products of the light reception signals respectively outputted from the predetermined number of photoelectrical converters and specified weight coefficients for the approximation to a spectral responsitivity of a specified color matching function, and calculating characteristics of the monochromatic light based on the composite intensity.
Accordingly, color luminance values and the like can be measured with high precision by maximally suppressing errors in the measurements of lights in a narrow band such as monochromatic lights.
These and other objects, features and advantages of the present invention will become more apparent upon a reading of the following detailed description and accompanying drawings.
Hereinafter, embodiments of the present invention are described with reference to the accompanying drawings.
(Description of the Overall Construction of Light Measuring Apparatus)
The main unit 10 includes a power switch 101, an operable section 102 comprised of a plurality of operation keys and a cross-shaped button, a display 103 formed of a LCD or the like arranged on a side surface to display measurement results and the like, a measurement angle changeover switch 104 for switching a measurement angle in accordance with a measurement object, a connector portion 105 such as a USB terminal, and a finder 106 for enabling the visual confirmation of a measurement area. The objective lens unit 11 includes an objective lens 111.
The objective lens 111 is adapted for introducing measurement light emitted from a specified light measurement area (light source) to an incidence slit 40 of the polychrometer 4 in cooperation with the light introducing lens 13 arranged therebehind. The measurement light (incident light) may be light reflected by a sample, light having transmitted through the sample, or light irradiated from the sample. In this embodiment, the measurement light is irradiated light. If this irradiated light is irradiated from a light source having a specified narrow wavelength band, particularly monochromatic light emitted from one of various displays such as an LCD (liquid crystal display) and a PDP (plasma display panel) and light sources such as a LED (semiconductor light emitting device), there is an advantage of being able to conduct a color luminance value measurement having less errors. This point is described in detail later.
The aperture mirror 12 is a mirror arranged on an incident light path of the measurement light on the polychrometer 4 and formed with an aperture for permitting the passage of fluxes of light from the light measurement area. Out of the measurement light incident from the objective lens 111, the fluxes of light from the light measurement area propagate straight to the light introducing lens 13 arranged behind the aperture mirror 12 after passing the aperture of the aperture mirror 12, but the fluxes of light outside the light measurement area are reflected by the aperture mirror 12 and are incident on the finder optical system 15 by way of an optical lens 141 and a reflection mirror 142.
The light introducing lens 13 is an optical system including a specified positive lens for causing the fluxes of light (measurement light) from the light measurement area having passed the aperture of the aperture mirror 12 to be incident on the incidence slit 40 of the polychrometer 4.
The finder optical system 15 is an optical system including the optical lens 141, the reflection mirror 142, a specified number of positive and negative lenses and a protection filter 151 for introducing the fluxes of light reflected by the aperture mirror 12 to the pupils of an observer. Since the fluxes of light having those from the light measurement area removed are incident on this finder optical system 15, a finder image is such that the light measurement is left out in black. Accordingly, the light measurement area can be confirmed without parallax. The protection filter 151 is formed of a light amount attenuating filter or the like for protecting the observer's eyes by suppressing the glare of the light and also functions to prevent the entrance of dusts into the finder optical system 15.
The shutter 3 is an opening/closing device having a function of shutting off a light path and, for example, is a disk-shaped member provided with a passing portion and a shutting portion. When the passing portion is disposed (shutter is open) at an intermediate position of the light path from the objective lens 111 to the incidence slit 40 of the polychrometer 4, the measurement light is incident on the polychrometer 4. On the other hand, when the shutting portion is arranged at an intermediate position of the light path (shutter is closed), the measurement light is not incident on the polychrometer 4. The shutter driving motor 31 opens and closes the shutter 3 as above in accordance with a control signal outputted from the operation control unit 6.
The polychrometer 4 forms the spectral optical system and includes the incidence slit 40, a collimating lens 41 as a focusing optical system, a grating (diffraction grating) 42 as a wavelength dispersing means, and a light receiving sensor array 43 (photoelectric conversion means) comprised of n (n is an integer equal to or greater than 2) light receiving sensors.
The incidence slit 40 is a circular slit formed in a side wall of the polychrometer 4, and the measurement light is introduced to the inside of the polychrometer 4 through the incidence slit 40 as described above. The collimating lens 41 introduces the measurement light having passed the incidence slit 40 to the grating 42 after making it into a parallel light, and causes a dispersed image of the incidence slit 40 dispersed by the grating 42 to be focused on a sensing surface of the light receiving sensor array 43. The grating 42 has a function of reflecting and dispersing the incident measurement light according to wavelength, thereby enabling the dispersed image of the incidence slit 40 to be focused on the light receiving sensor array 43.
The light receiving sensor array 43 is provided with n light receiving channels (pixels) CH1 to CHn arrayed at a specified interval. The light receiving channels CHk (k is a pixel or channel number and=0 to n) are formed, for example, by a silicon photodiode array in which silicon photodiodes are arrayed in one row (may be arrayed in a plurality of rows) as shown in
The signal processing circuit 5 amplifies and digitally converts the analog light reception signals outputted from the respective light receiving channels CH1 to CHn.
The dispersed lights (dispersed image of the incidence slit) incident on the respective light receiving channels CH1 to CHn of the sensor array 43 are converted into currents by the photoelectrically converting function of the respective photodiodes, and these currents are converted into voltage signals by the current-to-voltage converting circuits Ak connected with the respective pixels. The voltage signals of the respective pixels outputted from the current-to-voltage converting circuits Ak are successively selected and extracted in the multiplexer 51 controlled by the operation control unit 6. The voltage signals are amplified in the variable gain amplifier 52 and converted in the analog-to-digital converter 53 into digital signals, which are then sent to the operation control unit 6.
The operation control unit 6 includes, for example, a CPU (central processing unit) and an EEPROM (electrically erasable programmable storage medium) and controls the overall operation of the color luminance meter 1. Specifically, the operation control unit 6 controls the operations of the respective parts to measure the spectral intensity of the measurement light and calculates the color luminance values using specified spectral responsitivities.
The measurement controller 61 sends control signals to the respective parts to control a measuring operation upon measuring the characteristics of the measurement light. For example, a shutter movement controller 611 of the measurement controller 61 drives the shutter driving motor 31 during the measurement to open the shutter 3, thereby permitting the measurement light having passed the objective lens 111 and the light introducing lens 13 to be incident on the slit 40 of the polychrometer 4. The shutter movement controller 611 drives the shutter driving motor 31 to close the shutter 3, thereby shutting off the incidence of the measurement light, at a time of an offset signal measurement to be described later. The measurement controller 61 gives a control signal to the signal processing circuit 5 for controlling the selecting operation of the multiplexer 51.
The calculator 62 calculates the spectral intensity of the measurement light incident on the light receiving sensor array 43 and obtains the characteristics of the measurement light based on a composite spectral responsitivity approximate to a specified spectral responsitivity by adding products of the spectral responsitivities of the respective light receiving channels CH1 to CHn and specified weighting coefficients. This calculator 62 is comprised of a spectral intensity calculating section 621 and a composite intensity calculating section 622.
The spectral intensity calculating section 621 receives the outputs (light reception signals) of the respective light receiving channels CH1 to CHn of the light receiving sensor array 43 on which the dispersed image of the incidence slit 40 is focused as the digital signal data from the signal processing circuit 5 during the measurement (shutter 3 is “open”), and temporarily saves them in a RAM or the like of the memory section 63. Subsequently, the spectral intensity calculating section 621 receives the digital signal data (offset signals) from the signal processing circuit 5 with the measurement light shut out during the offset signal measurement (shutter 3 is “closed”). The spectral intensity calculating section 621 calculates pixel output data Ik of the measurement light having offsets removed by subtracting the offset signals from the measurement data previously received.
The composite intensity calculating section 622 calculates the characteristics of the measurement light based on the specified spectral responsitivity obtained by multiplying the spectral intensity data Ik obtained in the spectral intensity calculating section 621 by the weighting coefficients set beforehand and adding the resulting products. Specifically, data X′, Y′, Z′ (composite intensities) approximate to tristimulus values X, Y, Z given by CIE color matching functions x(═), y(λ), z(λ) are obtained by multiplying the spectral intensity data Ik by the weighting coefficients Gxk, Gyk, Gzk for compositing the CIE color matching functions x(λ), y(λ), z(λ), which are saved in a weighting coefficient storage 631 of the memory section 63 to be described later, and adding the resulting products.
A calculating technique (spectral fitting method) used in the composite intensity calculating section 622 is described with reference to
In order to obtain the composite intensity values X′, Y′, Z′ given by the composite spectral responsitivities xg(λ), yg(λ), zg(λ) approximate to the color matching functions x(λ), y(λ), z(λ) shown in
If a target spectral responsitivity α0(λ) (corresponding to the color matching functions x(λ), y(λ), z(λ)) is determined as shown by dotted line in
For example, the composition of the color matching function z(λ) is specifically described. The spectral responsitivities Rk(λ) of the respective light receiving channels CHk shown in
Referring back to
(Concerning the Half Power Band Width of the Light Receiving Sensor and the Center Wavelength Interval)
In the color luminance meter 1 thus constructed, the half power band widths of the spectral responsitivities of the respective light receiving sensors and the center wavelength interval (wavelength pitch) of the respective light receiving sensors (wavelength intervals) are selected from specific ranges in order to enable a light measurement of high precision. Specifically, when it is assumed that A denotes the half power band width of the spectral responsitivities of the respective light receiving sensors and B denotes an interval (wavelength pitch) between center wavelengths λk and λk+1 of the spectral responsitivities Rk(λ) and Rk+1(λ) of adjacent light receiving sensors (light receiving channels CHk and CHk+1) as shown in
the center wavelength interval B≧5 nm, and
the half power band width A/center wavelength interval B=1.5 to 4.0.
In this case, it is desirable that the center wavelength interval B lies within a range between 5 nm and 35 nm (5 nm≦B≦35 nm).
It is particularly desirable that the center wavelength interval B and a ratio of the half power band width A to the center wavelength interval B are:
7 nm≦B≦13 nm, and
A/B=2.5 to 3.5.
For example, the center wavelength interval B (wavelength pitch) of 10 nm and the half power band width A of 30 nm (A/B=3) as shown in
The half power band width A can be adjusted by suitably setting the diameter of a spot of light incident on each light receiving sensor of the light receiving sensor array 43.
The half power band width A is described with reference to
An area of the incidence slit 40 is in proportion to an amount of light usable in the polychrometer 4. Specifically, the larger the area of the incidence slit 40, the larger amount of light (higher sensitivity) the polychrometer 4 has. On the other hand, the shape of the incidence slit 40 influences a sensor output intensity distribution (spectral responsitivity). In other words, a desired half power band width A can be obtained by suitably selecting the shape of the incidence slit 40.
More specifically, since the incidence slit 40b is rectangular in
In this way, the incidence slit 40 can take various shapes and a suitable shape can be selected depending on a desired half power band width A. However, in view of precise fitting in the spectral fitting method, it is desirable to adopt the circular slit 40c of
The half power band width A can be suitably set not only by adjusting the shape and size of the incidence slit 40 as described above, but also by adjusting the optical characteristics and the optical arrangements of the incidence slit 40, the collimating lens 41, the grating 42 and the light receiving sensor array 43.
(Description of the Operation)
Next, the operation of the color luminance meter 1 constructed as above is described.
During the spectral responsitivity calibration in Step S1, which spectral responsitivities Rk(λ) the respective light receiving channels CHk(λ) of the sensor array 43 possess is obtained and temporarily saved in the memory section 63 or the like. During the sensitivity calibration of Step S2, calibration coefficients Ck for calibrating the sensitivities of the respective light receiving channels CHk are obtained and saved in the memory section 63 or the like. The weighting coefficient setting in Step S3 is a unique step required to carry out the spectral fitting method. In this step, the weighting coefficients Gxk, Gyk, Gzk used to composite the CIE color matching functions x(λ), y(λ), z(λ) are calculated using the spectral responsitivities Rk(λ) calculated in Step S1 and saved in the weighting coefficient storage 631. During the measurement (Step S4), the tristimulus values X′, Y′, Z′ approximate to the tristimulus values X, Y, Z given by the CIE color matching functions x(λ), y(λ), z(λ) are calculated using the weighting coefficients Gxk, Gyk, Gzk. The respective steps are successively described below.
<Step S1: Spectral Responsitivity Calibration>
In this case, as shown in the flowchart of
Such measurements are carried out in the wavelength region of 350 to 750 nm, the respective pixel outputs Pk(λc) at the respective center wavelengths λc are plotted on an axis of wavelength, whereby the relative spectral responsitivities Rk(λ) of the respective light receiving channels CHk are obtained (Step S16). In this way, the spectral responsitivities Rk(λ) of all the light receiving channels CHk (k=1 to 31 in this example) in the entire wavelength region λ of 350 to 750 nm are obtained. Then, the relative spectral responsitivities Rk(λ) based on the respective pixel outputs Pk(λc) are saved as the spectral responsitivities Rk(λ) of the respective light receiving channels CH1 to CHn in the memory section 63 (Step S17). The weighting coefficients Gxk, Gyk, Gzk to be described later are calculated as weights for the spectral responsitivities Rk(λ). Barycentric wavelengths (center wavelengths) Wck of the respective pixels are also calculated from the thus calculated relative spectral responsitivities Rk(λ).
<Step S2: Sensitivity Calibration>
Upon such a sensitivity calibration, since a single wave output of a reference monochrometer as used in the spectral responsitivity calibration in Step S1 cannot be generally said to possess an equal energy having a sufficient precision, a white light source for calibration having a known spectral intensity A(λ) is used for the sensitivity calibration. A white light is caused to be incident on the incidence slit 40 of the polychrometer 4 from such a light source for calibration, and the respective pixel outputs Pk in the sensor array 43 at this time are measured (Step S21).
On the other hand, pixel output reference values Bk assumed from the saved spectral responsitivities Rk(λ) of the respective pixels and the known spectral intensity A(λ) of the white light source for calibration are calculated by the following equation (1) (Step S22).
Bk=∫A(λ)·Rk(λ)dλ (1)
Then, the calibration coefficients Ck to be multiplied with the outputs Pk of the respective pixels are calculated for the respective wavelengths by the following equation (2) so that the pixel output reference values Bk and the calibrated outputs Ck·Pk of the respective pixels when the measurements are actually made using the white light source for calibration coincide (Step S23).
Ck=Bk/Pk (2)
The calibration coefficients Ck thus calculated are saved in the memory section 63 and used as data correction factors upon actually measuring the color values.
<Step S3: Weighting Coefficient Setting>
The weighting coefficients Gxk, Gyk, Gzk are set after the spectral responsitivity calibration and the sensitivity calibration in Step S1 and S2 at the time of production. The weighting coefficients Gxk, Gyk, Gzk are obtained as coefficients, for example, for approximating the composite spectral responsitivities obtained by the addition of the weighted spectral responsitivities Rk(λ) of the respective light receiving channels CHk obtained in Step S1 to the spectral responsitivities x(λ), y(λ), z(λ) of the color matching functions of 10° field of view specified by the CIE shown in
At the beginning of the calculation of the weighting coefficients, a calculation routine (least squares method loop) is entered after Gxk, Gyk, Gzk are respectively set at 0 (Step S32). The weighting coefficients Gxk, Gyk, Gzk are calculated by the least squares method so that the composite spectral responsitivities xg(λ), yg(λ), zg(λ) given by the following equations (3-1), (3-2), (3-3) and related to the spectral responsitivities Rk(λ) of the respective pixels obtained in Step S1 to the spectral responsitivities x(λ), y(λ), z(λ) of the color matching functions shown in
xg(λ)=ΣkGxk·Rk(λ) (3-1)
yg(λ)=ΣkGyk·Rk(λ) (3-2)
zg(λ)=Σk·Gzk·Rk(λ) (3-3)
Specifically, the weighting coefficients Gxk, Gyk, Gzk are suitably corrected and the composite spectral responsitivities xg(λ), yg(λ), zg(λ) at this time are set by the equations (3-1), (3-2), (3-3) (Step S33). Sums Exg, Eyg, Ezg of squares of differences between the composite spectral responsitivities xg(λ), yg(λ), zg(λ) and the target spectral responsitivities x(λ), y(λ), z(λ) of the color matching functions at the respective wavelengths are calculated as evaluation coefficients by the following equations (4-1), (4-2), (4-3) (Step S34).
Exg=Σλ[xg(λ)−x(λ)]2 (4-1)
Eyg=Σλ[yg(λ)−y(λ)]2 (4-2)
Ezg=Σλ[zg(λ)−z(λ)]2 (4-3)
Subsequently, the sums of squares Exg, Eyg, Ezg calculated in Step S34 and the marginal error Egt set in Step S31 are compared (Step S35). If the values of the sums of squares Exg, Eyg, Ezg exceed the marginal error Egt (NO in Step S35), the weighting coefficients Gxk1, Gyk1, Gzk1, used in the above calculation are further corrected to set new weighting coefficients Gxk2, Gyk2, Gzk2 (Step S36) and Step S33 follows to repeat the least squares method loop.
Specifically, using the sums of squares Exg, Eyg, Ezg given by the equations (4-1), (4-2), (4-3) as the evaluation coefficients, the weighting coefficients Gxk, Gyk, Gzk are so calculated by the least squares method that the evaluation coefficients Exg, Eyg, Ezg fall below the marginal error Egt as the threshold value (are minimized). If such weighting coefficients Gxk, Gyk, Gzk are calculated (YES in Step S35), they are saved in the weighting coefficient storage 631 of the memory section 63. The above operation steps are carried out at the time of producing the color luminance meter 1 (at the time of initial setting). It should be noted that different marginal errors Egt may be set for the targeted color matching functions x(λ), y(λ), z(λ).
<Step S4: Flow During the Measurement>
Next, operations carried out in the case of actually measuring the color luminance values (spectral intensities) using the color luminance meter 1 according to this embodiment are described.
In the spectral intensity calculating section 621, the respective pixel outputs Qk are obtained by the flow in Step S2 and corrected to calibrated pixel output data Ik by the following equation (5) using the calibration coefficients Ck saved in the memory section 63 (Step S42).
Ik=Ck·Qk (5)
Further, data interpolation is applied using the light receiving channels CHk (k=1 to n) or a correspondence table of pixels PDk and the barycentric wavelengths Wck to convert the calibrated pixel output data Ik into spectral intensity I(λ) (Step S43).
Subsequently, the composite intensity calculating section 622 calculates the tristimulus values using the calibrated pixel output data Ik calculated in the spectral intensity calculating section 621 and the weighting coefficients Gxk, Gyk, Gzk saved in the weighting coefficient storage 631 beforehand (Step S44). Specifically, the composite intensity calculating section 622 calculates data X′, Y′, Z′ approximate to the tristimulus values X, Y, Z given by the CIE color matching functions x, y, z in accordance with the following equations (6-1), (6-2), (6-3) for adding products of the weighting coefficients Gxk, Gyk, Gzk and the calibrated pixel output data Ik calculated by the equation (5) (Step 542), and outputs the calculated data.
X′=ΣkGxk·Ik (6-1)
Y′=ΣkGyk·Ik (6-2)
Z′=ΣkGzk·Ik (6-3)
The characteristics of the measurement light, i.e., the color luminance values of the measurement light are obtained based on the composite intensities X′, Y′, Z′.
Several specific examples according to the embodiment and comparative examples to be compared with the specific examples are shown concerning the half power band width A and the center wavelength interval B of the light receiving sensors of the above light receiving sensor array 43. In the examples (comparative examples), the approximation precision of the composite spectral responsitivities to the CIE color matching function z(λ) and S/N ratios based on the weighting coefficients for the respective wavelengths were evaluated when the center wavelength interval (wavelength pitch) of the light receiving sensors was set at 10 nm and the half power band width A was set at 10 nm, 15 nm, 20 nm, 30 nm, 40 nm and 50 nm.
FIGS. 15 to 18 are graphs showing data according to the examples of the embodiment, whereas
Since the color matching function z(λ) whose spectral responsitivity peaks in the neighborhood of 460 nm as shown in
The approximation precision to the CIE color matching function z(λ) and the evaluation values of the S/N ratios based on the weighting coefficients for the respective wavelengths are shown in TABLE-1 below for the examples and the comparative examples shown in FIGS. 15 to 20.
In examples in which the center wavelength interval was 10 nm and a ratio of the half power band width A to the center wavelength interval B lied within a range of 1.5 to 4.0, satisfactory results could be obtained for the approximation precision to the color matching function z(λ) and the S/N ratios. Specifically, as is clear from
On the other hand, in the comparative example in which the ratio of the half power band width A to the center wavelength interval B is set at 1, the approximation precision is poor in the neighborhood of the peak value of the color matching function z(λ), wherein “undulation” is found in the composite spectral responsitivity as is also clear from
Even in the case that the composite spectral responsitivity locally undulates as above, errors can be evened if the measurement light is a light having a broad wavelength spectrum such as the one emitted from a white light source. Thus, there are relatively few problems. However, if the measurement light is a monochromatic light having a line spectrum, errors are large at wavelengths where the spectral responsitivity is largely deviated from the color matching function z(λ) due to the “undulation”. This causes an inconvenience of being unable to measure the color luminance values with high precision. Accordingly, with the ratio of the half power band width A to the center wavelength interval B set at 1, the comparative example 1 may not be proper for the light measurement, particularly the measurement of the monochromatic light.
Although the approximation precision is relatively good in the comparative example 2 where the ratio of the half power band width A to the center wavelength interval B is set at 5, some of the weighting coefficients take negative values as is clear from
The above S/N ratio is additionally described. First, when the ratio of the half power band width A to the center wavelength interval B increases, curves of the spectral responsitivities of the respective light receiving sensors become wider than the curve of the color matching function z(λ), with the result that it is necessary to narrow the curve of the composite spectral responsitivity by letting the weighting coefficients of some of the light receiving sensors take negative values in order to approximate the wider parts of the curves to the color matching function z(λ). Here, the outputs of the light receiving sensors include random noise components in addition to signal components. The noise components are added in every composition regardless of whether the composition is to add or subtract the outputs of the respective light receiving sensors. On the other hand, the signal components become smaller in the case of the subtractive composition, wherefore the noise components become relatively larger as compared to the signal components. Thus, if the outputs of the light receiving sensors to be composited include many of those multiplied by the weighting coefficients taking negative values, a ratio of the noise components accordingly increases, leading to the deterioration of the S/N ratio. For such a reason, the comparative example in which the ratio of the half power band width A to the center wavelength interval B is set at 5 is not proper in terms of the S/N ratio.
As described above, a light measuring apparatus comprises: a spectral device for dispersing measurement light at each of specified wavelengths; a predetermined number of photoelectrical converters for receiving the dispersed measurement light and outputting light reception signals corresponding to the predetermined number of wavelengths; a storage device for storing the predetermined number of weight coefficients set beforehand to obtain a specified spectral responsitivity in accordance with the light reception signals outputted from the respective photoelectrical converters; and a calculator for calculating characteristics of the measurement light based on the specified spectral responsitivity using the respective light reception signals and the respective weight coefficients.
The spectral responsitivities and wavelength pitch of the respective photoelectrical converters are so determined as to satisfy the following conditions (1), (2):
(1) B≧5 nm,
(2) A/B=1.5 to 4.0.
wherein A denotes a half power band width of the spectral responsitivities of the respective photoelectrical converters and B denotes a center wavelength interval of the spectral responsitivities of the respective photoelectrical converters.
With this construction, the half power band width A of the spectral responsitivities of the respective photoelectrical converters and the center wavelength interval B of the spectral responsitivities of the respective photoelectrical converters are optimized, and approximation precision can be improved in the spectral fitting method of measuring light by calculating a composite spectral responsitivity approximate to a specified spectral responsitivity. Thus, a highly precise light measurement can be conducted based on the composite spectral responsitivity, enabling a luminance measurement and a chromaticity measurement with high precision.
If the center wavelength interval B is shorter than 5 nm, the wavelength pitch is too short, making it less necessary to improve the approximation precision of the composite spectral responsitivity to the specified spectral responsitivity by this technology (construction analogous to the aforementioned “spectral luminance meter”). If the ratio of the half power band width A to the center wavelength interval B falls below 1.5, the half power band width A of the spectral responsitivities of the respective photoelectrical converters becomes too short as compared to the wavelength pitch, with the result that the pointedness of the spectral responsitivity becomes conspicuous. Thus, even if an attempt is made to approximate the spectral responsitivity to a target curve of, e.g., a color matching function by the least squares method or the like, there is a tendency to make approximation having a good fitting property difficult due to the pointedness of the spectral responsitivity. Further, if the ratio of the half power band width A to the center wavelength interval B exceeds 4.0, the half power band width A of the spectral responsitivities becomes too long as compared to the wavelength pitch, with the result that a larger percentage of the weighting coefficients to be multiplied with the spectral responsitivities of the respective photoelectrical converters take negative values. This increases an extent of negatively weighted signal components, thereby presenting a tendency to deteriorate a S/N ratio. Therefore, there can be provided a light measurement apparatus capable of highly precisely measuring lights of all spectra from broad lights to monochromatic lights.
The measurement light may be preferably introduced to the spectral device via a circular specified incidence slit. With this construction, the shape of the spectral responsitivity can be approximated to a Gaussian curve. Therefore, more precise fitting can be carried out in the spectral fitting method.
The center wavelength interval B may preferably lie in a range of 5 nm or more to 35 nm or less. With this construction, the wavelength pitch of the photoelectrical converters is restricted to at longest 35 nm, with the result that the light reception signals can be obtained at relatively narrower wavelength pitches although the wavelength pitch is not as narrow as in conventional spectral luminance meters. Thus, the number of data used for the approximation is increased, leading to a higher approximation precision. Therefore, there can be provided a light measuring apparatus capable of conducting luminance measurement, chromaticity measurements or like light measurements with higher precision.
The center wavelength interval B may preferably lie in a range of 7 nm or more to 13 nm or less. The ratio of the half power band width A to the center wavelength interval B may preferably lie in a range of 2.5 or more to 3.5 or less. With this construction, the aforementioned pointedness of the spectral responsitivity and the S/N ratio can be optimized, thereby further improving the approximation precision and bettering the S/N ratio. Therefore, there can be provided a light measuring apparatus capable of conducting luminance measurement, chromaticity measurements or like light measurements with even higher precision.
It may be preferable that any of the predetermined number of photoelectrical converters receives light in a visible wavelength region, the weight coefficients are for generating a composite spectral responsitivity approximated to a spectral responsitivity of a specified color matching function, the calculator calculates the characteristics of the measurement light based on the composite spectral responsitivity, and the light measuring apparatus is used as a spectral calorimeter.
With this construction, the approximation precision to the color matching function can be improved, with the result that there can be provided a spectral color meter adopting the spectral fitting method and capable of conducting color measurements such as luminance measurements and chromaticity measurements with high precision. Therefore, there can be provided a spectral color meter adopting the spectral fitting method, which meter is capable of conducting luminance measurement, chromaticity measurements or like light measurements with high precision.
Further, a method for measuring a monochromatic light using the above-mentioned light measuring apparatus, comprises: calculating a composite intensity by adding products of the light reception signals respectively outputted from the predetermined number of photoelectrical converters and specified weight coefficients for the approximation to a spectral responsitivity of a specified color matching function; and calculating characteristics of the monochromatic light based on the composite intensity.
In the case that a measurement light is a light having a broad light emission wavelength band such as the one emitted from a white light source, errors can be substantially evened and do not particularly stand out as a whole due to a broad light emission wavelength band even if the half power band width A and the center wavelength interval B are not optimized and a spectral responsitivity does not partially or locally precisely approximate to a specified spectral responsitivity. However, in the case that measurement light is a monochromatic light such as one of those emitted from various displays and LEDs, there is a possibility of obtaining a measurement result including a large error due to a narrow light emission wavelength band if the monochromatic light is measured at the spectral responsitivity having an insufficient approximation precision. Accordingly, the approximation precision is improved entirely as well as partially by optimizing the half power band width A and the center wavelength interval B as described above, with the result that the monochromatic can also be measured with high precision. Therefore, lights of all spectra from broad lights to monochromatic lights can be highly precisely measured by one light measuring apparatus.
Although the present invention has been fully described by way of example with reference to the accompanied drawings, it is to be understood that various changes and modifications will be apparent to those skilled in the art. Therefore, unless otherwise such changes and modifications depart from the scope of the present invention hereinafter defined, they should be construed as being included therein.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2005-000583 | Jan 2005 | JP | national |
