JP2008011153A - Image analysis method and color sample - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image analyzing method and a color sample that can facilitate the selection operation of a spectrum cut-off filter. <P>SOLUTION: The image analyzing method comprises photographing a spectral sensitivity chart 10, having patches of a plurality of raw materials differing in spectral reflection characteristics containing lights other than visible light (first stage); measuring a light source used for the photography (second stage); finding X, Y, and Z stimulus values, by using spectral sensitivities of the patches 1 to 8 of the spectral sensitivity chart 10, the spectral reflectivity of the light source used for the photography, and X, Y, and Z isochromatic functions (third stage); and then comparing the X, Y, and Z stimulation values with the image data obtained by the photography in the first stage and adjusting the filter, imaging element, etc., used in a digital camera, based on the numerically obtained result (fourth stage). <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an image analysis method and a color sample.

2. Description of the Related Art Conventionally, color samples called color charts and color tables are known as photographing color samples used for camera color reproduction evaluation. As a color sample, a material having a spectral reflectance in the human color gamut is generally used (see Japanese Patent Application Laid-Open No. 2000-112649).
JP 2000-112649 A

  However, there are a large number of materials having various spectral reflectances in the natural world, and many materials have a spectral reflectance peak in light having a wavelength longer than 700 nm, which is light other than visible light.

  In addition, an image sensor used in a digital camera has spectral sensitivity even for light having a long wavelength of 700 nm or more. Therefore, in the digital camera, the light on the long wavelength side is blocked using a spectral cut filter before the light enters the image sensor.

  The operation for selecting the spectral cut filter according to the spectral sensitivity characteristic of the image sensor is performed by numerical calculation by actually photographing a color chart and multiplying the spectral transmittance of the spectral cut filter by the spectral sensitivity of the image sensor.

  However, since there are a large number of materials having a large amount of spectral reflectance in nature, the work of selecting a material for a color chart has been troublesome. Therefore, a great deal of labor has been required for selecting the spectral cut filter.

  The present invention has been made in view of such circumstances, and an object thereof is to provide an image analysis method and a color sample capable of simplifying a spectral cut filter selection operation.

  In order to solve the above-mentioned problem, the invention described in claim 1 includes a first step of photographing a color sample having a plurality of material patches having different spectral reflectances including light other than visible light, and after the first step, A second step of measuring a light source used for photographing, and an XYZ stimulus using the spectral reflectance of each patch of the color sample, the spectral reflectance of the light source used for photographing, and the XYZ color matching function after the second step. A third step of obtaining a value; and a fourth step of comparing the XYZ stimulation value with the image data acquired in the first step after the third step.

  According to a second aspect of the present invention, in the image analysis method according to the first aspect, after the fourth step, a filter and an image sensor used for the photographing are selected.

  A third aspect of the present invention is the image analysis method according to the first aspect, wherein the color sample is a color chart.

  According to a fourth aspect of the present invention, in the image analysis method according to the third aspect, the color chart is stored in a predetermined storage medium as image data.

  According to a fifth aspect of the present invention, there is provided a color sample having a plurality of patches of materials having different spectral reflectances including light other than visible light.

  According to this invention, the selection work of the spectral cut filter can be simplified.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  1 is a diagram showing a spectral reflectance chart, FIG. 2 is a diagram showing the spectral reflectance of each patch in FIG. 1, and FIG. 3 is a diagram showing an example of a spectral distribution of a light source.

  In FIG. 2, the vertical axis and the horizontal axis indicate the reflectance (%) and the wavelength (nm), respectively.

  A spectral reflectance chart (color chart) 10 as one color sample is configured from eight types of patches 1 to 8. The patches 1 to 8 are arranged in the same size. The patches 1 to 8 of the spectral reflectance chart 10 are imaged by an imaging device (not shown), and the obtained chart image data is stored in a storage medium (storage medium) such as a memory card or a disk medium (not shown). ).

  The spectral reflectances of the patches 1 to 8 in FIG. 1 correspond to (a) to (h) in FIG. As can be seen from FIG. 2, patches 1 to 8 include those whose reflectance peak values are other than visible light (wavelength 700 nm or more).

  Here, assuming that the frequency of the light source is f (λ) and the spectral reflectance of each of the patches 1 to 8 of the spectral reflectance chart 10 is g (λ), coordinates of the color matching space using the XYZ color matching function (Equation 1). Ask for.

X = ∫x (λ) g (λ) dλ
Y = ∫y (λ) g (λ) dλ (Formula 1)
Z = ∫z (λ) g (λ) dλ
By multiplying f (λ) and g (λ), an XYZ stimulus value that is a color actually reflected in human eyes can be obtained. The XYZ stimulus value represents a color actually reflected in the human eye using X, Y, and Z instead of R (red), G (green), and B (blue). Here, the numerical value of X represents a red component, the numerical value of Y represents a green component, and the numerical value of Z represents a blue component.

  Note that by using spectral sensitivity functions of digital cameras, scanners, etc. instead of the XYZ color matching functions, it is possible to simulate the color reproducibility of light having different wavelengths of the patches 1 to 8.

  Next, a method of using the spectral sensitivity chart 10 in color design of a digital camera will be described.

  First, a spectral reflectance chart 10 having a plurality of patches 1 to 8 of materials having different spectral reflectances including light other than visible light is photographed (first step). At this time, a stable light source such as an artificial solar lamp is used as the light source.

  After this first step, the spectral reflectance of patches 1 to 8 is measured with a measuring instrument using a white reference plate as the light source used for photographing, and the spectral reflectance of the light source (see FIG. 3) is obtained (second). Process).

  After this second step, an XYZ stimulus value is obtained using the spectral reflectance of the patches 1 to 8 of the spectral reflectance chart 10, the spectral reflectance of the light source used for photographing, and the XYZ color matching function (third step).

  After this third step, the XYZ stimulus value is compared with the image data acquired in the first step, and the colorimetric value and digital value of the color (determined by the color matching function and the XYZ stimulus value) when viewed with the naked eye The difference from the RGB value of the camera is evaluated (fourth step).

  Based on the numerically obtained evaluation results, a spectral cut filter (filter), an image sensor, and the like used for the digital camera are selected.

  By the above process, the difference between the color actually seen by human eyes and the color (RGB) obtained by the image sensor of the digital camera can be obtained numerically. In particular, if the spectral reflectance chart 10 is prepared in advance as in the above embodiment, the color reproducibility of the digital camera can be easily brought close to the spectral sensitivity of the human eye.

  According to this embodiment, since the color reproducibility of a digital camera can be easily brought close to the spectral sensitivity of the human eye, a spectral cut filter (filter) for light having a wavelength of 700 nm or more that cannot be identified by the human eye. The operation of adjusting the characteristics (not shown) and selecting the image sensor can be simplified. As a result, the color reproducibility of the digital camera can be performed efficiently.

  In addition, since a single chart having a plurality of patches 1 to 8 having different peak values of reflectance is recorded as image data on a recording medium, patches having the same spectral reflectance (stable spectral sensitivity) are always used. Selection work can be performed.

  Of course, the number of patches in the chart is not limited to eight, but may be nine or more.

FIG. 1 is a diagram showing a spectral reflectance chart. FIG. 2 is a diagram showing the spectral reflectance of each patch in FIG. FIG. 3 is a diagram showing an example of the spectral distribution of the light source.

Explanation of symbols

  1-8: Patch, 10: Spectral sensitivity chart (color chart).

Claims (5)

A first step of photographing a color sample having patches of a plurality of materials having different spectral reflectances including light other than visible light;
After the first step, a second step of measuring a light source used for the photographing,
After the second step, a third step of obtaining an XYZ stimulus value using a spectral reflectance of each patch of the color sample, a spectral reflectance of a light source used for the photographing, and an XYZ color matching function;
An image analysis method comprising: after the third step, a fourth step of comparing the XYZ stimulus value with the image data acquired in the first step.   The image analysis method according to claim 1, wherein after the fourth step, a filter and an image sensor used for the photographing are selected.   The image analysis method according to claim 1, wherein the color sample is a color chart.   The image analysis method according to claim 3, wherein the color chart is stored as image data in a predetermined storage medium.   A color sample comprising a plurality of patches of materials having different spectral reflectances including light other than visible light.

Images