JP2007147507A - Spectrometry and optical spectrometer - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide spectrometry capable of measuring a large number of bands, and having short measurement time and without displacement in the spectroscopic image of each band. <P>SOLUTION: The object to be measured 100 is imaged, by using a standard camera C1 and reference cameras C2, C3 to generate standard image data and reference image data by using respective cameras. The reference image data produced by the reference cameras are subjected to projective transformation, as though photographing has been performed from the position of the standard camera C1 to produce the transformed image data, such that the coordinates, indicating arbitrary point on the object to be measured in the image indicates the coordinates on the same point as the object to the measured of the standard image data photographed by the standard camera. The coordinates are made common, between the standard image data photographed by the standard camera and the transformed image data, and spectroscopic data of an arbitrary point on the object to be measured on the coordinates are generated. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a spectroscopic measurement method and a spectroscopic measurement apparatus that perform image inspection in an inspection field by spectroscopic analysis in an industrial production process, and in particular, a spectroscopic measurement method and spectroscopic method that perform image inspection by spectroscopic analysis at each wavelength of ultraviolet, visible, and infrared. It relates to a measuring device.

  In industrial production processes, image inspection is widely performed in which the surface of a product is photographed with a camera and the surface properties are inspected. The image inspection may be performed by spectroscopic analysis measured using a plurality of wavelengths having different wavelength characteristics. Here, in order to obtain a plurality of images having different wavelength characteristics, the following methods are conventionally performed.

  As the first method, for example, as disclosed in Japanese Patent Application Laid-Open No. 2003-309858 (Patent Document 1), a multi-plate camera having a plurality of (for example, three) solid-state image sensors in one camera is used. This is a method of providing a band-pass filter that transmits only light of a specific wavelength on an optical path incident on each solid-state imaging device. In this method, a single camera is used, and a spectral image can be obtained with the number of bands equal to the number of solid-state imaging elements mounted on the camera. In addition, since the light incident on each image sensor is incident on the camera through a common optical axis, the image of each band does not shift in position due to the shift of the optical axis, and the position of the image data in each band. Can be obtained. Therefore, since the coordinates of the measurement point on the specific image can be shared with the image of the measurement point of another image, the spectral data can be easily used.

  The second method includes a filter switching mechanism including a plurality of filters having different wavelength characteristics in front of one camera, and repeatedly performs imaging while switching a bandpass filter that transmits only light of a specific wavelength during imaging. It is. According to this method, spectral data can be obtained by the number of filters provided in the filter switching mechanism, and spectral data having a large number of bands can be obtained.

The third method is disclosed, for example, in JP-A-10-300580 (Patent Document 2). This method uses a single camera, but the data imaged on the built-in CCD has wavelength information in the X-axis direction and position information of the measurement object in the Y-axis direction. In this case, the number of spectral data is determined by the spectral capability of the built-in diffraction grating and the number of arrays in the X-axis direction of the CCD, but the data obtained in a single measurement is the surface information only from the line area of the measurement object. In order to obtain image data, the camera itself is scanned in the direction perpendicular to the line information, and measurement is continued, so that image data having a resolution determined by the number of times of measurement and the moving distance can be obtained.
JP 2003-309858 A Japanese Patent Laid-Open No. 10-300580

  However, the first method has a limit on the number of solid-state imaging devices that can be mounted on one camera, and currently, for example, only spectral data up to three wavelengths of RGB and at most four wavelengths can be obtained. . If the number of solid-state imaging elements is increased further, the structure of the optical system becomes complicated, which is not practical.

  In addition, the second method requires repeated imaging for the number of bands, so the measurement time becomes very long, and the images of the obtained number of bands are not taken at the same timing. There's a problem. Therefore, it is disadvantageous for the analysis of a moving object. Further, since a mechanical part for switching the bandpass filter on the optical axis of the camera is essential, there are problems that long-term reliability is low and cost is high.

  Further, the third method dramatically increases the number of spectral data compared to the first and second methods. It is also possible to acquire several hundreds to thousands of spectral data. However, in terms of high-speed measurement of the object surface, scanning in one axis direction is necessary, so that it does not reach the first method. However, since the data acquisition in the line direction is at the camera rate, it is fast.

  In comparison with the second method, if the number of bands of the second method is the same as that of the third method, the wavelength switching mechanism system of the second method and the uniaxial direction of the third method It is a comparison of which is simpler and more reliable about the scanning mechanism. The second method has few problems of image positional deviation, but the third method has a problem that pixel deviation and wavelength deviation in the scanning direction coexist. Further, there is no guarantee that a reference sample having a known spectral intensity distribution pattern does not change with time. Although correction is performed using the pattern, it is difficult to completely eliminate the shift correction because the interpolation is performed between the patterns. Further, since the spectral image of the optical system is shifted after the correction work until the next correction, it is not a fundamental solution. On the other hand, if the correction interval is shortened, measurement cannot be performed during that time, so that the measurement efficiency decreases.

  Therefore, the technical problem to be solved by the present invention is a spectroscopic measurement method and a spectroscopic measurement apparatus that can measure a large number of bands and that the measurement time is short, and that there is no positional deviation of spectral images of each band. Is to provide.

  In order to achieve the above object, the present invention is configured as follows.

According to the first aspect of the present invention, an object to be measured is photographed using a standard camera and a reference camera having a wavelength sensitivity characteristic different from that of the standard camera, and standard image data and reference image data are created by each camera. ,
The reference image data created by the reference camera is converted as if it was photographed from the position of the standard camera, and the coordinates indicating an arbitrary point on the object to be measured in the image are covered by the standard image data photographed by the standard camera. Create converted image data that shows the coordinates on the same point on the measurement object,
Spectral measurement characterized in that the coordinates of the reference image data photographed by the reference camera and the converted image data are shared to generate spectral data of an arbitrary point on the object to be measured on the coordinates. Provide a method.

According to the second aspect of the present invention, the object to be measured is photographed using a plurality of reference cameras each having different wavelength sensitivity characteristics, and reference image data is created by each camera,
A plurality of reference image data created by the plurality of reference cameras is converted as if taken from the position of a virtual reference camera, and coordinates indicating arbitrary points on the object to be measured in the image are positions of the virtual reference camera. Create converted image data that shows the coordinates on the same point on the measured object of the virtual reference image data taken from
There is provided a spectroscopic measurement method characterized by sharing the coordinates with the converted image data and generating spectroscopic data of an arbitrary point on the object to be measured on the coordinates.

  According to the third aspect of the present invention, the reference camera and the reference camera start acquisition of an image by a trigger signal that is continuously input every predetermined time, and coordinates on the same point on the object to be measured. The converted image data indicating is the data at the same time, and the spectral data of an arbitrary point on the object to be measured on the coordinates is spectral data at the previous trigger time, The spectroscopic measurement method according to the first or second aspect is provided.

According to the fourth aspect of the present invention, when each of the objects to be measured is imaged using the reference camera and the reference camera, the image is taken with a reference point attached to the surface of the object to be measured,
In the creation of the converted image data, the conversion formula of the converted image data is calculated from the reference image data using the coordinates of the reference point in the standard image data and the reference image data, and each coordinate of the reference image data is converted into the converted image data. A spectroscopic measurement method according to any one of the first to third aspects is provided, wherein converted image data is created by substituting into an equation.

  According to a fifth aspect of the present invention, the reference point is attached to the surface of the object to be measured by projecting light including the reference point onto the surface of the object to be measured from a light projecting device. The spectroscopic measurement method according to the fourth aspect is provided.

  According to a sixth aspect of the present invention, in any one of the first to fifth aspects, the reference camera is composed of the reference camera and a plurality of cameras having different wavelength sensitivity characteristics. A spectroscopic measurement method is provided.

  According to a seventh aspect of the present invention, the reference camera and the reference camera have bandpass filters that transmit only light of a specific wavelength, and have different wavelength sensitivity characteristics. The spectroscopic measurement method according to any one of the sixth aspect is provided.

  According to an eighth aspect of the present invention, in any one of the first to seventh aspects, the reference camera and the reference camera have wavelength sensitivity characteristics approximate to a color matching function of an XYZ color system. One spectroscopic method is provided.

According to the ninth aspect of the present invention, the wavelength sensitivity characteristic is different from that of the reference camera that images the object to be measured and outputs the reference image data, and the reference camera that images the object to be measured and outputs the reference image data. A reference camera;
An image data storage unit for storing the standard image data and the reference image data;
The reference image data stored in the image data storage unit is converted so as to be photographed from the position of the standard camera, and coordinates indicating an arbitrary point on the measurement object in the reference image data are photographed by the standard camera. A converted image creating unit that creates converted image data indicating coordinates on the same point on the object to be measured of the reference image data that has been obtained,
A spectrum data creating unit that creates the spectral data of an arbitrary point on the object to be measured on the coordinates by sharing the coordinates of the reference image data and the converted image data captured by the reference camera; A spectroscopic measurement device is provided.

According to the tenth aspect of the present invention, a plurality of reference cameras having different wavelength sensitivity characteristics that shoot the object to be measured and generate reference image data, respectively,
An image data storage unit for storing the reference image data,
A plurality of reference image data created by the plurality of reference cameras are converted as if taken from the position of the virtual reference camera, and coordinates indicating arbitrary points on the object to be measured in the image are taken by the virtual reference camera. A converted image creating unit that creates converted image data indicating coordinates on the same point on the object to be measured of the virtual reference image data that has been obtained,
A spectroscopic measurement device is provided, comprising: a spectral data creation unit that creates the spectral data of an arbitrary point on the object to be measured on the coordinates by sharing the coordinates with the converted image data.

  In the present invention, when the same image is measured by the area cameras having different wavelength sensitivity characteristics, the positional deviation of the images coming from different camera positions is corrected, and the data from the same point of the measurement object is the same additional group number. The image data is digitized and converted so that it can be stored in a memory. Since the image data photographed by each camera indicates spectral data at one point of the object to be measured, the spectral data can be easily measured at high speed by sharing luminance data for each coordinate of each image. be able to.

  In addition, by using a trigger signal as means for transmitting to each camera that the time is the same, the data of each camera at the measurement point at the same time and the same location can be used, and the measurement point can be inspected more accurately. be able to.

  Hereinafter, a spectrometer according to an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing a schematic configuration of a spectrometer according to an embodiment of the present invention. The spectroscopic measurement apparatus 1 shoots a device under test 100 and outputs a plurality of cameras 11, 12, and 13 that can output image data, and a processing calculation unit 20 that processes and calculates image data output from the cameras 11, 12, and 13. A light source 30 that irradiates the device under test 100, an operation unit 40 that operates the processing arithmetic unit 20, and an output unit 50 that includes an output device such as a display or a printer.

  The plurality of cameras 11, 12, and 13 are cameras having different spectral characteristics from each other, and a camera having optical characteristics at a specific wavelength from infrared to ultraviolet is appropriately selected and used as many as the required number of bands. be able to. In addition, information about the relative positions of the plurality of cameras 11, 12, and 13 is known, and the information is stored in the processing calculation unit 20. The plurality of cameras 11, 12, and 13 function as reference cameras or reference cameras. The reference camera is a camera based on the camera position in the projective transformation described later, and the reference camera is a camera in which the captured image data is projectively converted as if it was taken from the camera position of the standard camera in the projective transformation. It is. That is, the difference between the two is the difference between the reference and the reference in the projective transformation process, and no substantial difference in configuration is observed.

  The light source 30 is a light source for irradiating the measurement object 100 with a reference point necessary for image processing in the processing calculation unit 20, and specifically, it is configured to project an arbitrary image such as a liquid crystal projector. preferable. Information about the projected image is stored in the processing calculation unit 20. The operation unit 40 includes a key input switch operated by a measurer. Note that the processing operation unit 20 can actually be configured using a personal computer, and can execute various processes described later by causing the computer to execute a predetermined program.

  FIG. 2 is a block diagram showing a functional configuration of a main part of the processing operation unit 20 in FIG. Image data output from the cameras 11, 12, and 13 is converted into a digital signal by the A / D conversion unit 21 for each light receiving element and input to the projection conversion unit 22. A camera position information memory 26 that stores the relative position information of each camera necessary for calculation in the projective conversion is connected to the projective conversion unit 22, and each camera is processed by a process as described later based on the position information. Projective transformation is performed on the image data captured.

  In the projective transformation process, a projection image irradiated from the light source 30 onto the object to be measured 100 is used as the reference point of each image. FIG. 3 is a diagram illustrating an example of a projected image projected on the surface 60 of the DUT 100. The projected image 62 is a lattice pattern, and an arbitrary point in the vertical line and horizontal line cross point 62 can be used as a reference point. However, in each image, the reference point needs to be at the same position in the projection image. Further, at least seven reference points are necessary, and projective transformation is performed using the coordinates of the reference points on the image.

  The spectral image corrected by the projection conversion unit 22 is input to the coordinate calculation unit 23. The coordinate calculation unit 23 integrates the luminance at the same coordinates on the image with respect to the image information from each of the cameras 11, 12, and 13 subjected to the projective transformation to obtain spectral data. The spectral data created by the coordinate calculation unit 23 is converted into a synthesized spectral image by the image synthesis unit 24 and output to the output unit 50. Hereinafter, processing operations performed in the processing operation unit 20 of the spectroscopic measurement apparatus having the above-described configuration and application examples of the spectroscopic measurement apparatus will be described in detail.

(Example 1) Visible light spectral spectrum area measuring apparatus using seven cameras As shown in FIG. 4, one camera is arranged at the center and six cameras around it, and numbers C1 to C7 are assigned respectively. Each camera uses seven black and white digital video camera modules XCL-V500 (manufactured by Sony Corporation). Since they are installed in a dense state, they are arranged with a positional relationship as shown in FIG. This camera is capable of digital output at 60 frames per second.

  Each of the seven cameras is equipped with the same type of lens that can focus the measurement surface, and filters corresponding to C1 to C7 are attached in front of the lenses. The filter is a bandpass filter, and its spectral characteristics are shown in FIG. The seven cameras C <b> 1 to C <b> 7 start photographing the objects to be measured 100 all at once in response to a synchronization signal generated from the processing calculation unit 20. The data is transmitted to the processing calculation unit 20 as described above, and the following processing is performed.

Hereinafter, in the image data calculated by the processing calculation unit 20, the pixel (I, J) of the camera K is represented as (I, J, K, L) in the Lth measurement data, and the image data is S _{Expressed as ijkl} . Since data is acquired by all the cameras at the same time by the synchronization signal, in the following, since l is described as the same data, description of l is omitted as S _ijk l = S _ijk .

  First, the surface of the object to be measured is irradiated with a lattice pattern from the light source 30 and measured with seven cameras. The grating pattern is irradiated from a light source including light having a wide wavelength that can be sufficiently measured in the range of 400 nm to 700 nm. As shown in FIG. 4, the respective cameras C <b> 1 to C <b> 7 are slightly different in position from the measurement surface, and thus the obtained image data is different.

Using the C1 camera installed at the center of FIG. 4 as a reference camera, the pixels of the 25 cross points 62 of the lattice pattern shown in FIG. 5 are obtained by a computer. Now focus only on the first camera (C2). The number h-th pixel (X, Y) at the cross point 62 of the camera is defined as (X _1-h , Y _1-h ). Next, the number h-th pixel (X, Y) of the cross point 12 of the second camera is set to (X _2-h , Y _2-h ).

Hereafter,
[X ₁ ] = [P ₂ ] [X ₂ ] and is represented by a matrix. Here, [P ₂ ] is a constant matrix for projectively transforming the image of the camera C2 to the position of the camera C1, which is the reference camera.

From the above formula (1),
[P ₂ ] = ([X ₁ ] [X ₂ ] ^t ) ([X ₂ ] [X ₂ ] ^t ) ⁻¹
[P ₂ ] is obtained as follows. Using [P ₂ ] obtained in this way, the coordinates on the image of the reference camera C2 can be projectively transformed into the coordinates on the image of the camera C1. This is the simplest conversion for matching the image of the camera C2 with the camera C1. According to the above equation, only rotation and linear distortion can be corrected.

Note that, in the correction of the image between the camera C1 and the camera C2, [P ₂ ] is calculated using Equation (2) when parallel movement is required between the images. If non-linear conversion is also necessary, a parameter of a quadratic equation may be introduced and [P ₂ ] may be calculated using equation (3).

[P ₃ ] to [P ₇ ] are obtained from the third camera C3 to the seventh camera C7 in the same manner as described above. In the case of the simplest conversion example, the (I, J) pixel data of the camera K of the object to be measured is obtained as the following equation (4).

  Here, since I and J are the coordinates of the image data taken by the reference cameras C2 to C7, 1 to N and M (N and M are the numbers of pixels in the X-axis and Y-axis directions of the image data, respectively. It is a natural number. I 'and J' include a decimal point as a calculation result, but are rounded to a natural number.

S _IJK1 is stored in the memory address M (I ′, J ′, K, 1) of I ′, J ′ obtained in this way.
Since the camera C1 is a reference camera, only each pixel of the image data photographed by the camera C1 is M (I, J, 1,1) = S _IJ11 . This is performed for each of the cameras 1 to 7 from 1 to N and J from 1 to M for each pixel (I, J).

Similarly, S _IJK2 is stored in M (I ′, J ′, K, 2) at the next synchronization timing.

Thereby, the spectral data of 7 wavelengths corresponding to the pixel (I, J) at the measurement symmetry position at time t is as _follows : S _IJ1t S _IJ2t S _IJ3t S _IJ4t S _IJ5t S _IJ6t S _IJ7t

  According to the present embodiment, spectral data of 7 bands corresponding to the pixel (I, J) at the measurement symmetry position at time t can be easily obtained by only accessing data on the memory of the computer.

(Example 2) Example of water content measurement system of measurement symmetry plane by three infrared cameras and one visible camera Each camera arrangement is shown in FIG. A visible camera C1 is installed at the center. Three infrared cameras of the same type are installed symmetrically, and each camera is focused on the surface of the object to be measured.

  In general, an infrared camera has fewer pixels than a visible camera and can only measure a blurred image. Moisture has absorption in the infrared spectral region. Even with water that is transparent in the visible light region, the infrared camera can measure and detect the amount of water due to its strong absorption. Industrial measurement lines often require product water detection. This example shows an example of a spectroscopic measurement device that measures the moisture content of a surface treatment film on a plated steel plate in the manufacture of a steel plate.

  Infrared cameras are usually expensive, and there are many single-plate cameras in terms of sensitivity. However, in the case of a single-panel camera, only a fixed sensitivity pattern in the infrared spectral region can be measured, so accurate measurement is possible due to fluctuations in the intensity of the light source projecting on the measurement target or fluctuations in the sensitivity of the infrared camera sensor. Is difficult. When measuring moisture normally, select a wavelength with a strong absorption in the absorption spectrum of water (measurement wavelength below) and a wavelength with little or no absorption, where the wavelength is longer or shorter than the measurement wavelength. Hereinafter, these will be referred to as reference wavelength 1 and reference wavelength 2, respectively (see FIG. 11). Using these, a baseline is obtained, and a calculation (hereinafter referred to as a three-wavelength calculation) is performed so that only the amount of water absorption can be accurately obtained. In this embodiment, the measurement wavelength is 2.8 μm, the reference wavelength 1 is 2.3 μm, and the reference wavelength 2 is 3.9 μm.

  The infrared camera C2 has a bandpass filter that matches the measurement wavelength, the infrared camera C3 has a bandpass filter that matches the reference wavelength 1, and the infrared camera C4 has a bandpass filter that matches the reference wavelength 2. Install a filter. Note that the infrared camera used in this example is model number PV320-LD manufactured by Electrophysics.

  The distortion of images obtained from the four cameras is performed by the same method as the processing method shown in the first embodiment. However, the difference is that the number of pixels differs between the visible camera and the infrared camera. The number of pixels is usually larger for a visible camera. Here, the visible camera C1 installed in the center is used as a reference camera, and a memory suitable for the pixel resolution is prepared. Since the infrared camera has a small number of pixels, a memory address M (I, J, K, L) corresponding to the infrared camera has no data. For this reason, the infrared camera data may not be obtained at the location specified by the visible camera pixel (I, J), so it is calculated by interpolation from the memory address data at the nearest point where the data exists. The value is obtained from and stored. Although it is not actual measured data, an approximate value is calculated.

  The visible camera C1 has a faster response speed than the infrared camera. Therefore, in the process measurement, regarding the measurement object flowing on the belt conveyor or the like, the shape corresponding to the marker captured from the visible camera is determined, the position is determined, and the measurement timing is generated. In synchronization with the generation timing, photographing is performed with four cameras including a visible camera, and the data is stored in the memory. The image distortion correction in that case is the same as the method described in the first embodiment.

  Next, M (I, J, 1, t), M (I, J, 2, t), M (I, J, 3, t) of infrared cameras C2 to C4 (K = 1,2,3) ) The following calculation is performed on the data. M (I, J, 1, t) is the camera with the measurement wavelength, M (I, J, 2, t) is the camera with the reference wavelength 1, and M (I, J, 3, t) is the reference wavelength 2. It is a camera.

  First, the data when there is no moisture is measured in advance, and M (I, J, 1,0), M (I, J, 2,0), M (I, J, 3,0) Keep it as

next,
T (I, J, 1, t) = M (I, J, 1, t) / M (I, J, 1,0)
T (I, J, 2, t) = M (I, J, 2, t) / M (I, J, 2,0)
T (I, J, 3, t) = M (I, J, 3, t) / M (I, J, 3,0)
After asking
A (I, J, 1, t) = −LOG ₁₀ [T (I, J, 1, t)]
A (I, J, 2, t) = −LOG ₁₀ [T (I, J, 2, t)]
A (I, J, 3, t) = −LOG ₁₀ [T (I, J, 3, t)]
Is calculated and the following equation is calculated.
S (I, J, t) = A (I, J, 1, t) − {A (I, J, 2, t) + A (I, J, 3, t)} / 2

  This S (I, J, t) is proportional to the amount of water corresponding to the pixel (I, J) at the measurement symmetrical position. As a result, image data of the amount of moisture can be obtained. This image data is used to adjust the dryer in the previous stage according to the increase or decrease of the moisture content.

Example 3 Tristimulus Value Direct-Reading Color Measurement System Example Using Multiple CCD Cameras Spectral sensitivity corresponding to the human eye representing the sensitivity to distinguish colors is called a color matching function. The spectroscopic measurement apparatus according to this embodiment is a stimulus value direct-reading type color measuring machine, and measures a sample with sensors x (λ), y (λ), and z (λ) having response sensitivities approximated to color matching functions. Then, three values of X, Y, and Z that are directly called “tristimulus values” are measured.

  Here, the X, Y, and Z are obtained by the following equations (5), (6), and (7), respectively.

  K is obtained by the following equation (8). Here, K is a definition formula indicating the normalization of brightness.

  The spectral sensitivity (color matching function) corresponding to the human eye is as shown in FIG. In this embodiment, four CCD cameras x1, x2, y, and z having spectral responsivities approximating the color matching function are prepared, and the outputs are set as tristimulus values X, Y, and Z. Calculate chromaticity coordinates. Each camera is provided with a filter and has the spectral characteristics shown in FIG. Further, the four CCD cameras x1, x2, y, and z are arranged in a grid as shown in FIG. Each camera takes a picture at the same time in synchronization with a signal generated from the processing calculation unit 20 and stores the data in a memory. The image distortion correction in that case is the same as the method described in the first embodiment.

  As shown in the figure, because the position of the field of view varies from camera to camera due to the CCD camera arrangement, as shown in FIG. 10, a grid pattern 66 (calibration plate) is arranged in advance for correction, and an image is captured by each camera. A common area 67 in the field of view of each camera is registered in the processing calculation unit 20. Then, the common area 67 is extracted from the captured images of the cameras, and these are combined to obtain a measurement image.

  The stimulus value direct reading type color measuring machine according to the present embodiment has the following effects. That is, the spectral responsivity approximated to the color matching function has many portions that overlap each response sensitivity, and therefore, incident light cannot be divided by a dichroic mirror employed in a general three-plate camera optical system. Further, if the incident light is divided by the beam splitter, there is a problem that the amount of light after the division becomes as small as 1/2 or 1/4 of the incident light and the dynamic range is lowered. By using an independent CCD camera as in this embodiment, it is possible to directly receive incident light and improve the dynamic range. Further, it is not necessary to adjust the pixel position for each CCD sensor that is performed when the three-plate camera system is assembled.

  In addition, this invention is not limited to the said embodiment, It can implement in another various aspect. For example, the number of cameras used in the spectroscopic measurement device is not particularly limited, and an appropriate number may be used according to the number of bands required. Further, in the above-described embodiment, the projective transformation is performed so that an image captured by each camera is captured from the position of one of the cameras (reference camera), but this is assumed to be a virtual reference camera, It is also possible to perform projective transformation as if it was taken from the shooting position.

It is a block diagram which shows schematic structure of the spectrometer which concerns on embodiment of this invention. It is a block diagram which shows the function structure of the process calculating part in FIG. It is a figure which shows the example of the projection image of the lattice pattern projected on the surface of the to-be-measured object. FIG. 3 is a diagram illustrating a camera arrangement in the first embodiment. It is a figure which shows the spectral characteristic of each camera of FIG. FIG. 10 is a diagram illustrating a camera arrangement in a second embodiment. It is a figure which shows the spectral sensitivity corresponding to a human eye. FIG. 10 is a diagram illustrating spectral characteristics of a camera in Example 3. FIG. 10 is a diagram illustrating a camera arrangement in Example 3. FIG. 10 is an explanatory diagram of a common area extraction process in the third embodiment. It is a figure which shows the absorption spectrum of water.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11, 12, 13 Camera 20 Processing calculation part 21 A / D conversion part 22 Projection conversion part 23 Coordinate calculation part 24 Image composition part 26 Memory 30 Light source 40 Operation part 50 Output part 60 Measuring object surface 62 Projection image 100 Measuring object

Claims (10)

Shoot a measurement object using a reference camera and a reference camera having a wavelength sensitivity characteristic different from that of the reference camera, and create reference image data and reference image data with each camera,
The reference image data created by the reference camera is converted as if it was photographed from the position of the standard camera, and the coordinates indicating an arbitrary point on the object to be measured in the image are covered by the standard image data photographed by the standard camera. Create converted image data that shows the coordinates on the same point on the measurement object,
Spectral measurement characterized in that the coordinates of the reference image data photographed by the reference camera and the converted image data are shared to generate spectral data of an arbitrary point on the object to be measured on the coordinates. Method. Photograph the object to be measured using multiple reference cameras with different wavelength sensitivity characteristics, create reference image data with each camera,
A plurality of reference image data created by the plurality of reference cameras is converted as if taken from the position of a virtual reference camera, and coordinates indicating arbitrary points on the object to be measured in the image are positions of the virtual reference camera. Create converted image data that shows the coordinates on the same point on the measured object of the virtual reference image data taken from
A spectroscopic measurement method characterized by sharing the coordinates with the converted image data and creating spectroscopic data of an arbitrary point on the object to be measured on the coordinates.   The reference camera and the reference camera start acquisition of an image by a trigger signal that is continuously input every predetermined time, and converted image data indicating coordinates on the same point on the object to be measured are at the same time. The spectroscopic measurement method according to claim 1, wherein the spectroscopic data of an arbitrary point on the object to be measured on the coordinates is spectroscopic data at the previous trigger time. . When each of the objects to be measured is photographed using the reference camera and the reference camera, it is photographed with a reference point attached to the surface of the object to be measured.
In the creation of the converted image data, the conversion formula of the converted image data is calculated from the reference image data using the coordinates of the reference point in the standard image data and the reference image data, and each coordinate of the reference image data is converted into the converted image data. 4. The spectroscopic measurement method according to claim 1, wherein converted image data is created by substituting into an equation.   The spectroscopic measurement method according to claim 4, wherein the reference point is attached to the surface of the object to be measured by projecting light including the reference point onto the surface of the object to be measured from a light projecting device. .   The spectroscopic measurement method according to claim 1, wherein the reference camera includes the reference camera and a plurality of cameras each having different wavelength sensitivity characteristics.   The said reference camera and the said reference camera are equipped with the band pass filter which permeate | transmits only the light of a specific wavelength, respectively, and each has a different wavelength sensitivity characteristic, The one characterized by the above-mentioned. Spectroscopic measurement method.   The spectroscopic measurement method according to claim 1, wherein the reference camera and the reference camera have wavelength sensitivity characteristics approximate to a color matching function of an XYZ color system. A reference camera that shoots an object to be measured and outputs standard image data; a reference camera that shoots the object to be measured and outputs reference image data;
An image data storage unit for storing the standard image data and the reference image data;
The reference image data stored in the image data storage unit is converted so as to be photographed from the position of the standard camera, and coordinates indicating an arbitrary point on the measurement object in the reference image data are photographed by the standard camera. A converted image creating unit that creates converted image data indicating coordinates on the same point on the object to be measured of the reference image data that has been obtained,
A spectrum data creating unit that creates the spectral data of an arbitrary point on the object to be measured on the coordinates by sharing the coordinates of the reference image data and the converted image data captured by the reference camera; A spectroscopic measurement device. A plurality of reference cameras having different wavelength sensitivity characteristics that shoot a measurement object and create reference image data, respectively,
An image data storage unit for storing the reference image data,
A plurality of reference image data created by the plurality of reference cameras are converted as if taken from the position of the virtual reference camera, and coordinates indicating arbitrary points on the object to be measured in the image are taken by the virtual reference camera. A converted image creating unit that creates converted image data indicating coordinates on the same point on the object to be measured of the virtual reference image data that has been obtained,
A spectroscopic measurement apparatus comprising: a spectral data creating unit that creates a spectral data of an arbitrary point on the measurement object on the coordinates by sharing the coordinates with the converted image data.

Images