JP4859502B2 - Imaging device - Google Patents

Description

  The present invention relates to an imaging apparatus that generates a color image based on an image signal read from an imaging element, and more particularly to color reproducibility.

  In an imaging apparatus such as a digital camera, a primary color filter or a complementary color filter is arranged on the light receiving surface of an imaging element, and color elements such as R, G, and B are arranged in a mosaic pattern so as to correspond to each pixel position. Has been. When the reflected light from the subject passes through the color filter, an image signal composed of color signals corresponding to each color element is generated and read out from the image sensor. The read image signal is subjected to color conversion processing by matrix calculation or the like in order to generate an image signal conforming to a color space standardized based on colorimetry. And it outputs to external devices, such as a monitor, as predetermined | prescribed video signals, such as a NTSC signal, R, G, and B component signal.

In digital cameras, various color filter arrangement methods and color conversion processing methods are applied to achieve more faithful color reproducibility. For example, a filter in which color elements (G ′) having spectral sensitivity characteristics close to green (G) are arranged to obtain information on a spectral region in which the red (R) spectral sensitivity curve is negative in the color matching function. Is used to generate a corrected red (R ′) signal (see Patent Document 1). By converting to a new red (R ′) signal based on the color element (G ′), a wide range of the color gamut is reproduced. In addition to red (R), green (G), and blue (B), a four-color filter with color elements having different spectral transmission characteristics is used. R, G, B primary color signals are generated (see Patent Document 2). By using a four-color filter, colors are reproduced while reducing noise.
Japanese Patent No. 2872759 (pages 6 to 7, FIG. 2) JP 2003-284084 A (page 5, right column, 2nd line to 11th line, FIG. 18)

  Also in the imaging devices of Patent Documents 1 and 2 described above, the spectral transmission characteristics of the color filters do not completely match the ideal visibility distribution characteristics, so there is a problem in color reproducibility. In Patent Document 1, color elements of a filter that improve the reproducibility of green and blue intermediate colors are prepared. However, it is difficult to sufficiently reproduce other intermediate colors. Further, in Patent Document 2, R, G, and B primary color signals are generated based on a predetermined matrix coefficient. Therefore, for a subject having a specific spectral distribution characteristic (for example, a bluish subject). Although the color can be reproduced appropriately, appropriate color reproduction cannot be realized for an object having other spectral distribution characteristics (for example, a reddish object).

  The image pickup apparatus of the present invention is applicable to a digital camera or the like, and is an image pickup apparatus that can provide an image with excellent color reproducibility for any subject, and arranges at least four color elements. R, G, B primary color signals suitable for human visual sensitivity are generated using the filter.

  An image pickup apparatus according to the present invention includes an image pickup element, a color filter configured from at least four types of color elements and disposed on a light receiving surface of the image pickup element, and a series of colors corresponding to at least four types of color elements from the image pickup element. Signal reading means for reading a signal. The spectral transmission characteristics of the color elements of the color filter are different from each other, and are represented by a continuous spectral distribution curve having a peak at a predetermined wavelength. In addition, the spectral transmission characteristics of the color elements have a wavelength peak and a spectral distribution determined in consideration of the color matching function, and this spectral transmission characteristics include the imaging device and the infrared cut filter. Characterize spectral sensitivity characteristics. As the imaging method, a single plate type or a three plate type can be applied. In the case of a single plate simultaneous type, a single color filter (on-chip color filter) in which color elements are arranged in a checkered pattern is applied. The color filter may be composed of at least four types of color elements, and may be composed of either a primary color filter including R, G, B corresponding to the color matching function or a complementary color filter such as Y, Cy, Mg, G. . A series of color signals is composed of color component signals corresponding to the respective color elements. For example, when four types of color elements are arranged in a mosaic pattern, four adjacent pixels are determined as a block, and signal processing is performed in units of image signals composed of first to fourth color component signals.

The imaging apparatus of the present invention includes signal processing means for generating R, G, B primary color signals corresponding to tristimulus values, and the signal processing means is a spectral of an object included in a non-selected color signal among a series of color signals. Based on the distribution characteristic (spectral reflectance) information, an appropriate color signal is determined from the remaining plurality of selected color signals, and R, G, B primary colors corresponding to the tristimulus values based on the non-selected color signal and the appropriate color signal. Generate a signal. The non-selected color signal is a signal corresponding to a predetermined spectral sensitivity distribution curve among standard spectral sensitivity distribution curves based on the color space in consideration of the color matching function. For example, the color filters are R, G, B In the case of a primary color filter including a color filter, color signals obtained from color elements having spectral transmission characteristics corresponding to R and B are color signals corresponding to R and B spectral sensitivity curves of a standard spectral sensitivity distribution curve, that is, non-color filters. It is defined as a selection color signal. On the other hand, color signals obtained by color elements having spectral transmission characteristics (for example, G) correlated with the remaining spectral sensitivity distribution curves among the color matching functions are determined as a plurality of selection color signals. For example, a color signal obtained from at least two types of color elements having spectral transmission characteristics correlated with G is the selected color signal. For example, when the color element has a spectral transmission characteristic correlated with G, the color element has a spectral transmission characteristic having a wavelength peak between G and B, and has a spectral transmission characteristic having a wavelength peak between G and R. It consists of color elements. As a color space to be applied, a color space standardized laterally, for example, an XYZ system, an L ^* a ^* b ^* space, an L ^* u ^* v ^* space, an sRGB color space, or the like is applied. A standard spectral sensitivity distribution curve based on the color space corresponds to a color matching function.

  The generated color signal is affected by the spectral transmission characteristics of the color elements and the spectral distribution characteristics of the subject. The non-selected color signal includes information on the spectral distribution characteristic of the subject, that is, the reflectance characteristic, and the spectral distribution characteristic of the subject becomes clear by detecting the value of the non-selected color signal. For example, when photographing a subject with a bluish tint, the spectral distribution characteristic is relatively large in the spectrum of the wavelength region corresponding to blue. On the other hand, when shooting a reddish subject, the spectral distribution characteristic is relatively large in the spectrum of the wavelength region corresponding to red. When the wavelength region of light transmitted by a color element is in common with the wavelength region having a large spectrum in the spectral distribution characteristic of the subject, the color signal obtained from the color component sufficiently includes the color information of the subject. In order to determine the color information of the subject, for example, the color signal value corresponding to red (R) and blue (B) is determined to be larger or smaller than a predetermined value (threshold). Or you may judge by ratio of the color signal according to red (R) and blue (B).

  In the present invention, the non-selected color signal corresponds to the entire predetermined spectral sensitivity distribution curve. That is, the spectral sensitivity curve corresponding to the non-selected color signal among the sensitivity distribution characteristics of the imaging system substantially coincides or has a linear relationship over the entire spectral distribution of the predetermined spectral sensitivity distribution curve (for example, R, B). Have. On the other hand, the selected color signal corresponds to a part of the remaining spectral sensitivity distribution curve. That is, the spectral sensitivity curve (for example, G) corresponding to the selected color signal in the sensitivity distribution characteristics of the imaging system matches in a part of the remaining spectral sensitivity distribution curve (for example, only in the short wavelength region or only in the long wavelength region). Or have a linear relationship. Furthermore, in the present invention, the signal processing means determines a color signal that matches the remaining spectral sensitivity curve in a wavelength region having a relatively large spectrum in the spectral distribution characteristics of the subject to be imaged. That is, a color signal for deriving the spectral sensitivity characteristic of the imaging system substantially along the remaining spectral sensitivity curve of the color matching function is determined based on the plurality of selected color signals. For example, in the case of a subject with a bluish tint, a color signal that is as close as possible to the spectral sensitivity distribution curve in the vicinity of the short wavelength region is determined as the compatible color signal. The primary color signal generation means may select one of the plurality of selection color signals as the selection color signal, or may calculate and determine the plurality of selection color signals. For example, when a subject is bluish, a color signal obtained from a color element having a spectral transmission characteristic that relatively transmits light in a short wavelength region is selected as a compatible color signal, and the subject is reddish. The color signal obtained from the color element having the spectral transmission characteristic that relatively transmits light in the long wavelength region may be selected as the compatible color signal.

  Since the information on the spectral distribution characteristics of the subject is obtained directly from the predetermined color signal read from the image sensor, R, G, and B primary color signals are obtained based on the color signal suitable for the spectral distribution characteristics of the subject. .

  As processing for generating the R, G, and B primary color signals, first, matrix conversion for color conversion may be performed, and then a suitable color signal may be determined from the selected color signal. Alternatively, the matrix calculation may be performed after the suitable color signal is first determined. In the former case, the signal processing means includes a color conversion processing means for performing a matrix operation for color conversion on a series of color signals, and a spectral distribution characteristic of a subject included in a non-selected color signal that has been subjected to matrix operation among the series of color signals. Based on this information, a primary color that determines a compatible color signal from a plurality of selection color signals subjected to matrix operation and outputs R, G, B primary color signals based on the non-selected color signal and the compatible color signal subjected to matrix operation Signal generating means. In the latter case, the signal processing means includes a suitable color signal determining means for determining a suitable color signal from a plurality of selected color signals based on information on the spectral distribution characteristics of the subject included in the non-selected color signal among the series of color signals. And color conversion processing means for performing matrix calculation for color conversion on the non-selected color signal and the suitable color signal and outputting R, G, B primary color signals.

  In order to be able to cope with the spectral distribution characteristics of various subjects from the short wavelength range to the long wavelength range within the visible light wavelength range, the spectral transmission characteristics corresponding to at least four types of color elements are in the visible wavelength range. It is preferable that the spectral distribution has peak values at substantially equal intervals. A color signal corresponding to the spectral distribution characteristic can be obtained for a reddish subject, a bluish subject, and an intermediate color subject.

  The imaging method of the present invention is a series of colors corresponding to at least four types of color elements from an imaging element in which a color filter composed of at least four types of color elements having different spectral transmission characteristics is arranged on the light receiving surface. Based on information on the spectral distribution characteristics of the subject included in the non-selected color signal corresponding to a predetermined spectral sensitivity distribution curve out of a standard spectral sensitivity distribution curve based on the color space. A proper color signal is determined from at least two selected color signals correlated with the remaining spectral sensitivity distribution curve, and R, G, B primary color signals corresponding to the tristimulus values are generated based on the non-selected color signal and the proper color signal. The spectral transmission characteristic characterizes the spectral sensitivity characteristic of the imaging system, and the non-selected color signal corresponds over the entire wavelength region of the predetermined spectral sensitivity distribution curve, while the selected color signal is remaining A color signal that corresponds to a part of the wavelength range of the spectral sensitivity distribution curve and that matches the remaining spectral sensitivity distribution curve in a wavelength region in which the spectrum of the subject's spectral distribution characteristic is relatively large is determined as a compatible color signal. And

  The primary color signal generation processing apparatus according to the present invention includes at least four types of color elements that are read out from an image sensor on which a color filter composed of at least four types of color elements having different spectral transmission characteristics is arranged on the light receiving surface. A primary color signal generation processing apparatus that performs signal processing on a series of color signals according to a predetermined spectral sensitivity distribution curve among standard spectral sensitivity curves based on a color space in the series of color signals. Based on the information of the spectral distribution characteristics of the subject included in the non-selected color signal, an appropriate color signal is determined from at least two selected color signals correlated with the remaining spectral sensitivity distribution curves, and based on the non-selected color signal and the appropriate color signal Signal processing means for generating R, G, B primary color signals corresponding to the tristimulus values, the spectral transmission characteristic characterizes the spectral sensitivity characteristic of the imaging system, and the non-selected color signal has a predetermined spectral sensitivity component. Corresponding over the entire wavelength region of the curve, the selected color signal corresponds in a part of the wavelength region of the remaining spectral sensitivity distribution curve, and the signal processing means relatively compares the spectrum in the spectral distribution characteristics of the subject. A color signal that matches the remaining spectral sensitivity distribution curve in a large wavelength region is defined as a compatible color signal.

  According to the present invention, an image that faithfully reproduces the color of a subject can be obtained.

  Below, the digital camera which is this embodiment is demonstrated with reference to drawings.

  FIG. 1 is a block diagram of a digital camera.

  The digital camera 10 includes a photographing optical system 12 and a CCD 16, and light reflected from a subject passes through the photographing optical system and reaches the light receiving surface of the CCD 16. Thereby, a subject image is formed on the light receiving surface of the CCD 16. Here, a single plate simultaneous type is applied as an imaging method, and a color filter 14 in which four color elements are arranged in a checkered pattern is provided on the light receiving surface of the CCD 16. When the light from the subject reaches the light receiving surface of the CCD 16, a pixel signal corresponding to each color element of the color filter 14 is generated, and the pixel signal is read from the CCD 16 according to a clock pulse signal from a CCD drive circuit (not shown). The read pixel signal is amplified in an initial circuit (not shown), and sent to the matrix circuit 18 after a predetermined process.

  In the matrix circuit 18, color conversion processing is performed on a series of pixel signals corresponding to the four color elements of the color filter 14, and the first selection according to the four colors, that is, red (R) and green (G). A series of color signals corresponding to the color and the second selected color, blue (B) is generated. Then, in the G signal output circuit 20, the color corresponding to green (G) from the color signal corresponding to the first selection color and the second selection color based on the color signal corresponding to red (R) and blue (B). A signal is determined. As a result, R, G, and B primary color signals corresponding to the tristimulus values are generated and transmitted to the image signal processing circuit 22.

  In the image signal processing circuit 22, processing such as white balance adjustment and gamma correction is performed on the primary color signals of R, G, and B, and a video signal according to the sRGB color space is generated. Here, color conversion processing is performed so that colors specified in the standard monitor and the standard viewing environment are accurately reproduced in colorimetry. When the video signal is sent to the LCD driver 24, the LCD driver 24 drives the LCD 26 based on the video signal. As a result, the captured image is displayed on the LCD 26 as a moving image.

  When a release button (not shown) is pressed halfway and the half-push switch 42 is turned on, the distance to the subject is measured by the distance measuring sensor 30 and the brightness of the subject is detected by the photometric sensor 32. Then, the system control circuit 34 calculates the shutter speed and aperture value based on the measured subject information. Further, the photographing optical system 12 is driven for focus adjustment by the lens driving circuit 28 in accordance with the measured distance to the subject. When the release button is fully pressed and the full-press switch 44 is turned on, a shutter and a diaphragm (not shown) are driven, and a pixel signal for one frame is generated. The pixel signal for one frame read from the CCD 16 is processed in the matrix circuit 18, the G signal output circuit 20, and the image signal processing circuit 22. Then, after being compressed in the system control circuit 34, it is recorded in the memory card 36. Further, when a transmission button (not shown) is operated and the transmission switch 46 is turned on, the image data recorded on the memory card 36 is output to the external computer 40 via the interface circuit (I / F) 38. The

  FIG. 2 is a diagram showing a part of the color filter 14, and FIG. 3 is a diagram showing spectral sensitivity characteristics of the input system in the digital camera. The color arrangement of the filter will be described with reference to FIGS.

  As shown in FIG. 2, the color filter 14 is a Bayer-array filter in which blocks composed of four kinds of color elements are regularly arranged. The four types of color elements transmit blue light in a short wavelength region including blue light (hereinafter referred to as B0), from the blue (B) wavelength region to the green (G) wavelength region. Color element (hereinafter referred to as X), color element that transmits light from the green (G) wavelength range to the red (R) wavelength range (hereinafter referred to as Z), and a long wavelength range including red light The color element (hereinafter referred to as R0) that transmits the light of the above. The four color elements (B0, X, Z, R0) are arranged in a checkered pattern, and are arranged so that each color element faces each pixel on the light receiving surface of the CCD 16.

  FIG. 3 shows the spectral sensitivity characteristic in the input system of the digital camera 10 in which the spectral transmission characteristic of the color filter 14, the spectral sensitivity characteristic of the photographing optical system 12 or the infrared cut filter, and the spectral sensitivity characteristic of the CCD 16 are combined. The spectral sensitivity characteristics of the input system that are characterized by the spectral transmission characteristics of the color elements B0, X, Z, and R0 are all expressed by curves close to a Gaussian distribution, and are represented by “L1”, “L2”, “L3”, Let it be “L4”. The spectral sensitivity curves L1, L2, L3, and L4 are represented by spectral distributions having peak wavelengths λ1 (450 nm), λ2 (500 nm), λ3 (550 nm), and λ4 (590 nm), respectively.

  The spectral sensitivity curve L1 is a spectral sensitivity curve mainly corresponding to blue (B) light, and the spectral distribution (spectral sensitivity) region is in the range of 360 nm to 540 nm. The spectral sensitivity curve L4 is a spectral sensitivity curve mainly corresponding to red (R) light, and the spectral distribution region is in the range of about 520 nm to 680 nm. The spectral sensitivity curve L2 and the spectral sensitivity curve L3 draw lines that are substantially evenly spaced between the spectral sensitivity curve L1 and the spectral sensitivity curve L4, and the peak wavelengths λ1, λ2, λ3, and λ4 are substantially equidistant. Thus, spectral sensitivity curves L1, L2, L3, and L4 are determined.

  FIG. 4 is a diagram showing spectral sensitivity characteristics of an ideal imaging system, that is, an input system. FIG. 5 is a diagram showing spectral sensitivity characteristics of color signals corresponding to the color filters R0 and B0 among color signals obtained by matrix calculation. FIG. 6 is a diagram showing spectral sensitivity characteristics of color signals corresponding to the color filters X and Z among color signals obtained by matrix calculation. The spectral sensitivity characteristics will be described with reference to FIGS.

  The spectral sensitivity characteristic shown in FIG. 4 is a spectral sensitivity characteristic that can realize color reproduction appropriately in colorimetry, is represented by a color matching function, and is a spectroscopically defined spectral color according to human visual sensitivity. Corresponds to sensitivity characteristics. Here, spectral sensitivity curves corresponding to the color matching functions are represented as standard spectral sensitivity curves as Rs, Gs, and Bs. Here, Rs, Gs, and Bs indicate spectral sensitivity curves based on the sRGB color space as the color space.

The four color signals (hereinafter referred to as Bin, Xin, Zin, and Rin) obtained based on the spectral sensitivity characteristics of the input system shown in FIG. 3 are spectral sensitivity curves in the sRGB color space shown in FIG. In order to obtain a colorimetric value (device-independent color) based on the above, a matrix operation is performed in the matrix circuit 18 according to the following equation. As can be seen from the following equation, the matrix coefficient in the first row and third column is set to a relatively small value, and the matrix coefficient in the third row and second column is set to a relatively large value. In the above formula, each coefficient of the matrix is determined so as to satisfy the router condition as much as possible.

  The four color signals Rin, Xin, Zin, and Rin are subjected to color conversion processing to generate four color signals Rout, G1out, G2out, and Bout. Here, the color signals Rout and Bout correspond to X and Z among the tristimulus values X, Y and Z obtained from the color matching function, and are output as R and B primary color signals as they are. On the other hand, the color signals G1out and G2out correspond to Y among the tristimulus values, and a G primary color signal is output based on the color signals G1out and G2out. Hereinafter, the color signals G1out and G2out are referred to as a first selection color signal and a second selection color signal, respectively.

As shown in FIG. 5, the spectral sensitivity curves R ′s and B ′s after matrix calculation generally coincide with the standard spectral sensitivity curves Rs and Bs according to the sRGB space. That is, the spectral sensitivity distribution curves L1 and L4 of the spectral sensitivity characteristics of the input system shown in FIG. 3 are in a substantially linear relationship with the spectral sensitivity curves R ′s and B ′s. On the other hand, as shown in FIG. 6, the first selected color signal G1out, spectral sensitivity curve G1s after the matrix calculation in response to the second selection color signal G2out, G2s will vary as a boundary peak wavelength lambda _G. That is, in a short region of wavelengths than the peak wavelength lambda _G, while the spectral sensitivity curves G1s the first selection color signal G1out is substantially coincident with the standard spectral sensitivity curve Gs according to sRGB space, a long region wavelength than the peak wavelength lambda _G is The spectral sensitivity curve G2s of the second selection color signal G2out substantially coincides with the spectral sensitivity curve Gs. This means that the spectral sensitivity curve G1s has a linear relationship with the standard spectral sensitivity curve Gs in the short wavelength region, and the spectral sensitivity curve G2s has a linear relationship with the spectral sensitivity curve Gs in the long wavelength region. In the present embodiment, as shown below, the spectral sensitivity curve Gs based on the sRGB space is adapted based on the information on the spectral distribution characteristics (reflectance characteristics) of the subject included in the color signals Rout and Bout after the matrix calculation. A color signal G is determined.

  FIG. 7 is a flowchart showing a process of determining a color signal corresponding to G.

  In step S101, it is determined whether or not the value of the color signal Bout is greater than or equal to the boundary value B0 and the value of the color signal Rout is smaller than the boundary value R0. In the case of a subject having a spectral distribution characteristic in which the spectral component in the short wavelength region corresponding to blue is relatively large, the value of the color signal Bout is relatively larger than the value of the color signal Rout. Conversely, in the case of a subject having a spectral distribution characteristic in which the spectral component in the long wavelength region corresponding to red is relatively large, the value Rout of the color signal R is relatively larger than the value Bout of the color signal B. Here, for subjects with a spectral distribution characteristic with a large spectrum in the short wavelength region including blue light, subjects with a spectral distribution characteristic with a large spectrum in the long wavelength region including red light, and subjects with other spectral distribution characteristics. The boundary values B0 and R0 for classification and identification are determined in advance. The boundary values B0 and R0 are set to predetermined values based on the spectral sensitivity curves Rs and Bs, matrix coefficients, and the like.

  If it is determined in step S101 that the value of the color signal Bout is equal to or greater than the boundary value B0 and the value of the color signal Rout is smaller than the boundary value R0, the process proceeds to step S102, and the color signal G1out corresponding to the spectral sensitivity curve G1s. Is selected. As a result, the color signals Rout, G1out, and Bout are output as R, G, and B primary color signals.

  On the other hand, if it is determined in step S101 that the value of the color signal Bout is not greater than or equal to the boundary value B0, or the value of the color signal Rout is not smaller than the boundary value R0, the process proceeds to step S103. In step S103, it is determined whether or not the value of the color signal Bout is smaller than the boundary value B0 and the value of the color signal Rout is greater than or equal to the boundary value R0.

  If it is determined in step S103 that the value of the color signal Bout is smaller than the boundary value B0 and the value of the color signal Rout is greater than or equal to the boundary value R0, the process proceeds to step S104, and the color signal G2out corresponding to the spectral sensitivity curve G2s. Is selected. As a result, the color signals Rout, G2out, and Bout are output as R, G, and B primary color signals.

  On the other hand, if it is determined in step S103 that the value of the color signal Bout is not smaller than the boundary value B0 or the value of the color signal Rout is not equal to or greater than the boundary value R0, the process proceeds to step S105. In step S105, the color signals G1out and G2out are added and divided by 2 to calculate the color signal G ', and the color signal G' is output as the G primary color signal.

  As described above, according to the present embodiment, the color filter in which the color elements R0 and B0 corresponding to R and B and the color elements X and Z correlated with G are arranged in a checkered pattern is arranged on the light receiving surface of the CCD 16. The In the matrix circuit 18, the four color signals read from the CCD 16 are subjected to matrix calculation as color conversion processing based on the sRGB space. Of the four color signals Rout, Bout, G1out, and G2out generated by the matrix operation, the spectral distribution characteristics (reflectance characteristics) of the subject are detected from the color signals Rout and Bout. That is, the color signals Rout and Bout are compared with the predetermined values R0 and B0, and the subject having a spectral distribution characteristic having a relatively large spectrum in the blue short wavelength region or the spectrum having a relatively large spectrum in the red long wavelength region. It is estimated whether the subject has a distribution characteristic. Then, color signals that match the spectral sensitivity curve corresponding to the spectral distribution characteristics of the subject, that is, the color matching function G, are selected or calculated from the color signals G1out and G2out, and R, G, and B primary color signals are selected. Is generated.

  The color signals G1out and G2out may be processed only by selection without calculation. The spectral transmission characteristics of the color elements R0, X, Z, B0 of the color filter 14 may be characteristics other than those shown in FIG. Instead of setting a color signal corresponding to G, a B color signal is set based on information on the spectral distribution characteristics of the subject included in the color signals corresponding to R and G, or a color corresponding to G and B is set. The R color signal may be set based on the spectral distribution characteristic information of the subject included in the signal. Alternatively, primary color signals corresponding to R, G, and B may be generated by matrix calculation using a complementary color filter.

  Next, a second embodiment will be described. In the second embodiment, first, one color signal is selected from two color signals corresponding to green (G) out of the four color signals, and then matrix calculation processing is performed to obtain R, G, and B primary colors. A signal is generated. About another structure, it is the same as 1st Embodiment.

  FIG. 8 is a block diagram of a digital camera according to the second embodiment. Components common to the digital camera of the first embodiment are denoted by the same reference numerals.

  The digital camera 10 'includes a selection circuit 19, a first matrix circuit 21A, and a second matrix circuit 21B. The system control circuit 34 controls the first matrix circuit 21A and the second matrix circuit 21B based on the signal from the selection circuit 19.

  Of the color signals corresponding to the four color elements R0, X, Z, and B0 read from the CCD 16, two color signals R′in and B′in corresponding to R0 and B0 are sent to the selection circuit 19. The selection circuit 19 selects one of the color signals X′in and Z′in corresponding to the two color elements X and Z based on the color signal corresponding to R0 and B0.

  In the first matrix circuit 21A, color signals corresponding to the three color elements R0, X, and B0 are subjected to matrix conversion, and R, G, and B primary color signals are generated. On the other hand, in the second matrix circuit 21B, the color signals corresponding to the three color elements R0, Z, and B0 are subjected to matrix conversion to generate R, G, and B primary color signals. The system control circuit 34 switches between the first matrix circuit 21 </ b> A and the second matrix circuit 21 </ b> B based on the signal sent from the selection circuit 19. R, G, and B primary color signals including the selected signal are transmitted to the image signal processing circuit 22, the LCD driver 24, and the system control circuit 34.

  FIG. 9 is a flowchart showing a process of selecting a color signal corresponding to G in the second embodiment.

In step S201, it is determined whether or not the value of the color signal B′in is greater than or equal to the boundary value B0 ′ and the value of the color signal R′in is smaller than the boundary value R0 ′. However, the boundary values B0 ′ and R0 ′ are determined by a spectral sensitivity curve or the like. If it is determined in step S201 that the value of the color signal B′in is greater than or equal to the boundary value B0 ′ and the value of the color signal R′in is smaller than the boundary value R0 ′, the process proceeds to step S202 and the color element X is set. The corresponding color signal X′in is selected. Thereby, in the first matrix circuit 21A, the color signals Rout, Gout, and Bout are generated as primary color signals of R, G, and B by the following equations. However, the color signals input to the first matrix circuit 21A are represented as R′in, X′in, and B′in.

On the other hand, if it is determined in step S201 that the value of the color signal B′in is not greater than or equal to the boundary value B0 ′, or the value of the color signal R′in is not smaller than the boundary value R0 ′, the process proceeds to step S203. A color signal Z′in corresponding to the element Z is selected. Thereby, in the second matrix circuit 21A, the color signals Rout, Gout, and Bout are generated as primary color signals of R, G, and B by the following formula.

  FIG. 10 is a diagram showing a spectral sensitivity distribution curve based on R, G, B primary color signals obtained by the first matrix circuit 21A. FIG. 11 is a diagram showing a spectral sensitivity distribution curve based on R, G, B primary color signals obtained by the second matrix circuit 21B.

  As shown in FIG. 10, in the case of a subject having a spectral distribution characteristic having a large spectrum in a wavelength region corresponding to blue, the color of the subject is faithfully reproduced. On the other hand, as shown in FIG. 11, in the case of a subject having a spectral distribution characteristic having a large spectrum in the wavelength region corresponding to red, the color of the subject is faithfully reproduced.

  As in the first embodiment, not only one of the color signals X and Z may be selected, but the color signal may be set by calculation.

  When determining the color information of the subject, it may be determined by the ratio of the color signals instead of comparing the color signals corresponding to red and blue with predetermined values as shown in FIGS. For example, in step S101 in FIG. 7, it may be determined whether Bout / Rout is greater than a predetermined value (eg, 1.1). In step S103, it may be determined whether Bout / Rout is smaller than a predetermined value (for example, 0.59). Similarly, in step S201 of FIG. 9, the determination may be made based on the ratio of B'in / R'in.

It is a block diagram of a digital camera. It is the figure which showed a part of color filter. It is the figure which showed the spectral sensitivity characteristic of the input system. It is the figure which showed the spectral sensitivity characteristic of the ideal imaging system. It is the figure which showed the spectral sensitivity characteristic of the color signal according to color filter R0, B0 among the color signals obtained by matrix calculation. It is the figure which showed the spectral sensitivity characteristic of the color signal according to the color filters X and Z among the color signals obtained by matrix calculation. 6 is a flowchart showing a process for determining a color signal according to G. It is a block diagram of the digital camera which is 2nd Embodiment. It is the flowchart which showed the process which determines the color signal according to G in 2nd Embodiment. It is the figure which showed the spectral sensitivity distribution curve by the primary color signal of R, G, B obtained by the 1st matrix circuit. It is the figure which showed the spectral sensitivity distribution curve by the primary color signal of R, G, B obtained by the 2nd matrix circuit.

Explanation of symbols

10, 10 'digital camera 14 color filter 16 CCD (imaging device)
18 matrix circuit 19 selection circuit 20 G signal output circuit 21A first matrix circuit 21B second matrix circuit

Claims (9)

An image sensor;
Four color elements each having different spectral transmission characteristics are arranged in a checkered pattern on the light receiving surface of the image sensor, and according to R and B of R, G, and B of color matching functions A color filter comprising two types of non-selective color elements having spectral transmission characteristics and two types of color elements correlated with G, wherein the spectral transmission characteristics characterize the spectral sensitivity characteristics of the imaging system;
A series of color signals corresponding to the four types of color elements from the image sensor, which are correlated with the non-selection color signals according to R and B based on the two types of non-selection color elements, and the G. Signal reading means for reading a series of color signals composed of two selection color signals conforming to G based on two types of color elements;
Signal processing means for generating R, G, B primary color signals according to tristimulus values from the series of color signals,
The signal processing means is
Color conversion processing means for performing a matrix operation on the series of color signals for color conversion;
A wavelength region having a relatively large spectrum in the spectral distribution characteristic of the subject out of the two selected color signals subjected to the matrix calculation based on information on the spectral distribution characteristic of the subject included in the non-selected color signal that has been subjected to matrix calculation The color matching the spectral sensitivity distribution curve corresponding to G of the color matching function is selected as a matching color signal, and primary colors for generating R, G, B primary color signals based on the non-selected color signal and the matching color signal An image pickup apparatus comprising: a signal generation unit. An image sensor;
Four color elements each having different spectral transmission characteristics are arranged in a checkered pattern on the light receiving surface of the image sensor, and according to R and B of R, G, and B of color matching functions A color filter comprising two types of non-selective color elements having spectral transmission characteristics and two types of color elements correlated with G, wherein the spectral transmission characteristics characterize the spectral sensitivity characteristics of the imaging system;
A series of color signals corresponding to the four types of color elements from the image sensor, which are correlated with the non-selection color signals according to R and B based on the two types of non-selection color elements, and the G. Signal reading means for reading a series of color signals composed of two selection color signals conforming to G based on two types of color elements;
Signal processing means for generating R, G, B primary color signals according to tristimulus values from the series of color signals,
The signal processing means is
Based on the information on the spectral distribution characteristics of the subject included in the non-selected color signals corresponding to R and B in the series of color signals, the spectral distribution characteristics of the subject among the two selected color signals corresponding to G A matching color signal determining means for selecting, as a matching color signal, one that matches a spectral sensitivity distribution curve corresponding to G of the color matching function in a wavelength region having a relatively large spectrum in
An image pickup apparatus comprising: color conversion processing means for performing a matrix operation for color conversion on the non-selected color signal and the compatible color signal and outputting R, G, B primary color signals.   The spectral sensitivity distribution curve of one color signal of the two selected color signals subjected to matrix calculation is adapted to the spectral sensitivity distribution curve corresponding to G of the color matching function in a short wavelength region shorter than the peak wavelength, 2. The imaging apparatus according to claim 1, wherein the spectral sensitivity distribution curve of the color signal is adapted to a spectral sensitivity distribution curve corresponding to G of the color matching function in a wavelength range longer than the peak wavelength.   The spectral sensitivity distribution curve of one color signal of the two selected color signals is adapted to the spectral sensitivity distribution curve corresponding to G of the color matching function in a short wavelength region shorter than the peak wavelength, and the spectral of the other color signal. The imaging apparatus according to claim 2, wherein the sensitivity distribution curve is adapted to a spectral sensitivity distribution curve corresponding to G of the color matching function in a wavelength range longer than a peak wavelength.   The primary color signal generating means compares the color signal corresponding to R and the color signal corresponding to B with the predetermined value in the non-selected color signal subjected to matrix calculation, and the spectrum in the short wavelength region corresponding to blue is relatively 2. The method according to claim 1, wherein it is determined whether the subject has a particularly large spectral distribution characteristic or a subject having a relatively large spectral distribution characteristic in a long wavelength region corresponding to red. Imaging device. The adaptive color signal generating unit compares the color signal corresponding to R and the color signal corresponding to B with the predetermined value in the non-selected color signal, and the spectrum in the short wavelength region corresponding to blue is relatively large. imaging of claim 2, wherein the subject der having spectral distribution characteristic Luke, the or the spectrum in the long wavelength region corresponding to red is determined whether the object having a relatively large spectral power distribution apparatus.   Based on the ratio of the color signal corresponding to R and the color signal corresponding to B in the matrix-calculated non-selected color signal, the primary color signal generation means has a relatively short spectrum in a short wavelength region corresponding to blue. The imaging according to claim 1, wherein it is determined whether the subject has a large spectral distribution characteristic or the subject has a relatively large spectral distribution characteristic in a long wavelength region corresponding to red. apparatus. The adaptive color signal generating means has a spectral distribution having a relatively large spectrum in a short wavelength region corresponding to blue based on a ratio of a color signal corresponding to R and a color signal corresponding to B in the non-selected color signal. subject der Luke having properties or imaging apparatus according to claim 2, characterized in that to determine whether the subject spectrum in the long wavelength region corresponding to red has a relatively large spectral distribution characteristics.   2. The imaging according to claim 1, wherein spectral transmission characteristics corresponding to the four types of color elements are determined so as to have a spectral distribution having peak values at substantially equal intervals in a wavelength range of visible light. apparatus.

Images