JP2011227388A - Imaging device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an imaging device capable of performing a highly accurate shading correction by means of a simple calculation.SOLUTION: The imaging device comprises image pick-up means provided with plural focus detection pixels for photoelectrically converting an image formed of beams divided by means of exit pupil division from a beam passing through a lens unit having an iris and an imaging optical system, focus detection means for acquiring a pair of image signals from the plural focus detection pixels and for detecting a focus state of the imaging optical system, and control means for controlling the iris in such a manner that an aperture shape of the exit pupil at the position of the focus detection pixels depends on an aperture shape of the iris, when the focus state is detected by the focus detection means.

Description

  The present invention relates to an imaging apparatus including an imaging device having a plurality of focus detection pixels.

  As one of methods for detecting the focus state of the photographing lens, Patent Document 1 discloses an apparatus that performs pupil division type focus detection using a two-dimensional sensor in which a microlens is formed in each pixel of the sensor. Yes. Further, the present applicant discloses a solid-state imaging device that performs pupil division type focus detection using a CMOS image sensor (solid-state imaging device) used in a digital still camera. In the focus detection method of the pupil division method, the focus state of the photographic lens is detected by performing a correlation operation on two images that are transmitted through different areas of the pupil of the photographic lens.

  When the light quantity distribution in the correlation direction of the two images differs due to pupil vignetting, an asymmetry occurs in the output waveform of the corresponding subject image of the two images, and an error occurs in the correlation calculation. In order to avoid such a phenomenon, a method is used in which the light quantity distribution in the correlation direction between two images is predicted in advance, and the obtained image is subjected to shading correction. Further, a method of calculating the amount of focus shift by calculating the baseline length in advance is used. In Patent Document 3, a method for predicting the shape of pupil vignetting from parameters such as the type of lens and image height is taken. In Patent Document 4, a specific filter stored in a camera is deformed by an aperture ratio (F value), an exit pupil position, and an image shift amount, and the deformation filter (image correction filter) is applied to a subject image. A technique for detecting an image state is disclosed.

  Patent Document 5 discloses a method in which two apertures other than the normal aperture can be switched at the aperture position in the lens so that the pupil vignetting shape matches the pupil aperture shape of the lens. Patent Document 6 discloses a method of changing the shape of the aperture opening into two apertures by using the aperture blades of the lens so that the shape of pupil vignetting matches the shape of the pupil aperture of the lens.

JP 58-24105 (2nd page, FIG. 1) Japanese Patent Laying-Open No. 2005-106994 (page 7, FIG. 3) JP 2004-191629 A (pages 8 to 12, FIGS. 7 to 10) JP-A-5-127074 (page 15, FIG. 34) Japanese Patent Laying-Open No. 2001-083407 (pages 10 to 11, FIG. 27) Japanese Patent Laid-Open No. 2002-350718 (page 3, FIG. 1)

  A camera capable of exchanging lenses has a wide variety of pupil vignetting shapes with respect to lens types and image heights, requires a large amount of data and a large number of parameters for shading correction, and requires a great amount of computing power. Further, in the case of a complicated vignetting shape, an error is likely to occur between the predicted value and the actual light amount value, and it becomes difficult to ensure the accuracy of desired shading correction.

  In this regard, Patent Documents 5 and 6 disclose the contents of providing a diaphragm mechanism that becomes a special lens opening (two openings) for each lens. However, in Patent Documents 5 and 6, the vignetting shape is simplified for a special lens, but in a lens that acquires a photographed image by a diaphragm aperture that is generally close to a circle at the center of the optical axis, it is patented. Two openings as in References 5 and 6 cannot be configured. For this reason, in Patent Documents 5 and 6, the vignetting shape cannot be simplified with respect to a generally commercially available lens.

  The present invention provides an imaging apparatus capable of performing highly accurate shading correction with a simple calculation.

  An image pickup apparatus according to an aspect of the present invention includes an image pickup device having a plurality of focus detection pixels that photoelectrically convert an image formed by dividing an exit pupil of a light beam that has passed through a lens unit having a diaphragm and an imaging optical system. A focus detection unit that acquires a pair of image signals from the plurality of focus detection pixels to detect a focus state of the imaging optical system, and the focus detection unit detects the focus state by the focus detection unit. Control means for controlling the diaphragm so that the shape of the exit pupil opening at the position of the focus detection pixel depends on the shape of the aperture of the diaphragm.

  An imaging method according to another aspect of the present invention includes a step of photoelectrically converting an image formed by dividing an exit pupil out of a light beam that has passed through a lens unit having a diaphragm and an imaging optical system, using a plurality of focus detection pixels. Obtaining a pair of image signals from the plurality of focus detection pixels and detecting a focus state of the imaging optical system; and at the time of detecting the focus state, at the position of the focus detection pixel And controlling the aperture so that the shape of the exit pupil aperture depends on the aperture aperture shape.

  Other objects and features of the present invention are illustrated in the following examples.

  According to the present invention, it is possible to provide an imaging apparatus capable of highly accurate shading correction with a simple calculation.

It is a block diagram of the camera (imaging device) provided with the focus detection apparatus in a present Example. It is a circuit block diagram of the image pick-up element in a present Example. It is sectional drawing of the pixel part of the image pick-up element in a present Example. 3 is a timing chart of the image sensor in the present embodiment. In this embodiment, (1) a plan view and a sectional view of an imaging pixel, (2) a plan view and a sectional view of a focus detection pixel for performing pupil division in the x direction of the photographing lens, and (3) a photographing lens. It is the top view and sectional drawing of the pixel for a focus detection for performing pupil division in the y direction. It is explanatory drawing of the pupil division of the image pick-up element in a present Example. In an Example, it is explanatory drawing of the image acquired at the time of focus detection, and a focus detection area | region. In this embodiment, (1) a schematic diagram showing an incident angle characteristic of a focus detection pixel at the center of the image sensor, (2) a pupil intensity distribution diagram of the focus detection pixel, and (3) focus detection when a lens is attached. It is a pupil intensity distribution figure of a pixel for use. In this embodiment, the incident angle characteristic of the focus detection pixel is represented in one dimension. It is explanatory drawing of the vignetting of the light beam in a present Example. It is a figure which shows the pupil area | region on the pupil surface in a present Example. It is a figure which shows the image height of the image pick-up element in a present Example. FIG. 6 is a one-dimensional representation of the pupil intensity distribution of an incident light beam on the pupil plane of a focus detection pixel having a central image height of the image sensor in the present embodiment. In this embodiment, the vignetting state of the focus detection range at the center image height is shown. In this embodiment, it is a diagram showing a light amount change within a focus detection range of a pixel at a central image height. In this embodiment, (a) a pupil vignetting shape at the time of normal image shooting at the center position of each image height, and (b) a pupil vignetting shape at each image height when performing focus detection. In this embodiment, it is a relationship diagram between the lens aperture position and the focus detection position at the pupil plane position Me in the focus detection means. In the present embodiment, (1) an example of a gain map curve obtained by correcting the amount of light, and (2) an example of a graph of the reciprocal of the center-of-gravity shift amount of the pupil intensity distribution of the pixel at the focus detection position. It is a flowchart in the case of the focus detection in a present Example.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In each figure, the same members are denoted by the same reference numerals, and redundant description is omitted.

  FIG. 1 is a configuration diagram of a camera 200 (imaging device) provided with a focus detection device in the present embodiment. In FIG. 1, reference numeral 100 denotes a lens unit having a stop and an imaging optical system (imaging optical system). The lens unit 100 may be detachable from the camera body or not. Reference numeral 101 denotes a first lens group disposed at the tip of the photographing optical system, and is held so as to be able to advance and retract in the optical axis direction. Reference numeral 102 denotes an aperture / shutter (aperture), which adjusts the light amount at the time of shooting by adjusting the aperture diameter, and functions as an exposure time adjustment shutter at the time of still image shooting. Reference numeral 103 denotes a second lens group. The diaphragm / shutter 102 and the second lens group 103 integrally move forward and backward in the optical axis direction, and have a zooming function in conjunction with the forward / backward movement of the first lens group 101.

  Reference numeral 105 denotes a third lens group, which performs focus adjustment by moving back and forth in the optical axis direction. Reference numeral 106 denotes an optical low-pass filter, which is an optical element for reducing false colors and moire in a captured image. Reference numeral 107 denotes an image sensor (imaging means) composed of a C-MOS sensor and its peripheral circuits. The image sensor 107 is a two-dimensional single-plate color sensor in which a Bayer array primary color mosaic filter is formed on-chip on light receiving pixels of m pixels in the horizontal direction and n pixels in the vertical direction. As will be described later, the image sensor 107 photoelectrically converts an image formed by the light beam from the imaging optical system, and an image formed by the divided light beam among the light beams from the imaging optical system. It has a plurality of focus detection pixels for photoelectric conversion.

  Reference numeral 111 denotes a zoom actuator, which rotates a cam cylinder (not shown) to drive the first lens group 101 to the third lens group 103 forward and backward in the optical axis direction to perform a zooming operation. Reference numeral 112 denotes a diaphragm actuator (control means). When the focus is detected by the focus detection unit (CPU 121), the aperture actuator 112 stops the aperture so that the shape of the exit pupil opening at the position of the focus detection pixel depends on the shape of the aperture of the aperture (the aperture / shutter 102). To control. Reference numeral 114 denotes a focus actuator, which performs focus adjustment by driving the third lens group 105 back and forth in the optical axis direction. Reference numeral 115 denotes an electronic flash for illuminating a subject at the time of photographing. A flash illumination device using a xenon tube is preferably used, but an illumination device including an LED that emits light continuously may be used.

  Reference numeral 121 denotes a CPU that performs various controls of the camera body. The CPU 121 includes a calculation unit, data storage means ROM, RAM, an A / D converter, a D / A converter, a communication interface circuit, and the like. The CPU 121 drives various circuits of the camera 200 based on a predetermined program stored in the ROM, and executes a series of operations such as AF, shooting, image processing, and recording. The CPU 121 corresponds to the calculation means, focus detection means, and data storage means of the present invention. The CPU 121 as focus detection means acquires a pair of image signals from the focus detection pixels (pixels SHA and SHB) and detects the focus state of the imaging optical system. The CPU 121 as a focus detection unit includes a focus shift amount calculation unit that calculates a focus shift amount from a pair of image signals obtained from a plurality of focus detection pixels.

  An electronic flash control circuit 122 controls lighting of the illumination unit 115 in synchronization with the photographing operation. Reference numeral 124 denotes an image sensor drive circuit that controls the image capturing operation of the image sensor 107 and A / D-converts the acquired image signal and transmits it to the CPU 121. An image processing circuit 125 performs various processes such as γ conversion, color interpolation, and JPEG compression of an image acquired by the image sensor 107.

  A focus driving circuit 126 controls the focus actuator 114 based on the focus detection result, and performs focus adjustment by driving the third lens group 105 back and forth in the optical axis direction. Reference numeral 128 denotes an aperture shutter drive circuit that controls the aperture of the aperture / shutter 102 by drivingly controlling the aperture actuator 112. Reference numeral 129 denotes a zoom drive circuit that drives the zoom actuator 111 in accordance with the zoom operation of the photographer.

  Reference numeral 131 denotes a display device such as an LCD, which displays information related to the shooting mode of the camera 200, a preview image before shooting and a confirmation image after shooting, a focus state display image when focus is detected, and the like. An operation switch group 132 includes a power switch, a release (shooting trigger) switch, a zoom operation switch, and a shooting mode selection switch. Reference numeral 133 denotes a detachable flash memory that records a photographed image.

  FIG. 2 is a circuit configuration diagram of the image sensor in the present embodiment. FIG. 2 shows a range of 2 columns × 4 rows of pixels of a two-dimensional C-MOS area sensor. However, when used as an image sensor, a large number of pixels shown in FIG. Acquisition is possible. In FIG. 2, 1 is a photoelectric conversion portion of a photoelectric conversion element comprising a MOS transistor gate and a depletion layer under the gate, 2 is a photogate, 3 is a transfer switch MOS transistor, 4 is a reset MOS transistor, and 5 is a source follower amplifier MOS. It is a transistor. Also, 6 is a horizontal selection switch MOS transistor, 7 is a source follower load MOS transistor, 8 is a dark output transfer MOS transistor, 9 is a light output transfer MOS transistor, 10 is a dark output storage capacitor CTN, and 11 is a light output storage. Capacitance CTS. Further, 12 is a horizontal transfer MOS transistor, 13 is a horizontal output line reset MOS transistor, 14 is a differential output amplifier, 15 is a horizontal scanning circuit, and 16 is a vertical scanning circuit.

  FIG. 3 is a cross-sectional view of the pixel portion of the image sensor in the present embodiment. In FIG. 3, 17 is a P-type well, 18 is a gate oxide film, 19 is a first-layer poly-Si, 20 is a second-layer poly-Si, and 21 is an n + floating diffusion portion (FD portion). The FD unit 21 is connected to another photoelectric conversion unit via another transfer MOS transistor. In FIG. 3, the drain of the two transfer switch MOS transistors 3 and the FD portion 21 are made common to improve the sensitivity by miniaturization and the capacity reduction of the FD portion 21, but even if the FD portion 21 is connected by Al wiring. Good.

  Next, the operation of the image sensor will be described using the timing chart of FIG. FIG. 4 is a timing chart of the image sensor in the present embodiment, and this timing chart shows a case of all pixel independent output. First, according to the timing output from the vertical scanning circuit 16, the control pulse φL is set high to reset the vertical output line. Further, the control pulses φR0, φPG00, and φPGe0 are set high, the reset MOS transistor 4 is turned on, and the first-layer poly Si 19 of the photogate 2 is set high. At time T0, the control pulse φS0 is set high, the selection switch MOS transistor 6 is turned on, and the pixel portions of the first and second lines are selected. Next, the control pulse φR0 is set to low, the reset of the FD unit 21 is stopped, the FD unit 21 is brought into a floating state, and the source-follower amplifier MOS transistor 5 is made through between the gate and the source. Thereafter, at time T1, the control pulse φTN is set to high, and the dark voltage of the FD unit 21 is output to the storage capacitor CTN10 by the source follower operation.

  Next, in order to perform photoelectric conversion output of the pixels of the first line, the control pulse φTX00 of the first line is set high, the transfer switch MOS transistor 3 is turned on, and then the control pulse φPG00 is lowered low at time T2. At this time, it is preferable that the potential well that has spread under the photogate 2 is raised to have a voltage relationship that allows the photogenerated carriers to be completely transferred to the FD portion 21. Therefore, if complete transfer is possible, the control pulse φTX may be a predetermined fixed potential instead of a pulse.

  At time T2, the electric charge from the photoelectric conversion unit 1 of the photodiode is transferred to the FD unit 21, so that the potential of the FD unit 21 changes according to light. At this time, since the source follower amplifier MOS transistor 5 is in a floating state, the potential of the FD portion 21 is output to the storage capacitor CTS11 with the control pulse φTs being high at time T3. At this time, the dark output and the light output of the pixels on the first line are stored in the storage capacitors CTN10 and CTS11, respectively. The horizontal output line reset MOS transistor 13 is turned on by setting the control pulse φHC at time T4 to be temporarily high to reset the horizontal output line. Outputs light output. At this time, if the differential output VOUT is obtained by the differential amplifier 14 of the storage capacitors CTN10 and CTS11, a signal having a good S / N from which random noise and fixed pattern noise of the pixel are removed can be obtained. Further, the photoelectric charges of the pixels 30-12 and 30-22 are accumulated in the respective storage capacitors CTN10 and CTS11 simultaneously with the pixels 30-11 and 30-21. The timing pulse from the horizontal scanning circuit 15 is delayed by one pixel, read out to the horizontal output line, and output from the differential amplifier 14. In the present embodiment, a configuration in which the differential output VOUT is performed in the chip is shown. However, the same is true even if a conventional CDS (Correlated Double Sampling) circuit is used outside without being included in the chip. An effect is obtained.

  After outputting a bright output to the storage capacitor CTS11, the control pulse φR0 is set to high to make the reset MOS transistor 4 conductive, and the FD portion 21 is reset to the power supply VDD. After the horizontal transfer of the first line is completed, the second line is read. In reading the second line, the control pulses φTXe0 and φPGe0 are driven in the same manner, and high pulses are supplied to the control pulses φTN and φTS, respectively. . With the above driving, the first and second lines can be read independently. Thereafter, by scanning the vertical scanning circuit and reading out the second n + 1 and second n + 2 (n = 1, 2,...) In the same manner, all pixel independent outputs can be performed. That is, when n = 1, first, the control pulse φS1 is set to high, and then φR1 is set to low. Subsequently, the pixel signals of the pixels 30-31 and 30-32 are read with the control pulses φTN and φTX01 being high, the control pulse φPG01 being low, the control pulse φTS being high, and the control pulse φHC being temporarily high. Subsequently, the control pulses φTXe1 and φPGe1 and the control pulses are applied in the same manner as described above, and the pixel signals of the pixels 30-41 and 30-42 are read out.

  FIG. 5 is a diagram illustrating the structure of the imaging pixels and the focus detection pixels. In this embodiment, out of 4 pixels in 2 rows × 2 columns, pixels having G (green) spectral sensitivity are arranged in 2 diagonal pixels, and R (red) and B (blue) are arranged in the other 2 pixels. A Bayer arrangement in which one pixel each having a spectral sensitivity of 1 is arranged is employed. Focus detection pixels, which will be described later, are distributed in a predetermined rule between the Bayer arrays.

  FIG. 5A is a plan view and a cross-sectional view of the imaging pixel. FIGS. 5A and 5A are plan views of 2 × 2 imaging pixels located in the center of the imaging device. In the Bayer array, G pixels are arranged diagonally, and R and B pixels are arranged in the other two pixels. The 2 rows × 2 columns structure is repeatedly arranged. 5 (1) and 5 (b) are cross-sectional views taken along line AA in FIGS. 5 (1) and 5 (a). ML is an on-chip microlens disposed on the forefront of each pixel, CFR is an R (Red) color filter, and CFG is a G (Green) color filter. PD schematically shows the photoelectric conversion unit of the C-MOS sensor described in FIG. 3, and CL is a wiring layer for forming signal lines for transmitting various signals in the C-MOS sensor. . TL schematically shows the photographing optical system.

  Here, the on-chip microlens (microlens ML) of the imaging pixel and the photoelectric conversion unit PD are configured to capture the light beam that has passed through the imaging optical system (imaging lens TL) as effectively as possible. In other words, the exit pupil EP of the photographing lens TL and the photoelectric conversion unit PD are substantially conjugated with each other by the microlens ML, and the effective area of the photoelectric conversion unit PD is designed to be large. 5 (1) and 5 (b) describe the incident light beam of the R pixel, the G pixel and the B (Blue) pixel have the same structure. Accordingly, the exit pupil EP corresponding to each of the RGB pixels for imaging has a large diameter, and the S / N of the image signal is improved by efficiently capturing the light flux from the subject.

  FIG. 5B is a plan view and a cross-sectional view of focus detection pixels for performing pupil division in the x direction of the photographing lens. FIGS. 5A and 5A are plan views of pixels of 2 rows × 2 columns including focus detection pixels located at the center of the image sensor. When obtaining an imaging signal, the G pixel is a main component of luminance information. Since human image recognition characteristics are sensitive to luminance information, image quality degradation is likely to be noticed when G pixels are missing. On the other hand, the R pixel or the B pixel is a pixel that acquires color information, but since humans are insensitive to color information, the pixel that acquires color information is less likely to notice deterioration in image quality even if some loss occurs. Therefore, in this embodiment, among the pixels in 2 rows × 2 columns, the G pixel is left as an imaging pixel, and focus detection pixels (pixels SHA and SHB) are provided at a ratio of pixels in positions corresponding to R and B. Arranged.

  FIGS. 5 (2) and 5 (b) are cross-sectional views taken along line BB in FIGS. 5 (2) and 5 (a). The microlens ML and the photoelectric conversion unit PD have the same structure as that of the imaging pixel shown in FIGS. In this embodiment, since the signal of the focus detection pixel is not used for image creation, a transparent film CFW (White) is arranged instead of the color separation color filter. Further, since pupil division is performed by the image sensor, the opening of the wiring layer CL is deviated in the x direction with respect to the center line of the microlens ML. Specifically, since the opening OPHA of the pixel SHA is deviated in the −x direction, the light beam that has passed through the exit pupil EPHA on the left side of the photographing lens TL is received. Similarly, since the opening OPHB of the pixel SHB is biased in the + x direction, the light beam that has passed through the right exit pupil EPHB of the photographic lens TL is received. In this embodiment, the pixels SHA are regularly arranged in the x direction, and the subject image acquired by these pixel groups is defined as an A image. Further, the pixels SHB are also regularly arranged in the x direction, and the subject image acquired by these pixel groups is defined as a B image. At this time, by detecting the relative positions of the A image and the B image, it is possible to detect the focus shift amount (defocus amount) of the subject image.

  In the pixels SHA and SHB, focus detection is possible for a subject having a luminance distribution in the x direction on the photographing screen, for example, a line in the y direction. However, focus detection cannot be performed on a line in the x direction having a luminance distribution in the y direction. Therefore, in the present embodiment, pixels that perform pupil division are also provided in the y direction of the photographing lens so that the focus can be detected for the latter.

  FIG. 5C is a plan view and a cross-sectional view of focus detection pixels for performing pupil division in the y direction of the photographing lens. 5A and 5A are plan views of pixels of 2 rows × 2 columns including focus detection pixels located in the center of the image sensor. As in FIG. 5 (2) (a), the G pixel is left as an image pickup pixel, and focus detection pixels (pixels SVC and SVD) are arranged at a fixed ratio to pixels at positions corresponding to the R and B pixels. Yes.

  FIGS. 5 (3) and 5 (b) are CC cross-sectional views in FIGS. 5 (3) and 5 (a). 5 (2) and 5 (b) have a structure in which the pupil is separated in the x direction, whereas the pixels in FIGS. 5 (3) and 5 (b) have the y direction in the pupil separation direction. The structure is the same. That is, since the opening OPVC of the pixel SVC is biased in the −y direction, the light beam that has passed through the exit pupil EPVC in the + y direction of the photographic lens TL is received. Similarly, since the opening OPVD of the pixel SVD is biased in the + y direction, the light beam that has passed through the exit pupil EPVD of the photographing lens TL in the −y direction is received. In the present embodiment, the pixels SVC are regularly arranged in the y direction, and the subject image acquired by these pixel groups is defined as a C image. The pixels SVD are also regularly arranged in the y direction, and the subject image acquired by these pixel groups is defined as a D image. At this time, by detecting the relative positions of the C image and the D image, it is possible to detect the focus shift amount (defocus amount) of the subject image having the luminance distribution in the y direction.

  FIG. 6 is an explanatory diagram of pupil division of the image sensor in the present embodiment. TL is a photographing lens, 107 is an image sensor, OBJ is a subject, and IMG is a subject image. As described with reference to FIG. 5A, the imaging pixel receives the light beam that has passed through the entire exit pupil EP of the photographing lens TL. On the other hand, the focus detection pixel has a pupil division function as described with reference to FIGS. 5 (2) and 5 (3). Specifically, the pixel SHA in FIG. 5B receives the light beam that has passed through the pupil on the + X direction side, that is, the light beam LHA that has passed through the pupil EPHA in FIG. Similarly, the pixels SHB, SVC, and SVD receive the light beams LHB, LVC, and LVD that have passed through the pupils EPHB, EPVC, and EPVD, respectively. Then, by distributing the focus detection pixels over the entire area of the imaging element 107, it is possible to detect the focus in the entire imaging area.

  FIG. 7 is an explanatory diagram of an image acquired during focus detection and a focus detection area. In FIG. 7, the subject image formed on the imaging surface includes a person at the center, a close-up tree on the left side, and a distant mountain range on the right side. In the present embodiment, the pixel for focus detection includes a pixel pair for detecting x-direction misalignment (pixels SHA and SHB) and a pixel pair for detecting y-direction misalignment (pixels SVC and SVD) over the entire imaging region. Are arranged at a high density. When detecting the x-direction deviation, a pair of image signals obtained from the pixels SHA and SHB for detecting the x-direction deviation are used as AF pixel signals for phase difference calculation. When detecting the y-direction deviation, a pair of image signals obtained from the pixels SVC and SVD for detecting the y-direction deviation are used as AF pixel signals for phase difference calculation. Therefore, it is possible to set a focus detection area for detecting an x-direction shift and a y-direction shift at an arbitrary position in the imaging area.

  In FIG. 7, a human face exists in the center of the screen. When the presence of a face is detected by a known face recognition technique, a focus detection area AFARh (x1, y1) for detecting an x-direction shift around the face area, and a focus detection area AFARv (for detecting a y-direction shift) x3, y3) are set. Here, h represents the x direction, and (x1, y1) and (x3, y3) represent the coordinates of the upper left corner of the focus detection area. The A image signal for phase difference detection obtained by connecting the pixels SHA for detecting the x-direction deviation included in each section of the focus detection area AFARh (x1, y1) over 30 sections is AFSIGh (A1). . Similarly, a B image signal for phase difference detection obtained by connecting the pixels SHB for detecting x-direction misalignment of each section over 30 sections is AFSIGh (B1). By calculating a relative x-direction shift amount between the A image signal AFSIGh (A1) and the B image signal AFSIGh (B1) by a known correlation calculation, a focus shift amount (defocus amount) of the photographing lens can be obtained. .

  Similarly, for the focus detection area AFARv (x3, y3), the C image signal for phase difference detection obtained by connecting the pixels SVC for detecting the y-direction deviation is AFSIGv (C3). Further, the D image signal for phase difference detection obtained by connecting the pixels SVD for y direction deviation detection is AFSIGv (D3). By calculating the relative y-direction shift amount between the C image signal AFSIGv (C3) and the D image signal AFSIGv (D3) by correlation calculation, the focus shift amount (defocus amount) of the photographing lens can be obtained. Then, two focus shift amounts detected in the focus detection areas of the x-direction shift and the y-direction shift may be compared and a highly reliable value may be adopted.

  On the other hand, since the tree trunk on the left side of the screen is mainly composed of the y-direction component, that is, has a luminance distribution in the x-direction, it is determined as a subject suitable for x-direction deviation detection, and focus detection for x-direction deviation detection An area AFARh (x2, y2) is set. Further, the mountain ridge portion on the right side of the screen is mainly composed of the x-direction component, that is, has a luminance distribution in the y-direction, so that it is determined as a subject suitable for y-direction deviation detection, and focus detection for y-direction deviation detection An area AFARv (x4, y4) is set.

  In this embodiment, since the focus detection area for detecting the x-direction shift and the y-direction shift can be set at any position on the screen, the focus is always maintained even if the projection position of the subject and the directionality of the luminance distribution are various. Detection is possible. The principle is the same except that the x-direction deviation and the y-direction deviation are different in direction. For this reason, in the following, detection of the x-direction deviation will be described, and description of the y-direction deviation detection will be omitted.

  FIG. 8A is a schematic diagram illustrating the incident angle characteristic of the focus detection pixel at the center of the image sensor. 8A and 8B show the incident angle characteristics of the pixel SHA, and FIGS. 8A and 8B show the incident angle characteristics of the pixel SHB. The x-axis and y-axis in FIG. 8A represent the incident angles of the pixel in the x-direction and y-direction, respectively. FIG. 8 (1) shows that the received light intensity is higher as the color is darker. In FIG. 5 (2), for ease of explanation, the exit pupil of the pixel SHA is shown separately as EPHA, and the exit pupil of the pixel SHB is shown separately as EPHB. However, as shown in FIG. 8 (1), the exit pupils of the pixel SHA and the pixel SHB are actually a partial region in order to improve the influence of diffraction due to the openings of the opening OPHA and the opening OPHB and SN. There are overlapping parts.

  FIG. 9 shows the incident angle characteristic of the focus detection pixel represented in one dimension. The horizontal axis represents the incident angle, and the vertical axis represents the sum of the light receiving sensitivities in the θy direction of FIG. 8A, and the origin is the optical axis. The characteristic of the pixel SHA is indicated by a solid line, and the characteristic of the pixel SHB is indicated by a broken line. As shown in FIG. 9, in the focus detection pixel at the center of the image sensor, the incident angle characteristics of the pixel SHA and the pixel SHB are substantially symmetric with respect to the sensitivity centroid.

  FIG. 10 is an explanatory diagram of vignetting of light flux. 10A shows a light beam incident on the central pixel of the image sensor 107, and FIG. 10B shows a light beam incident on a pixel at a position having an image height from the center of the image sensor 107. FIG. A light beam limited by several components such as a lens holding frame of the photographing lens (first lens group 101) and a diaphragm / shutter 102 is incident on the image sensor 107. Here, in order to simplify the description, it is assumed that there are two members that limit the luminous flux at any image height. Iw1 and Iw2 are windows that restrict the light flux, and the light flux passes through the inside. Me represents a pupil plane (ML pupil) set by the configuration of the microlens ML.

  With reference to FIG. 10A, vignetting of the light beam incident on the center pixel of the image sensor 107 will be described. L1rc and L1lc represent the outer periphery of the light beam emitted from the window Iw1, L1rc represents the right end in FIG. 10, and L1lc represents the left end in FIG. L2rc and L2lc represent the outer periphery of the projected light beam from the window Iw2 up to the pupil position of the microlens ML, L2rc represents the right end in FIG. 10, and L2lc represents the left end in FIG. As shown in FIG. 10A, the pupil region on the pupil plane Me of the light beam incident on the center pixel of the image sensor 107 is indicated by a light beam having L2lc and L2rc as outer circumferences, that is, an arrow Area1.

  Next, with reference to FIG. 10B, vignetting of the light beam incident on the pixel at the position having the image height from the center of the image sensor 107 will be described. L1rh and L1lh represent the outer periphery of the luminous flux emitted from the window Iw1, L1rh represents the right end in FIG. 10, and L1lh represents the left end in FIG. L2rh and L2lh represent the outer periphery of the projected light beam of the window Iw2 up to the pupil position of the microlens ML, L2rh represents the right end in FIG. 10, and L2lh represents the left end in FIG. As shown in FIG. 10B, the pupil region on the pupil plane Me of the light beam incident on the pixel at the image height position from the center of the image sensor 107 is a light beam having L1lh and L2rh as outer circumferences, This is indicated by an arrow Area2.

  FIG. 11 is a diagram illustrating a pupil region on the pupil plane Me. FIG. 11A shows a pupil region of a pixel at the center of the image sensor, and FIG. 11B shows a pupil region of a pixel at a position having an image height from the center of the image sensor. As described with reference to FIG. 10, since the light beam limited by the same window Iw2 is incident on the central pixel of the image sensor, as shown in FIG. 11A, the pupil area Area1 has the window Iw2. The shape is projected as it is. Since the window for limiting the luminous flux is circular, the shape of the pupil area Area1 is also circular. On the other hand, since the light beam limited by Iw1 and Iw2 is incident on the pixel at the image height from the center of the image sensor, the pupil area Area2 has a shape as shown in FIG.

  FIG. 8B is a pupil intensity distribution diagram of focus detection pixels. 8A and 8B show the characteristics of the pixel SHA, and FIGS. 8B and 8B show the characteristics of the pixel SHB. The microlens ML of the pixel at a position having an image height from the center of the image sensor is made eccentric so that the optical axis center passes through the optical axis center at a predetermined pupil (ML pupil) distance. Therefore, the incident angle characteristic of the focus detection pixel at the center of the image sensor shown in FIG. 8 (1) is equivalent to the projection onto the ML pupil, and the vertical and horizontal axes in FIG. 8 (2) are the coordinates on the pupil. It is developed. This pupil intensity distribution has the same characteristics for pixels at positions having an image height from the center of the image sensor. Note that the position dimensions of the lens holding frame and the diaphragm described above differ depending on the type of lens, and thus the shape of the pupil region differs depending on the type of lens even at the same image height position.

  FIG. 12 is an explanatory diagram of the focus detection position in the image sensor and shows the image height of the image sensor 107. In FIG. 12, Img0 indicates the center image height, Img1 indicates the horizontal 40% image height, and Img2 indicates the diagonal 40% image height. FIG. 8 (3) is a pupil intensity distribution diagram of focus detection pixels when a lens is attached. 8 (3) and (a) show the pupil intensity distribution characteristics of the pixel SHA, and FIGS. 8 (3) and (b) show the pupil intensity distribution characteristics of the pixel SHB. FIG. 8 (3) is an overlay of the pupil vignetting shape described in FIG. 11 (a) and the pupil intensity distribution diagram of the focus detection pixel in FIG. 8 (2). , A light beam that has passed through the inside of the shape indicated by Area 1 is incident in the illustrated pupil intensity distribution. FIG. 13 is a one-dimensional representation of the pupil intensity distribution of the incident light beam on the pupil plane Me of the focus detection pixel having the central image height Img0 of the image sensor. The horizontal axis represents coordinates in the x direction on the pupil plane Me, and the vertical axis represents the intensity of each coordinate. The intensity of each coordinate is obtained by adding the pupil intensity in the y direction in FIG. The characteristic of the pixel SHA is indicated by a solid line, and the characteristic of the pixel SHB is indicated by a broken line.

  FIG. 14 shows the vignetting state of the focus detection range at the central image height Img0. FIGS. 14A, 14B, and 14C show vignetting states at the left end, center, and right end of the focus detection region, respectively. As shown in FIG. 14, the vignetting is round at the center (FIG. 14B), whereas at the left end (FIG. 14A) and the right end (FIG. 14C) of the focus detection area. Another lens frame appears, and the round shape of the vignetting is broken. FIG. 15 is a diagram illustrating a light amount change within the focus detection range of the pixels SHA and SHB at the central image height Img0. In both the pixels SHA and SHB, the amount of light changes according to the distance measurement area. If such a light amount deviation is not corrected, an error occurs in the correlation calculation in accordance with the state of the subject even in a focused state.

  FIG. 16A shows a pupil vignetting shape at the time of normal image shooting at the center positions of the image heights Img0, Img1, and Img2 shown in FIG. As shown in FIG. 16 (a), the pupil vignetting at each image height varies greatly depending on the focus detection position. Particularly, when the image height is high, the pupil vignetting affects between the A image signal and the B image signal. The difference in the amount of light is large, and factors that cause focus detection errors increase. In this embodiment, the vignetting shape is determined only by the shape of the diaphragm by controlling the diaphragm when performing the focus detection operation.

  FIG. 16B shows pupil vignetting shapes at image heights Img0, Img1, and Img2 when focus detection is performed in this embodiment. As shown in FIG. 16 (b), in each focus detection region at image heights Img0, Img1, and Img2, the circular aperture shape formed by the aperture is unified, and vignetting caused by other lens frames is caused. Not affected. For this reason, it is possible to eliminate the appearance error of the lens frame due to the manufacturing error of the lens frame and the manufacturing error of the optical axis on the camera side. In addition, since the light quantity change is caused only by the aperture frame, no sudden light quantity change occurs in the focus detection area. For this reason, the correction curve has comparatively gentle characteristics, and correction errors are unlikely to occur.

  FIG. 17 is a relationship diagram between the lens aperture position and the focus detection position at the pupil plane position Me in the focus detection means. FIG. 17A is a diagram showing the shape of the pupil vignetting on the pupil plane in the pixel SHA, and FIG. 17B is a diagram showing how the center of the aperture-shaped circle is shifted depending on the focus detection position. When the distance between the imaging element 107 and the aperture position of the lens is P1, the distance between the imaging element 107 and the ML pupil plane Me is Pme, and the aperture value is Fno, the aperture shape on the ML pupil plane Me is circular. The diameter is represented by the formula (1).

D1 = Pme / Fno (1)
On the other hand, the misalignment position (dx, dy) at the center of the circle can be expressed as in equations (2) and (3), where the image height at the focus detection position is (xs, ys).

dx = (P1-Pme) × xs / P1 (2)
dy = (P1-Pme) * ys / P1 (3)
FIG. 18A is an example of a gain map curve obtained by performing light amount correction (shading correction) based on the light amount at a specific aperture value Fno. In FIG. 18 (1), the horizontal direction and the vertical direction represent the center shift of the aperture-shaped circle. The dx and dy in the above formulas (2) and (3) correspond to horizontal and vertical, respectively. The gain characteristic shown in FIG. 18A corresponds to a gain obtained by taking the reciprocal of the amount of light at each image height. By storing the gain map data (gain map curve) as shown in FIG. 18 (1) in the memory in the camera in advance, no matter what lens is mounted, it is only necessary to control the aperture at the time of focus detection. Gain characteristics can be unified. At this time, since a sudden light quantity change due to a plurality of lens frame vignetting is eliminated, a light quantity correction error is also reduced.

  Although the light amount correction has been described here, the vignetting shape is uniformly calculated from the relationship between the lens aperture value and the focus detection position at the pupil plane position Me of the focus detection unit shown in FIG. 17 regardless of the lens. Is done. For this reason, if the parameter (correction value) related to focus detection specified by determining the vignetting shape is converted into data beforehand, it is not necessary to have a parameter (correction value) related to focus detection for each lens. Such a correction value relating to focus detection corresponds to the deviation of the aperture center on the exit pupil plane of the imaging optical system and is stored in the data storage means. Here, the correction value related to focus detection is a value corresponding to a gain correction value for correcting the output of the focus detection pixel. The correction value related to focus detection may be a value corresponding to an image shift conversion coefficient of a pair of image signals obtained from the focus detection pixels. Further, the correction value related to focus detection may be a value corresponding to an image correction filter of a pair of image signals obtained from the focus detection pixels.

  FIG. 18B is an example of a graph regarding the reciprocal number (image shift conversion coefficient) of the center-of-gravity shift amount (baseline length) of the pupil intensity distribution of the pixels SHA and SHB at the focus detection position. A value obtained by multiplying the image shift value obtained by the correlation calculation at the focus detection position by the image shift conversion coefficient in FIG. 18B corresponds to the amount of focus shift at the focus detection position. In FIG. 18 (2), the horizontal direction and the vertical direction represent the misalignment of the center of the aperture-shaped circle, and dx and dy in the above formulas (2) and (3) correspond to the horizontal and vertical directions. By storing the data shown in FIG. 18 (2) in the memory in the camera in advance, the aperture value (F number) is controlled at the time of focus detection by the focus detection means no matter what lens is mounted. Thus, the data of the focus detection position and the image shift conversion coefficient can be unified. Specifically, the aperture value control means performs control so that the opening of the exit pupil at the position of the focus detection pixel has a round shape when the focus state is detected by the focus detection means. Here, the “round shape” is not limited to a strict circle, but includes a shape that can be evaluated as a substantially round single shape. According to the present embodiment, since the vignetting shape is determined only by the aperture shape, the filter shape can be determined regardless of the exit pupil position of the lens. As a result, the filter shape data can be used in a unified manner for a plurality of lens conditions.

  FIG. 19 shows a flowchart relating to focus detection of this embodiment, which is part of autofocus. Since the main flow is the same as the flow of a general camera, description here is omitted. Note that the operation in the flow of FIG. 19 is executed by the CPU 121 which is a calculation unit and a focus detection unit of the present embodiment.

  First, in step S001, the aperture value at each focus detection position and the aperture position (distance P1 between the image sensor and the aperture) are read from lens information such as the lens type, zoom position, aperture value, and the like. Proceed to step S002. In step S002, the focus detection position set by the user is read. Here, the user selects the focus detection position, but a method such as detecting the face by image processing and automatically selecting the focus detection position may be employed. After selection is completed, the process proceeds to step S003.

  In step S <b> 003, the CPU 121 calculates the center deviation (dx, dy) of the circular diaphragm shape on the ML pupil of each focus detection pixel corresponding to the focus detection position. In step S004, the gain correction coefficient is read from the stored ROM in the CPU 121 based on the center deviation (dx, dy) of the round diaphragm shape calculated in step S003 and the diaphragm value Fno.

  In step S005, an appropriate image correction filter is read from the ROM in the CPU 121 based on the round diaphragm shape calculated in step S003. For the image correction filter, for example, a known method disclosed in Patent Document 4 is adopted. In step S006, based on the center deviation (dx, dy) and the aperture value Fno of the round aperture shape calculated in step S003, the baseline length data is read from the ROM in the CPU 121 where it is stored. In step S007, an aperture value is determined based on the information obtained in steps S001 and S002, and the aperture drive circuit 128 controls the aperture value to a predetermined value.

  In step S008, the image signal of the focus detection pixel at the focus detection position is read. In step S009, the subject A image and the subject B image are each corrected by the gain correction data captured in step S004. In step S010, the subject A image and the subject B image are corrected by the image correction filter captured in step S005.

  In step S011, based on the baseline length data obtained in step S006, the image shift amount is obtained by a known correlation calculation method using the subject image A image and the subject image B image after image correction, and the defocus amount is calculated. To do. After the calculation of the defocus amount is completed, the process proceeds to step S012. In step S012, it is determined from the calculated defocus amount whether or not it is in focus. If it is determined that the subject is not in focus, the process proceeds to step S013. In step S013, the third lens group 105 is advanced or retracted according to the calculated defocus amount, and the process returns to step S008. On the other hand, if it is determined that the subject is in focus, the process proceeds to step S014. In step S014, a series of focus detection flow ends, focus display is performed, and the process returns to the main flow.

  With the configuration as described above, the correction coefficient corresponding to the vignetting state of the light fluxes at the focus detection positions Img1, 2 and 3 can be read, the image can be restored by simple calculation, and the focusing accuracy can be improved. Accordingly, processing by a simple arithmetic processing circuit is possible, and low cost and space saving can be ensured. In the present embodiment, the diaphragm shape is described as being circular. In general, interchangeable lenses on the market actually have multiple shapes such as a hexagonal diaphragm with six blades, but it can be confirmed that there are few errors even if it is assumed to be circular in advance, so this example can be applied as it is It is. For a lens with an actual aperture shape and a large error, data may be stored in advance as a special lens.

  According to the present embodiment, individual shading correction coefficients can be calculated from a common shading correction data by a simple calculation process even for a commercially available interchangeable lens. For this reason, it is possible to ensure the desired accuracy of shading correction in spite of small-scale arithmetic processing. Further, processing by a simple arithmetic processing circuit is possible, and the focus detection speed can be secured at low cost.

  The embodiment of the present invention has been specifically described above. However, the present invention is not limited to the matters described as the above-described embodiments, and can be appropriately changed without departing from the technical idea of the present invention.

101 First Lens Group 102 Shutter / Shutter 103 Second Lens Group 105 Third Lens Group 107 Image Sensor 121 CPU

Claims (6)

An imaging unit having a plurality of focus detection pixels for photoelectrically converting an image formed by dividing an exit pupil of a light beam that has passed through a lens unit having a diaphragm and an imaging optical system;
Focus detection means for acquiring a pair of image signals from the plurality of focus detection pixels and detecting a focus state of the imaging optical system;
Control means for controlling the diaphragm so that the shape of the exit pupil opening at the position of the focus detection pixel depends on the shape of the aperture of the diaphragm when the focus state is detected by the focus detection means; An imaging device comprising: The imaging apparatus further includes data storage means,
The imaging apparatus according to claim 1, wherein the data storage unit stores a correction value related to focus detection corresponding to an aperture center shift on an exit pupil plane of the imaging optical system.   The imaging apparatus according to claim 2, wherein the correction value related to the focus detection is a value corresponding to a gain correction value for correcting an output of the focus detection pixel.   The imaging apparatus according to claim 2, wherein the correction value related to the focus detection is a value corresponding to an image shift conversion coefficient of the pair of image signals obtained from the focus detection pixels.   The imaging apparatus according to claim 2, wherein the correction value related to the focus detection is a value corresponding to an image correction filter of the pair of image signals obtained from the focus detection pixels. A step of photoelectrically converting an image formed by dividing an exit pupil of a light beam that has passed through a lens unit having an aperture and an imaging optical system by a plurality of focus detection pixels;
Obtaining a pair of image signals from the plurality of focus detection pixels and detecting a focus state of the imaging optical system;
And controlling the stop so that the shape of the opening of the exit pupil at the position of the focus detection pixel depends on the shape of the opening of the stop when the focus state is detected. Imaging method.