JP2013037296A - Image pickup apparatus and image pickup device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To retain sufficient focus accuracy, while retaining the quantity of light of a subject image even when a subject is present outside the optical axis.SOLUTION: The image pickup apparatus includes: an image pickup device; a pupil position determination part for determining a pixel for focus detection used for detecting a focus by a comparison between an exit pupil position and a plurality of design pupil positions; a correction part that, when the exit pupil position and the design pupil positions are different, corrects an image signal from the pixel for the focus detection used for detecting the focus on the basis of information of the exit pupil position, information of the design pupil positions and information of the pixel for the focus detection used for detecting the focus, and obtains a corrected image signal; and a focus detecting part for detecting the focus using the corrected image signal corrected by the correction part.

Description

  The present invention relates to an imaging apparatus and an imaging element having an autofocus function.

  In recent years, many portable devices (imaging devices) with a photographing function such as a digital camera have an autofocus function. This type of imaging apparatus employs an imaging element that incorporates a focus detection pixel (hereinafter referred to as an AF pixel) in addition to an imaging pixel (normal pixel) for image configuration, and uses a pupil division phase difference method. There are some that perform autofocus. In this method, it is necessary to divide the pupil in the left-right direction, and to configure an AF pixel having an imaging unit that separately receives light beams transmitted through the left and right pupils as an imaging element. A high-speed autofocus is possible by generating an AF signal for focusing and performing focusing on an image signal by these various AF pixels (hereinafter referred to as AF calculation or correlation calculation).

  By the way, there is an image sensor in which a micro lens is arranged in each pixel. By decentering the microlens according to the pixel position of the image sensor, the light from the photographing lens is appropriately guided to the light receiving area of each pixel. For example, Patent Document 1 discloses a technique in which a microlens is decentered according to a pixel position and a light receiving region is set on a straight line connecting the center of an exit pupil and the top of the microlens. As a result, the light beam transmitted through the left and right pupils can be reliably incident on the left and right AF pixels, and the AF accuracy can be improved.

    JP 2009-290157 A

  However, in the proposal of Patent Document 1, reduction in the amount of light due to so-called shading, which is caused by the inclination of the angle when the light beam from the outside of the optical axis of the photographing lens enters the image sensor, is not considered.

  An object of the present invention is to provide an imaging device and an imaging element that can maintain sufficient focus accuracy while maintaining the amount of light of a subject image even for a subject outside the optical axis.

An image pickup device according to the present invention is an image pickup device in which a plurality of pixels each having an image pickup pixel and a focus detection pixel are arranged in a matrix, and the pixel is a light receiving region where light from a photographing lens enters. And a microlens that guides light from the photographing lens to the light receiving region, and the focus detection pixel includes a light receiving region of the focus detection pixel, an assumed design pupil position, and the micro The position of one end of the region is defined corresponding to the positional relationship with the lens.
An imaging apparatus according to the present invention determines whether the imaging pupil according to claim 1 and an exit pupil position coincide with any of the plurality of designed pupil positions, and is used for focus detection based on a determination result. A pupil position determination unit that determines the pixel of the first pixel, and a focus detection unit that performs focus detection using the focus detection pixels used for the focus detection.

  According to the present invention, it is possible to maintain sufficient focusing accuracy while maintaining the amount of light of a subject image even for a subject outside the optical axis.

1 is a block diagram illustrating an imaging apparatus according to an embodiment of the present invention. Explanatory drawing which shows the structure of the pixel of the optical axis vicinity of a photographic lens among each pixel which comprises an image pick-up element. Explanatory drawing which shows the structure of the pixel of the optical-axis periphery part of an imaging lens among each pixel which comprises an image pick-up element. Explanatory drawing which shows the positional relationship among an effective light-receiving area | region, a micro lens, and an exit pupil. FIG. 4 is an explanatory diagram showing the relationship between the focal length and the position of the exit pupil. Explanatory drawing for demonstrating the amount of eccentricity according to the exit pupil position. Explanatory drawing for demonstrating the amount of eccentricity according to the exit pupil position. Explanatory drawing for demonstrating the amount of eccentricity according to the exit pupil position. The graph which shows the change of the amount of eccentricity according to image height and exit pupil position, with the image height on the horizontal axis and the amount of eccentricity ε on the vertical axis. Explanatory drawing for demonstrating the correction method of an image signal. Explanatory drawing for demonstrating the correction method of an image signal. Explanatory drawing for demonstrating the correction method of an image signal. Explanatory drawing which shows an example of the pixel arrangement | sequence of the light-receiving part 14a in FIG. 6 is a flowchart for explaining camera control in the present embodiment. 14 is a flowchart specifically illustrating AF processing in FIG. 13.

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

  FIG. 1 is a block diagram showing an imaging apparatus according to an embodiment of the present invention.

  First, with reference to FIGS. 2A and 2B to FIG. 11, an image sensor and an image signal correction method employed in the present embodiment will be described. FIG. 2A is an explanatory diagram illustrating the configuration of pixels in the vicinity of the optical axis of the photographing lens among the pixels constituting the image sensor, and FIG. 2B is a pixel in the periphery of the optical axis of the photographing lens among the pixels constituting the image sensor. It is explanatory drawing which shows this structure.

In the conventional technology, the relationship between the microlens and the light receiving region at each pixel position is optimized only for one exit pupil position. For this reason, when shooting is performed at a pupil position that is deviated from the optimized pupil position, the light beam that has passed through the left and right pupils does not uniformly enter the left and right AF pixels, and the AF accuracy deteriorates. For example, when an interchangeable lens with a different pupil position is used, or when a zoom lens with a moving pupil position is used, there is a problem that the AF accuracy is deteriorated.
In this embodiment, an imaging device that can maintain sufficient focus accuracy while maintaining the amount of light of an image even when the pupil position changes will be described.
The imaging lens of the imaging apparatus forms an optical image incident on the imaging apparatus from the subject via each optical path on the light receiving surface of the imaging element. In the pupil division phase difference method, two imaging units (for example, an R imaging unit and an L imaging unit) are configured as AF pixels, and each optical path is divided into, for example, a right direction and a left direction at the exit pupil, and the right direction Light (right light) and light from the left direction (left light) are incident on the R imaging unit and the L imaging unit, respectively.

  When the subject is in focus, light from substantially the same point of the subject enters the R and L imaging units. Accordingly, the image signal obtained by the plurality of R imaging units arranged in the horizontal direction is the same as the image signal obtained by the plurality of L imaging units. When the focus is shifted, light from substantially the same point of the subject enters the R imaging unit and the L imaging unit at positions shifted according to the focus shift amount. Therefore, the image signals obtained by the plurality of R imaging units arranged in the horizontal direction and the image signals obtained by the plurality of L imaging units are out of phase, and the amount of phase deviation corresponds to the amount of focus deviation. Autofocus can be realized by driving a focus adjustment lens based on the phase difference (correlation) between the image signals obtained by the R and L imaging units.

  Note that in order to share the readout circuit between the AF pixel and the normal pixel, a pixel having only the R imaging unit (hereinafter referred to as an R pixel) and an L pixel without forming both the R and L imaging units in one pixel. An AF pixel may be configured by a pixel having only an imaging unit (hereinafter referred to as an L pixel). Further, the L pixel is omitted, and only the R pixel is used as the AF pixel, and the focus is determined by the correlation between the image signal obtained by a plurality of imaging pixels (hereinafter also referred to as N pixels) and the image signal obtained by the R pixel. The amount of deviation may be obtained.

  2A and 2B show an example in which the pixels 51 (51 ') and 52 (52') are an R pixel and an L pixel, respectively. The pixels including the pixels 51 (51 ′) and 52 (52 ′) are, in order from the top, the microlens 53 (53 ′), the color filter 54 (54 ′), and a light shielding film for preventing color mixture of the color pixels. 55 (55 ′), a smoothing layer 56 (56 ′) for flattening the surface on which the color filter layer is placed, and a photoelectric conversion region (hereinafter referred to as a light receiving region) are arranged. The light receiving area is shielded by the light shielding portions 58 (58R, 58L) (shaded portions). The light receiving regions 57R and 57R 'constitute an R imaging unit and the pixels 51 and 51' become R pixels by the light shielding by the light shielding units 58 and 58R. In addition, the light receiving regions 57L and 57L 'constitute an L imaging unit and the pixels 52 and 52' become L pixels by the light shielding of the light shielding units 58 and 58L.

  A broken line in FIG. 2A indicates an optical axis of each microlens 53. The light receiving regions 57R and 57L constituting the R imaging unit and the L imaging unit are shielded from light by the light shielding unit 58, and the position of one end of the effective light receiving region for effectively receiving the light is as shown in FIG. It matches the optical axis.

Since the pixels 51 and 52 are pixels in the vicinity of the optical axis of the photographic lens, the optical axis of the microlens 53 of the pixels 51 and 52 is substantially parallel to the optical axis of the photographic lens (not shown). It substantially coincides with the optical axis of the taking lens. Accordingly, the right light beam, which is obtained by equally dividing the light beam passing through the exit pupil into the left and right, is incident on the effective light receiving region of the light receiving region 57R, and the light beam passing through the exit pupil is incident on the effective light receiving region of the light receiving region 57L. Of the left side of the light is evenly divided into left and right. That is, the light fluxes that pass through the exit pupil are incident on the light receiving areas 57R and 57L after being equally divided into left and right.
On the other hand, the screen periphery is designed so that the effective light receiving region is shifted by δ with respect to the optical axis (one-dot chain line) of the microlens 53 ′ as a countermeasure against shading. When expressed in a simplified manner, the periphery of the screen is configured as shown in FIG. 2B.

In addition, when the pixel is arranged at a position shifted from the vicinity of the optical axis of the photographing lens, the principal ray does not become parallel to the optical axis of the microlens 53. In this case, in order to make the light beam passing through the exit pupil equally split and enter the L imaging unit or the R imaging unit, one end of the effective light receiving region constituting the L and R imaging units is connected to the microlens 53. It must be located on an axis passing through the top of the surface and the center of the exit pupil.
Therefore, as shown in FIG. 2B, light shielding portions 58R and 58L are provided on the light receiving regions 57R and 57L, respectively, and the light shielding regions by these light shielding portions 58R and 58L are shifted by εR and εL, respectively. As a result, the balance of the R and L outputs when a light beam parallel to the optical axis of the micro lens 53 is incident on the pair of AF pixels 51 and 52 and the pair of AF pixels 51 ′ and 52 ′ are microscopic. The balance of the R and L outputs when a light beam not parallel to the optical axis of the lens 53 ′ is incident can be made uniform.

  FIG. 3 is an explanatory diagram showing the positional relationship among the effective light receiving region, the microlens, and the exit pupil. FIG. 4 is an explanatory diagram showing the relationship between the focal length and the position of the exit pupil. In the following description, the light receiving area indicates an effective light receiving area determined by the light shielding portion and the light receiving area.

  FIG. 3 shows a microlens 61 that constitutes one of the pixels other than the pixels in the vicinity of the optical axis of the photographic lens and a light receiving region 62R that is an R imaging unit. FIG. 3 shows exit pupils 63T, 63S, and 63W (hereinafter, representatively referred to as exit pupils 63) at three positions corresponding to changes in the position of the photographing lens. FIG. 4 illustrates a state in which the light beam from the subject forms an image plane with the photographing lens. FIG. 4A illustrates an example in which the photographing lens is located on the tele side (subject side), and FIG. ) Shows an example in which the photographing lens is located at the reference position, and FIG. 4C shows an example in which the photographing lens is located on the wide side (image side). As shown in FIG. 4, generally, the longer the focal length, the greater the distance from the image plane to the exit pupil position.

  As shown in FIG. 3, in the light receiving region 62R, one end of the light receiving region 62R is located at a position where the principal ray 65 passing through the center of the exit pupil 63S at the reference position and the top of the micro lens 61 intersects the light receiving surface. It is configured. That is, the light receiving region 62R is decentered with respect to the optical axis of the microlens 61 by a predetermined amount of eccentricity corresponding to the inclination of the principal ray 65. This amount of eccentricity changes according to the pixel position, and increases as the image height increases.

  Of the light flux from the subject passing through the exit pupil 63S, the right half light flux 64S of the exit pupil 63S is incident on the light receiving area 62R constituting the R imaging unit. That is, by decentering each light receiving region constituting the L imaging unit and the R imaging unit to a position where the principal ray 65 passing through the center of the exit pupil 63S and the top of the micro lens 61 intersects the light receiving surface, the L imaging unit. In addition, the light beam passing through the exit pupil can be equally divided into the left and right sides and incident on the R imaging unit.

  However, as shown in FIG. 3, with respect to the exit pupil 63W when the photographing lens moves to the wide side (image side), a light flux 64W larger than half of the light flux passing through the exit pupil 63W constitutes the R imaging unit. The light enters the light receiving region 62R. Further, regarding the exit pupil 63T when the photographic lens moves to the tele side (subject side), a light beam 64T smaller than half of the light beam passing through the exit pupil 63T enters the light receiving region 62R.

  That is, when the position of the light receiving region 62R is determined according to the principal ray 65 passing through the center of the exit pupil 63S, when the exit pupil 63 is positioned other than the position of the exit pupil 63S, the L imaging unit and the R imaging The light beam incident on the part is not uniform. Then, when the exit pupil 63 is located at a position other than the reference position, the AF accuracy is significantly deteriorated.

  Therefore, in the present embodiment, the amount of eccentricity of the light receiving area of the AF pixel is set according to the exit pupil position. That is, even with a plurality of AF pixels in which the inclination of the principal ray passing through the center of the exit pupil and the top of the microlens at the reference exit pupil position is substantially the same, the inclination of the principal ray changes depending on the exit pupil position. A plurality of types of eccentricity corresponding to each are set.

  For example, assuming three exit pupil positions, a tele-side position, a reference position, and a wide-side position, the light-receiving area is configured with an eccentric amount corresponding to the chief ray inclination when the exit pupil position is located on the tele side. Pixels, the pixels constituting the light receiving area with the amount of eccentricity according to the inclination of the principal ray when the exit pupil position is at the reference position, and the eccentricity according to the inclination of the principal ray when the exit pupil position is located on the wide side Three types of pixels, which constitute the light receiving region by the amount, are formed on the image sensor.

  5 to 7 are explanatory views for explaining the amount of eccentricity according to the exit pupil position. The pixels shown in FIG. 5 to FIG. 7 are three pixels having substantially the same principal ray inclination connecting the center of the exit pupil 73S at the reference position and the tops of the microlenses 70M to 73M, that is, the horizontal on the image sensor. Three pixels having substantially the same position in the direction are shown.

  FIG. 5 shows a pixel 70 in which the light receiving area is configured with an eccentric amount corresponding to the inclination of the principal ray 75T when the exit pupil position is located on the tele side. The pixel 70 includes a microlens 70M and a light receiving region 70RT that constitutes an R imaging unit. The light receiving region 70RT is formed such that the end of the light receiving region 70RT is positioned at a position where the principal ray 75T connecting the center of the exit pupil 73T and the top of the micro lens 70M intersects the light receiving surface. The amount of eccentricity between the optical axis of the micro lens 70M and the end of the light receiving region 70RT is εt.

  In the pixel 70, when the exit pupil position is at the position of the exit pupil 73T, the right half of the light flux 74T passing through the exit pupil 73T is incident on the light receiving region 70RT. That is, the pixel 70 is a pixel capable of obtaining sufficient AF accuracy when the exit pupil position is at the position of the exit pupil 73T on the tele side.

  FIG. 6 shows a pixel 71 in which the light receiving area is configured with an eccentric amount corresponding to the inclination of the principal ray 75S when the exit pupil position is located at the reference position. The pixel 71 includes a microlens 71M and a light receiving region 71RS that forms an R imaging unit. The light receiving region 71RS is formed so that the end of the light receiving region 71RS is located at a position where the principal ray 75S connecting the center of the exit pupil 73S and the top of the micro lens 71M intersects the light receiving surface. The amount of eccentricity between the optical axis of the micro lens 71M and the end of the light receiving region 71RS is εs.

  In the pixel 71, when the exit pupil position is at the position of the exit pupil 73S, the right half of the light beam 74S passing through the exit pupil 73S enters the light receiving region 71RS. That is, the pixel 71 is a pixel that can obtain sufficient AF accuracy when the exit pupil position is at the exit pupil 73S as the reference position.

  FIG. 7 shows a pixel 72 in which the light receiving area is configured with an eccentric amount corresponding to the inclination of the principal ray 75W when the exit pupil position is located on the wide side. The pixel 72 includes a microlens 72M and a light receiving region 72RW that forms an R imaging unit. The light receiving region 72RW is formed so that the end of the light receiving region 72RW is positioned at a position where the principal ray 75W connecting the center of the exit pupil 73W and the top of the micro lens 72M intersects the light receiving surface. The amount of eccentricity between the optical axis of the micro lens 72M and the end of the light receiving region 72RW is εw.

  In the pixel 72, when the exit pupil position is at the exit pupil 73W, the right half of the light beam 74W passing through the exit pupil 73W enters the light receiving region 72RW. That is, the pixel 72 is a pixel that can obtain sufficient AF accuracy when the exit pupil position is at the position of the wide exit pupil 73W.

  FIG. 8 is a graph showing the change in the amount of eccentricity according to the image height and the exit pupil position, with the image height on the horizontal axis and the amount of eccentricity ε on the vertical axis. As shown in FIG. 8, the amount of eccentricity increases as the image height increases. Further, in the present embodiment, the amount of eccentricity varies depending on the exit pupil position, and pixels having three types of eccentricity are configured even at the same image height. In FIGS. 5 to 8, three types of exit pupil positions are described. However, assuming two or more types of exit pupil positions, pixels whose eccentricity amount is changed for each exit pupil position. May be configured.

  By the way, in the case of using a photographing lens in which a plurality of exit pupil positions are fixed, it is possible to obtain sufficient AF accuracy by forming pixels that constitute a light receiving area with an eccentric amount corresponding to each exit pupil position. Is possible. However, when the exit pupil position is not fixed, it is conceivable that the actual exit pupil position and the exit pupil position assumed to determine the amount of eccentricity (hereinafter referred to as the designed pupil position) are different. Therefore, in the present embodiment, when the actual exit pupil position and the designed pupil position are deviated, the image signal obtained from the AF pixel is corrected according to the deviation amount.

  9 to 11 are explanatory diagrams for explaining such a method of correcting an image signal. As described above, the distance (eccentricity ε) from the optical axis of the micro lens of each AF pixel to the end of the L or R imaging unit increases as the image height increases. 9 to 11, the distance from the optical axis of the photographing lens will be described as the image height h.

  FIG. 9 shows an example where the actual exit pupil position is on the wide side with respect to the designed pupil position. In the pixel constituted by the microlens 81 and the light receiving region 82 constituting the R imaging unit, the distance from the optical axis of the photographing lens to the optical axis of the microlens 81 is h. The position of the end of the light receiving region 82 is set based on a principal ray 85 that connects the center of the exit pupil at the design pupil position and the top of the microlens 81.

  On the other hand, when the exit pupil position in actual photographing is on the wide side, as shown in FIG. 9, the right light beam 84R is larger in size than the left light beam 84L at the exit pupil position. In this case, the light beam is not equally divided into the right and left sides, but the large light beam 84R is incident on the light receiving region 82. Therefore, if the correlation calculation is performed using the image signal obtained from the light receiving region 82 as it is, AF accuracy is increased. Will fall.

  Therefore, in order to correct the amount of light received by the L and R imaging units resulting from the area difference of the exit pupil divided asymmetrically, calculation using the exit pupil position EPZtemp and the pupil diameter EXP is performed. As is apparent from FIG. 9, the distance d1-temp from the center of the actual exit pupil position to the left and right pupil division positions is given by the following equation (1) with the design pupil position as EPZAF1.

d1-temp = h. (EPZAF1-EPZtemp) / EPZAF1 (1)
Left and right pupil division ratio EXPtemp1-l: EXPtemp1-r is
EXPtemp1-l: EXPtemp1-r = (EXP / 2) + d1-temp: (EXP / 2) -d1-temp
... (2)
It becomes.

Using the pupil division ratio of the above equation (2), it is possible to correct the nonuniformity of the left and right pupil divisions. The following equation (3) represents a function for obtaining a corrected image signal with RawAFsignal as the uncorrected image signal and CorAFsignal as the corrected image signal.
CorAFsignal = RwaAFsignal · g (EXP, EXPtemp1-l, EXPtemp1-r) (3)
For example, specifically, the corrected value CorAFsignalr of the image signal obtained from the R imaging unit and the corrected value CorAFsignall of the image signal obtained from the L imaging unit are obtained by the following equation (4).
CorAFsignalr = RwaAFsignalr (EXPtemp1-l / EXP)
CorAFsignall = RwaAFsignall (EXPtemp1-r / EXP) (4)
This formula (4) can correct the non-uniformity of the left and right pupil divisions due to the deviation between the actual exit pupil position and the designed pupil position, and sufficient accuracy can be obtained by the correlation calculation using the corrected image signal. Can achieve the focus.

  FIG. 10 shows an example in which the actual exit pupil position is on the tele side with respect to the designed pupil position. In the pixel constituted by the microlens 86 and the light receiving region 87 constituting the R imaging unit, the distance from the optical axis of the photographing lens to the optical axis of the microlens 86 is h. The position of the end portion of the light receiving region 87 is set based on a principal ray 89 that connects the center of the exit pupil at the design pupil position and the surface top of the microlens 86.

  On the other hand, when the exit pupil position in actual photographing is on the tele side, as shown in FIG. 10, the right light beam 84R is smaller in size than the left light beam 84L at the exit pupil position. In this case, since the light beam is not equally divided into the left and right sides and the small light beam 84R is incident on the light receiving region 87, if the correlation calculation is performed using the image signal obtained from the light receiving region 87 as it is, the AF accuracy is improved. Will fall.

  Therefore, in order to correct the amount of light received by the L and R imaging units resulting from the area difference of the exit pupil divided asymmetrically, calculation using the exit pupil position EPZtemp and the pupil diameter EXP is performed. As is apparent from FIG. 10, the distance d2-temp from the center of the actual exit pupil position to the left and right pupil division positions is given by the following equation (5), where the design pupil position is EPZAF2.

d2-temp = h. (EPZtemp-EPZAF2) / EPZAF2 (5)
Left and right pupil division ratio EXPtemp2-l: EXPtemp2-r is
EXPtemp2-l: EXPtemp2-r = (EXP / 2) + d2-temp: (EXP / 2) -d2-temp (6)
It becomes.

The following equation (7) shows a function for obtaining a corrected image signal using the pupil division ratio of the above equation (6).
CorAFsignal = RwaAFsignal · g (EXP, EXPtemp2-l, EXPtemp2-r) (7)
For example, specifically, the corrected value CorAFsignalr of the image signal obtained from the R imaging unit and the corrected value CorAFsignall of the image signal obtained from the L imaging unit are obtained by the following equation (8).
CorAFsignalr = RwaAFsignalr (EXPtemp2-l / EXP)
CorAFsignall = RwaAFsignall · (EXPtemp2-r / EXP) (8)
This equation (8) can correct the non-uniformity of the left and right pupil division due to the deviation between the actual exit pupil position and the designed pupil position, and sufficient accuracy can be obtained by the correlation calculation using the corrected image signal. Can achieve the focus.

  Note that each of the above equations shows the calculation for correcting the image signal using the ratio of the left and right light beam diameters by pupil division at the actual exit pupil position. Based on this, the image signal may be corrected. In this case, there is an advantage that the AF accuracy can be further improved.

  In this way, by correcting the image signal obtained from one AF pixel set corresponding to the designed pupil position by the above formula (4) or (8), the left and right pupils are evenly distributed at the actual exit pupil position. An image signal similar to that obtained when the division is performed can be obtained. Furthermore, one corrected image signal may be obtained using image signals obtained from a plurality of AF pixels set corresponding to a plurality of adjacent design pupil positions.

  FIG. 11 is a diagram for explaining a correction signal generation method in this case. Of the two AF pixels (not shown) arranged relatively close to each other, one pixel is formed assuming that the position of the exit pupil 91 is the design pupil position, and the other pixel is the position of the exit pupil 92. It is assumed that it is formed assuming the design pupil position. Assume that the actual exit pupil position is the position of the exit pupil 93.

  The light beams 94L, 94R, 95L, and 95R in FIG. 11 are shown by shifting the left and right light beams resulting from the division of the left and right pupils of the exit pupil 93 from the position of the exit pupil 93 in order to make the drawing easier to see. When the AF pixel whose design pupil position is the position of the exit pupil 91 is an R pixel, the light beam 94R enters the light receiving region, and when the AF pixel is an L pixel, the light beam 94L enters the light receiving region. When the AF pixel whose design pupil position is the position of the exit pupil 92 is an R pixel, the light beam 95R is incident on the light receiving region, and when the AF pixel is an L pixel, the light beam 95L is incident on the light receiving region.

  The exit pupil position of the exit pupil 93, the design pupil positions of the exit pupils 91 and 92, and the distance (h) between the optical axis of the micro lens of the two AF pixels and the optical axis of the photographing lens corresponding to the design pupil position are as follows. If known, the image signals obtained from these two AF pixels can be corrected by the above equations (4) and (8) as described in FIGS. 9 and 10 and the above equations.

The corrected image signal obtained by the calculation of the above expression (4) for the image signal from the AF pixel whose design pupil position is the position of the exit pupil 91 is CorAFsignal-1, and the AF pixel whose design pupil position is the position of the exit pupil 92. The corrected image signal obtained by the calculation of the above equation (8) with respect to the image signal from is called CorAFsignal-2. In the example of FIG. 11, these two corrected image signals are weighted based on the distances a and b between the design pupil position and the actual exit pupil position and added to obtain one corrected image signal. ing. For example, the corrected image signal CorAFsignalAVE by such weighted addition is obtained by the following equation (9).
CorsingalAVE = {wa (a, b) · CorAFsignal-1 + wb (a, b) · CorAFsignal-2} / {wa (a, b) + wb (a, b)} (9)
However, wa (a, b) and wb (a, b) are weighting coefficients, for example, wa (a, b) = b / (a + b), wb (a, b) = a / (a + b). .

  For example, when the actual exit pupil position is not in the vicinity of the design pupil position, the corrected image signals of the two AF pixels corresponding to the two design pupil positions in the vicinity of the actual exit pupil position are weighted and added. A highly reliable image signal can be obtained. Thereby, even when the actual exit pupil position deviates relatively greatly from the designed pupil position, sufficient AF accuracy can be ensured.

  In the above description, an example in which L pixels or R pixels are used as AF pixels has been described. However, when there are many horizontal lines in the captured image, it is conceivable that the images obtained by the L and R imaging units match even if the focus is shifted. In this case, for example, the exit pupil may be divided into upper and lower parts, and a U imaging unit and a D imaging unit that receive light from above and light from below may be configured. Then, by comparing the phase of the image signal obtained by the plurality of U imaging units and the phase of the image signal obtained by the plurality of D imaging units, it is possible to detect the amount of focus shift and to focus. .

  Even in this case, the correction shown in FIGS. 9 to 11 is possible, and the distance from the optical axis of the photographing lens may be set to the image height h.

(Circuit configuration)
As shown in FIG. 1, a main body circuit unit 10 is provided in a main body housing (not shown) of the imaging apparatus 1, and an interchangeable lens circuit unit 21 is provided in an interchangeable lens unit 20 that is detachably attached to the main body housing. Yes. The main body circuit unit 10 is provided with a communication unit 12, and the interchangeable lens circuit unit 21 is provided with a communication unit 23. The communication unit 12 of the main body circuit unit 10 can exchange information with the communication unit 23 of the interchangeable lens circuit unit 21.

  The interchangeable lens 20 has a photographic lens 26. The photographic lens 26 has an autofocus function that enables focusing by being driven by the drive unit 25. The photographing lens 26 is driven by the drive unit 25 and has a zoom function. As the interchangeable lens 20, a lens having a single focus photographing lens may be adopted.

  The lens control unit 22 of the interchangeable lens 20 controls the driving unit 25 by a signal from the main body circuit unit 10 or a signal from the operation unit 24 based on a user operation to drive the photographing lens 26. Aperture, focus, zoom, etc. can be controlled.

  The communication unit 23 of the lens control unit 22 transmits and receives information to and from the communication unit 12 of the main body circuit unit 10 via a predetermined transmission path. When communication with the communication unit 12 of the main body circuit unit 10 is established, the lens control unit 22 can cause the main body circuit unit 10 to transmit information on the lens stored in the lens memory 27 by the communication unit 23.

  In the present embodiment, the lens memory 27 includes, for example, information on the lens position, the aperture diameter, the exit pupil position, the exit pupil diameter, the focus lens position, the vignetting according to the image height and direction, and the like. It comes to memorize. Accordingly, the main body circuit unit 10 can recognize the zoom magnification, focal length, brightness number, and the like of the interchangeable lens 20 and can acquire information necessary for correcting the image signal of the AF pixel. It can be done.

  The main body circuit unit 10 includes an imaging unit 14 configured by an imaging element such as a CMOS sensor. The imaging unit 14 includes a light receiving unit 14 a that receives light of a subject from the interchangeable lens 20. The optical image of the subject from the interchangeable lens 20 is formed on the light receiving unit 14a.

  FIG. 12 is an explanatory diagram showing an example of a pixel array of the light receiving unit 14a in FIG. In the present embodiment, an example in which a Bayer array is adopted as the pixel array will be described. FIG. 12 shows one pixel by one frame, and rough hatching in the frame indicates a red pixel in which a red filter is arranged, and dense hatching indicates a blue pixel in which a blue filter is arranged. Solid color indicates a green pixel in which a green filter is arranged. The character in the frame indicates an AF pixel, the character W indicates a pixel whose design pupil position is set to the wide side, and the character S is a pixel whose design pupil position is set to the reference position side. The letter T indicates a pixel whose design pupil position is set to the tele side.

  Further, a pixel in which L is included in a subscript of a character indicating an AF pixel indicates an L pixel, and a pixel in which R is included in a subscript of a character indicating an AF pixel is an R pixel. The number included in the subscript of the letter indicating the AF pixel corresponds to the horizontal distance h between the optical axis of the micro lens and the optical axis of the photographing lens of each AF pixel, and the same number is the same distance. h.

  When the correction method of FIG. 11 is employed, the correction accuracy is improved when two pixels corresponding to design pupil positions close to each other among a plurality of types of design pupil positions are arranged at close positions. Is considered high. For this reason, it is better to arrange AF pixels corresponding to a plurality of design pupil positions to be set. For example, when setting three types of design pupil positions of the wide side, the reference position, and the tele side, the AF pixels corresponding to these design pupil positions are set to the wide side, the reference position and the tele side in order or the tele side, Arrange in order of position and wide side.

  The light receiving unit 14a illustrated in FIG. 12 has three types of pixels set as three design pupil positions as AF pixels. That is, AF pixels W1L, W2L,..., W1R, W2R,... With the designed pupil position set to the wide side, AF pixels S1L, S2L,..., S1R, S2R,. And there are three types of AF pixels T1L, T2L,..., T1R, T2R,.

  The imaging unit 14 is driven and controlled by the signal processing and control unit 11. The light receiving unit 14a of the imaging unit 14 photoelectrically converts an optical image from the subject and outputs an image signal. The signal processing and control unit 11 is provided with a signal extraction unit 11 a, and the signal extraction unit 11 a takes in an image signal from the imaging unit 14. The signal extraction unit 11a outputs the captured image signal to the image processing unit 11b, and outputs the image signal of the AF pixel to the correction unit 11c.

  The determination result of the pupil position determination unit 11d is given to the correction unit 11c. The correction unit 11c is caused by the difference between the design pupil position set in the AF pixel and the actual exit pupil position, and the right and left pupil division is unbalanced, and the sizes of the incident light beams of the L and R imaging units are different from each other. The resulting image signal error is corrected. The correction unit 11c uses the determination result of the pupil position determination unit 11d for image signal correction processing.

  Information regarding the lens is given to the pupil position determination unit 11d via the communication unit 12, and information about the AF pixel is acquired from the main body memory 19. The main body memory 19 stores information on each AF pixel, for example, information on the distance h between the optical axis of the microlens and the photographing lens optical axis of each AF pixel.

  The pupil position determination unit 11d acquires information regarding the actual exit pupil position from the lens memory 27, acquires information information regarding the design pupil position set for each AF pixel from the main body memory 19, and uses AF pixels for AF calculation. And a correction method of the image signal is determined. That is, when there is a design pupil position that matches the actual exit pupil position, the pupil position determination unit 11d outputs a determination result for performing AF calculation using the AF pixel corresponding to the design pupil position as it is. Moreover, the pupil position determination part 11d outputs the determination result for selecting the AF pixel corresponding to the design pupil position close to the exit pupil position when the actual exit pupil position and the design pupil position do not match. At the same time, it is determined which of the correction methods shown in FIGS. 9 to 11 is used, and the determination result is output to the correction unit 11c.

  The correction unit 11c acquires information on the actual exit pupil position and exit pupil diameter from the lens memory 27 via the communication unit 12, and acquires information on the distance h and the design pupil position of each AF pixel from the main body memory 19, According to the determination result of the pupil position determination unit 11d, the image signal is output to the AF signal calculation unit 11e without being corrected, or the image signal of the AF pixel is selected by using any one of the correction methods shown in FIGS. It corrects and outputs the corrected image signal to the AF signal calculation unit 11e.

  The AF signal calculation unit 11e is supplied with the corrected image signal of the AF pixel, calculates the correlation of the corrected image signal, and calculates the AF signal by the pupil division phase difference method. The signal processing and control unit 11 performs focusing by supplying the AF signal calculated by the AF signal calculating unit 11e to the driving unit 25 via the communication units 12 and 23.

  The image processing unit 11b of the signal processing and control unit 11 performs predetermined signal processing, such as color signal generation processing, matrix conversion processing, and other various digital processing, on the image signal from the signal extraction unit 11a. The signal processing and control unit 11 can output image information, audio information, and the like compressed by performing an encoding process when recording an image signal, an audio signal, and the like.

  The main body circuit unit 10 is also provided with a clock unit 15 and an operation determination unit 16. The clock unit 15 generates time information used by the signal processing and control unit 11. The operation determination unit 16 generates an operation signal based on a user operation with respect to various switches (not shown) such as an imaging start / end button and a shooting mode setting provided in the imaging apparatus 1 and outputs the operation signal to the signal processing and control unit 11. It has become. The signal processing and control unit 11 controls each unit based on the operation signal.

  The main body circuit unit 10 is provided with a recording / reproducing unit 17 and a display unit 18. The recording / reproducing unit 17 can record image information and audio information from the signal processing and control unit 11 on a recording medium (not shown). For example, a card interface can be employed as the recording / reproducing unit 17, and the recording / reproducing unit 17 can record image information, audio information, and the like on a memory card or the like. Further, the recording / reproducing unit 17 can read out image information and audio information recorded on the recording medium and supply them to the signal processing and control unit 11. The signal processing and control unit 11 can decode the image information and the audio information from the recording / reproducing unit 17 to obtain an image signal and an audio signal.

  The display unit 18 is supplied with the captured image from the imaging unit 14 and the reproduced image from the recording / reproducing unit 17 from the signal processing and control unit 11 and can display these images. Further, the display unit 18 can be controlled by the signal processing and control unit 11 to display a menu display or the like for operating the photographing apparatus 1.

  When the generated video signal is supplied to the recording / playback unit 17 for recording, the signal processing and control unit 11 converts the recorded video signal into a file based on a user operation. Note that, by making a file, various processes based on a user operation, such as a playback process, can be performed on the video signal recorded in the recording / playback unit 17 (hereinafter referred to as a video file).

  Next, the operation of the embodiment configured as described above will be described with reference to FIGS. 13 and 14. FIG. 13 is a flowchart for explaining camera control in the present embodiment, and FIG. 14 is a flowchart specifically showing the AF processing in FIG.

  When the image pickup apparatus 1 is turned on, the signal processing and control unit 11 determines whether or not the shooting mode is instructed in step S21 of FIG. If the shooting mode is not instructed, the signal processing and control unit 11 determines in step S22 whether or not the playback mode is instructed. When the reproduction mode is instructed, the signal processing and control unit 11 displays a list of thumbnails in step S23. When the user selects an image with reference to the thumbnail list, the process proceeds from step S24 to step S25, and the signal processing and control unit 11 reproduces the selected image. If no file is selected, it is determined in step S26 that the playback mode has ended.

  On the other hand, when the shooting mode is instructed, the signal processing and control unit 11 causes the display unit 18 to display the captured image (through image) in live view based on the image signal from the imaging unit 14 in step S31. In this case, the signal processing and control unit 11 thins the captured image from the imaging unit 14 in accordance with the number of display pixels of the display unit 18 and then gives the image to the display unit 18.

  In the next step S32, focusing is performed by AF processing. That is, the signal processing and control unit 11 acquires lens information stored in the lens memory 27 via the communication units 12 and 23 in step S51 of FIG. Further, in step S52, the signal processing and control unit 11 acquires information regarding the AF pixels stored in the main body memory 19.

  Next, the pupil position determination unit 11d of the signal processing and control unit 11 determines whether or not the actual exit pupil position matches the design pupil position in step S53, and outputs the determination result to the correction unit 11c. . The pupil position determination unit 11d acquires information on the design pupil position set for each AF pixel from the main body memory 19, and uses the information on the actual exit pupil position included in the lens information to determine the exit pupil position and the design. The coincidence with the pupil position is determined. For example, when a single focal lens having a fixed focal length is employed, the exit pupil position may coincide with the designed pupil position.

  In this case, the correction unit 11c reads out the image signal of the AF pixel corresponding to the design pupil position that matches the actual exit pupil position, and outputs it as it is to the AF signal calculation unit 11e (step S54). The AF signal calculation unit 11e generates an AF signal by correlation calculation using the read image signal. The AF signal is supplied to the drive unit 25 via the communication units 12 and 23, and focusing is performed (step S60).

  If the actual exit pupil position does not match the design pupil position, the pupil position determination unit 11d moves the process to step S55, and determines whether the difference between the exit pupil position and the design pupil position is within a predetermined threshold. It determines whether or not, and outputs the determination result to the correction unit 11c. When the difference between the exit pupil position and the designed pupil position is within a predetermined threshold, the correction unit 11c corrects the image signal by using the correction method shown in FIG. 9 or FIG. 10 (step S56). That is, the correction unit 11c corrects the image signal by, for example, the calculation of the above formula (4) or (8) using the information about the exit pupil, the information about the design pupil position, and the information about the distance h of the AF pixel. Next, in step S60, an AF signal is generated and focusing is performed.

  When the pupil position determination unit 11d gives a determination result indicating that the difference between the exit pupil position and the design pupil position exceeds a predetermined threshold, the correction unit 11c employs the correction method illustrated in FIG. Then, the image signal is corrected. That is, the correction unit 11c first reads out image signals of a plurality of pixels respectively corresponding to a plurality of design pupil positions close to the actual exit pupil position (step S57). Next, the correction unit 11c corrects each image signal by the calculation of the above expression (4) or (8) (step S58). Next, the correcting unit 11c adds the weight based on the distance between the exit pupil position and the designed pupil position to the corrected image signal and calculates the average. Thus, one image signal is generated from the plurality of AF pixels. Next, in step S60, an AF signal is generated and focusing is performed.

  By steps S61 and S62, the focus control is continued until focusing is achieved.

  Next, in step S33 in FIG. 13, when shooting is instructed by a shutter release operation, the signal processing and control unit 11 performs predetermined signal processing on the image signal from the imaging unit 14 to record an image for recording. Is generated. The signal processing and control unit 11 records the generated captured image in a memory (not shown) (step S34). For the AF pixel in the recorded captured image, the signal processing and control unit 11 obtains an image signal when the AF pixel is configured by the imaging pixel by correction processing using pixels around the AF pixel. Thus, an image signal at the AF pixel position is set (step S36).

  The signal processing and control unit 11 gives the captured image in which the image signal of the AF pixel is corrected to the recording / reproducing unit 17 to create a file (step S36). In addition, the signal processing and control unit 11 gives the recorded captured image to the display unit 18 to perform REC view display (step S37).

  In step S41, the signal processing and control unit 11 determines whether or not a power-off operation has been performed. When the power-off operation is not performed, the signal processing and control unit 11 receives the photographing / playback mode changing operation (step S42), and then returns the process to step S21. When the power-off operation is performed, the signal processing and control unit 11 turns off the power (step S43).

  In the AF process of FIG. 14, an example has been described in which the correlation calculation is performed using only the AF pixels corresponding to the matched design pupil position when the exit pupil position matches the design pupil position. An AF pixel corresponding to a design pupil position that has not been corrected may be corrected by the correction method shown in FIGS. 9 to 11 and used for correlation calculation. Further, the correction method shown in FIG. 9 or 10 or the correction method shown in FIG. 11 may be adopted regardless of whether or not the difference between the exit pupil position and the designed pupil position is within a predetermined threshold.

  As described above, in the present embodiment, the image sensor is formed with a plurality of types of AF pixels each having a light receiving area set corresponding to a plurality of designed pupil positions. Thereby, even when the exit pupil position changes, the left and right pupil divisions can be performed equally, and the AF accuracy can be prevented from deteriorating. Further, by performing correction using the image signal of the AF pixel corresponding to the designed pupil position that is relatively close to the exit pupil position, the correction accuracy of the AF signal can be improved, and sufficient AF accuracy can be obtained. Is possible.

    DESCRIPTION OF SYMBOLS 10 ... Main body circuit part, 11 ... Signal processing and control part, 11a ... Signal extraction part, 11c ... Correction | amendment part, 11d ... Pupil position determination part, 11e ... AF signal calculation part, 12, 23 ... Communication part, 14 ... Imaging part , 14a ... light receiving part, 17 ... recording / reproducing part, 18 ... display part, 19 ... main body memory, 27 ... lens memory.

Claims (6)

An imaging device in which a plurality of pixels having imaging pixels and focus detection pixels are arranged in a matrix,
The pixel includes a light receiving region where light from a photographing lens is incident,
A microlens that guides light from the photographing lens to the light receiving region;
As the focus detection pixel, the light receiving area of the focus detection pixel is defined such that the position of one end of the area is defined in accordance with the positional relationship between the assumed design pupil position and the microlens. An image sensor. The image sensor according to claim 1,
A pupil position determination unit that determines which of the plurality of designed pupil positions the exit pupil position matches, and determines a focus detection pixel used for focus detection based on the determination result;
A focus detection unit that performs focus detection using the focus detection pixels used for the focus detection;
An imaging apparatus comprising: The image sensor according to claim 1,
A pupil position determination unit that determines a focus detection pixel used for focus detection by comparing the exit pupil position and the plurality of designed pupil positions;
When the exit pupil position is different from the designed pupil position, the image signal from the focus detection pixel used for the focus detection is used as the information about the exit pupil, the information about the designed pupil position, and the focus detection used for the focus detection. A correction unit that performs correction based on pixel information for obtaining a corrected image signal;
A focus detection unit that performs focus detection using the corrected image signal corrected by the correction unit;
An imaging apparatus comprising: The correction unit corrects an image signal from a focus detection pixel used for focus detection based on an uneven left / right pupil division ratio based on a difference between the exit pupil position and the designed pupil position. The imaging apparatus according to claim 3. The correction unit obtains a new corrected image signal by weighted addition of a plurality of corrected image signals for a plurality of image signals from a plurality of focus detection pixels corresponding to a plurality of designed pupil positions. The imaging device according to claim 3 or 4. A communication unit that acquires information on the exit pupil position and information on the design pupil position;
The image pickup apparatus according to claim 2, further comprising: a memory that stores information on a pixel for focus detection used for the focus detection.

Images