JP2008015353A - Imaging device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform a focus detecting operation even by a small number of focus detection pixels provided on an imaging surface. <P>SOLUTION: This imaging device is configured so that the group of imaging pixels forms an image signal of a received luminous flux transmitted through a photographing lens; the focus detection elements are provided in the group of the imaging pixels and each includes a pupil-formation microlens and a photoelectric conversion area; the pupil-formation microlens forms the real image of the exit pupil of the photographing lens per pixel unit; in the photoelectric conversion area, the rear image is pupil-divided and a pair of pupil division signals are photoelectrically converted; and a focus calculating part obtains a signal difference from the pair of pupil division signals and obtains the focusing state of the photographing lens from the obtained signal difference. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an imaging apparatus.

Conventionally, a pupil division phase difference method is known as one of focus detection techniques. Patent Document 1 is known as a solid-state imaging device to which the principle of the pupil division phase difference method is applied. In Patent Document 1, focus detection pixels are arranged on an imaging surface. Each of these focus detection pixels is provided with two photoelectric conversion areas. In accordance with the array pattern of the focus detection pixels, a light / dark pattern of the pupil divided image is detected. A pattern shift amount (phase difference) that minimizes the difference is detected while shifting this light / dark pattern in the pupil division direction. Corresponding to this pattern shift amount, the defocus amount of the photographing lens can be obtained.
Japanese Patent Laid-Open No. 2003-244712 (FIG. 2 etc.)

In the prior art of Patent Document 1 described above, a large number of focus detection pixels must be arranged on the imaging surface in order to satisfactorily detect the bright and dark pattern of the divided image.
Since many of these focus detection pixels are different from normal imaging pixels, it is difficult to obtain an image signal equivalent to the imaging pixels. For this reason, by disposing a large number of focus detection pixels on the imaging surface, the quality of the image signal to be captured is degraded.
Accordingly, an object of the present invention is to enable focus detection even with few focus detection pixels on the imaging surface.

<< 1 >> The imaging device of the present invention includes a group of imaging pixels, a focus detection pixel, and a focus calculation unit.
A plurality of groups of imaging pixels are provided on the imaging surface, and a received light beam transmitted through the photographing lens is photoelectrically converted pixel by pixel to generate an image signal.
The focus detection pixel is provided in the group of imaging pixels, and includes a pupil forming microlens and a photoelectric conversion area. This pupil forming microlens forms a real image of the exit pupil of the photographic lens from the received light beam in pixel units. On the other hand, in the photoelectric conversion area, a real image of the exit pupil is divided into pupils and each photoelectrically converted to generate a set of pupil division signals.
The focus calculation unit acquires a set of pupil division signals from the focus detection pixels, detects a signal difference, and estimates a focus adjustment state of the photographing lens according to the signal difference.
<< 2 >> Preferably, there is provided a changing unit that changes the optical distance between the photographic lens and the imaging surface. The focus calculation unit estimates the defocus (focus shift) amount or focus position of the photographing lens based on a change in signal difference accompanying a change in optical distance.
<< 3 >> Preferably, the focus calculation unit obtains a plurality of sample values of the signal difference as the optical distance varies. The focus calculation unit estimates the in-focus position where the signal difference becomes zero by performing an extension approximation or interpolation approximation on the sample values of the plurality of signal differences.
<< 4 >> Preferably, the focus calculation unit determines whether the photographing lens is in the rear pin state or the front pin state based on the increase / decrease direction of the signal difference accompanying the change in the optical distance.
<< 5 >> Also preferably, the photoelectric conversion area of the focus detection pixel divides the real image of the exit pupil into pupils concentrically. In the photoelectric conversion area, a photoelectric conversion signal in the central area of the real image and a photoelectric conversion signal in the peripheral area of the real image are generated and used as a set of pupil division signals.
<< 6 >> Preferably, a plurality of focus detection pixels are provided. The focus calculation unit calculates the sum of absolute values from the individual signal differences obtained from the plurality of focus detection pixels, and obtains the signal difference.
<< 7 >> Preferably, a focus control unit that focuses the photographic lens by controlling the focus of the photographic lens in a direction to reduce the signal difference detected by the focus calculation unit is provided.

  In the present invention, the focus adjustment state is estimated based on the signal difference (level difference) between a set of pupil division signals. In this case, if at least one focus detection pixel is on the imaging surface, the focus adjustment state can be estimated. Therefore, the number of focus detection pixels arranged on the imaging surface can be appropriately reduced as compared with the conventional example in which a large number of focus detection pixels for phase difference detection are arranged. As a result, the image quality of the captured image can be improved.

<< First Embodiment >>
[Description of electronic camera configuration]
FIG. 1 is a block diagram showing an electronic camera 10 of the present embodiment.
In FIG. 1, a photographing lens 12 is attached to the electronic camera 10. The photographing lens 12 is driven at a focus adjustment position by a focus control unit 12a. In the image space of the photographic lens 12, the imaging surface of the solid-state imaging device 11 is arranged. This imaging surface is slightly displaced in the optical axis direction by a variable portion 11a such as a piezo element.
The imaging control unit 14 reads the image signal by controlling the solid-state imaging device 11. The image signal is processed through the signal processing unit 15 and the A / D conversion unit 16 and then temporarily stored in the memory 17.
This memory 17 is connected to a bus 18. The bus 18 is also connected with a variation unit 11a, a focus control unit 12a, an imaging control unit 14, a microprocessor 19, a focus calculation unit 20, a recording unit 22, an image compression unit 24, an image processing unit 25, and the like. The focus calculation unit 20 may be realized by software using the microprocessor 19.
The microprocessor 19 is connected to an operation unit 19a such as a release button. A recording medium 22a is detachably attached to the recording unit 22.

[Description of pixel layout]
FIG. 2 is a diagram illustrating a pixel layout of the solid-state imaging device 11.
A group of imaging pixels 31 is arranged on the imaging surface of the solid-state imaging device 11. Each imaging pixel 31 is provided with a photoelectric conversion area 32 that photoelectrically converts a received light beam in units of pixels. A microlens 33 for condensing the received light beam on the photoelectric conversion area 32 is provided on the upper layer of the photoelectric conversion area 32 through a planarization layer. The group of imaging pixels 31 generates an image signal by photoelectrically converting a subject image projected onto the imaging surface via the imaging lens 12 in units of pixels.
A plurality of focus detection areas Z are arranged on the imaging surface. In this focus detection area Z, focus detection pixels 37 are arranged in a checkered pattern so as to sew between groups of imaging pixels 31.
A pupil forming microlens 36 is formed in the focus detection pixel 37. The pupil forming microlens 36 condenses the received light beam in pixel units to form a real image of the exit pupil of the photographing lens 12.
In the focus detection pixel 37, two photoelectric conversion areas 34a and 34b are arranged in order to divide the real image of the exit pupil concentrically. The photoelectric conversion area 34a generates a photoelectric conversion signal in the peripheral area of the real image. On the other hand, the photoelectric conversion area 34b generates a photoelectric conversion signal in the central area of the real image.

[Circuit explanation]
FIG. 3 is a diagram illustrating an equivalent circuit of the solid-state imaging device 11.
The solid-state imaging device 11 is roughly composed of a vertical transfer circuit 3, a horizontal transfer circuit 4, a correlated double sampling circuit 5, a group of imaging pixels 31, and a group of focus detection pixels 37.
First, the circuit configuration of the imaging pixel 31 will be described. The imaging pixel 31 is provided with a floating diffusion FD. This floating diffusion FD is connected to the power supply line VDD via a reset transistor QR. A transfer transistor QT is disposed between the floating diffusion FD and the photoelectric conversion area 32. A control signal φTGa is supplied from the vertical transfer circuit 3 to the gate of the transfer transistor QT.

  The voltage of the floating diffusion FD is applied to the gate of the amplifying element QA. The source of the amplifying element QA is connected to the vertical readout line 2 by turning on the row selection transistor QS. The current source Is is connected to the source of the amplification element QA via the row selection transistor QS, so that the amplification element QA constitutes a source follower circuit. As a result, a source follower voltage corresponding to the voltage of the floating diffusion FD is output to the vertical readout line 2.

Next, structural features of the focus detection pixel 37 will be described. The focus detection pixel 37 is provided with photoelectric conversion areas 34 a and 34 b that are concentrically divided instead of the photoelectric conversion area 32 of the imaging pixel 31.
The photoelectric conversion area 34a is connected to the floating diffusion FD via the transfer transistor QTa. A control signal φTGa is supplied from the vertical transfer circuit 3 to the gate of the transfer transistor QTa.

The other photoelectric conversion region 34b is connected to the floating diffusion FD via the transfer transistor QTb. A control signal φTGb is supplied from the vertical transfer circuit 3 to the gate of the transfer transistor QTb.
Note that the photoelectric conversion area 34b is surrounded by a doughnut-shaped photoelectric conversion area 34a. This makes it difficult to connect the photoelectric conversion area 34b to the peripheral circuit (transfer transistor QTb or the like) described above. In such a case, the photoelectric conversion area 34a may be formed in a C shape, and the photoelectric conversion area 34b and the peripheral circuit may be connected through the gap.

[Optical Action of Focus Detection Pixel 37]
FIG. 4 is a diagram for explaining the optical action of the focus detection pixel 37.
In the focus detection pixel 37, a pair of photoelectric conversion areas 34a and 34b are arranged concentrically. As shown in FIG. 4, a light beam that has passed through the periphery of the exit pupil of the photographic lens 12 enters the photoelectric conversion area 34a. On the other hand, as shown in FIG. 4, the light beam that has passed through the center of the exit pupil of the photographing lens 12 enters the photoelectric conversion area 34b. The set of photoelectric conversion areas 34a and 34b photoelectrically convert the respective received light beams to generate a set of pupil division signals Sa and Sb.

FIG. 5 is a diagram illustrating a state of a received light beam incident on a pair of photoelectric conversion areas 34a and 34b.
In the in-focus state shown in FIG. 5A, the received light beam incident on the pair of photoelectric conversion areas 34a and 34b is a light beam emitted from the same location of the focused subject. Accordingly, the pair of pupil division signals Sa and Sb have the same signal level if the light receiving efficiency of the photoelectric conversion areas 34a and 34b is gain-corrected. That is, if the correction coefficients of the light receiving efficiency of the photoelectric conversion areas 34a and 34b are α and β,
In-focus signal difference D = | αSa−βSb | = 0
It becomes.

On the other hand, in the front pin state shown in FIG. 5B, the received light beams incident on the pair of photoelectric conversion areas 34a and 34b intersect in front of the subject. Therefore, as shown in FIG. 5B, light beams emitted from different concentric areas on the subject side are incident on the pair of photoelectric conversion areas 34a and 34b, respectively. If the subject has an appropriate luminance pattern, the light intensities of these light beams emitted from different regions are different. Therefore, the set of pupil division signals Sa and Sb have different signal levels. That is,
Signal difference D of front pin state D = | αSa−βSb |> 0
It becomes.

In the rear pin state shown in FIG. 5C, the received light beams incident on the set of photoelectric conversion areas 34a and 34b virtually intersect behind the subject. Therefore, as shown in FIG. 5C, light beams emitted from different concentric areas on the subject side are incident on the pair of photoelectric conversion areas 34a and 34b, respectively. If the subject has an appropriate luminance pattern, the light intensities of these light beams emitted from different regions are different. Therefore, the set of pupil division signals Sa and Sb have different signal levels. That is,
Rear pin state signal difference D = | αSa−βSb |> 0
It becomes.
It is possible to estimate the focus adjustment state of the photographic lens 12 by using such an optical action of the focus detection pixel 37.

[Description of focus detection operation]
FIG. 6 is a flowchart for explaining the operation of the electronic camera 10. Hereinafter, this operation will be described in the order of the step numbers shown in FIG.

Step S1: First, the focus detection area Z is selected by a known function (manual selection, automatic selection, etc.) on the electronic camera 10 side.

Step S2: The microprocessor 19 instructs the solid-state imaging device 11 to read out the pupil division signal from the selected area via the imaging control unit 14 in synchronization with the half-press operation of the release button.
In accordance with this instruction, the vertical transfer circuit 3 in the solid-state imaging device 11 controls the control signal φRS (n), the control signal φTGa (n), and the control signal φTGb (n) in the nth row where the focus detection pixels 37 in the selected area exist. ) And launch. Thus, the pair of photoelectric conversion areas 34a and 34b is reset via the transfer transistor QTa, the floating diffusion FD, and the reset transistor QR.

Thereafter, the vertical transfer circuit 3 causes the control signal φTGa (n) and the control signal φTGb (n) in the n-th row to fall, and changes the transfer transistors QTa and QTb to non-conduction. From this point of time, the pair of photoelectric conversion areas 34a and 34b starts to accumulate signal charges.
The potential of the n-th row floating diffusion FD is reset by continuing the conduction state of the n-th row reset transistor QR. After this reset, the vertical transfer circuit 3 changes the reset transistor QR to non-conduction. As a result, the floating diffusion FD returns to the floating state and holds the voltage (reset voltage) at that moment.
In this state, the vertical transfer circuit 3 sets the control signal φL (n) to a high level, and turns on the row selection transistor QS of the nth row. As a result, the reset voltage in the n-th row is output as a source follower to the vertical readout line 2 via the amplifier element QA.
The reset voltage output to the vertical readout line 2 is held in the correlated double sampling circuit 5 (capacitor group in the circuit) by the fall of the control signal φSH.

Next, the vertical transfer circuit 3 conducts the transfer transistor QTa in the n-th row using the control signal φTGa (n). By this conduction, the signal charge accumulated in the photoelectric conversion area 34a connected to the transfer transistor QTa is transferred to the floating diffusion FD. Along with this transfer operation, the voltage of the floating diffusion FD changes relative to the reset voltage by the amount of signal charge transferred. The signal voltage of the nth row is output as a source follower to the vertical readout line 2 via the amplifying element QA.
Thus, the signal voltage of the photoelectric conversion area 34 a read out via the vertical readout line 2 is applied to the correlated double sampling circuit 5. The correlated double sampling circuit 5 outputs a pupil division signal Sa corresponding to the difference between the signal voltage and the reset voltage.
In this state, the horizontal transfer circuit 4 sequentially reads the pupil division signal Sa of the photoelectric conversion area 34a connected to the transfer transistor QTa from Vout using the control signal φH1 of the column in which the focus detection pixels 37 are present. .
Subsequently, the readout operation described above is repeated by replacing the control signal φTGa (n) with the control signal φTGb (n), whereby the pupil division signal Sb of the photoelectric conversion area 34b is also sequentially read from Vout.

By repeating the above operation for only the focus detection pixels 37 in the selected area, it is possible to read out the pupil division signals Sa and Sb in a short time without reading out all the pixels.
The pupil division signals Sa and Sb read in this way are digitized via the signal processing unit 15 and the A / D conversion unit 16 and then temporarily stored in the memory 17.

Step S3: The focus calculation unit 20 reads a set of pupil division signals Sa and Sb from the memory 17 and obtains the following signal difference D for each focus detection pixel 37.
Signal difference D = | αSa−βSb |
The correction coefficients α and β are coefficients corresponding to the ratio of the light receiving efficiencies of the photoelectric conversion areas 34a and 34b, and are set in advance to coefficient values at which the signal difference D becomes zero in the focused state.

Step S4: When there are a plurality of focus detection pixels 37 in the selected area, the focus calculation unit 20 calculates the sum of the absolute values of the signal differences obtained in Step S3 and sets the signal difference D. Instead of taking the sum, an average value, an intermediate value, a maximum value, or the like may be obtained and used as the signal difference D.

Step S5: The focus calculation unit 20 determines whether or not the signal difference D falls within the focus allowable value Dth. This in-focus tolerance value Dth defines an upper limit value of the signal difference D that can be substantially regarded as an in-focus state, and is a value determined in advance from an actual measurement result or the like.
Here, when the signal difference D falls within the allowable focus value Dth, the focus calculation unit 20 determines that the photographing lens 12 is in focus and shifts the operation to step S6.
On the other hand, when the signal difference D exceeds the in-focus allowable value Dth, the focus calculation unit 20 determines that the photographing lens 12 is in an out-of-focus state, and moves the operation to step S7.

Step S6: The microprocessor 19 in the electronic camera 10 instructs the solid-state imaging device 11 to read out the image signal via the imaging control unit 14 in synchronization with the full pressing operation of the release button. The read image signal is temporarily stored in the memory 17 via the signal processing unit 15 and the A / D conversion unit 16.
Note that image signals are missing at the locations where the focus detection pixels 37 are arranged. This missing image signal can be interpolated using surrounding image signals. Further, the image signal of the missing portion may be generated using the signals of the photoelectric conversion areas 34 a and 34 b of the focus detection pixel 37.
The image signal stored in the memory 17 is processed through the image processing unit 25 and the image compression unit 24. The processed image signal is recorded and stored in the recording medium 22 a via the recording unit 22. With the above operation, the imaging operation of the electronic camera 10 is completed.

Step S7: The microprocessor 19 drives the changing unit 11a to displace the imaging surface of the solid-state imaging device 11 in the optical axis direction. In this displaced state, the electronic camera 10 performs the same processing as steps S <b> 2 to S <b> 4 to obtain a change in the signal difference D.

Step S8: The focus calculation unit 20 determines whether the photographing lens 12 is in the rear pin / front pin state based on the increase / decrease direction of the signal difference D accompanying the displacement of the imaging surface.
FIG. 7 is an explanatory diagram showing the relationship between the change in the signal difference D and the focus adjustment state of the photographic lens 12.
When the signal difference D decreases as the optical distance between the photographic lens 12 and the imaging surface increases, the in-focus position exists in the direction in which the optical distance is increased. That is, it can be determined that the photographic lens 12 is in the rear pin state.
Conversely, when the signal difference D increases with an increase in the optical distance between the photographic lens 12 and the imaging surface, the in-focus position exists in the direction in which the optical distance decreases. That is, it can be determined that the photographic lens 12 is in the front pin state.

Step S9: The focus calculation unit 20 estimates the defocus amount or the focus position of the photographic lens 12 based on the change in the signal difference D accompanying the displacement of the imaging surface.
For example, in the relationship as shown in FIG. 7, the larger the change in the signal difference D, the farther from the in-focus position, the greater the defocus amount can be estimated. Conversely, it can be estimated that the smaller the change in the signal difference D, the closer to the in-focus position and the smaller the defocus amount. It is possible to estimate the defocus amount from the signal difference D by presetting the correspondence between the change amount of the signal difference D and the defocus amount (or in-focus position) through an experiment.
For example, the focus calculation unit 20 obtains a plurality of sample values of the signal difference D with the displacement of the imaging surface. The focus calculation unit 20 obtains an approximation function as shown in FIG. 7 by performing extension approximation or interpolation approximation on the sample values of the plurality of signal differences D. In this case, as the approximation function, either a linear function or a curve function can be used. The focus calculation unit 20 estimates the in-focus position where the signal difference D is zero based on this approximate function. The focus calculation unit 20 obtains the defocus amount of the photographing lens 12 based on the estimation result of the focus position.

Step S10: The focus calculation unit 20 outputs the focus adjustment state of the photographing lens 12 obtained in steps S5, S8, and S9 to the microprocessor 19.
The microprocessor 19 performs focus control of the photographing lens 12 via the focus control unit 12a according to the focus adjustment state.

  For example, the focus control unit 12a scans the focus adjustment position of the photographing lens 12 between the closest side and the infinity side, and focuses so that the signal difference D sequentially detected by the focus calculation unit 20 is set to near zero. Control may be implemented.

  For example, the focus control unit 12a may determine the driving direction of the focus adjustment position of the photographic lens 12 according to the determination result of the front pin / rear pin determined in step S8. This operation can speed up the focus control operation.

For example, the focus control unit 12a may determine a target position for focus adjustment of the photographing lens 12 according to the defocus amount or the focus position obtained in step S9. This operation can further speed up the focus control operation.
After such a focus control operation, if it is necessary to make another in-focus determination, the microprocessor 19 shifts the operation to step S2.
Note that if the focus determination has already been made in the focus control process in step S10, the operation may be shifted directly to the imaging operation in step S6.

[Effects of First Embodiment]
As described above, in the first embodiment, the focus adjustment state is determined based on the signal difference (level difference) between a set of pupil division signals. This focus detection is different from the conventional pupil division phase difference method. Here, in order to express the difference clearly, the method of the present invention is referred to as a pupil division level difference method. In this pupil division level difference method, it is possible to detect the focus adjustment state by arranging at least one focus detection pixel 37 on the imaging surface. Therefore, unlike the conventional example in which a large number of focus detection pixels are arranged in order to obtain a bright and dark pattern of the pupil-divided image, the number of focus detection pixels on the imaging surface can be significantly reduced. Therefore, it is possible to suppress the deterioration of the quality of the captured image.

  Furthermore, in the first embodiment, the optical distance between the photographic lens 12 and the imaging surface is changed using the changing unit 11a. The focus calculation unit 20 estimates the defocus (focus shift) amount and focus position of the photographic lens 12 based on the change in the signal difference D accompanying this distance variation.

  In particular, in the first embodiment, the focus calculation unit 20 obtains a plurality of sample values of the signal difference D as the optical distance varies. The focus calculation unit 20 performs the extension approximation or the interpolation approximation on the sample value of the signal difference D. This approximate calculation makes it possible to accurately estimate the in-focus position where the signal difference D is near zero.

  In the first embodiment, the rear pin / front pin of the photographing lens 12 can be accurately determined based on the increase / decrease direction of the signal difference D accompanying the change in the optical distance.

  Incidentally, in the first embodiment, concentric photoelectric conversion areas 34 a and 34 b are provided in the focus detection pixel 37. This concentric pupil division has a feature that there is no direction dependency in the pupil division direction. For this reason, focus detection can be performed on subject edges in all directions without being limited to specific edge directions such as vertical edges and horizontal edges.

  Furthermore, this concentric pupil division has an excellent feature that there is almost no false detection of focus by a periodic subject. For example, the signal difference D is almost zero in the vertical stripe periodic subject shown in FIG. 5 only in the in-focus state of FIG. 5A. In the out-of-focus state of FIGS. 5B and 5C, there is no singular point where the signal difference D becomes zero.

  In the first embodiment, a plurality of focus detection pixels 37 are arranged in one focus detection area. An absolute value sum of the signal differences obtained for each of these focus detection pixels 37 is calculated to obtain a signal difference D. In this case, focus detection is possible if a change in brightness (such as an edge) of the subject is projected onto any one focus detection pixel 37 in the selected area. Therefore, the certainty of focus detection can be further improved.

  Furthermore, by taking the sum of absolute values of a plurality of signal differences, it is possible to reduce the influence of noise of the individual pupil division signals Sa and Sb. As a result, accurate focus detection is possible even in a situation where noise increases, such as in a dark shooting environment.

<< Second Embodiment >>
FIGS. 8A and 8B are diagrams illustrating an example of arrangement of photoelectric conversion areas of focus detection pixels. Other configurations are the same as those in the first embodiment, and thus redundant description is omitted here.
In FIG. 8A, the focus detection pixel 57 includes a pair of photoelectric conversion areas 54a and 54b. The pair of photoelectric conversion areas 54a and 54b are arranged so as to divide the real image of the exit pupil symmetrically in the pupil division direction (vertical, horizontal, diagonal, etc.). Note that a plurality of focus detection pixels 57 having different pupil division directions are provided in one focus detection area. With this configuration, focus detection can be performed on subject edges in various directions.

  On the other hand, in FIG. 8B, the focus detection pixel 77 is provided with photoelectric conversion areas 74a and 74b each having a donut divided into two. The pair of photoelectric conversion areas 74a and 74b are arranged so as to divide the real image of the exit pupil symmetrically in the pupil division direction (vertical, horizontal, diagonal, etc.). A plurality of focus detection pixels 77 with different pupil division directions are provided in one focus detection area. With this configuration, focus detection can be performed on subject edges in various directions.

<< Additional items of embodiment >>
In the above-described embodiment, the effect that the focus detection pixels can be roughly arranged on the imaging surface is described. However, the present invention is not limited to this. If necessary, focus detection pixels may be densely arranged on the imaging surface as in the conventional case.

  In the above-described embodiment, the case where the focus detection pixels are arranged in a checkered pattern in the focus detection area has been described. However, the present invention is not limited to this. For example, in a color image sensor such as a Bayer array, it is preferable to disperse the focus detection pixels appropriately so that the same color component is not lost intensively from the image signal.

  In the above-described embodiment, the case where the signal difference D is an even-order convergence curve (see FIG. 7) has been described. However, the present invention is not limited to this. For example, as shown in FIG. 9, even when the signal difference D indicates an odd-order convergence curve, it is possible to determine the rear pin / front pin and to roughly estimate the defocus amount. Further, by approximating the sample value of the signal difference D using an odd-order function, it is possible to estimate the in-focus position (defocus amount) and the in-focus position.

  In the above-described embodiment, the imaging pixels and the focus detection pixels have the same size. However, the present invention is not limited to this. For example, one focus detection pixel may have a pixel size corresponding to a plurality of imaging pixels. In this case, the light detection efficiency of the focus detection pixels can be increased, and the focus detection accuracy can be improved. Further, the size of one focus detection pixel may be reduced so as to fit within the gap in the arrangement of the imaging pixels.

  In the above-described embodiment, the case of the solid-state imaging device of the XY address method (CMOS type or the like) has been described. However, the present invention is not limited to this. The present invention may be applied to a CCD type solid-state imaging device or the like.

  Further, in the above-described embodiment, the solid-state imaging device 11 and the focus calculation unit 20 are provided separately. However, the present invention is not limited to this. For example, the focus calculation unit 20 may be a built-in circuit of the solid-state imaging device 11. In this case, the focus calculation unit 20 can be configured as an analog circuit such as a difference circuit or an absolute value circuit. Further, the focus calculation unit 20 may be configured as a digital circuit.

  In the above-described embodiment, the optical distance between the imaging lens and the imaging surface is changed by displacing the imaging surface in the optical axis direction by the changing unit. However, the present invention is not limited to this. For example, the optical distance between the photographic lens and the imaging surface may be changed by driving the focus adjustment position of the imaging lens. Further, for example, the optical distance between the photographing lens and the imaging surface may be changed by inserting and removing optical members having different refractive indexes on the optical path between the photographing lens and the imaging surface.

  In the above-described embodiment, the pupil forming microlens 36 is constituted by a single-layer lens. However, the present invention is not limited to this. The pupil forming microlens 36 may be composed of a plurality of layers of lenses. In this case, it is possible to improve the imaging performance of the real image of the exit pupil by the pupil forming microlens 36, and the focus detection accuracy can be further improved.

  In the first embodiment described above, the outer shapes of the photoelectric conversion areas 34a and 34b are formed in a circular shape. However, the present invention is not limited to this. The photoelectric conversion areas 34a and 34b may be concentric, and are not limited to the outer shape. For example, the photoelectric conversion area 34b may be rectangular and the optical conversion area 34a may be formed in a mouth shape or a U shape.

  As described above, the present invention is a technique that can be used for an imaging apparatus having a focus detection function.

1 is a block diagram showing an electronic camera 10. FIG. 2 is a diagram illustrating a pixel layout of the solid-state imaging device 11. FIG. 2 is a diagram illustrating an equivalent circuit of the solid-state imaging device 11. FIG. It is a figure explaining the optical effect | action of the pixel 37 for focus detection. It is a figure which shows the mode of the received light beam which injects into a pair of photoelectric conversion area 34a, 34b. 3 is a flowchart for explaining the operation of the electronic camera 10. It is explanatory drawing which shows the relationship between the change of the signal difference D, and the focus adjustment state of the imaging lens 12. FIG. It is a figure which shows the example of arrangement | positioning of the photoelectric conversion area of the pixel for focus detection. It is explanatory drawing which shows the relationship between the change of the signal difference D, and the focus adjustment state of the imaging lens 12. FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Electronic camera, 11 ... Solid-state imaging device, 11a ... Change part, 12 ... Shooting lens, 12a ... Focus control part, 17 ... Memory, 19 ... Microprocessor, 20 ... Focus calculating part, 31 ... Imaging pixel, 32 ... Photoelectric conversion area 33 ... micro lens 34a ... photoelectric conversion area 34b ... photoelectric conversion area 36 ... pupil forming microlens 37 ... focus detection pixel 57 ... focus detection pixel 77 ... focus detection pixel

Claims (7)

A group of imaging pixels that are provided on the imaging surface and photoelectrically convert the received light flux that has passed through the imaging lens in units of pixels to generate an image signal;
Provided in the group of pixels for imaging, pupil forming microlens that forms a real image of the exit pupil of the photographic lens from a received light beam in pixel units, and a pupil split of the real image of the exit pupil, respectively, photoelectrically converted, A focus detection pixel having a photoelectric conversion region for generating a set of pupil division signals;
A focus calculation unit that acquires a set of the pupil division signals from the focus detection pixels, detects a signal difference, and estimates a focus adjustment state of the photographic lens according to the signal difference. An imaging device. The imaging apparatus according to claim 1,
A variable unit that varies the optical distance between the imaging lens and the imaging surface;
The focus calculation unit estimates an amount of defocus (focus shift) or a focus position of the photographing lens based on a change in the signal difference accompanying a change in the optical distance. The imaging device according to claim 3.
The focus calculation unit obtains a plurality of sample values of the signal difference as the optical distance varies, and an in-focus position at which the signal difference becomes zero by extending or interpolating the sample values of the plurality of signal differences. An imaging apparatus characterized by estimating In the imaging device according to claim 2 or 3,
The imaging apparatus according to claim 1, wherein the focus calculation unit determines whether the photographing lens is in a rear pin state or a front pin state based on an increase / decrease direction of the signal difference accompanying the change in the optical distance. The imaging apparatus according to any one of claims 1 to 4,
The photoelectric conversion area of the focus detection pixel is:
The real image of the exit pupil is concentrically divided into pupils, and a photoelectric conversion signal in a central region of the real image and a photoelectric conversion signal in a peripheral region of the real image are generated to form a set of pupil division signals. Imaging device. The imaging device according to any one of claims 1 to 5,
A plurality of the focus detection pixels are provided,
The focus calculation unit calculates an absolute value sum from individual signal differences obtained from the plurality of focus detection pixels, and sets the difference as the signal difference. The imaging device according to any one of claims 1 to 6,
An imaging apparatus comprising: a focus control unit that focuses the photographic lens by controlling the focus of the photographic lens in a direction to reduce the signal difference detected by the focus calculation unit.

Images