JP2016206441A - Control device, imaging device, control method, program, and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize high-speed focus control while speeding up display update of a display device.SOLUTION: The control device comprises: acquisition means for acquiring a first image signal for forming a photographic image, which is output by adding a signal from a plurality of photoelectric conversion units in each of a plurality of first pixels included in a first image region, and acquiring a second image signal which is output mutually independently of the plurality of photoelectric conversion units in each of a plurality of second pixels included in a second image region; first computation means for performing correlation operation using a phase difference system on the basis of a second pixel signal and determining correlation data; second computation means for performing shift addition of adding the second pixel signal while performing shift for the plurality of photoelectric conversion units corresponding to the respective second pixels between the second pixels and determining addition data; and control means for performing focus control on the basis of the correlation data and the addition data.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a control device that performs focus control of an interchangeable lens mounted on an imaging device.

  An imaging device such as an interchangeable lens digital camera is equipped with a focus control device that performs focus control (autofocus: AF) of the interchangeable lens, but various aberrations inherent to the interchangeable lens are related to the accuracy of focus control, that is, focusing. May affect accuracy.

  In Patent Document 1, a combination of focus detection pixels that are discretely arranged between a plurality of imaging pixels is switched between a case where signals of each pixel of an imaging element are read out and a case where reading is performed without thinning out. An imaging device is disclosed. As a result, the image data read out by thinning out the signals of all the pixels does not include the focus detection pixels, and the adverse effect that the focus detection by the phase difference method cannot be performed can be solved.

  Patent Document 2 discloses an automatic predetermined detection apparatus having a mode for reading signals at different thinning rates or different thinning phases for reading out focus detection pixels discretely arranged between a plurality of imaging pixels. Has been. In this apparatus, optimum exposure is realized by switching modes according to the state of the imaging apparatus, and deterioration of image quality is suppressed.

JP 2004-347665 A JP 2010-181751 A JP 2012-252280 A Japanese Patent No. 4235422 Japanese Patent No. 4185541

  However, Patent Documents 1 and 2 are related to increasing the speed of live view display in an image sensor having a plurality of photoelectric conversion units in one pixel and correcting a focus shift mainly caused by optical characteristics of a photographing lens. Not disclosed. Moreover, in the automatic focus detection apparatus in Patent Document 2, it takes time to change modes and change the exposure associated with the switching.

  A control device according to one aspect of the present invention performs focus control using a signal from an imaging device that includes a plurality of photoelectric conversion units for one pixel and the pixels are two-dimensionally arranged. The control device adds a signal from a plurality of photoelectric conversion units to each of a plurality of first pixels included in the first pixel region, and outputs the first image signal for forming a captured image. Acquisition means for acquiring second image signals output independently from each other from the plurality of photoelectric conversion units in each of the plurality of second pixels included in the second pixel region, and the second pixel A plurality of photoelectric conversions corresponding to each of the plurality of second pixels between a plurality of second pixels and a first calculation means for performing correlation calculation by a phase difference method based on the signal and determining correlation data; Shift addition for adding the second pixel signals while shifting each other with respect to each other, and a second calculation means for determining the addition data, and a control means for performing focus control based on the correlation data and the addition data. Having And features.

A control method according to another aspect of the present invention performs focus control using a signal from an image sensor that has a plurality of photoelectric conversion units for one pixel and the pixels are arranged two-dimensionally. In the control method, a first image signal for forming a captured image output by adding signals from a plurality of photoelectric conversion units in each of a plurality of first pixels included in a first pixel region. And obtaining a second image signal output independently from each other from the plurality of photoelectric conversion units in each of the plurality of second pixels included in the second pixel region, And performing correlation calculation using a phase difference method to determine correlation data, and shifting between the plurality of second pixels with respect to the plurality of photoelectric conversion units corresponding to each of the plurality of second pixels. Performing shift addition for adding the second pixel signal to determine addition data; performing focus control based on the correlation data and the addition data;
It is characterized by having.

  In the present invention, in an image sensor having a plurality of photoelectric conversion units in one pixel, high-speed focus control is realized while updating the display in the display device at high speed.

It is a block diagram of the imaging device in Example 1 and 2 of this invention. FIG. 6 is an explanatory diagram of a one-dimensional refocus process using an A image signal and a B image signal acquired by an image sensor in Examples 1 and 2. FIG. 3 is a schematic diagram of an image sensor included in the imaging device according to the first and second embodiments. FIG. 3 is a schematic diagram illustrating a structure of each pixel constituting the image sensor in Examples 1 and 2. 6 is a timing chart showing the operation of the vertical scanning circuit during normal operation in Examples 1 and 2. 3 is a flowchart illustrating an operation in the first embodiment. 3 is a timing chart illustrating an operation of a vertical scanning circuit when performing phase difference AF in Embodiment 1. 3 is a timing chart illustrating an operation of a vertical scanning circuit when performing scan AF in Embodiment 1. 6 is a flowchart illustrating an operation of phase difference AF processing in the first embodiment. It is a figure explaining contrast AF in Example 1. FIG. 6 is a flowchart illustrating an operation of RFAF processing in the first embodiment. FIG. 6 is a flowchart illustrating an operation of calculating a BP correction value interpolation coefficient in the first embodiment. It is a figure which shows the position which acquires BP correction value in Example 1. FIG. 6 is a flowchart illustrating an operation for determining a BP correction value in the first embodiment. 10 is a flowchart illustrating an operation in the second embodiment. FIG. 6 is a diagram illustrating the structure of an image sensor in Example 2. 10 is a flowchart illustrating operations of phase difference AF processing, RFAF processing, and lens driving in the second embodiment.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 shows a configuration of a digital camera (hereinafter simply referred to as a camera) 1 as an imaging apparatus according to an embodiment of the present invention.

  The camera 1 is provided with a photographic lens barrel 100 including a zoom lens group 2, a focus lens group 3, and a diaphragm 4. Reference numeral 5 denotes a solid-state imaging device (hereinafter referred to as an image sensor) that converts light collected by the imaging optical system in the taking lens barrel 100 into an electrical signal. Reference numeral 6 denotes an imaging circuit (acquisition means) that performs various signal processing on the electrical signal output from the image sensor 5 to generate an analog image signal. Reference numeral 7 denotes an A / D conversion circuit that converts an analog image signal generated by the imaging circuit 6 into a digital image signal. Reference numeral 8 denotes a buffer memory that temporarily stores the digital image signal output from the A / D conversion circuit 7. Or the like. Reference numeral 9 denotes a D / A conversion circuit that reads out a digital image signal stored in the VRAM 8 and converts it into an analog signal and converts it into a display image signal in a form suitable for display. Reference numeral 10 denotes an image display unit (hereinafter referred to as LCD) configured by a liquid crystal display element or the like that displays a display image signal from the D / A conversion circuit 9 as a live view image.

  An image memory 12 includes a semiconductor memory and stores image data. Reference numeral 11 denotes a compression / decompression circuit that performs compression processing and encoding processing for reading an image signal temporarily stored in the VRAM 8 and converting it into recording image data in a form suitable for storage in the image memory 12. . The compression / decompression circuit 11 performs a decoding process and an expansion process for converting the recording image data stored in the image memory 12 into a reproduction image signal in a form suitable for reproduction and display. As the image memory 12, not only a fixed type semiconductor memory but also a card type or stick type semiconductor memory that is detachable from the camera 1, an optical disk, a hard disk, and other various storage media can be used. .

  Reference numeral 13 denotes an AE processing circuit that performs automatic exposure (AE) processing using the digital image signal output from the A / D conversion circuit 7. A scan AF processing circuit 14 generates an AF evaluation value for performing a scan AF process (contrast AF) using the digital image signal output from the A / D conversion circuit 7.

  Reference numeral 15 denotes a CPU (control means) that controls the entire camera 1 and incorporates a memory for calculation. The CPU 15 constitutes a focus detection unit together with the scan AF processing circuit 14, the phase difference AF processing circuit 37 and the RFAF processing circuit 31.

  Reference numeral 16 denotes a timing generator (hereinafter referred to as TG) that generates a timing signal, and reference numeral 17 denotes a sensor driver that drives the image sensor 5.

  Reference numeral 21 denotes an aperture drive motor that drives the aperture 4, and reference numeral 18 denotes a first motor drive circuit that drives and controls the aperture drive motor 21. Reference numeral 22 denotes a focus drive motor that drives the focus lens group 3, and reference numeral 19 denotes a second motor drive circuit that drives and controls the focus drive motor 22. Reference numeral 23 denotes a zoom drive motor that drives the zoom lens group 2, and reference numeral 20 denotes a third motor drive circuit that drives and controls the zoom drive motor 23.

  An operation switch unit 24 is a main switch for turning on / off the power of the camera 1, an operation switch unit for starting AE / AF processing and shooting / recording processing, and a playback switch for starting playback operation of a shot image. Etc. The operation switch unit for starting the AE / AF process and the photographing / recording process includes a two-stage switch including a first stroke switch (hereinafter referred to as SW1) and a second stroke switch (hereinafter referred to as SW2). The first stroke switch generates an instruction signal for starting the AE / AF processing, and the second stroke switch generates an instruction signal (recording instruction) for starting the photographing / recording operation.

  A camera EEPROM 25 stores various control computer programs, various operation data, and the like in a rewritable manner. 26 is a battery.

  Reference numeral 28 denotes a strobe light emitting unit. Reference numeral 27 denotes a switching circuit for controlling the flash emission of the strobe light emitting unit 28, and reference numeral 29 denotes a display element constituted by an LED or the like for performing a warning display. Reference numeral 30 denotes a speaker for voice guidance and warning.

  31 calculates the sum (shift addition) of both while relatively shifting the reference signal (B image signal described later) with respect to the reference signal (A image signal described later), and calculates the relative shift amount and the sum of the two. This is an RFAF processing circuit for recording the signal.

  Reference numeral 33 denotes an AF auxiliary light composed of a light source such as an LED that illuminates all or a part of the subject when obtaining an AF evaluation value, and reference numeral 32 denotes an AF auxiliary light driving circuit for driving the AF auxiliary light 33. is there.

  Reference numeral 35 denotes a shake detection sensor that detects camera shake and the like. Reference numeral 34 denotes a shake detection circuit that processes a signal from the shake detection sensor 35. Reference numeral 36 denotes a face detection circuit that detects the position and size of a human face from the digital image signal from the A / D conversion circuit 7.

  37 generates a reference signal (A image signal) and a reference signal (B image signal) for performing phase difference AF processing using the digital image signal output from the A / D conversion circuit 7, and these A image signals And a phase difference AF processing circuit for performing a correlation operation on the B image signal.

  Reference numeral 38 denotes a first thermometer that is installed on the outer periphery of the camera 1 and measures the ambient temperature. Reference numeral 39 denotes a second thermometer that is installed inside the photographing lens barrel 100 and measures the temperature inside the photographing lens barrel 100.

  The operations of the camera 1 and the photographic lens barrel 100 configured as described above will be described below.

  When the light flux from the subject enters the imaging optical system of the taking lens barrel 100, the imaging optical system forms a subject image on the light flux. The image sensor 5 outputs an electrical signal generated by photoelectrically converting a subject image formed on the light receiving surface to the imaging circuit 6. The imaging circuit 6 performs various signal processing on the input electrical signal as described above to generate an analog image signal. The A / D conversion circuit 7 converts the analog image signal into a digital image signal, and then temporarily stores it in the VRAM 8. The image signal stored in the VRAM 8 is output to the D / A conversion circuit 9 and converted into an analog image signal, further converted into a display image signal, and displayed on the LCD 10 as a live view image.

  On the other hand, the image signal stored in the VRAM 8 is also output to the compression / decompression circuit 11. The compression / decompression circuit 11 converts the image signal into image data for recording through compression processing and encoding processing, and stores the image signal in the image memory 12. When the reproduction switch provided in the operation switch unit 24 is turned on, the recording image data stored in the compressed state in the image memory 12 is output to the compression / decompression circuit 11, and is decoded and decompressed to the VRAM 8. Recorded temporarily. Further, this image data is output to the D / A conversion circuit 9, converted into a reproduction image signal, and displayed on the LCD 10 as a reproduction image.

  In addition to the VRAM 8, the image signal digitized by the A / D conversion circuit 7 includes an AE processing circuit 13, a scan AF processing circuit 14, a phase difference AF processing circuit 37, an RFAF (refocus AF) processing circuit 31, Also output to the face detection circuit 36. The AE processing circuit 13 extracts luminance components from the input digital image signal, performs cumulative addition for one frame, and performs AE processing for generating an AE evaluation value signal indicating the brightness of the subject. The AE evaluation value signal is output to the CPU 15.

  The scan AF processing circuit 14 extracts a high-frequency component of the selected AF area from the input digital image signal using a high-pass filter (HPF), and performs cumulative addition for one frame. Thereby, an AF evaluation value signal indicating the contrast of the subject image in the AF area is generated. Since the contrast of the subject image corresponds to the focus state of the imaging optical system, generation of the AF evaluation value signal corresponds to detection of the focus state. Therefore, the AF evaluation value signal (or the AF evaluation value that is the value) is information obtained by detecting the focus state of the imaging optical system. The AF area may be one area of an arbitrary part of the shooting screen, or may be a plurality of areas including an arbitrary part and its surrounding area.

  The phase difference AF processing circuit 37 extracts an A image signal and a B image signal, which are a pair of image signals corresponding to the A image and the B image of the pair of subject images included in the AF area, from the input digital image signal. Then, a correlation calculation between the standard image signal (A image signal) and the reference image signal (B image signal) is performed to calculate an image shift amount (phase difference) between the A image and the B image.

  In this embodiment, pixels for phase difference AF are arranged on the imaging surface of the image sensor 5. Therefore, as in the case of performing phase difference detection using a secondary imaging optical system, a field lens that corrects image distortion caused by image height, a diaphragm for a light beam incident on a phase difference AF sensor, A mask that blocks unnecessary light flux cannot be used. For this reason, the signal for phase difference AF needs to be corrected because the shading offset differs for each pixel. Therefore, the phase difference AF processing circuit 37 has an image correction function and a function for obtaining an image shift amount by correlation calculation.

  The CPU 15 performs the phase difference AF by obtaining the defocus amount of the imaging optical system and the driving amount of the focus lens group 3 from the image shift amount calculated by the phase difference AF processing circuit 37.

  Here, the calculation of the defocus amount using the refocus method by the RFAF processing circuit 31 and the CPU 15 will be described with reference to FIG.

  FIG. 2 is an explanatory diagram of a refocusing process in a one-dimensional direction (row direction, horizontal direction) using a reference image signal (A image signal) and a reference image signal (B image signal) acquired by the image sensor 5 of the present embodiment. It is. In FIG. 2, i is an integer, and the A image signal of the i-th pixel in the row direction of the image sensor 5 disposed on the imaging surface 800 is schematically represented as Ai, and the B image signal is represented as Bi. The A image signal Ai is a light reception signal of a light beam incident on the i-th pixel at the principal ray angle θa. The B image signal Bi is a light reception signal of a light beam incident on the i-th pixel at the principal ray angle θb.

  The A image signal Ai and the B image signal Bi include not only light intensity distribution information but also incident angle information. Therefore, the A image signal Ai is translated along the angle θa to the virtual imaging plane 810, the B image signal Bi is translated along the angle θb to the virtual imaging plane 810, and these are added to obtain a virtual A refocus signal at the image plane 810 can be generated. Translating the A image signal Ai along the angle θa to the virtual imaging plane 810 corresponds to a +0.5 pixel shift in the row direction, and the B image signal Bi along the angle θb to the virtual imaging plane 810. Translating corresponds to a -0.5 pixel shift in the row direction. Accordingly, the A image signal Ai and the B image signal Bi are relatively shifted by +1 pixel, and the A image signal Ai and the B image signal (Bi + 1) are added in correspondence with each other, thereby refocusing on the virtual image plane 810. A signal can be generated.

  Similarly, by shifting the A image signal Ai and the B image signal Bi by an integer number of pixels and adding them, it is possible to generate a shift addition signal (sum signal sequence) on each virtual imaging plane according to the integer shift amount. .

  The RFAF processing circuit 31 creates a sum signal sequence by the above method, and records the sum signal sequence and the relative shift amount in a recording area in the circuit. When the generation and recording of the sum signal sequence is completed for the signal acquisition region on the image sensor 5 corresponding to the scan AF processing circuit 14 or the phase difference AF processing circuit 37, the sum signal sequence is input to the scan AF processing circuit 14 and AF evaluation is performed. A value signal is generated. The acquired AF evaluation value signal is associated with the relative shift amount and recorded in a recording area in the RFAF processing circuit.

  The CPU 15 obtains a relative shift amount at which the AF evaluation value signal of the sum signal becomes a maximum value, and obtains a focus correction position using the value. The focus correction position is a focus position finally obtained, and is a position in which an error in the position of the focus lens group 3 caused mainly by the optical characteristics of the photographing lens is corrected. For example, the position of the focus lens group 3 at which the AF evaluation value by the scan AF process is maximized or the position of the focus lens group 3 at which the defocus amount by the phase difference AF process is zero maximizes the resolution of the recorded image. It may be different from the position of the focus lens group 3. Hereinafter, the correction value for obtaining the focus correction position is referred to as a BP correction value.

  In other words, the difference between the position of the focus lens group 3 at which the AF evaluation value obtained by the scan AF process is maximized and the position of the focus lens group 3 at which the resolution of the recorded image is maximized is BP correction in scan AF. Value. Further, the difference between the position of the focus lens group 3 where the defocus amount obtained by the phase difference AF process is zero and the position of the focus lens group 3 where the resolution of the recorded image is maximum is the BP correction in the phase difference AF. Value.

  The face detection circuit 36 searches a part characterizing the face such as eyes and eyebrows on the image from the input digital image signal, and obtains the position of the person's face. Furthermore, the size and inclination of the face are obtained from the positional relationship of the parts characterizing the face.

  A timing signal from the TG 16 is output to the CPU 15, the imaging circuit 6, and the sensor driver 17. The CPU 15 performs various controls in synchronization with this timing signal. The imaging circuit 6 performs various signal processing such as color separation on the image signal from the image sensor 5 in synchronization with the timing signal. Further, the sensor driver 17 drives the image sensor 5 in synchronization with the timing signal.

  The CPU 15 sends instructions to the first motor drive circuit 18, the second motor drive circuit 19, and the third motor drive circuit 20, and the diaphragm drive motor 21, the focus drive motor 22, and the zoom drive motor 23 cause the diaphragm 4 and the focus lens to move. The group 3 and the zoom lens group 2 are driven. Specifically, the CPU 15 obtains an exposure time and an aperture value at which an appropriate exposure amount is obtained based on the AE evaluation value calculated by the AE processing circuit 13 and drives the aperture drive motor 21 by the first motor drive circuit 18. By doing so, the aperture amount of the aperture 4 is adjusted to be appropriate.

  Further, as the AF control, the CPU 15 drives the focus driving motor 22 by the second motor driving circuit 19 based on the processing results of the scan AF processing circuit 14, the phase difference AF processing circuit 37, and the RFAF processing circuit 31. Move group 3 to the in-focus position.

  When a zoom switch (not shown) in the operation switch unit 24 is operated, the CPU 15 moves the zoom lens group 2 by driving the zoom drive motor 23 by the third motor drive circuit 20, and the imaging optical system. Performs zooming operation.

  Next, the configuration of the pixels included in the image sensor 5 of FIG. 1 will be described with reference to FIG.

  A light receiving pixel 201 photoelectrically converts incident light from the taking lens barrel 100 and outputs it as an electrical signal. The light receiving pixel 201 includes a photodiode 202, a transfer transistor 203, a signal amplification amplifier 204, and a signal resetting transistor 205 as one unit (pixel). The transfer transistor 203 and the reset transistor 205 operate in accordance with a signal from the driving circuit 206 connected to each pixel. Here, the driving circuit 206 includes a shift register, a signal generation circuit that drives each pixel such as the transfer transistor 203, and the like. Then, the drive circuit 206 controls the transfer transistor 203 and the reset transistor 205 in accordance with the timing signals (TX1 to 4, RS1 to 4 and the like) generated by the TG 16, thereby resetting and reading the charge of the photodiode. Do. Thereby, the exposure time can be controlled.

  The light receiving pixels 201 having the above functions are two-dimensionally arranged on the image sensor 5 as shown in FIG. 4B, and serve as imaging pixels and phase difference AF pixels. FIG. 4A is a cross-sectional view of the image sensor 5, and two rectangular photodiodes 202 are arranged for one microlens 212. One photodiode is for the reference pixel and the other is for the reference pixel.

  In the image sensor 5, a row of phase difference AF pixels (second pixel region) and a row of display pixels (imaging pixels) for use in LCD display (first pixel region) are set every two rows. Has been. When performing LCD display or normal imaging, the image sensor 5 adds and reads the outputs of the pair of photodiodes. In order to match the number of displayable pixels of the LCD to be displayed and the number of pixels of the display image, it is also possible to add a plurality of pixel groups having the same color filter in the horizontal direction. Further, in the vertical direction, it is possible to add signals of rows having the same color filter arrangement or to thin out rows.

  When performing the phase difference AF, the reference pixel photodiode and the reference pixel photodiode are read out separately. With respect to the vertical direction, it is also possible to add the signals of the rows having the same color filter arrangement and to thin out the rows.

  A horizontal scanning circuit 209 includes a shift register, a column amplifier circuit 210, a signal output selection switch 211, and an output circuit (not shown) to the outside. The horizontal scanning circuit 209 can amplify the signal read from the pixel by changing the setting of the column amplifier circuit 210 according to the signal from the sensor driver 17.

  Next, a method for acquiring an image signal in the camera 1 will be described.

  FIG. 5 is a timing chart of signals generated from the driving circuit 206 when a normal image is acquired. The image sensor 5 performs exposure and signal reading in accordance with a vertical synchronization signal generated by the sensor driver 17. At this time, the charge of the photodiode 202 is reset and the exposure is started by the rise of the TX signal and the RS signal. This operation is performed in order according to the conditions set by the TG 16. The TX signal rises again after the exposure time elapses from the fall, and reads the charge of the photodiode 202 to the signal amplification amplifier 204. A video signal is generated from the signal from the signal amplification amplifier 204 and output through the horizontal scanning circuit 209. This operation is also performed under the conditions set by the TG 16.

  The image sensor 5 mounted on the camera 1 of the present embodiment is a CMOS image sensor. The image sensor 5 can select which row of the transfer transistors 203 is driven in what order by setting the shift register of the driving circuit 206, and can repeatedly select the same row and read the signal. Further, the image sensor 5 can select which column signal is output from the same row signal depending on which column selection switch 211 is operated by setting the shift register of the horizontal scanning circuit 209. In this way, it is possible to specify from which pixel in the screen and in what order.

  Next, the photographing operation of the camera 1 in this embodiment will be described with reference to FIGS.

  When the main power switch of the camera 1 is on and the operation mode of the imaging apparatus is in the shooting (recording) mode, a shooting process sequence is executed and power is supplied to the image sensor 5 and the like.

  In step S <b> 1 shown in FIG. 6, the CPU 15 sets the drive mode of the image sensor 5 in order to acquire an image to be displayed on the LCD 10 and perform phase difference AF.

  In the drive mode of the present embodiment, the image sensor 5 adds the signals of the reference pixel and the reference pixel for reading the signal of the image displayed on the LCD 10 and does not add the pixel for the phase difference AF. Therefore, the exposure time for performing the phase difference AF operation is longer than the exposure time for obtaining an image to be displayed on the LCD 10 as shown in FIG. Accordingly, the readout period for the pixel signal related to the phase difference AF operation is lengthened, and the readout rate is lowered. Further, the number of pixels in the horizontal direction to be read for obtaining an image to be displayed on the LCD 10 is obtained by performing phase difference AF operation in order to perform addition of adjacent pixels in the horizontal direction and addition of a plurality of pixel groups having the same color filter. Therefore, it is smaller than the number of pixels to be read without adding.

  FIG. 7 shows that the exposure time is sufficiently shorter than the readout time, and the addition of a plurality of pixels having the same color filter in the horizontal direction is not performed in the readout for image display of the LCD 10, but only the addition of adjacent pixels in the horizontal direction is performed. The case where it went is shown. In FIG. 7, the readout time for the phase difference AF operation is twice the readout time for image display on the LCD 10. The actual readout period is a product of the exposure time and the number of readout pixels, and the difference between the readout time of the image display and the phase difference AF operation may be larger than the example shown in FIG. As for the signal for image display, in order to save power, the signal may be acquired in the first half of the period for outputting the signal for one screen, and the drive waveform of the image sensor 5 may be stopped in the second half. In this case, the readout rate of the image signal displayed on the LCD 10 is the same as the readout rate of the signal for the phase difference AF operation.

  In step S2, the CPU 15 sets exposure conditions by AE processing. For the image display on the LCD 10, exposure conditions (mainly exposure time) are set in consideration of the optimal exposure amount for the subject to view the subject, panning or followability to movement of the subject, and the like. For the phase difference AF operation, exposure conditions (mainly exposure time) are set in consideration of the exposure amount optimal for phase difference AF, the AF time, the movement of the subject within the AF time, the influence on AF due to camera shake, and the like.

  As described above, there is a difference in whether or not pixels are added between the imaging pixel and the phase difference AF pixel. Further, at a position away from the center of the image sensor 5, the light beam incident from the exit pupil of the taking lens barrel 100 becomes oblique, so that the amount of incident light decreases. Therefore, it is desirable that the phase difference AF pixel has an exposure amount that is equal to or greater than the ratio of the number of added pixels as compared with the imaging pixel.

  When the exposure condition (exposure time) is set, the TX signal and the RS signal are controlled as shown in FIG. FIG. 7 is a timing chart of signals for image display or phase difference AF operation of the LCD 10 generated by the drive circuit in the drive mode set in step 1.

  By driving the image sensor 5 with a signal shown in the timing chart of FIG. 7, an image signal displayed on the LCD 10 and a signal for phase difference AF operation are acquired under different exposure conditions. Specifically, the exposure conditions of the image signal and AF operation signal displayed on the LCD 10 are set for every two rows, and the timing of charge reset and output transfer of the image sensor 5 is made different.

  That is, the image signal displayed on the LCD 10 and the signal for the phase difference AF operation cause the image sensor 5 to reset the charge of the photodiode 202 of each pixel and start exposure by the rise of the TX signal and the RS signal. This operation is performed in the order set by the TG 16.

  In the image display row of the LCD 10, the TX1 signal and the TX2 signal sequentially rise after the exposure time elapses from each fall. As a result, the charge of the photodiode 202 is output to the imaging circuit 6 by the horizontal scanning circuit 209 via the signal amplification amplifier 204. Thereafter, the RX1 signal and the RX2 signal sequentially rise, and the image display row of the LCD 10 is reset. In this way, the operation of outputting the image signal displayed on the LCD 10 to the imaging circuit 6 is repeated. With regard to the phase difference AF operation, the TX3 signal and the TX4 signal sequentially rise after the exposure time has elapsed from their respective fall. Thereby, the charge of the photodiode 201 is output to the imaging circuit 6 by the horizontal scanning circuit 209 via the signal amplification amplifier 204, and an image for phase difference AF operation is acquired.

  In step S <b> 3, the CPU 15 displays an image acquired by the image sensor 5 on the LCD 10.

  Next, in step S4, the CPU 15 determines whether or not the SW1 of the operation switch unit 24 has been turned on (on / off), and if it has been turned on, the process proceeds to step S5. Repeat the determination.

  In step S5, the CPU 15 causes the phase difference AF processing circuit 37 to perform phase difference AF processing.

  Next, in step S6, the CPU 15 determines the result of the phase difference AF process, and if it is determined that focusing is possible only with the phase difference AF, the process proceeds to step S21. If it is determined that the scan AF process is necessary for focusing, the process proceeds to step S8. If it is determined that focusing is impossible, the process proceeds to step S7.

  In step S21, the CPU 15 determines whether or not a BP correction value in the shooting environment has already been calculated. This determination method will be described later.

  If it is determined that the BP correction value has been calculated, the process proceeds to step S22. If it is determined that the BP correction value has not been calculated, the process proceeds to step S13. In step S22, the CPU 15 causes the second motor driving circuit 19 to drive the focus lens group 3. Specifically, the CPU 15 drives the focus lens group 3 to a position obtained by adding the BP correction value of the phase difference AF to the position where the defocus amount obtained by the phase difference AF process becomes zero, and then, step S16. Proceed to In step S <b> 16, the CPU 15 indicates to the user that AF has been completed by a process such as displaying a green frame on the LCD simultaneously with the lighting of the display element 29. When the process of step S16 ends, the process proceeds to step S17.

  If it is determined in step S6 that focusing cannot be performed and the process proceeds to step S7, the CPU 15 changes the drive mode of the image sensor 5 to a drive mode capable of high-speed reading for scan AF processing shown in FIG. Change the drive timing. At this time, the aperture is not changed. If the frame rate is increased to speed up the scan AF process, the exposure time is shortened. For this reason, the signal level is secured by increasing the amplification factor accordingly. However, if the amplification factor is too large, good scan AF cannot be expected. Therefore, by setting the upper limit of the amplification factor, the exposure time is adjusted to ensure an appropriate exposure amount. Also in this case, the exposure time is shorter than that in FIG.

  For example, when the upper limit of the amplification factor is five steps, if the optimal exposure time for an image displayed on the LCD 10 is 1/30 second and the amplification factor is 0, the exposure time for scan AF is 1/120 second The amplification factor becomes two stages. However, if the exposure time is low and the optimum exposure time for an image displayed on the LCD 10 is 1/30 second and the amplification factor is four, the exposure time for scan AF is 1/60 second and the amplification factor is five. It becomes.

  On the other hand, the signal for scan AF processing can be thinned out in the vertical direction by adding adjacent pixels in the horizontal direction sharing the microlens and adding a plurality of pixel groups having the same color filter. is there. By this addition and thinning, a high frame rate can be realized.

  During the scan AF process, an image signal for display on the LCD 10 is used as a scan AF signal. After changing the drive mode and drive timing in step S7, the CPU 15 proceeds to step S9.

  When it is determined in step S6 that the scan AF process is necessary for high-precision focusing and the process proceeds to step S8, the CPU 15 causes the second motor drive circuit 19 to drive the focus lens group 3 to the scan AF start position. The start position of the scan AF is set to the current position by a half of the range in which the AF evaluation value is acquired by the scan AF from the position of the focus lens group 3 where the defocus amount obtained by the phase difference AF process in step S5 becomes zero. Close position. In this case, the drive timing and drive mode of the image sensor 5 are not changed as in step S7. Since the range for acquiring the AF evaluation value is limited to the vicinity of the substantially in-focus position obtained by the phase difference AF process, the number of times of acquiring the AF evaluation value is small and the time required for the scan AF process is short. As a result, the scan AF processing time can be shortened by starting the scan AF process without changing the drive mode of the image sensor 5. After driving the focus lens group 3 in step S8, the CPU 15 proceeds to step S9.

  In step S9, the CPU 15 performs a scan AF process and proceeds to step S10. This process will be described later.

  In step S10, if the drive timing and drive mode of the image sensor 5 have been changed in the scan AF process in step S7, the CPU 15 returns to the original drive timing and drive mode. When the process of step S10 is completed, the process proceeds to step S11.

  In step S11, the CPU 15 evaluates the result of the scan AF process. If it is determined that focusing is impossible, the process proceeds to step S15. If it is determined that focusing is possible, the process proceeds to step 12. In step S15, the CPU 15 causes the second motor driving circuit to drive the focus lens group 3 to a predetermined position (fixed point), and proceeds to step S16. In step S16, the CPU 15 performs processing such as displaying a yellow frame on the LCD 10 simultaneously with the blinking of the display element 29, and indicates to the user that AF has not been completed.

  When it is determined in step S11 that focusing is possible and the process proceeds to step S12, the CPU 15 determines whether or not a BP correction value in the shooting environment has already been calculated. This determination method is the same as step S21, and the contents will be described later.

  If it is determined that the BP correction value has been calculated, the process proceeds to step S14. If it is determined that the BP correction value has not been calculated, the process proceeds to step S13. In step S <b> 14, the CPU 15 causes the second motor driving circuit 19 to drive the focus lens group 3. Specifically, the CPU 15 drives the focus lens group 3 to a position where the BP correction value is added to the position where the AF evaluation value becomes a maximum in the scan AF process, and then proceeds to step S16. In step S <b> 16, the CPU 15 indicates to the user that AF has been completed by a process such as displaying a green frame on the LCD 10 simultaneously with the lighting of the display element 29. When the process of step S16 ends, the process proceeds to step S17.

  In step S13, the CPU 15 performs the above-described RFAF processing.

  The RFAF processing circuit 31 calculates the sum of both signals while shifting the pixel position of the reference image signal (B image signal) with respect to the standard image signal (A image signal), creates a sum signal sequence, The relative shift amount is recorded in a recording area in the circuit. Then, the CPU 15 inputs the sum signal sequence to the scan AF processing circuit 14 to obtain an AF evaluation value signal, obtains a relative shift amount that maximizes the AF evaluation value, and corresponds to the obtained relative shift amount. The position to be used is the in-focus position. Further, the CPU 15 sets a difference value between the focus position and the result of the scan AF process or the result of the phase difference AF process as a BP correction value, and stores the value together with a shooting plan environment such as a zoom position (focal length) and an aperture value. To do. If the BP correction value can be calculated in an environment other than the shooting environment, the CPU 15 calculates the BP correction value. When the process of step S13 ends, the process proceeds to step S14.

  In step S14, the CPU 15 causes the second motor driving circuit 19 to drive the focus lens group 3 as described above, and proceeds to step S16. In step S <b> 16, the CPU 15 indicates to the user that AF has been completed by a process such as displaying a green frame on the LCD 10 simultaneously with the lighting of the display element 29. When the process of step S16 ends, the process proceeds to step S17.

  Next, in step S17, the CPU 15 determines whether or not the SW2 of the operation switch unit 24 is turned on. If the switch 15 is turned on, the process proceeds to step S18. If the switch is not turned on, the determination in this step is performed. repeat.

  In step S18, the CPU 15 executes an exposure process according to the exposure conditions set in the AE process in step S2, and ends the actual photographing operation of the camera 1.

  Next, the phase difference AF process in step S5 will be described in detail with reference to FIG.

  The image signal to be displayed on the LCD 10 and the signal for performing the phase difference AF operation are acquired by the image sensor 5 under different exposure conditions and different readout rates. Therefore, the exposure conditions for the image display signal and the AF operation signal of the LCD 10 are set for every two lines, and the reset timing and output transfer timing of the image sensor 5 are made different. As a result, the image display signal of the LCD 10 is also obtained while the phase difference AF signal is being acquired, so the CPU 15 displays the image acquired during the phase difference AF process on the LCD 10. The signal used for the phase difference AF process is a signal output independently without performing pixel addition obtained from the row of pixels for phase difference AF of the image sensor 5.

  First, in step S901, the phase difference AF processing circuit 37 records the image signal for phase difference AF output from the A / D conversion circuit 7 in a recording area (not shown) in the circuit.

  In step S902, the phase difference AF processing circuit 37 corrects the recorded image. In this embodiment, pixels for phase difference AF are arranged on the imaging surface of the image sensor 5. Therefore, a field lens that corrects image distortion caused by the image height as in the case of re-imaging with the secondary imaging optical system, a diaphragm for the light beam incident on the phase difference AF sensor, and an unnecessary light beam. A mask for blocking the image cannot be disposed between the imaging surface and the image sensor 5. Therefore, the signal for phase difference AF has a different shading offset for each pixel and needs to be corrected. This shading differs depending on the pixel position (image height) from the center of the optical axis, the exit pupil position of the photographing lens, and the stop. The phase difference AF processing circuit 37 corrects the pixel image for each pixel for phase difference AF using the image correction amount for each of these factors.

  The offset differs depending on the amplification factor of the signal for phase difference AF and the characteristics of the column amplifier of each pixel. The phase difference AF processing circuit 37 performs pixel image correction for each pixel for phase difference AF using the image correction amount for each of these factors. Details of the image correction method are described in Patent Document 3 and the like.

  Next, in step S903, the phase difference AF processing circuit 37 rearranges the pixels for phase difference AF whose image has been corrected, and generates a reference image and a reference image.

  In the image sensor 5, two rectangular photodiodes (photoelectric conversion units) are arranged for one microlens, and one photodiode 202 is for a reference pixel and the other is for a reference pixel. The reading order is “reference pixel” → “reference pixel” → “reference pixel” → “reference pixel” →... → “reference pixel” → “reference pixel”. The phase difference AF processing circuit 37 also records the standard image and the reference image after the image correction in step S902 in a recording area (not shown) in the circuit in an order corresponding to the reference image. At this time, the phase difference AF processing circuit 37 extracts only the pixels constituting the reference image, and records the signals of the reference image in the extracted order. Similarly, the phase difference AF processing circuit 37 extracts only the pixels constituting the reference image, and records the reference image signal in the extracted order.

  Thereafter, in step S904, the phase difference AF processing circuit 37 sets an initial value for performing the correlation calculation. In step S905, the phase difference AF processing circuit 37 calculates a correlation value (correlation data) between the reference image and the reference image by performing a correlation calculation according to the equation (1) using the initial value set in step S904.

U _k = Σmax (a _{j + 1} , b _{j + k} ) −Σmax (a _j , b _{j + k + 1} ) (1)
Here, a _i and b _i represent output values of the A and B images of the i-th pixel, where i is an integer. max (a, b) means the larger value of a and b. Further, k is an image shift amount for performing correlation calculation, and j is a pixel number (position) for performing correlation calculation. Σ in equation (1) means the total sum when j is changed in order from 1. These are initialized in step S904.

  In step S <b> 906, the CPU 15 acquires the correlation value between the standard image and the reference image from the phase difference AF processing circuit 37.

  If the correlation value is already temporarily recorded, the CPU 15 determines whether the acquired correlation value is equal to the sign, and replaces the acquired correlation value with the recorded correlation value. If there is no correlation value already recorded, the acquired correlation value is recorded.

  In step S907, if it is determined that the sign of the acquired correlation value is inverted with respect to the code of the already recorded correlation value, or if the acquired correlation value is zero, the process proceeds to step S910, where the signs are equal. In this case, the process proceeds to step S908.

  In step S908, the CPU 15 determines whether or not the shift amount has reached the end value of the calculation for obtaining the correlation value. If it is not the end value, the process proceeds to step S920, and if it is the end value, the process proceeds to step S909.

  In step S920, the CPU 15 updates the shift amount k to k ← k + 1.

  In step S909, the CPU 15 determines that phase difference AF is not possible (NG), and ends the process.

  In step S910, the CPU 15 calculates an image shift amount that causes the correlation amount to be zero. Since the correlation value is calculated by shifting one pixel at a time, the correlation amount calculated in the phase difference AF processing circuit 37 is rarely zero. Therefore, the CPU 15 obtains an image shift amount for which the correlation amount becomes zero from the two correlation amounts having different signs and the shift amount that gives the correlation amount.

As a result of calculating the correlation amount by the equation (1), if the sign of the correlation amount U _k is inverted between K = 1 and K = 1 + 1, the image shift amount δ at which the correlation amount becomes zero by linear interpolation is expressed by the following equation: It is obtained by.
δ = l + | U _l | ÷ [| U _l | + | U _{l + 1} |]
However, | z | means the absolute value of z.

Next, in step S911, the CPU 15 obtains the prediction amount P from the image shift amount δ by the following equation.
P = δ−Δ
Δ means an image shift amount at the time of focusing.

Then, the CPU 15 obtains the defocus amount d (the moving amount and direction of the focus lens group) from the prediction amount P by using the base line length determined from the characteristics of the photographing lens barrel 100, and ends the processing. .
d = K · P
K is a sensitivity related to the focus, and is a value depending on the focal length of the photographing lens barrel 100, the value of the diaphragm 4, the image height, and the like. Therefore, a table using these as parameters is prepared in the EEPROM 25, and values are obtained by referring to the table.

  Next, the scan AF process will be described with reference to FIG.

  In the following description, the operation of acquiring the AF evaluation value while driving the focus lens group 3 to a predetermined position is referred to as scanning, and the interval of the position of the focus lens for acquiring the AF evaluation value is referred to as scanning interval. The number of AF evaluation values acquired is called the number of scan points, the range of acquiring AF evaluation values is called the scan range, and the area for acquiring the image signal for detecting the in-focus position is called the AF frame.

  The CPU 15 performs scan AF under the exposure conditions determined by the AE processing circuit 13 in step S9, and when focus adjustment is possible, the focus lens group is moved to the in-focus position by the second motor drive circuit 19 according to steps S11 to S14. 3 is driven. If focus adjustment is impossible, the CPU 15 causes the second motor drive circuit 19 to follow the steps S11 and S15 in order to cause the second motor drive circuit 19 to include the infinity called the hyperfocal position on the far side of the depth of field. The focus lens group 3 is driven to the focus lens position.

  When the drive mode of the image sensor 5 is changed in step S7, the image sensor 5 during the scan AF process is driven according to the timing chart shown in FIG. Here, an image signal for display on the LCD 10 is used as a signal for the scan AF operation. The image display signal and the scan AF operation signal of the LCD 10 are acquired from the image sensor 5 under the same exposure conditions and the same readout rate. On the other hand, signals for performing the phase difference AF operation are acquired under different exposure conditions and acquired from the image sensor 5 at different readout rates.

  For this purpose, the exposure conditions of the scan AF operation and the image display signal of the LCD 10 and the signal for the phase difference AF operation are set for every two rows, and the reset timing and output transfer timing of the image sensor 5 are made different. In the scan AF process, a signal for performing the phase difference AF operation is not used.

  As for the exposure time for the scan AF operation, 1/32 second is set as the limit on the long time side, and the minimum exposure time determined by the performance of the sensor is set as the limit on the short time side. The exposure time is determined so as to obtain an appropriate exposure amount by adjusting the value of the diaphragm 4 or the column amplifier circuit 210 with reference to the AE processing result in step S2. In the LCD display, the exposure amount during the scan AF may become inappropriate, but the scan AF performance is prioritized in this process.

  By performing processing according to the timing chart of FIG. 8, the exposure time for the scan AF operation is further shorter than the exposure time for the image for the phase difference AF operation, and the readout rate is also increased. For example, the exposure time for the scan AF operation is 1/128 second, the readout rate is 128 FPS, the exposure time for the image for the phase difference AF operation is 1/8 second, and the readout rate is 8 FPS. At this time, the value of the diaphragm 4 is maintained as it is without being changed.

  If it is determined in step S6 that it is necessary to perform scan AF processing for focusing (substantially focusing is possible), the drive mode of the image sensor 5 has not been changed, and scanning is performed according to the timing chart of FIG. AF processing is performed.

  The TX signal and the RS signal are controlled as shown in FIG. 8 according to the determined exposure condition (exposure time). In acquiring the scan AF operation signal and the phase difference AF operation signal, the charge of the photodiode 202 of each pixel is reset and the exposure is started in response to the rise of the TX signal and the RS signal. This operation is sequentially performed in a predetermined order of the light receiving pixels 201 under the conditions set by the TG 16.

  The TX1 and TX2 signals sequentially rise after a predetermined exposure time has elapsed from the start of exposure. In response to this, the charge of the photodiode 202 in the row for the scan AF operation is read by the signal amplification amplifier 204 and output through the horizontal scanning circuit 209. Next, RX1 and RX2 sequentially rise again, and the charge of the row for the scan AF operation signal is reset accordingly. By repeating these operations, a scan AF operation signal is acquired.

  Similarly, the TX3 and TX4 signals sequentially rise after the exposure time has elapsed from the start of exposure. In response to this, the charge of the photodiode 201 for phase difference AF operation is read by the signal amplification amplifier 204 and output through the horizontal scanning circuit 209. Next, RX3 and RX4 sequentially rise again, and the charge of the row for the scan AF operation signal is reset accordingly. By repeating these operations, a signal for phase difference AF is acquired.

  By performing these operations after the focus lens group 3 is driven to a predetermined position, a scan AF operation signal is acquired, and the scan AF processing circuit 14 calculates an AF evaluation value. When the desired AF evaluation value is acquired, the focus lens group 3 is driven to the next predetermined position, the above-described scan AF operation signal acquisition operation is performed again, and the AF evaluation value acquisition is repeated.

  Next, the scan AF operation in the case where it is determined that focusing cannot be performed in the phase difference AF process in step S6 in FIG. 6 will be described with reference to FIG.

  If it is determined in step S6 that the focus is substantially in focus, the scan range is limited to the vicinity of the substantially in-focus position obtained as a result of the phase difference AF process. Excluding this difference, the same operation as that performed when it is determined in step S6 that focusing cannot be performed is performed.

  In the scan AF, the driving of the focus lens group 3 and the acquisition of the AF evaluation value are repeated as described above, and the high frequency component extracted from the signal for the scan AF operation set for every two rows generated by the image sensor 5 is obtained. The position of the focus lens group 3 that is the largest is obtained.

  First, the CPU 15 drives the focus lens group 3 from a position corresponding to infinity (“A” in FIG. 10) to a position corresponding to the closest distance set in each shooting mode (“B”). The focus drive motor 22 is controlled via the two-motor drive circuit 19. The CPU 15 acquires the output (AF evaluation value signal) of the scan AF processing circuit 14 while performing this driving. Then, when the driving of the focus lens group 3 is completed, a position (“C” in FIG. 10) where the acquired AF evaluation value signal is maximum is obtained, and the focus lens group 3 is driven to that position.

  If it is determined that the in-focus state is substantially in focus in the phase difference AF process in step S6, an AF evaluation value signal) is acquired only in the vicinity of the in-focus position. For example, when the phase difference AF process is performed at the position P in FIG. 10 and the position C ′ in FIG. 10 is calculated as a substantially in-focus position, the scan range is divided by two from the position C ′ to the position P. 1 position ("A '" in FIG. 10) is obtained. Next, the AF evaluation value signal is acquired by performing the scan AF process from the position A ′ to the position half the scan range (“B ′” in FIG. 10) on the opposite side, and the acquired AF evaluation value signal Is determined to be the position ("C" in FIG. 10). Then, the focus lens group 3 is driven to a position where the AF evaluation value signal becomes maximum.

  The acquisition of the AF evaluation value signal may be performed every predetermined step without performing the stop positions of all the focus lens groups 3 in order to perform the scan AF at high speed. For example, the AF evaluation value signal may be acquired at the positions a1, a2, and a3 shown in FIG. In such a case, the in-focus position C is obtained by calculation from the point where the AF evaluation value signal becomes the maximum value and the points before and after the point.

  When the position of the focus lens group 3 is X1 (a2 in FIG. 10), the AF evaluation value becomes the maximum value Y1. When the AF evaluation values acquired at the front and rear positions X2 and X3 are Y2 and Y3 (a1 and a3 in FIG. 10), the position X0 of the focus lens group 3 at the in-focus position C is obtained by the following expression.

X0 = {(Y3-Y2) .X1 + (Y3-Y1) .X2 + (Y2-Y1) .X3} / {2. (Y3-Y1)}
However, Y1> Y3 and Y1 ≧ Y2.

  Further, the reliability of the AF evaluation value signal may be evaluated before performing the interpolation calculation and obtaining the point (“C” in FIG. 10) where the AF evaluation value signal becomes the maximum value. Specific methods are described in Patent Document 4 and Patent Document 5.

  Next, the RFAF process performed in step S13 will be described with reference to FIG.

  The RFAF process corrects the difference between the position of the focus lens group 3 where the resolution of the recorded image is the maximum and the position resulting from the scan AF process or the phase difference AF process, which mainly occurs due to the optical characteristics of the photographing lens. Process. This difference value is used as a BP correction value and stored in the EEPROM 25 together with the shooting environment such as the image height, zoom position, aperture value, temperature in the taking lens barrel 100, temperature of the camera 1, and camera posture. If the BP correction value can be calculated in an environment other than the shooting environment, the BP correction value in that environment is calculated and stored together with the shooting environment. In such RFAF processing, the position of the focus lens group 3 where the AF evaluation value obtained as a result of the scan AF processing is maximized, or the position of the focus lens group 3 where the defocus amount obtained as a result of the phase difference AF processing is zero. Only in the vicinity of The signal used in the RFAF processing is a signal output independently without performing pixel addition obtained from the phase difference AF pixel row of the image sensor 5.

  First, in step S1101 in FIG. 11, the CPU 15 determines whether or not the focus lens group 3 is located at the in-focus position obtained in the phase difference AF process (step S5) or the scan AF process (step S9). If it is determined that the current position is the in-focus position, the process advances to step S1102, and the CPU 15 determines whether the phase difference AF process has already been performed.

  Here, as a result of performing the phase difference AF process in step S5, it is determined in step S6 that focus is possible, and the position of the focus lens group 3 at that time is the focus position. In this case, a signal of a pixel for phase difference AF at the in-focus position is acquired, and image correction or rearrangement processing is performed for the phase difference AF. At this time, the CPU 15 determines that the phase difference AF processing has already been performed in step S1102, and proceeds to step S1104.

  Further, as a result of performing the phase difference AF processing in step S5, it is determined in step S6 that the in-focus state is substantially possible or impossible, and as a result of the scan AF processing being performed in step S9, the focus lens group 3 is set as the in-focus position. It is assumed that the moved position is a position where the phase difference AF processing is performed. In this case, a pixel signal for phase difference AF at that position is acquired, and image correction or rearrangement processing is performed for the phase difference AF. Also at this time, the CPU 15 determines that the phase difference AF process has already been performed, and proceeds to step S1104.

  In step S1104, the CPU 15 reads the standard image signal and the reference image signal recorded in the predetermined recording area of the phase difference AF processing circuit 37, and then proceeds to step S1105.

  On the other hand, if it is determined in step S1101 that it is not in the in-focus position, the process proceeds to step S1103, and the CPU 15 moves the focus lens group 3 to the in-focus position and proceeds to step S901.

  In step S901, the phase difference AF processing circuit 37 records the image signal for phase difference AF output from the A / D conversion circuit 7 in a predetermined recording area of the phase difference AF processing circuit 37, and proceeds to step S902. In step S902, the phase difference AF processing circuit 37 corrects the recorded image and proceeds to step S903.

  In step S903, the CPU 15 rearranges the pixels for phase difference AF whose image has been corrected, generates a reference image and a reference image, and records them in a predetermined recording area of the phase difference AF processing circuit 37, and then proceeds to step S1105. move on. A series of these operations is the same as the phase difference AF process performed in step S5.

  In step S <b> 1105, the CPU 15 sets an initial value for obtaining the sum signal of the standard image signal and the reference image signal in the RFAF processing circuit 31.

  In step S1106, the RFAF processing circuit 31 to which the initial value is set obtains a sum signal according to the equation (2).

Add (j, k) = aj + bj + k (2)
Here, k is an image shift amount when obtaining the sum signal, j is a pixel number (position) on which the correlation calculation is performed, and is initialized in step S1105.

For example, when initialization is performed with k = κ (kappa) and the range of j is 0 to J, signals of Add (0, κ) to Add (j, κ) are obtained as Add (j, k). It is done. Its value is
Add (0, κ) = a0 + b0 + κ
Add (1, κ) = a1 + b1 + κ
・
・
・
・
Add (j, κ) = aJ + bJ + κ
It becomes.

  After obtaining the sum signal in this way, the CPU 15 proceeds to step S1107, inputs this sum signal to the circuit for obtaining the AF evaluation value of the scan AF processing circuit 14, and obtains the AF evaluation value Tes (k) at the image shift amount k. Ask.

  In step S1108, the CPU 15 records the AF evaluation value Tes (k) obtained in step S1107 together with the image shift amount k in a predetermined recording area, and proceeds to step S1109.

  In step S1109, the CPU 15 determines whether or not the image shift amount k has reached the end value of the calculation for obtaining the sum signal. If the end value is not reached, the CPU 15 proceeds to step S1120 and updates the image shift amount k to k ← k + 1.

  In step S1106, the RFAF processing circuit 31 obtains a sum signal according to the equation (2).

As described above, initialization is performed with k = κ (kappa), and when j is in the range of 0 to J, Add (0, κ + 1) to Add (J, κ + 1) is added as Add (j, k). A signal is required. Its value is Add (0, κ + 1) = a0 + b0 + κ + 1
Add (1, κ + 1) = a1 + b1 + κ + 1
・
・
・
・
Add (j, κ + 1) = aJ + bJ + κ + 1
It becomes.

  Thereafter, as described above, the CPU 15 inputs this sum signal to the circuit for obtaining the AF evaluation value of the scan AF processing circuit 14 in step S1107, and obtains the AF evaluation value Tes (k) at the image shift amount k. Next, in step S1108, the CPU 15 records the AF evaluation value Tes (k) obtained in step S1107 together with the image shift amount k in a predetermined recording area. In step S1109, the CPU 15 determines whether or not the shift amount has reached the end value of the calculation for obtaining the sum signal.

  The CPU 15 repeats the above operation until the shift amount reaches the end value.

When the CPU 15 reaches the end value, the CPU 15 proceeds to step S1110 and obtains a shift amount and a BP correction value that maximizes the AF evaluation value obtained by the RFAF processing. Here, the in-focus position in the RFAF processing is obtained by calculation from the point where the AF evaluation value signal becomes the maximum value and the points before and after that point. When the shift amount is Kd, the AF evaluation value becomes maximum, and the AF evaluation value is set to Tes (Kd). Assuming that the AF evaluation values acquired with the previous and subsequent shift amounts are Tes (Kd−1) and Tes (Kd + 1), the shift amount Kδ that gives the in-focus position is expressed by the following equation.
Kδ = [{Tes (Kd + 1) −Tes (Kd−1)} kd + {Tes (Kd + 1) −Tes (Kd)} (kd−1) + {Tes (Kd−1) −Tes (Kd)} (kd + 1) ] / 2 · {(Tes (Kd + 1) −Tes (Kd))}
However, Tes (Kd)> Tes (Kd + 1) and Tes (Kd) ≧ Tes (Kd−1).

  As described above, interpolation calculation may be performed before obtaining the point at which the AF evaluation value signal has the maximum value, and the reliability of the AF evaluation value signal may be evaluated. In this case, when the reliability is sufficient, a point at which the AF evaluation value signal becomes the maximum value is obtained.

  In the RFAF processing, a signal that does not perform pixel addition obtained from the phase difference AF pixel row of the image sensor 5 is used. On the other hand, in the scan AF process, pixel addition is performed to increase the signal readout rate from the image sensor 5 and the number of readout pixels is reduced, so that the high frequency component included in the recorded image may not be included. Then, when the imaging position of the subject for each spatial frequency is different depending on the optical characteristics of the photographing lens, the imaging position of the recorded image signal and the signal used for the scan AF process are different from each other. Therefore, the resolution of the recorded image may not be maximized at the position of the focus lens group 3 where the AF evaluation value obtained by the scan AF process is maximized.

  In this regard, in the RFAF processing, since a signal that does not perform pixel addition is used, the resolution of the recorded image is maximized at the position of the focus lens group 3 where the AF evaluation value obtained by the RFAF processing is maximized.

  In the phase difference AF processing, a signal that does not perform pixel addition is used, but in order to avoid a calculation error when performing a correlation calculation, a low-pass filter suppresses a high-frequency signal component, The high-frequency spatial frequency components included are different. As a result, the resolution of the recorded image may not be maximized at the position of the focus lens group 3 where the defocus amount obtained by the phase difference AF process is zero.

  In this regard, in the RFAF processing, since the processing by the low-pass filter is not performed, the resolution of the recorded image is maximized at the position of the focus lens group 3 obtained by the RFAF processing.

If the shift amount Kδ at which the obtained AF evaluation value is maximized is obtained by the RFAF processing, the amount of movement to the position of the focus lens group 3 at which the resolution of the recorded image is maximized is determined based on the shift amount Kδ. It calculates | requires according to Formula (3).
Lens movement (pulse)
= K · kδ ÷ Focus sensitivity ÷ Movement amount per pulse (3)
Here, K is a coefficient for converting the image shift amount on the image sensor into the defocus amount. Since K is a value that depends on the focal length of the taking lens barrel 100, the value of the diaphragm 4, the image height, and the like, a table using these as parameters in the EEPROM 25 is prepared in advance, and K is acquired by referring to the table.

  This lens movement amount is a focus correction amount (BP correction value) mainly resulting from the optical characteristics of the photographing lens. The position corrected with this correction amount is the focus correction position. This focus correction position is taken as the final focus position.

  In this way, the BP correction value is obtained, and the focus lens group 3 is moved to the final focus position.

  Thereafter, in step S1111, an interpolation coefficient for calculating a BP correction value in a case where environments such as the zoom position, the image height of the AF frame center coordinates, and the aperture value are different is obtained. The procedure for obtaining this coefficient will be described with reference to FIG.

  In step S1201, the CPU 15 reads the shooting environment. Here, the zoom position, the image height of the AF frame center coordinates, the aperture value, the lens barrel temperature, the ambient temperature, and the camera posture when the AF process before shooting is executed are read as the shooting environment.

  Next, in step S1202, the CPU 15 reads the BP correction value obtained in the processing up to step S1110 as the BP correction value in the shooting environment.

  In step S1203, the CPU 15 determines whether the currently read shooting environment (current shooting environment) is substantially equal to the past shooting environment recorded from the EEPROM 25. This determination method is the same as step S21, and detailed contents will be described later.

  If it is determined that the current shooting environment is substantially equal to the recorded shooting environment, the process advances to step S1204 to update the BP correction value corresponding to the environment with reference to the EEPROM 25.

  This BP correction value setting method may be a method different from the above method. For example, the recorded BP correction value may be reduced to a fraction, and a weighted average value with a new BP correction value may be obtained and replaced with that value.

  If it is determined that the BP correction value in the substantially equal shooting environment is not recorded, the CPU 15 proceeds to step S1205 and records the BP correction value together with the shooting environment.

  Thereafter, in step S1206, the CPU 15 determines whether interpolation calculation of the BP correction value in the image height direction is possible.

  This is determined to be that interpolation calculation is possible if the BP correction value is calculated at the center position of the outermost position screen where the AF frame is set and the intermediate position in an approximately equal shooting environment. That is, if nine BP correction values as shown in FIG. 13A or 13B are recorded in a predetermined shooting environment, it is determined that interpolation calculation of the BP correction values in the image height direction is possible. . The predetermined shooting environment includes, for example, the same zoom position, the same aperture value, substantially the same lens barrel temperature (for example, within ± 10 ° C. with respect to normal temperature), substantially the same ambient temperature (for example, within ± 10 ° C. with respect to normal temperature), and approximately Equal camera posture (for example, within ± 30 ° with respect to horizontal). These exemplified numerical values vary depending on characteristics of the taking lens barrel 100 and the like.

  In FIG. 13A and FIG. 13B, the dotted line represents the AF frame setting range, and the AF frame is freely set within this range. The center of the figure indicated by a cross near the outer periphery indicates the center coordinates of the AF frame when an AF frame having a standard size is set to be in contact with the dotted line range.

  The center cross represents the AF frame at the center of the screen, and the intersection is the center coordinate. Four crosses arranged at intermediate coordinates represent coordinates for acquiring a BP correction value used when performing interpolation calculation of the BP correction value in the image height direction.

  If it is determined in step S1206 that interpolation calculation of the BP correction value in the image height direction is possible, the CPU 15 performs interpolation calculation in step S1207 to create and record a coefficient for calculating the BP correction value. Thereafter, in the same shooting environment, the BP correction value is calculated from the recorded calculation coefficient, and the BP correction value is not calculated using the RFAF process.

  The interpolation calculation of the BP correction value in step S1207 is performed as follows.

  When the BP correction values are acquired at the nine points shown in FIG. 13A, the CPU 15 uses the five points arranged in a line in the horizontal direction as a quadratic function for the BP correction value for the image height in the horizontal direction. Approximate. Similarly, the CPU 15 approximates the BP correction value for the image height in the vertical direction to a quadratic function using five points arranged in a straight line in the vertical direction. Then, the CPU 15 records the approximated quadratic function coefficient in a predetermined area of the EEPROM 25.

  When obtaining the BP correction value at the image height with reference to the recorded interpolation coefficient of the BP correction value, the CPU 15 first obtains the horizontal BP correction value and then obtains the vertical BP correction value. A BP correction value in the image height direction is obtained from the values of these two approximated quadratic functions.

For example, consider a case where a horizontal BP correction amount BPh and a vertical BP correction amount BPv at an arbitrary image height (x, y) are expressed by the following quadratic function.
BPh = ah · x ² + bh · x + ch
BPv = av · x ² + bv · x + cv
In this case, the BP correction value BPz in the image height direction is as follows.
BPz = √ (BPh ² + BPv ² )
If it is determined in step S1206 that interpolation calculation of the BP correction value in the image height direction is impossible, the CPU 15 advances the process to step S1212.

  In step S1208, the CPU 15 determines whether interpolation calculation of the BP correction value based on the aperture value can be performed. For example, if the coefficient for interpolation calculation of the BP correction value in the image height direction is obtained when the aperture value of the aperture 4 is full, when the aperture value is F8, and when the aperture value is an intermediate aperture, interpolation calculation based on the aperture value is performed. Determine that it can be done. However, for a lens with an open F value of 5.6 or more, if the coefficient for interpolation calculation of the BP correction value in the image height direction when the aperture value is the open F value and when the aperture value is F8 is obtained, It is determined that interpolation calculation based on the aperture value can be performed. For a lens with an open F value of F8 or more, if the coefficient for interpolation calculation of the BP correction value in the image height direction when the aperture value is the open F value is obtained, interpolation calculation by the aperture value can be performed. judge.

  If it is determined in step S1208 that interpolation calculation of the BP correction value based on the aperture value is possible, the CPU 15 performs interpolation calculation in step S1209, and creates and records a coefficient for calculating the BP correction value. Thereafter, in the same shooting environment, the BP correction value is calculated from the recorded calculation coefficient, and the BP correction value is not calculated using the RFAF processing circuit 31.

  If it is determined in step S1208 that interpolation calculation of the BP correction value of the aperture value is impossible, the CPU 15 advances the process to step S1212.

  The interpolation calculation of the BP correction value in step S1209 is performed as follows.

  When BP correction values at three aperture values are required for coefficient calculation, a quadratic function approximated from the three values is obtained, and the coefficient of the quadratic function is recorded in a predetermined area of the EEPROM 25. This coefficient can be obtained using a second order Lagrangian interpolation polynomial. At this time, the average value of the nine AF frame position values shown in FIGS. 13A and 13B is used as the BP correction value at the three aperture values.

  For example, in the case of a lens with an open F value of F2.8, F3.5, and F4.0, a coefficient of a quadratic function is obtained from the open F value, the BP correction value at F5.6, and F8.0. When the open F value is F1.2 or F1.4, from the open F value, the BP correction value at F2.8 or F8.0, and when the open F value is F1.8 or F2.0, the open F value , F4.0, and F8.0, the coefficient of the quadratic function is obtained from the BP correction value.

  If the open F value is F5.6 or more and less than F8.0, the coefficient of the linear function is obtained from the open F value and the BP correction value at F8.0, and the coefficient of the linear function is recorded in a predetermined area of the EEPROM 25. To do. This coefficient can be determined using a first order Lagrangian interpolation polynomial.

  When the open F value is F8.0 or more, the BP correction value at the open F value (F8.0) is recorded in a predetermined area of the EEPROM 25 as the BP correction value for all aperture values.

  Further, referring to the interpolation coefficient of the recorded BP correction value, the BP correction value at the aperture value is obtained by the following method.

The BP correction value BPz of the aperture value closest to the aperture value for which the BP correction value is to be obtained is obtained by the method described above. When the aperture value (F value) for which the BP correction value is to be obtained is Fz, and the aperture value (F value) that gives the BP correction value BPz is Fo, the BP correction value BPz (av adj) calculated by the aperture is calculated. Is obtained by the following equation (4).
BPz (av adj) = BPz × {calculated value in Fz} ÷ {calculated value in Fo} (4)
However, {calculated value in Fz} is a calculated value in a quadratic expression or a primary expression obtained by using a coefficient for interpolating a BP correction value with the previously obtained aperture for Fz. {Calculated value in Fo} is a calculated value in a quadratic or linear expression obtained using a coefficient for interpolating a BP correction value with the previously obtained aperture.

  When the aperture value Fz to be obtained is F8.0 or more, the BP correction value BPz when the aperture value is F8.0 is set to BPz (av adj).

  When the CPU 15 performs interpolation calculation in step S1209 to create and record a coefficient for calculating the BP correction value, the CPU 15 determines in step S1210 whether interpolation calculation of the BP correction value based on the zoom position is possible.

  In this embodiment, if the coefficient for interpolation calculation of the BP correction value by the image height direction and the aperture value is obtained at the zoom position at the Wide end, the intermediate zoom position, and the Tele end, the interpolation calculation by the zoom position is performed. It is determined that you can.

  If it is determined in step S1210 that interpolation calculation of the BP correction value based on the zoom position is possible, the CPU 15 performs interpolation calculation in step S1211 to create and record a coefficient for calculating the BP correction value. Thereafter, in the same shooting environment, the BP correction value is calculated from the recorded calculation coefficient, and the BP correction value is not calculated using the RFAF process.

  The interpolation calculation of the BP correction value in step S1211 is performed as follows.

  A quadratic function approximated from three BP correction values is obtained at the Wide end, intermediate zoom position, and Tele end for coefficient calculation, and the coefficient of the quadratic function is recorded in a predetermined area of the EEPROM 25. This coefficient can be obtained using a second order Lagrangian interpolation polynomial. At this time, as the BP correction values at the three zoom positions, the average values at all the image heights and aperture values at the respective zoom positions are used. For the image height, the values of the nine AF frame positions shown in FIGS. 13A and 13B, and for the aperture value, the values of the open to F8 are set (for example, the open F value is F2.8, F3.5). In the case of a lens of F4.0, the average value of the open F value, the values at F5.6, F8.0) is used.

  The BP correction value at the zoom position is obtained as follows with reference to the interpolation coefficient of the recorded BP correction value.

  The BP correction value BPz of the aperture value closest to the aperture value for which the BP correction value is to be obtained at the closest zoom position among the three zoom positions for which the BP correction value is to be obtained is obtained by the method described above. . Then, the BP correction value BPz (av adj) of the aperture value (F value) Fz for which the BP correction value at the closest zoom position is to be obtained is obtained according to the equation (4).

  Further, the BP correction value BPz (av adj, zm adj) at the zoom position Zz for which the BP correction value is to be obtained is obtained by the following expression.

BPz (av adj, zm adj) = BPz (av adj, zm adj) × {calculated value in Zz} ÷ {calculated value in Zo}
However, {calculated value in Zz} is a quadratic expression obtained using a coefficient for interpolating the BP correction value by the previously obtained zoom position for the zoom position Zz for which the BP correction value is to be obtained. It is a calculated value. {Calculated value in Zo} is a quadratic or linear expression obtained by using a coefficient for interpolating the BP correction value with the previously obtained diaphragm for the diaphragm value (F value) FoFz giving the BP correction value BPz. Calculated value in the formula.

  In this way, an arbitrary image height, aperture value (F value) Fz, and BP correction value BPz (av adj, zm adj) at the zoom position Zz are obtained.

  When the CPU 15 performs interpolation calculation in step S1211, and creates and records a coefficient for calculating the BP correction value, the CPU 15 determines in step S1213 whether the temperature correction coefficient can be calculated. The same applies to the case where it is determined in step S1210 that interpolation calculation of the BP correction value of the aperture value is impossible.

  The temperature correction coefficient interpolates the BP correction amount by a quadratic function with the lens barrel temperature as a parameter for each ambient temperature and each zoom position. For this reason, calculation of the BP correction value is performed with an appropriate interval (for example, an interval of 10 ° C. or more) at an ambient temperature (for example, within ± 2 ° C.) that is substantially equal in the normal temperature range (for example, 10 ° C. to 50 ° C.). When performing, it is determined that temperature interpolation of the BP correction value is possible at the zoom position. When it is determined that temperature interpolation is possible, the CPU 15 calculates a temperature correction coefficient in step S1213.

  For example, in the zoom position Wide, when the ambient temperature is 20 ° C. and the lens barrel temperature is 23 ° C., the ambient temperature is 18 ° C. and the lens barrel temperature is 35 ° C., and the ambient temperature is 22 ° C. and the lens barrel temperature is 45 ° C. Consider a case where the value is open and the central BP correction value is calculated for the image height. A quadratic function is obtained from the BP correction values at these three lens barrel temperatures, and the obtained quadratic function is set as an interpolation function at an ambient temperature of 20 ° C. (average of ambient temperatures of 20 ° C., 18 ° C., and 22 ° C.). The coefficient of the next function (interpolation coefficient) is recorded in a predetermined area of the EEPROM 25. This coefficient can be obtained using a second order Lagrangian interpolation polynomial.

  If the BP correction value for the same peripheral image height is calculated at three lens barrel temperatures, the interpolation coefficient may be obtained using the average value.

  The interpolation coefficient is similarly obtained for other ambient temperatures, and the ambient temperature that is not obtained is obtained by linear interpolation from the coefficients of two adjacent ambient temperatures.

  Similarly, with respect to the aperture value, an interpolation coefficient is obtained for each aperture value, and an aperture value that has not been obtained is obtained by linear interpolation from the coefficients of two adjacent aperture values. However, a correction coefficient of F8.0 is used on the small aperture side at F8.0 or more.

  When the lens barrel temperature is higher or lower than the predetermined temperature, an interpolation coefficient for the predetermined temperature may be calculated or may not be interpolated.

  The interpolation using the temperature interpolation coefficient thus obtained is performed on the BP correction value obtained by image height interpolation, aperture value interpolation, and zoom interpolation.

  When the CPU 15 performs interpolation calculation in step S1213 to create and record a coefficient for calculating the BP correction value, the CPU 15 determines whether or not the attitude difference correction coefficient can be calculated in step S1214. The same applies when it is determined in step S1212 that the temperature correction coefficient cannot be calculated. The attitude difference correction coefficient means an interpolation coefficient using an attitude difference (for example, an elevation angle) as a parameter.

  Here, when the BP correction amounts of three predetermined elevation angles (for example, elevation angle +45 degrees, 0 degrees (horizontal), −45 degrees (decline angle +45 degrees)) are calculated in an environment where the other photographing environments are substantially equal. Then, it is determined that the attitude difference correction coefficient can be calculated.

  If it is determined that the posture difference correction coefficient can be calculated, the CPU 15 calculates the posture difference correction coefficient described above in step S1215.

  For example, consider a case where the aperture value is open for three elevation angles and the central BP correction value is calculated for the image height at the zoom position Wide. A quadratic function is obtained from the BP correction values at these three elevation angles, the obtained quadratic function is used as an interpolation function, and the coefficient of the quadratic function is recorded in a predetermined area of the EEPROM 25. This coefficient can be obtained if it has a second order Lagrangian interpolation polynomial. In the case of upward and downward from a predetermined elevation angle, an interpolation coefficient for the predetermined elevation angle is used.

  Similarly, for the aperture value, an interpolation coefficient is obtained for each aperture value, and for the aperture value that is not obtained, primary interpolation is obtained from the coefficients of two adjacent aperture values. However, a correction coefficient of F8.0 is used on the small aperture side at F8.0 or more.

  The interpolation using the posture difference correction coefficient obtained in this way is performed on the BP correction value obtained by image height interpolation, aperture value interpolation, and zoom interpolation. When the detection accuracy of the elevation angle is not high, the resolution may be coarsened according to the detection accuracy of the elevation angle, for example, the elevation angle is 0 degree within ± 5 degrees.

  When the CPU 15 performs interpolation calculation in step S1215 to create and record a coefficient for calculating the BP correction value, the flow of FIG. 12 ends. The same applies when it is determined in step S1214 that the attitude difference correction coefficient cannot be calculated.

  The values obtained by interpolation are used only for parameters not measured and calculated (image height, zoom position, aperture value). If the BP correction value obtained in the RFAF process in step S13 is different from the value obtained by interpolation, the measured and recorded BP correction value is used.

  When the capacity of the area for recording the BP correction value and the shooting environment is insufficient, the oldest one is deleted sequentially. When a certain period (predetermined time) has elapsed since the BP correction value calculation, the RFAF process is performed again to obtain the BP correction value.

  Next, a method for determining whether or not a BP correction value has been calculated will be described with reference to FIG.

  First, in step S1401, the CPU 15 reads a shooting environment (current shooting environment) for calculating a BP correction value. Next, in step S1402, the CPU 15 reads the shooting environment at the time of calculating the past BP correction value recorded in the EEPROM 25, and proceeds to step S1403.

  In step S1403, the CPU 15 determines whether the BP correction value has been calculated in the past at the same zoom position as the BP correction value, and if so, the process must proceed to step S1404. If yes, the process proceeds to step S1410.

  In step S1404, the CPU 15 determines whether BP correction value calculation has been performed in the past with an aperture value substantially equal to the aperture value at the time of calculating the BP correction value. If it has been performed, the process proceeds to step S1405. If not, the process proceeds to step S1410.

  At this time, if the difference between the aperture values is within a predetermined value, it is determined that the BP correction value has been calculated in the past with substantially the same aperture. Here, the smaller the aperture value is, the brighter the aperture value, the greater the degree of change in the BP correction value due to the change in aperture value. For this reason, when the aperture value when the BP correction value is calculated is small, the range of the predetermined value that is determined to be substantially equal is small. For example, F1.8 or less, within 0.1 stage, F1.8 or more to F2.8 or less, within 0.2 stage, F2.8 or more and F4.0 or less, within 0.3 stage, F4.0 or more, If it is within the range of 0.5 or less at F5.6 or less and within 0.7 or more than F8.0, it is determined that the aperture value is substantially equal to the past value.

  In step S1405, the CPU 15 determines whether the BP correction value has been calculated in the past at an image height substantially equal to the image height at the time of calculating the BP correction value, and if it has been performed, the process proceeds to step S1406. If not, the process proceeds to step S1410.

  At this time, if the difference between the center coordinates of the AF frames is within a predetermined value, it is determined that the BP correction value has been calculated in the past with a substantially equal image height. The further away from the center and the higher the image height, the greater the degree of change in the BP correction value due to the change in image height. For this reason, the higher the image height of the center coordinate of the AF frame for which the BP correction value is calculated, the smaller the range of predetermined values that are determined to be substantially equal. For example, if the image height is 30% or less, the image height is within 5% of the diagonal sensor length. If the image height is more than 30% and less than 50%, the image height is in the past within 3% of the diagonal sensor length. Judged to be approximately equal to Also, if the image height is more than 50% and less than 80%, it is within 2% of the diagonal sensor length. If the image height is more than 80%, it is within 1% of the diagonal sensor length. Judged to be approximately equal to

  In step S1406, the CPU 15 determines whether the BP correction value has been calculated in the past at an ambient temperature that is substantially equal to the ambient temperature when the BP correction value was calculated, and if it has been performed, the process proceeds to step S1407. If not, the process proceeds to step S1410.

  At this time, if the difference in ambient temperature is within a predetermined temperature (for example, within ± 2 ° C.), it is determined that the BP correction value has been calculated in the past at substantially the same ambient temperature.

  In step S1407, the CPU 15 determines whether the BP correction value has been calculated in the past at the lens barrel temperature substantially equal to the lens barrel temperature at which the BP correction value was calculated, and if so, the process proceeds to step S1408. If not, the process proceeds to step S1410.

  At this time, if the difference in the lens barrel temperature is within a predetermined temperature (for example, within ± 5 ° C.), it is determined that the BP correction value has been calculated in the past at the lens barrel temperature.

  In step S1409, the CPU 15 determines that the BP correction value has been calculated in the past in a shooting environment that is substantially equal to the shooting environment in which the BP correction value is calculated.

  In step S1410, the CPU 15 determines that the BP correction value has not been calculated in the past in substantially the same shooting environment.

  In this embodiment, when signals for phase difference AF, scan AF, and RFAF are generated, a difference in sensitivity due to a difference in color of a color filter formed on each pixel may be a problem. In the present embodiment, in order to solve this, the sensitivity difference is corrected according to the reciprocal of the sensitivity ratio of each light source color temperature for each color component.

  For example, since the red filter and the blue filter have correction coefficients for the green filter for each color temperature, the sensitivity difference is corrected by multiplying the coefficients of the red filter and the blue filter by the coefficients. This correction coefficient is the reciprocal of the sensitivity ratio of the red and blue filters to the green filter. For a color temperature having no correction coefficient, the correction coefficient is obtained by linear interpolation from the correction coefficients of the color temperatures on both sides.

  In this embodiment, the case where one AF frame is set has been described. However, when a plurality of AF frames are set, the same processing is repeated for the number of AF frames.

  In the present embodiment, the case where the signal read immediately after the switch SW1 is turned on is used for the phase difference AF processing. However, the phase difference AF signal acquired when performing the LCD display in step S3 is always recorded, and the signal immediately before SW1 is recorded. The signal acquired in step (2) may be used for the phase difference AF process.

  In some cases, for example, when the position (AF point) is selected on the screen for detecting the in-focus position by the photographer, the region where the phase difference AF is performed may be limited to a part on the screen. In this case, it is possible to read out only the signal in that region and shorten the acquisition time of the signal for performing phase difference AF, or read out only the signal in that region and stop driving the sensor for a certain period after the reading is completed. This may save power.

  As described above, the display quality can be improved by increasing the display update rate to the display device, and reading can be performed by reducing the number of rows for separately performing reading from the photodiode for the reference pixel and the reference pixel. The rate can be increased and AF can be performed at high speed.

  Further, in this embodiment, AF processing is performed using a signal that has not been added and read simultaneously with display on the display device, and the focus correction amount that is mainly caused by the optical characteristics of the photographing lens is obtained and recorded in the photographing environment. Thereby, even when the shooting environment such as the temperature or the posture of the imaging apparatus changes, the focus correction amount adapted to the shooting environment can be used. Therefore, AF accuracy can be improved regardless of the shooting environment. Also, since the already recorded focus correction amount can be used in the same shooting environment, it is possible to increase the AF speed.

  The second embodiment is different from the first embodiment in that the reference image signal and the reference image signal from the image sensor 5 for performing the phase difference AF process and the RFAF process are rearranged by the readout circuit of the image sensor 5. Different. Another difference is that, after the focus lens group 3 is controlled to a substantially in-focus position using the result of the phase difference AF processing before SW1, RFAF processing is performed with a signal that has undergone image correction.

  A line for generating a signal from the image sensor 5 for obtaining an image to be displayed on the LCD 10 and a line for generating two sets of signals of a reference image signal and a reference image signal from the image sensor 5 for performing a phase difference AF operation. And 2 are set in units of two lines in common with the first embodiment. Further, the embodiment 1 is also common in that the image displayed on the LCD 10 and a signal for performing the phase difference AF operation are obtained under different exposure conditions and obtained at different readout rates. In the following description, description of other parts common to the first embodiment is omitted.

  Hereinafter, the operation procedure in the present embodiment will be described with reference to FIG.

  First, in step S1501, the CPU 15 initializes variables used for processing such as an RFAF focusing flag and sets a sensor mode drive mode. The sensor drive mode that is set is the same sensor drive mode that is set in step S1 of the first embodiment.

  In step S1502, the CPU 15 performs AE processing and sets the exposure condition of the image sensor 5. When the exposure condition (exposure time) is set, the TX signal and the RS signal are controlled as shown in FIG.

  In step S1503, the CPU 15 displays the image acquired by the image sensor 5 on the LCD.

  In step S1504, the CPU 15 performs phase difference AF processing to obtain an approximate focus position. At this time, in order to perform high-speed processing with low power consumption, the circuit in the image sensor 5 rearranges the standard image signal and the reference image signal, and then performs phase difference AF processing without image correction and defocusing. Find the amount.

  Next, in step S1505, the CPU 15 determines whether or not the focus is achieved by the phase difference AF. If it is determined that the subject is in focus, the process proceeds to step S1511. If it is determined that the subject is not in focus, the process proceeds to step S1506.

  In step S1506, the CPU 15 determines whether or not the defocus amount Def obtained as a result of the phase difference AF process is equal to or smaller than a predetermined value NE · Fδ. If the defocus amount Def is less than or equal to the predetermined value NE · Fδ, the process proceeds to step S1507, and RFAF processing is performed. If the defocus amount Def is greater than the predetermined value NE · Fδ, the process proceeds to step S1510.

  In step S1510, the CPU 15 drives the focus lens group 3 by an amount corresponding to the defocus amount obtained in step S1504, thereby moving to a position that is considered to be the approximate focus position (focus position) at that time. . When the movement is finished, CPU 15 advances the process to step S1511.

  In step S1507, the CPU 15 corrects the image obtained by rearranging the standard image signal and the reference image signal in a circuit in the image sensor 5, and then performs RFAF processing to obtain a focus position. In step S1508, the CPU 15 determines whether or not the focus position has been obtained by RFAF processing in step S1507. If it is determined, the process proceeds to step S1509. If it is determined that the focus position has not been determined, the process proceeds to step S1511.

  In step S1509, the CPU 15 drives the focus lens group 3 to the obtained in-focus position, turns on the in-focus flag, and proceeds to step S1511.

  In step S <b> 1511, the CPU 15 confirms the state of the operation switch unit 24. If it is confirmed that the operation switch unit 24 has been operated by the photographer and SW1 (the first stroke of the operation switch unit 24) has been turned on, the process proceeds to step S1512. If the ON state of SW1 cannot be confirmed, the process returns to step S1502. .

  In step S1512, the CPU 15 determines whether or not the focus flag is on. If the focus flag is on, the process proceeds to step S1514. If the focus flag is off, the process proceeds to step S1513.

  In step S1513, the CPU 15 performs phase difference AF processing, RFAF processing, and lens driving based on the result, and proceeds to step S1514. In the phase difference AF processing and RFAF processing performed here, the circuit in the image sensor 5 rearranges the standard image signal and the reference image signal. Other than that, the processing is the same as that of the first embodiment. Note that since the scan AF is not performed, the sensor drive mode is not changed.

  In step S1514, when the CPU 15 determines that focusing is possible as a result of the phase difference AF process and the RFAF process, the CPU 15 displays that AF has been completed. If it is not determined that focusing is possible, a display indicating that AF has not been completed is displayed.

  Next, in step S1515, the CPU 15 determines whether or not SW2 (second stroke) of the operation switch unit 24 is turned on. If the switch is turned on, the process proceeds to step S1516. The process returns to S1502.

  In step S1516, the CPU 15 executes an exposure process under the exposure conditions set in the AE process in step S1502, and ends the photographing operation of the camera 1.

  In parallel with this processing, the CPU 15 determines whether or not the shooting scene has changed in step S1520. If it is determined that there is no scene change, this process is repeated. If it is determined that there is a scene change, the process advances to step S1501. In step S1501, the CPU 15 initializes the calculation result so far and performs the processing shown in FIG. 15 again.

  Next, the flow of processing performed in step S1513 will be described in detail with reference to FIG.

  In step S1801, the CPU 15 performs phase difference AF processing. In the phase difference AF process, after obtaining an image at the current focus lens position, the difference between the reference image signal (A image) and the reference image signal (B image) rearranged by the circuit in the image sensor 5 is obtained. At this time, image correction is performed by a multiplier in the image sensor 5 while shifting the pixel position of the reference image signal (B image). Then, a difference signal sequence is created from the absolute value of the difference between the A image and the B image, and the defocus amount is calculated from the shift amount that minimizes the integrated value of the difference signal and the sensitivity (K value) related to the focus. Thereby, an approximate focus position is obtained. If the defocus amount obtained here is larger than the predetermined value, the CPU 15 performs the phase difference AF process in step S1801 again.

  When the approximate focus position is obtained in step S1801, the CPU 15 starts RFAF processing in step S1802. In this RFAF process, after the focus lens group 3 is moved to the in-focus position obtained by the phase difference AF process, an image is acquired at the focus lens position. Then, the reference image signal (B image) is added to the reference image signal (A image) rearranged by the circuit in the image sensor 5. At this time, image correction is performed by a multiplier in the image sensor 5 while shifting the pixel position of the reference image signal (B image). Then, a sum signal sequence (addition data) is created from the sum of the A and B images, a signal indicating the focus level is generated from the signal sequence, and the relative shift amount at which the value indicating the focus level is maximized is determined. The position is determined as the final focus position.

  While performing this RFAF processing, the CPU 15 checks the state of SW2 in step S1803. If SW2 is turned on, the process proceeds to step S1804. If SW2 is not turned on, the process proceeds to step S1806.

  In step S1804, the CPU 15 determines whether the BP correction value is calculated in substantially the same shooting environment. The method is the same as in Example 1. If a BP correction value has already been calculated in substantially the same shooting environment, the process proceeds to step S1805, and if not, the process proceeds to step S1806.

  In step S1805, the CPU 15 interrupts the RFAF process, reads out the BP correction value in the substantially equal shooting environment recorded in the predetermined recording area, and proceeds to step S1808.

  In step S1806, the CPU 15 continues the RFAF process as it is.

  In step S1807, the CPU 15 determines whether or not the RFAF processing has ended. If it is determined that the RFAF processing has ended, the process proceeds to step S1808. If it is determined that the RFAF processing has not ended, the process returns to step S1803.

  In step S1808, the CPU 15 moves the focus lens group 3 to a position where the BP correction value read in step S1805 or the BP correction value obtained in the RFAF process is added to the result of the phase difference AF process obtained in step S1801. To drive.

  As a result, if SW1 is kept on, the BP correction amount corresponding to the shooting environment is obtained, and a highly accurate focusing operation is possible. When SW2 is turned on simultaneously with SW1, imaging can be performed by performing a high-speed focusing operation using a BP correction value obtained in the past in an environment substantially equal to the current shooting environment.

  Next, the phase difference AF process and the RFAF process performed in step S1504, step S1507, and step S1513 will be described in more detail with reference to FIGS. 16 (A) and 16 (B).

  The AF process is performed by the CPU 15 controlling the arithmetic circuit in the image sensor 5. FIG. 16A is a schematic diagram of the inside of the arithmetic circuit in the image sensor 5, and FIG. 16B is a more detailed view of the inside of the buffer circuit / multiplier and the arithmetic circuit & memory shown in FIG. FIG.

  The light flux of the subject that has passed through the microlens 1601 is photoelectrically converted by the photodiode 1602. In FIG. 16A and FIG. 16B, the pixel indicated as “A” outputs a reference image signal (A image), and the pixel indicated as “B” outputs a reference image signal (B image). A signal output from each pixel is transmitted to the first transfer path 1604 and the B image to the second transfer path 1605 via the transfer transistor 1603. In this way, using separate transfer paths, the separation (rearrangement) of the A and B images is facilitated.

  Thereafter, the A image of the first transfer path is output to the first buffer circuit 1701, and the B image of the second transfer path is output to the second buffer circuit 1702, and is temporarily recorded.

  The CPU 15 transmits a signal to the buffer circuits / multipliers 1606 and 1607 and performs calculation of a desired pixel by the multiplier 1703, the addition circuit 1608, and the subtraction circuit 1609.

  The signal added by the adding circuit 1608 is recorded in a recording area (memory) 1707 in the image sensor 5. The signal recorded in the recording area 1707 is input to a BPF (band pass filter) 1708. This BPF is set so as to extract a high frequency component such as a contour component of the input signal (subject image), and its output is in-focus as in the AF evaluation value obtained by the scan AF processing circuit 14 shown in FIG. A coefficient indicating the degree.

  The signal subtracted by the subtracting circuit 1609 is converted into an absolute value by the absolute value circuit 1610 and then recorded in the recording area 1710 in the image sensor 5. The signal recorded in the recording area 1710 is output to the adding circuit 1711, and the sum of the output of the adding circuit 1711 and the value recorded in the recording area 1712 is obtained and recorded in the recording area 1712. That is, a value to be recorded in the recording area 1712 is fed back. As a result, the integrated value of the output of the absolute value circuit 1610 is recorded in the recording area 1712.

  The values recorded in the recording areas 1709 and 1712 are appropriately transferred to the CPU 15 by DMA transfer.

  In the phase difference AF processing performed in step S1504, it is necessary to perform high-speed processing with low power consumption. Therefore, image correction is performed after the standard image signal and the reference image signal are rearranged in the circuit in the image sensor 5. Arithmetic processing is performed without performing it. That is, the CPU 15 can be set so that the signal passes through the multiplier 1704 and the multiplier 1705 without being processed.

  When starting the phase difference AF process in step S1504, the CPU 15 first performs an initialization process such as clearing the value of the recording area 1712. Thereafter, the CPU 15 outputs the accumulated charge from the image sensor 5 and transfers the output from the first transfer path 1604 and the second transfer path 1605 to the buffer circuits 1701 and 1702 for recording. Then, the CPU 15 sets the buffer circuit 1701 so as to output the base image portion used for the difference calculation of the central portion of the image, and sets the buffer circuit 1702 so as to output the reference image having the same number of pixels as the base image. .

  At this time, the CPU 15 sets so that the output signals of the pixels recorded in the buffer circuits 1701 and 1702 are output to the subtraction circuit 1609 in order from the top pixel. The output signal from the subtraction circuit 1609 is converted into an absolute value by the absolute value circuit 1610 and recorded in the recording area 1710. The signal recorded in the recording area 1710 is output to the adding circuit 1711, added with the signal recorded in the recording area 1712, and recorded in the recording area 1712.

  Since the value in the recording area 1712 is cleared in the initial state, the absolute value of the difference between the top pixels is recorded in the recording area 1712 first. Next, since the buffer circuits 1701 and 1702 output the output signals of the pixels next to the top pixel next to the top pixel to the subtraction circuit 1609, respectively, the absolute value of the difference between the pixels adjacent to the top pixel and between the top pixels The absolute value of the difference is added and the result is recorded in the recording area 1712. When this process is repeated and the integrated value of the absolute value of the difference between the A and B images for a predetermined pixel is recorded in the recording area 1712, DMA transfer is performed.

  When the integrated value of the absolute value of the difference between the A image and the B image calculated from the top pixel is transferred to the CPU 15, the CPU 15 calculates the integrated value and the pixel shift amount (-N pixel from the top pixel). Recording is performed in a recording area (not shown) of the CPU 15. This processing may be performed in synchronization with an interrupt signal for completion of DMA transfer. The CPU 15 records the integrated value together with the shift amount (−N + 1) pixel in a recording area (not shown) by similarly processing the output signal of the predetermined pixel from the pixel adjacent to the top pixel. The CPU 15 repeats this process to record the integrated value in a predetermined range (pixel shift amount—N pixels to N pixels) in the CPU 15 and obtain the minimum value.

Next, the CPU 15 calculates the defocus amount d (the amount and direction of movement of the focus lens group) from the shift amount giving the minimum value, the sensitivity (K value) concerning the focus, and the image shift amount Δ during focusing as follows. Obtained by the formula. A position where the defocus amount d is zero is set as a general focus position.
d = K value × (δ−Δ)
In the phase difference AF process in step S1513, image correction using the multiplier 1703 is added to the above process. This image correction is the same processing as described in the first embodiment, and is executed by the CPU 15 setting the image correction coefficient used in the first embodiment in the multiplication coefficient setting units 1706 and 1707 of the multiplier 1703. .

  Next, the RFAF process performed in steps S1507 and S1513 will be described in detail.

  When the CPU 15 starts the RFAF processing in step S1507 or step S1513, the CPU 15 performs initialization processing such as variable processing. Thereafter, the accumulated charge is output from the image sensor 5, and the output is transferred from the first transfer path 1604 and the second transfer path 1605 to the buffer circuits 1701 and 1702 and recorded. Then, the CPU 15 sets the buffer circuit 1701 so as to output the base image portion used for the addition calculation of the central portion of the image, and sets the buffer circuit 1702 so as to output the reference image having the same number of pixels as the base image. . Further, the CPU 15 sets a correction coefficient in the multiplication coefficient setting units 1706 and 1707 of the multiplier 1703 in order to perform image correction. This image correction is the same processing as that described in the first embodiment, and is performed by the CPU 15 setting the image correction coefficient used in the first embodiment in the multiplication coefficient setting units 1706 and 1707. Further, the CPU 15 performs setting so that the output signals of the pixels recorded in the buffer circuits 1701 and 1702 are output in order from the top pixel.

  Here, the reference image signal output from the buffer circuit 1701 is integrated with the correction coefficient of the multiplication coefficient setting unit 1706 by the multiplier 1704 and output to the addition circuit 1608. Similarly, the reference image signal output from the buffer circuit 1702 is integrated with the correction coefficient of the multiplication coefficient setting unit 1706 by the multiplier 1705 and output to the addition circuit 1608. The adder circuit 1608 adds the outputs of the multiplier 1704 and the multiplier 1705, and the addition result is recorded in the recording area 1707. The output from the recording area 1707 is recorded in the recording area 1709 via the BPF 1708.

  Since the BPF 1708 performs processing on a signal having a predetermined number of pixels, the BPF 1708 performs first processing after a signal having a predetermined number of pixels is input from the top pixel, and outputs it to the recording area 1709. Next, processing is performed on a signal having a predetermined number of pixels from the pixel adjacent to the top pixel, and the result is output to the recording area 1709. By repeating this process and recording the processing result by the BPF 1708 of the sum signal sequence of the A and B images for a predetermined pixel in the recording area 1709, DMA transfer is performed.

  When the processing result in the BPF 1708 of the sum signal sequence difference between the A and B images is transferred from the top pixel to the CPU 15, the CPU 15 detects the maximum value from the processed signal sequence. Then, the maximum value and the pixel shift amount (-N pixel in the case of the first pixel) are recorded in a recording area (not shown) of the CPU 15. This processing may be performed in synchronization with an interrupt signal for completion of DMA transfer.

  Next, the CPU 15 performs the same process on the output signal of a predetermined pixel from the pixel adjacent to the top pixel. As a result, the BPF processing result of the shift amount (−N + 1) pixels is transferred to the CPU 15, and the maximum value and the pixel shift amount are recorded in a predetermined recording area of the CPU 15. By repeating this process, the BPF processing result in a predetermined range (pixel shift amount—N pixels to N pixels) is transferred to the CPU 15. When the maximum value and the pixel shift amount are recorded for each pixel shift amount, the CPU 15 detects the maximum value of the entire recorded pixel shift amount and the pixel shift amount Kd that gives it.

When the pixel shift amount Kd that maximizes the BPF processing result obtained by the RFAF processing is obtained, the CPU 15 obtains the lens driving amount by the following equation.
Lens drive amount (pulse) = K · Kd ÷ Focus sensitivity ÷ Movement amount per pulse (3)
K is a sensitivity related to the focus, and is a value depending on the zoom position of the photographing lens barrel 100, the value of the diaphragm 4, the image height, and the like. Therefore, a table using these as parameters is prepared in the EEPROM 25 in advance, and values are obtained by referring to the table.

  As described above, the readout rate can be increased by reducing the number of rows that are separately read from the photodiode for the reference pixel and the reference pixel, and the display quality can be improved by increasing the display update rate of the display device. AF can be performed at high speed while improving.

  In this embodiment, if SW1 is kept on, AF processing is performed using a signal imaged by non-addition readout, and a focus correction amount generated mainly due to optical characteristics of the photographing lens is set in the photographing environment. Ask based. Thereby, even when the shooting environment such as the temperature or the posture of the imaging apparatus changes, the focus correction amount adapted to the shooting environment can be used. Therefore, AF accuracy can be improved regardless of the shooting environment. Also, since the already recorded focus correction amount can be used in the same shooting environment, it is possible to increase the AF speed.

  Further, by providing a circuit for performing the AF process inside the image sensor 5, the AF process can always be performed, and the release time lag can be shortened.

  In the first and second embodiments, a compact digital camera has been described as an example. However, the present invention can be applied to AF such as a digital video camera, a live view of a digital single-lens reflex camera, and a camera function of a mobile terminal. .

  Each embodiment described above is only a representative example, and various modifications and changes can be made to each embodiment in carrying out the present invention.

5 Solid-state image sensor (sensor)
6 Imaging circuit 15 CPU
31 RFAF processing circuit 37 Phase difference AF processing circuit

Claims (18)

A control device that has a plurality of photoelectric conversion units for one pixel and performs focus control using a signal from an image sensor in which the pixels are arranged two-dimensionally.
Obtaining a first image signal for forming a captured image output by adding signals from a plurality of photoelectric conversion units in each of the plurality of first pixels included in the first pixel region; Obtaining means for obtaining second image signals output independently from each other from the plurality of photoelectric conversion units in each of the plurality of second pixels included in the two pixel regions;
First calculation means for performing correlation calculation by a phase difference method based on the second pixel signal and determining correlation data;
Shift addition is performed to add the second pixel signal while shifting the plurality of photoelectric conversion units corresponding to each of the plurality of second pixels between the plurality of second pixels. A second computing means for determining;
And a control unit that performs the focus control based on the correlation data and the addition data.   The said control means correct | amends the said focus position by calculating | requiring a focus position by the phase difference system from the said correlation data, and calculating | requiring the maximum value of the said addition data in the said focus position. Control device. The control means determines whether or not a focus position can be obtained by the phase difference method,
If it is determined that the focus position can be obtained by the phase difference method, the focus position is obtained from the correlation data by the phase difference method, and the maximum value of the addition data at the focus position is obtained to correct the focus position. ,
When it is determined that the focus position cannot be calculated by the phase difference method, the focus position is obtained by the contrast method based on the first image signal, and the maximum value of the addition data at the focus position is obtained, whereby the focus The control device according to claim 2, wherein the position is corrected.   4. The control apparatus according to claim 2, further comprising recording means for recording a correction amount for correcting the focus position.   The recording means uses the correction amount as the image height when the image signal used for the focus control is acquired, the attitude of the imaging device, the temperature of the device, and the temperature of the imaging optical system that focuses on the imaging device. 5. The control apparatus according to claim 4, wherein recording is performed in association with an imaging environment that is at least one of an aperture value of the imaging optical system and a zoom position of the imaging optical system.   The control means acquires the correction amount corresponding to a current photographing environment with reference to the recording means, and corrects the focus position based on the correction amount acquired from the recording means. Item 6. The control device according to Item 5.   7. The control unit according to claim 6, wherein when the focus position is corrected based on the correction amount acquired from the recording unit, the focus position is not corrected based on a maximum value of the addition data. The control device described. The control means determines on / off of a switch for instructing recording of the captured image,
The focus position is corrected based on the maximum value of the added data without using the correction amount recorded in the recording unit when the switch is determined to be off. 8. The control device according to 7. The control means determines on / off of a switch for instructing recording of the captured image,
When it is determined that the switch is on, the focus position is corrected based on the correction amount from the recording unit without correcting the focus position based on the maximum value of the addition data. The control device according to any one of claims 4 to 8. When a predetermined time has elapsed since the correction amount was recorded in the recording means,
8. The control unit according to claim 4, wherein the control unit corrects the focus position based on a maximum value of the addition data without using the correction amount recorded in the recording unit. The control device according to item. The control means determines whether the shooting scene has changed,
If it is determined that the shooting scene has changed,
The first calculation unit, the second calculation unit, and the control unit initialize a calculation result and start an operation based on the second image signal acquired after the shooting scene changes. The control device according to claim 1, wherein:   The control device according to claim 1, wherein the first pixel and the second pixel have different charge accumulation times. An imaging device having a plurality of photoelectric conversion units for one pixel, and the pixels are arranged two-dimensionally;
An imaging device comprising: the control device according to claim 1.   The image pickup apparatus according to claim 13, wherein the image pickup device includes a rearranging unit that rearranges an output order of the second image signal. A control method in which focus control is performed using a signal from an image sensor in which a plurality of photoelectric conversion units are provided for one pixel and the pixels are two-dimensionally arranged.
A first image signal for forming a captured image output by adding signals from a plurality of photoelectric conversion units in each of the plurality of first pixels included in the first pixel region; Obtaining a second image signal output independently from each other from the plurality of photoelectric conversion units in each of the plurality of second pixels included in the pixel region;
Performing correlation calculation by a phase difference method based on the second pixel signal and determining correlation data;
Shift addition is performed to add the second pixel signal while shifting the plurality of photoelectric conversion units corresponding to each of the plurality of second pixels between the plurality of second pixels. A step to determine;
Performing the focus control based on the correlation data and the addition data;
A control method characterized by comprising:   The control method according to claim 15, further comprising a step of recording a correction amount for correcting a focus position by the focus control.   A program configured to cause a computer to execute the control method of the imaging apparatus according to claim 15 or 16.   A storage medium storing the program according to claim 17.