JP2020048135A - Imaging apparatus and control method of the same - Google Patents

Abstract

To improve low illuminance performance while suppressing subject blurring of a slow shutter phase difference detection pixel when lowering a frame rate in low light in video or live view.SOLUTION: In a slow shutter mode, an accumulation time of an image signal is extended, an accumulation time of a phase difference detection signal is not extended, the number of readout rows is increased, and the phase difference detection signal is added and calculated.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an imaging apparatus and a control method thereof, and more particularly to a focus detection technique.

  2. Description of the Related Art In an image pickup apparatus, a technique has been proposed that enables a focus shift amount of a photographic lens to be obtained by incorporating a phase difference detection function into an image pickup device.

  In Patent Literature 1, a pupil division function is provided by decentering a sensitivity region of a light receiving unit with respect to an optical axis of an on-chip micro lens in a part of light receiving elements of an imaging element. The pixels having these light receiving elements are arranged at predetermined intervals in the image pickup element to realize a phase difference detecting function.

  In Patent Literature 2, a plurality of photoelectric conversion elements of A pixels and B pixels are provided in each pixel corresponding to one microlens of an imaging element, and an A pixel output and an A + B pixel output are read. Then, a B pixel output is obtained by subtracting the A pixel output and the A + B pixel output, thereby realizing a phase difference detection function.

  When detecting these phase differences, a case in which an image is captured in an environment with low illuminance will be considered.

  As described on the premise of Patent Literature 3, there is a method of extending the storage time in a low illuminance environment like a slow shutter mode. This is a widely-used control in which, even if the subject is blurred in a video camera or the like, the subject only needs to be imaged for the time being.

  At this time, the accumulation time of the phase difference detection pixels is also prolonged, but the image is liable to shake the subject, which adversely affects the phase difference detection. On the other hand, if the accumulation time is kept short, the signal level is low in the first place, so that it is no longer a point of detection of the phase difference.

JP 2010-219958 A JP 2013-106194 A Japanese Patent No. 5004888

  Patent Document 3 attempts to solve the problem by applying a gain. However, applying the gain lowers the S / N, which also has a bad effect on phase difference detection.

  Even if the number of lines including the phase difference detection pixels is increased by the reading method as in Patent Document 1, the number of pixels increases to HDTV or 4K depending on the pixel reading speed of the sensor, and the frame rate also increases from 30 fps to 60 or 120 fps. There is a limit in increasing.

In order to solve the above-described problems, an imaging device according to the present invention includes:
Unit pixels for photoelectrically converting light from a subject are arranged in a matrix, a first signal for image generation based on charges generated in the unit pixels, and a position based on charges generated in a partial area of the unit pixels. An image sensor that outputs a second signal for detecting a phase difference;
A scanning circuit that controls scanning for reading out the first signal and the second signal for each row;
With
The scanning circuit,
A first scan for thinning out the first signal at a first cycle and reading the first signal;
A second scan for thinning out and reading out the second signal at a predetermined thinning rate every second cycle for a row from which the first signal is not read out in the first scan;
A third scan in which the third signal is read out by thinning out at a predetermined thinning rate every third cycle longer than the second cycle, for a row from which the first signal is not read out in the first scan;
In an imaging device that performs
In the case of the slow shutter mode, the first scanning and the second scanning are performed,
The accumulation time of the photoelectric conversion of the first signal is longer than in the slow shutter mode,
The accumulation time of the photoelectric conversion of the second signal is equal to that in modes other than the slow shutter mode,
Performing an addition operation in the second scanning row;
In modes other than the slow shutter mode,
Performing the first scan and the third scan,
It is characterized by the following.

  ADVANTAGE OF THE INVENTION According to the imaging device which concerns on this invention, the accumulation | storage time of an imaging pixel is extended in low illuminance, and the accumulation | storage time of a phase difference detection pixel increases the number of read-out rows of a phase difference detection pixel, reducing a subject blurring as it is. Accordingly, it is possible to provide an imaging apparatus with improved SN at low illuminance and improved phase difference detection performance, and a control method thereof.

It is a flowchart explaining operation | movement of the moving image photography process of this embodiment. It is a block diagram of an imaging device of this embodiment. FIG. 2 is a schematic diagram illustrating a pixel configuration of an image sensor according to the embodiment. FIG. 3 is a schematic diagram for explaining a method for reading out the image sensor according to the embodiment. FIG. 3 is a schematic diagram for explaining a lapse of time of driving of the image sensor according to the embodiment. 9 is a table illustrating exposure control in a moving image shooting process according to the embodiment. 5 is a graph illustrating a correlation amount calculation according to the embodiment.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

  FIG. 2 is a block diagram illustrating a configuration of a lens and a lens-interchangeable camera body according to the present embodiment. In the present embodiment, an example of a lens-interchangeable camera will be described. However, a lens-integrated camera may be used.

[Configuration of Imaging Device]
As shown in FIG. 2, the imaging device according to the present embodiment includes a lens 20 and a camera 21.

  The lens 20 is provided with a lens control unit 206 that controls the entire operation of the lens. The camera 21 is provided with a camera control unit 224 that controls the overall operation of the camera. The camera control unit 224 and the lens control unit 206 can communicate with each other. When the imaging apparatus is a lens-integrated type, all functions and means of the lens control unit 206 may be configured to be included in the camera control unit 224. The overall operation of the device is controlled.

  First, the configuration of the lens 20 will be described. The lens 20 includes a fixed lens 201, an aperture 202, a focus lens 203, an aperture control unit 204, a focus control unit 205, a lens control unit 206, and a lens operation unit 207. In the present embodiment, an imaging optical system includes the fixed lens 201, the aperture 202, and the focus lens 203. The fixed lens 201 is a fixed first group lens disposed closest to the subject of the lens 20. The diaphragm 202 is driven by a diaphragm control unit 204 and controls the amount of light incident on an image sensor 212 described later. The focus lens 203 is disposed closest to the image plane side of the lens 20 and is driven in the optical axis direction by the focus control unit 205 to adjust a focus for forming an image on an image sensor 212 described later. The lens control unit 206 controls the aperture control unit 204 to determine the opening amount of the aperture 202 and the focus control unit 205 to determine the position of the focus lens 203 in the optical axis direction. When a user operation is performed by the lens operation unit 207, the lens control unit 206 performs control according to the user operation. The lens control unit 206 controls the aperture control unit 204 and the focus control unit 205 according to a control command or control information from a camera control unit 224 described later. Further, the lens control unit 206 transmits lens control information (optical information) to the camera control unit 224.

  Next, the configuration of the camera 21 will be described. The camera 21 has the following configuration so that an imaging signal can be obtained from a light beam that has passed through the imaging optical system. That is, it includes a shutter 211, an image sensor 212, an analog front end (AFE) 213, a DSP (Digital Signal Processor) 214, a camera control unit 224, a timing generator 225, and a shutter control unit 226. The camera 21 includes a focus detection signal processing unit 215, a bus 216, a display control unit 217, a display unit 218, a recording medium control unit 219, a recording medium 220, a RAM 221, a ROM 222, a flash ROM 223, and a camera operation unit 227.

  The shutter 211 adjusts the exposure time of the image sensor 212 described below under the control of the camera control unit 224. The imaging element 212 is a photoelectric conversion element (photodiode PD) configured by a CCD or a CMOS sensor. The light beam (subject image) that has passed through the imaging optical system of the lens 20 is formed on the light receiving surface of the imaging element 212, and is converted by the photodiode PD (photoelectric conversion unit) into signal charges corresponding to the amount of incident light. The signal charge accumulated in each photodiode is converted from the image sensor 212 as a voltage signal (imaging signal, focus detection signal) corresponding to the signal charge based on a driving pulse given from the timing generator 225 according to a command from the camera control unit 224. Read sequentially.

  In the present embodiment, the image sensor 212 holds two photodiodes PD for one pixel unit as shown in FIG. 3B in order to perform focus detection by the phase difference detection method. The light beam that has passed through the entire area of the exit pupil EP of the imaging optical system TL is separated by the microlens ML, and the two photodiodes PD form an image, so that two signals for imaging and focus detection can be extracted. ing. The signal A + B obtained by adding the signals of the two photodiodes PD is an imaging signal, and the signals A and B of the individual photodiodes are two image signals (focus detection signals) for focus detection. At least a part of the exit pupil EP differs between one photodiode PD (first photoelectric conversion unit) and the other photodiode PD (first photoelectric conversion unit) of the two photodiodes PD (overlapping). (Not shown) each of which receives a light beam passing through the pupil region. That is, on the image sensor 212, the first photoelectric conversion unit and the first photoelectric conversion unit that photoelectrically convert a pair of light beams that have passed through pupil areas at least partially different from each other (not overlapping) among the exit pupils EP of the imaging optical system. A plurality of pixel units each having two photoelectric conversion units are two-dimensionally arranged. This enables focus detection over the entire area within the imaging screen.

  In the present embodiment, focus detection refers to detection of an image shift amount. Based on the focus detection signal, a focus detection signal processing unit 215, which will be described later, performs a correlation operation on the two (a pair of) image signals to calculate an image shift amount and various pieces of reliability information. FIG. 3A is an enlarged view of a part of a pixel portion of the image sensor 212 according to the present embodiment, and adopts a pixel configuration called a primary color Bayer array. Specifically, a two-dimensional single-chip CMOS color image sensor in which primary color filters of R (Red), G (Green), and B (Blue) are arranged in a Bayer array is used. Here, R shown in FIG. 3A represents an R pixel, B represents a B pixel, and Gr and Gb represent G pixels.

  An image signal and a focus detection signal read from the image sensor 212 are input to an analog front end (AFE) 213, where correlated double sampling (CDS) for removing reset noise, gain adjustment (AGC), and signal Digitization (AD) and also clamps the dark offset level.

  The DSP 214 performs various correction processing, development processing, and compression processing on the imaging signal (image signal) output from the AFE 213. Further, the DSP 214 performs a process of correcting various noises generated in the image sensor 212, a process of detecting a defective pixel, a process of correcting a defective pixel and a focus detection signal, a process of detecting a saturated pixel, and a process of correcting a saturated pixel.

  After that, the DSP 214 stores the imaging signal (image signal) in the RAM 221. The imaging signal (image signal) stored in the RAM 221 is displayed on the display unit 218 by the display control unit 217 via the bus 216. In addition, in the mode in which the recording of the image pickup signal is performed, the image is recorded on the recording medium 220 by the recording medium control unit 219. The ROM 222 connected via the bus 216 stores a control program executed by the camera control unit 224, various data necessary for control, and the like. The flash ROM 223 stores various kinds of setting information regarding the operation of the camera 21 such as user setting information. Further, the DSP 214 outputs a focus detection signal to the focus detection signal processing unit 215.

  The focus detection signal processing unit 215 performs a correlation operation based on two (a pair of) image signals for focus detection output from the DSP 214 to calculate an image shift amount. After that, the detected image shift amount is output to the camera control unit 224. Further, the camera control unit 224 notifies the focus detection signal processing unit 215 of a change in the setting for calculating these based on the acquired image shift amounts. For example, when the image shift amount is large, the area for performing the correlation calculation is set wide, or the type of the band-pass filter is changed according to the contrast information. In the present embodiment, the signal processing device includes at least the focus detection signal processing unit 215 and a camera control unit 224 described later.

  In the present embodiment, a total of three signals of the imaging signal and two (a pair of) image signals for focus detection are extracted from the imaging element 212, but the present invention is not limited to such a method. In consideration of the processing load of the image sensor 212, for example, a total of two of the image signal and one image signal of the focus detection signal are extracted, and another focus detection signal is generated by taking the difference between the image signal and the focus detection signal. Such control may be performed.

  The camera control unit 224 controls by exchanging information with the entire camera 21. Not only the processing in the camera 21 but also various operations operated by the user, such as power ON / OFF, release button, change of settings, start of recording, and confirmation of recorded video, in response to input from the camera operation unit 227. Execute the camera function. The release button includes a release switch SW1 that is turned on by a first stroke operation (half-press operation) of the release button operated by a user, and a release switch SW2 that is turned on by a second stroke operation (full-press operation) of the release button. Is connected. In addition, as described above, information is exchanged with the lens control unit 206 in the lens 20, a control command or control information is transmitted to the lens, and lens control information (optical information) in the lens is obtained.

  Here, a driving method of the image sensor 212 according to the present embodiment will be described with reference to FIG. FIG. 4A is a diagram illustrating a pixel region of the image sensor 212. The vertical line regions on the left and upper sides in the pixel regions shown in FIGS. 4A to 4E indicate optical black portions (OB portions) that are shielded from light. In addition, row numbers are shown on the left side of the pixel area in FIG.

  The image pickup signal is read out in the vertical direction while thinning out at the first cycle. This first scan is hereinafter referred to as “first scan”. Then, after the entire surface is scanned in the vertical direction, the vertical scanning is returned to the upper row in the vertical direction again. This time, the rows that have not been read in the first scan are again scanned while thinning out in the second cycle. This second scanning is hereinafter referred to as “second scanning”. In the first scan, an imaging signal for generating an image is read from the unit pixels of the target row. On the other hand, in the second scan, the focus detection signals of the A pixel and the B pixel in the unit pixel of the target row are sequentially read. Then, an image is generated using the read image signal, and the image shift amount is detected based on the focus detection signal.

  FIG. 4B is a schematic diagram of a readout row when performing the first scan and the second scan. In FIG. 4B, the row surrounded by a thick frame is a row to be read in the first scan, and the other rows are rows that are thinned out without being read in the first scan. The row with the diamond pattern in FIG. 4B is the row to be read in the second scan.

  In the example shown in FIG. 4B, after the image signal of the V0 line is read, the image signal is read from the V1 line after the first cycle (3 lines in FIG. 4). Subsequently, in the same first cycle, the imaging signals are sequentially read from the V2, V3, V4, V5, V6, and V7 rows. The reading from V0 to V7 is the first scan. The row read by the first scan outputs a charge signal generated in the entire pixel disposed below the same microlens ML suitable for image generation, and an image is generated from the imaging signals of the V0 to V7 rows. You.

  In the present embodiment, since reading is performed without thinning in the column direction (horizontal direction), the number of pixels read out in the horizontal direction and the vertical direction is different, and the aspect ratio of the image is different. However, the conversion of the aspect ratio may be performed at the subsequent stage, or the general thinning-out reading or the addition thinning-out reading may be performed at the same ratio in the horizontal direction, and the aspect ratio can be converted by an arbitrary method.

  After reading out to the V7th row in the first scan, the process returns to the V8th row, and among the pixels included in the V8th row, first, the signal A is read out from the A pixel. Subsequently, the signal B is read from the B pixel among the pixels included in the V8th row of the same row. After that, the signal A is read from the pixel A among the pixels included in the V9th row after the second cycle (one row after in FIG. 4). Subsequently, the signal B is read from the B pixel among the pixels included in the V9th row of the same row. After that, the V4 line already read out in the first scan is skipped, and the A pixel and the B pixel in the V10 line after the next second cycle (after one line), and further after the next second cycle ( The A pixel and the B pixel on the V11th row (after one row) are read out. Similarly, the V5 line already read in the first scan is skipped, and the V12 line and the next V13 line are sequentially read. The reading from V8 to V13 is the second scan.

  As described above, in the second scan, the signal A is read from the pixel A and the signal B is read from the pixel B for the same row. If the pixel area for detecting the focus is set in advance as the focus detection area, it is preferable that the narrowest row range including the focus detection area be the target row range for the second scan. For example, in FIG. 4B, when the focus detection area is set in advance in the range from the V8th row to the V13th row, the second to the V8th row to the V13th row are set as in the above-described operation. This is the range of rows to be scanned. The focus detection area may be specified by the user using an operation unit (not shown), may be automatically set by a known method such as subject detection, or may be a fixed area. There may be.

  When the pixel arrangement is rearranged in the order in which the data is read out and processed, a schematic diagram shown in FIG. 4C is obtained. As described above, since the imaging signals are read out in the first scanning in the V0 to V7 rows, normal image generation can be performed using the imaging signals. On the other hand, since the focus detection signals are read from the A pixel and the B pixel in the second scan in the V8 to V13 rows, an image shift is performed based on the pair of signals A and B obtained for each row. It is possible to detect the quantity.

  FIG. 4D illustrates a third scan in which the second cycle in the second scan described with reference to FIG. 4B is increased and the number of readout rows is reduced.

  Similarly, a row surrounded by a bold frame is a row to be read in the first scan, and the other rows are rows to be thinned out without being read in the first scan. The row with the diamond pattern in FIG. 4D is the row to be read in the third scan.

  As the first scan, after reading the image signal on the V0 line, the image signal is read from the V1 line after the first cycle, and the image signals are sequentially read from the V2, V3, V4, V5, V6, and V7 lines. . An image is generated from the imaging signals on the V0 to V7 rows.

  After the first scan is performed up to the V7th row, the process returns to the V8th row, and the signal A is read from the pixel A of the pixels included in the V8th row, and the signal B is read out from the B pixel as described above. Thereafter, the signal A is read from the pixel A among the pixels included in the V9th row after the third cycle (after the second row in FIG. 4), and the signal B is read from the B pixel. Thereafter, the V4th row already read out in the first scan is skipped, and the A pixel and the B pixel in the V10th row after the next third cycle (after the second row) are read out. The reading from V8 to V10 is the third scan.

  When the pixel arrangement is rearranged in the order in which the data is read and processed, a schematic diagram shown in FIG. As described above, since the imaging signals are read out in the first scanning in the V0 to V7 rows, normal image generation can be performed using the imaging signals. On the other hand, in the rows V8 to V10, since the focus detection signals are read from the A pixel and the B pixel in the second scan, the image shift is performed based on the signal A and the signal B, which are a pair of signals obtained for each same row. It is possible to detect the quantity. The calculation of the image shift amount will be described later with reference to FIG.

  As described above, in the second scanning, by reducing the second period, the number of rows for reading out the focus detection signal increases. By adding these rows, a noise reduction effect at the time of detecting a phase difference can be expected. This is particularly advantageous in low light situations. On the other hand, in the third scan, the number of readout rows of the focus detection signal is reduced by increasing the third cycle. However, the total number of rows is reduced as shown in FIG. It is shortened, and it is possible to cope with a high frame rate.

  Next, the time lapse of the driving of the above-described image sensor 212 will be described with reference to the schematic diagram of FIG.

  In FIG. 5, the horizontal axis represents the passage of time, each representing one cycle of reading, and this is repeated forever. Time a is twice as long as time b. Usually, such an image sensor reading requires a blanking period for performing processing of the next frame, but is not shown in this schematic diagram.

  The time during which the read operation is performed for each row is displayed in a frame corresponding to each row. The length of each row indicates the exposure period, and schematically indicates that resetting of the row is performed at the left end timing and reading is started at the right end timing.

  In FIG. 5B, the reset release of each row is performed so that the accumulation time of each line becomes the same in consideration of the time required for reading of the image sensor 212. FIG. 5A will be described later.

  The first scan and the second scan shown in FIGS. 4B and 4C correspond to FIG. 5A, and the first scan and the third scan shown in FIGS. 4D and 4E correspond to FIG. 3B. Corresponding to

  Note that the same reference numerals in FIG. 5 denote the same parts as in FIG. Therefore, FIG. 5A shows that rows are read from V0 to v13b, and FIG. 5B shows that rows are read from V0 to V12b in the order described in FIG.

  Here, the actual camera operation will be described with reference to FIGS.

  As an example, suppose that a moving image was shot at a frame rate of 60 frames per second. The same applies to live view display for still image shooting regardless of moving image shooting. FIG. 5B shows a state in which 60 frames are shot for the second. The reading in this case is as shown in FIGS. The time b of one frame at this time is as follows.

1000 (ms) /60≒16.7 (ms)
In the readout performance of the image sensor 212, the row reset release, exposure, and readout are performed with the same exposure time for each row so that the luminance in the screen is the same.

  In order to allow the exposure time to be as long as possible under these restrictions, in FIG. 5B, the number of rows of focus detection pixels to be read is reduced. This realizes 60 frames per second.

  At this time, it is assumed that the exposure time needs to be increased in a low illuminance environment or the like. However, 16.6 (ms) limit is used as described above, and the exposure time cannot be further extended.

  Here, it is assumed that the mode shifts to the slow shutter mode and shifts to a frame rate of 30 frames per second as in Patent Document 3. The time a for one frame in this case is as follows.

1000 (ms) /30≒33.3 (ms)
At this time, conventionally, as shown in FIG. 5C, both the image signal and the focus detection signal extend the exposure time by the same amount. Note that the read control is as shown in FIGS. This control is appropriate because the image signal is prioritized in the dark even if there is some subject blurring. However, if the exposure time of the focus detection signal is also extended, focus detection may not be possible due to subject shake. This cannot achieve the purpose of focus detection.

  Therefore, as shown in FIGS. 5A, 4B, and 4C, the exposure time of the image signal is extended to make the time a ≒ 33.3 (ms), but the exposure time of the focus detection signal is not extended. Then, since the SN ratio of the focus detection signal is lowered at low illuminance, the number of readout rows is increased. Thus, while the subject blur is equivalent to 60 frames per second, the number of readout rows can be increased and the SN ratio can be improved by the increased time b.

[Explanation of focus detection]
FIG. 7 is an image signal obtained from the focus detection areas V8 to V13 in FIG. s to t represent the focus detection range, and p to q are the ranges required for the focus detection calculation based on the shift amount. Further, x to y represent one divided focus detection area.

(A) is a diagram showing the image signal before the shift by a waveform. A solid line 701 is the image signal A, and a broken line 702 is the image signal B.

  (B) is a diagram shifted in the plus direction with respect to the image waveform before the shift in (A), and (C) is a diagram shifted in the minus direction with respect to the image waveform before the shift in (A). When calculating the correlation amount, 701 and 702 are shifted one bit at a time in the directions of the arrows. Next, a method of calculating the correlation amount COR will be described.

  First, as described in (B) and (C), the image signal A and the image signal B are shifted one bit at a time, and the sum of the absolute values of the differences between the image signals A and B at that time is calculated. At this time, the shift amount is represented by i, the minimum shift number is ps in FIG. 7, and the maximum shift number is qt in FIG. Further, x is the start coordinate of the focus detection area, and y is the end coordinate of the focus detection area. Using these, it can be calculated by the following equation (1).

  The smaller the correlation amount COR [i], the higher the degree of coincidence between the A image and the B image, and the shift amount at that time becomes the defocus amount, indicating that the focus can be detected.

  In the calculation of the correlation amount, the image signal A is obtained by adding V8a and V9a, which are the second scan, in the vertical direction in the slow shutter mode. By adding V8b and V9b in the vertical direction to form the image signal B, the S / N ratio of the image signal can be improved and the performance at low illuminance can be improved.

[Operation of imaging device]
Next, the operation of the imaging apparatus according to the present embodiment will be described with reference to FIG.

  FIG. 1 is a flowchart illustrating a processing procedure for capturing a moving image by the imaging apparatus. The same applies to the live view display for still image shooting. In this moving image shooting, S101 to S110 are repeated to drive the focus of the lens to continue the focus adjustment.

  In S101 to S103, control relating to moving image recording is performed. In step S101, it is determined whether or not the moving image recording switch is turned on. If the switch is turned on, the process proceeds to step S102 to start recording a moving image. If the switch is turned off, the process proceeds to step S103 to stop recording the moving image. In this embodiment, the recording of the moving image is started and stopped by pressing the moving image recording switch. However, the recording may be started and stopped by another method such as a changeover switch.

  In S104, the exposure is measured and controlled. FIG. 6 is a program diagram in the present embodiment. The table shows that, when the illuminance is EV13, the aperture is F1.4, the accumulation time is 1/4000 (indicated by the reciprocal in FIG. 6), and the gain is controlled to be 1 time. However, the frame rate of a moving image or a live view remains at 60 frames per second if the accumulation time is 1/60 or more.

  In the case of the program diagram, if the image is darker than EV6, the slow shutter mode is determined, and the frame rate is reduced to 30 frames per second. For example, in the case of EV5, it is determined that the shutter mode is the slow shutter mode, the frame rate is set to 30 frames per second, and the aperture F1.4, the accumulation time 1/30, and the control of applying a sensor gain of twice are performed.

  In S105, if the result of the exposure control in S104 is the slow shutter mode, the process proceeds to S107; otherwise, the process proceeds to S106.

  In S106, the exposure time of the next frame is set to the normal exposure mode as the normal shutter mode. If the imaging mode is 60 frames per second, 60 frames per second is used, and the exposure time is 16.7 (ms).

  In S107, the exposure time of the next frame is extended in the slow shutter mode. In the imaging mode of 60 frames per second, the slow shutter mode sets 30 frames per second, and the exposure time is 33.3 (ms).

  In S108, the image shift amount is detected based on the pair of focus detection signals. Further, a peak value of a signal A and a signal B which are a pair of focus detection signals is detected.

  In step S109, the amount of image blur detected in step S108 is converted into a driving amount of the lens (focus lens 203), and the lens is driven according to the calculated lens driving amount. In S110, it is determined whether or not the moving image shooting processing has been stopped. If the process is to be continued, the process proceeds to S101. If the process is to be stopped, the moving image shooting process ends.

[Effects of the present embodiment]
As described above, in the present embodiment,
According to the present invention, it is possible to increase the number of readout rows of the phase difference detection pixels while extending the accumulation time of the imaging pixels at low illuminance and reducing the subject blur while keeping the accumulation time of the phase difference detection pixels unchanged. It is possible to provide an imaging apparatus with improved SN in illuminance and improved phase difference detection performance, and a control method thereof.

[Modification]
In the present embodiment, when the slow shutter mode is entered under low illuminance, the first and second scans are always performed, but there is a method in which this is not always performed.

  If the subject is moving, the first and second scans are used to perform exposure and reading as shown in FIG. 5A, and the exposure time of the phase difference detection signal is not extended. If the subject is stationary, the first and third scans may be used to perform exposure and reading as shown in FIG. 5C to extend the exposure time of the phase difference detection signal. By doing this, the S / N ratio at low illuminance can be improved.

  For example, a method using a gyro (not shown) built in the imaging apparatus, a method for tracking and determining the motion of the subject, a method for detecting a motion vector of the entire screen, and the like can be considered to determine whether the subject is moving.

  In the calculation of the correlation amount in FIG. 7, in the slow shutter mode, the correlation amount calculated as V8a, which is the second scan, as the image signal A, and V8b as the image signal B, is COR8. Similarly, the correlation amount calculated as V9a as image signal A and V9b as image signal B is COR9.

  By adding the correlation amounts COR8 and COR9, the S / N ratio of the correlation amount can be improved and the performance at low illuminance can be improved.

  As described above, the preferred embodiments of the present invention have been described, but the present invention is not limited to these embodiments, and various modifications and changes can be made within the scope of the gist.

20 lenses, 21 cameras,
214 DSP (Digital Signal Processor),
215 focus detection signal processing unit, 224 camera control unit

Claims (11)

Unit pixels for photoelectrically converting light from a subject are arranged in a matrix, a first signal for image generation based on charges generated in the unit pixels, and a position based on charges generated in a partial area of the unit pixels. An image sensor that outputs a second signal for detecting a phase difference;
A scanning circuit that controls scanning for reading out the first signal and the second signal for each row;
With
The scanning circuit,
A first scan for thinning out the first signal at a first cycle and reading the first signal;
A second scan for thinning out and reading out the second signal at a predetermined thinning rate every second cycle for a row from which the first signal is not read out in the first scan;
A third scan in which the third signal is read out by thinning out at a predetermined thinning rate every third cycle longer than the second cycle, for a row from which the first signal is not read in the first scan;
In an imaging device that performs
In the case of the slow shutter mode, the first scanning and the second scanning are performed,
The accumulation time of the photoelectric conversion of the first signal is longer than in the slow shutter mode,
The accumulation time of the photoelectric conversion of the second signal is equal to that in modes other than the slow shutter mode,
Performing an addition operation in the second scanning row;
In modes other than the slow shutter mode, the first scan and the third scan are performed.
An imaging device characterized by the above-mentioned.   2. The slow shutter mode according to claim 1, wherein in shooting a moving image, a subject has low illuminance, a frame rate of an imaging device is reduced, and a shutter speed per frame is reduced. Imaging device.   The imaging apparatus according to claim 2, wherein the moving image shooting is a moving image for a viewfinder for shooting a still image.   In the slow shutter mode, the accumulation time of the photoelectric conversion of the second signal is extended as compared with that outside the slow shutter mode, but it is shorter than the extended accumulation time of the photoelectric conversion of the first signal. The imaging device according to any one of claims 1 to 3, wherein   The imaging apparatus according to any one of claims 1 to 4, wherein, in the slow shutter mode, a correlation operation is performed after adding the increased second scan.   5. The imaging apparatus according to claim 1, wherein in the slow shutter mode, an addition is performed after performing a correlation operation on the increased second scan. 6. The unit pixel has an A pixel and a B pixel as the partial region,
The first signal is a composite signal from the A pixel and the B pixel,
The imaging device according to claim 1, wherein the second signal is one of a signal from the A pixel and a signal from the B pixel.   The imaging apparatus according to claim 1, wherein the scanning circuit performs the second scan after performing the first scan. An optical system that forms an image of light from the subject on the image sensor,
Focus adjusting means for adjusting the focus of the optical system based on the second signal;
The imaging device according to any one of claims 1 to 8, further comprising:   Even in the slow shutter mode, unless the condition is satisfied, changing the accumulation time of photoelectric conversion of the first signal, not changing the accumulation time of photoelectric conversion of the second signal, 10. The imaging apparatus according to claim 1, wherein the number of scanning lines is not changed or added.   The imaging apparatus according to claim 10, wherein the condition is that a subject is moving.