JP7134712B2 - Imaging device and its control method - Google Patents

Description

The present invention relates to phase difference detection type focus detection and focus adjustment control.

In the imaging plane phase difference detection method, phase difference information can be obtained from a plurality of photoelectric conversion units that constitute an image sensor, but there is a problem when shooting an object with low brightness or low contrast. For example, when shooting a person's face in a low-brightness environment such as at night or indoors, the image signal level will be lower than when shooting a high-brightness or high-contrast subject, and focus detection accuracy may decrease. have a nature.

Patent Literature 1 discloses a method of changing the threshold value of the standard deviation of the defocus amount based on the brightness condition and the contrast condition. A process of calculating the final defocus amount based on the defocus amount calculated in the past is performed.

JP 2018-31916 A

Patent Document 1 does not mention adaptive processing based on the frequency band condition of the filter used for correlation calculation and the standard deviation of the defocus amount. and focusing performance may deteriorate.
SUMMARY OF THE INVENTION It is an object of the present invention to implement phase difference detection type focus detection and focus adjustment control with higher accuracy when photographing a low-brightness subject or the like.

An imaging apparatus according to an embodiment of the present invention includes acquisition means for acquiring a plurality of image signals having phase differences from an imaging device, and correlation calculation is performed on the plurality of image signals using a filter having a plurality of frequency bands. calculating means for calculating a plurality of pieces of focus detection information and standard deviations of the focus detection information; and control means for controlling focus adjustment using the plurality of pieces of focus detection information and standard deviations. When the first focus detection information calculated by the correlation calculation in the first frequency band is selected from among the plurality of frequency bands, the control means controls the focus detection information in the first frequency band. When the standard deviation of the first focus detection information calculated by the correlation calculation is larger than a threshold value , the focus adjustment is controlled using the first focus detection information, and out of the plurality of frequency bands, the When second focus detection information calculated by correlation calculation in a second frequency band lower than the first frequency band is selected, calculation by correlation calculation in the second frequency band The focus adjustment is controlled using the obtained second focus detection information, and the first focus detection information calculated by the correlation calculation in the first frequency band is selected from the plurality of frequency bands. and the standard deviation of the first focus detection information is equal to or less than the threshold, the first focus detection information and the first frequency prior to the first focus detection information The focus adjustment is controlled using the third focus detection information calculated from the focus detection information obtained by the correlation calculation in the band.

According to the present invention, it is possible to achieve more accurate phase difference detection type focus detection and focus adjustment control when photographing a low-brightness subject or the like.

1 is a block diagram showing a configuration example of an imaging system according to an embodiment of the present invention; FIG. It is a figure explaining the pixel structure of an image pick-up element. 5 is a flow chart for explaining an example of imaging processing according to the present embodiment; 4 is a flowchart for explaining automatic focus detection processing according to the embodiment; FIG. 10 is an explanatory diagram relating to the amount of correlation change between two images; FIG. 10 is a diagram showing a threshold value of standard deviation of defocus amount with respect to photographing sensitivity; FIG. 10 is a diagram showing a threshold value of standard deviation of defocus amount with respect to amplitude of an image signal; FIG. 10 is a diagram showing thresholds of standard deviations of defocus amounts with respect to contrast evaluation values; 7 is a flowchart for explaining frame addition determination calculation processing according to the embodiment; FIG. 10 is a flow chart showing processing subsequent to FIG. 9 ; FIG. FIG. 10 is a diagram showing a threshold of a blur evaluation value DFD with respect to a sharpness evaluation value;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In this embodiment, an interchangeable lens camera including a lens device and a camera body will be described, but the present invention can also be applied to an imaging device in which the lens is integrated with the camera body.

FIG. 1 is a block diagram showing a configuration example of an imaging system according to this embodiment. The imaging system of this embodiment is composed of a lens device 10 and a camera main body 20 . The lens control unit 106 that controls the operation of the lens device 10 and the camera control unit 212 that controls the operation of the entire imaging system can exchange information with each other.

First, the configuration of the lens device 10 will be described. The fixed lens unit 101, diaphragm 102, and focus lens unit 103 constitute an imaging optical system. A fixed lens portion 101 is a first group lens portion. A diaphragm 102 is driven by a diaphragm drive unit 104 to control the amount of light incident on an imaging device 201, which will be described later. A focus lens unit 103 is driven by a focus lens driving unit 105 to adjust the focus of the imaging optical system. A diaphragm drive unit 104 and a focus lens drive unit 105 are controlled by a lens control unit 106 .

The lens control unit 106 includes a CPU (Central Processing Unit), and when receiving a user's operation instruction from the lens operation unit 107, performs control according to the operation instruction. For example, the lens control unit 106 determines the opening amount of the diaphragm 102 and the position of the focus lens unit 103 according to control commands and control information received from the camera control unit 212 described later, and controls the diaphragm driving unit 104 and the focus lens driving unit. It controls the unit 105 . The lens control unit 106 also transmits lens control information to the camera control unit 212 . The lens operation unit 107 has operation members such as switches operated by the user.

Next, the configuration of the camera body section 20 will be described. The camera body 20 acquires an imaging signal from light that has passed through the imaging optical system. The imaging device 201 is a CCD (charge-coupled device) type image sensor or a CMOS (complementary metal oxide semiconductor) type image sensor. Light passing through the imaging optical system forms an image on the light-receiving surface of the imaging device 201, and is converted into signal charges according to the amount of incident light by a photoelectric conversion unit forming a pixel portion of the imaging device 201. FIG. The signal charges accumulated in the photodiodes constituting the photoelectric conversion unit are sequentially read out from the image sensor 201 as voltage signals corresponding to the signal charges based on the drive pulse given from the timing generator 215 in accordance with the command from the camera control unit 212. . The pixel configuration of the image sensor 201 will be described later with reference to FIG.

A CDS/AGC/AD converter (hereinafter referred to as an imaging signal processing unit) 202 acquires an imaging signal read from the imaging device 201 and an AF (automatic focusing) signal. Here, CDS (correlated double sampling), AGC (automatic gain control), and AD (analog-digital) conversion are performed to remove reset noise, and a digital signal is output. The imaging signal processing unit 202 outputs an imaging signal to the image input controller 203 and outputs a signal for imaging plane phase difference AF to the AF signal processing unit 204 .

The image input controller 203 acquires the imaging signal output from the imaging signal processing unit 202 and stores it in the memory 209 via the bus 21 . The memory 209 is, for example, SDRAM (Synchronous Dynamic Random Access Memory). A signal read from the SDRAM is sent to the display control unit 205 via the bus 21 and processed. The display control unit 205 outputs image signals for display to the display unit 206 . A display unit 206 includes a display device such as a liquid crystal panel, and displays an image on a screen according to an image signal.

Further, in the imaging signal recording mode, the recording medium control unit 207 acquires the signal read from the memory 209 via the bus 21 and controls recording in the recording medium 208 . A ROM (Read Only Memory) 210 connected to the bus 21 stores control programs executed by the camera control unit 212 and various data required for control. The flash ROM 211 stores various setting information (user setting information, etc.) related to the operation of the imaging apparatus.

The AF signal processing unit 204 performs a correlation operation on the AF image signal output from the imaging signal processing unit 202, and calculates an image shift amount and reliability information. The reliability information includes information indicating the degree of matching between two images, contrast information, saturation information, flaw information, and the like. The AF signal processing unit 204 outputs the calculated image shift amount and reliability information to the camera control unit 212 .

A camera control unit 212 is a central unit that controls the imaging system, and includes a CPU. Based on the image shift amount and the reliability information acquired from the AF signal processing unit 204, the camera control unit 212 notifies the AF signal processing unit 204 of changes in settings when calculating them. For example, when the amount of image shift is larger than a threshold value, a process of setting a wide area for correlation calculation, a process of changing the characteristics of the bandpass filter according to the contrast information of the captured image, and the like are performed. The camera control unit 212 transmits and receives information to and from the lens control unit 106 and executes various types of processing. The camera operation unit 214 includes operation members used for operations by the user, a touch panel on the display screen, and the like, and notifies the camera control unit 212 of operation instructions from the user. The camera control unit 212 executes processing related to camera functions such as power ON/OFF, setting change, recording start, AF control start, recorded image confirmation, etc., according to an input signal from the camera operation unit 214. do. The camera control unit 212 also transmits control commands and control information for the movable lens to the lens control unit 106 and receives information from the lens control unit 106 .

A camera control unit 212 includes an AF control unit 213 and cooperates with the lens control unit 106 to perform focus adjustment control of the imaging optical system. The AF control unit 213 is a characteristic component of the present invention, and performs focus control to bring a predetermined subject into focus.

The pixel configuration of the image sensor will be described with reference to FIG. FIG. 2A is a schematic diagram showing a normal pixel configuration that is not based on the imaging plane phase difference detection method. R, Gr, Gb, and B represent the colors of the color filter, and four pixels make up one set.

FIG. 2B is a schematic diagram showing the pixel configuration of the imaging plane phase difference detection method. The imaging device 201 includes two photodiodes for each pixel unit in order to perform imaging plane phase difference AF. The example of FIG. 2B shows a pixel portion having a photoelectric conversion portion divided into two with respect to the pixel configuration of FIG. 2A. This is an example in which two photodiodes are arranged for one microlens. The light that has passed through the imaging optical system is separated by a microlens and images are formed on the light receiving surfaces of the two photodiodes. The two photodiodes are distinguished by the addition of A and B symbols for R, Gr, Gb and B. A signal (A+B signal) obtained by adding the signals of the two photodiodes is an imaging signal. For example, an R signal for imaging can be acquired from the addition signal obtained from the signals of the photodiodes respectively corresponding to RA and RB. Signals (A signal and B signal) of individual photodiodes are two image signals for AF. For example, two image signals for AF can be obtained from a photodiode signal (A signal) corresponding to RA and a photodiode signal (B signal) corresponding to RB.

In this way, a signal for imaging can be obtained from the A+B signal, and two signals for AF can be obtained from the A signal and the B signal. The AF signal processing unit 204 performs a correlation operation on two image signals based on the AF signal, and calculates an image shift amount and reliability information.

In this embodiment, an example in which an imaging signal and two image signals for AF, that is, a total of three signals are extracted from the imaging element 201 is shown, but the present invention is not limited to such an example. For example, considering the processing load of the image sensor 201, the A+B signal and the A signal (or the B signal) may be acquired. In this case, two signals are read out from the imaging device, and the B signal (or A signal) can be generated by calculating the difference between the A+B signal and the A signal (or B signal). As the pupil-splitting type imaging device, an imaging device having a pixel structure having three or more photoelectric conversion units for one microlens may be used. In this case, phase difference information can be acquired from the output of a pair of photoelectric conversion units having parallax among the plurality of photoelectric conversion units.

The imaging process of this embodiment will be described with reference to FIG. FIG. 3 is a flow chart for explaining an outline of photographing processing. The following processing is realized by executing the control program read from the ROM 210 by the CPU of the camera control section 212 . First, in S501, the camera control unit 212 determines whether or not the first switch (denoted as SW1) for photographing is on. A camera operation unit 214 includes a two-stage switch operated by a release button. The user can issue an instruction to prepare for shooting by operating the first switch SW1, and can issue an instruction to start the imaging operation by operating the second switch (referred to as SW2). If it is determined that SW1 is on, the process proceeds to S502, and if it is determined that SW1 is off, the determination process of S501 is repeated in the standby state.

In S502, the camera control unit 212 performs automatic focus detection processing. Details of the automatic focus detection process will be described later with reference to FIG. In S503, the camera control unit 212 acquires focus detection information (image shift amount, defocus amount), and determines whether the focus detection information calculated in S502 is within a predetermined focus range. If it is determined that the focus detection information is within the focus range, the process proceeds to S505. If it is determined that the focus detection information is outside the in-focus range, the process proceeds to S504.

In step S<b>504 , the camera control unit 212 transmits a command to drive the focus lens unit 103 to the lens control unit 106 . The lens control unit 106 performs drive control of the focus lens unit 103 based on the focus detection information calculated in S502. After S504, the process returns to S502.

In S505, the camera control unit 212 determines whether the second switch SW2 is on. If it is determined that SW2 is on, the process proceeds to S506, and if it is determined that SW2 is off, the process proceeds to S507. After the imaging operation is performed in S506, processing for recording image data after imaging is performed in the next S508. After finishing the recording process, the process ends. Also, in S507, the camera control unit 212 determines whether SW1 is on. If it is determined that SW1 is on, the process returns to S505. If it is determined that SW1 is off, the process ends.

Next, automatic focus detection processing will be described with reference to FIG. FIG. 4 is a flowchart for explaining the processing shown in S502 of FIG. The following processing is performed by the AF control unit 213 and the AF signal processing unit 204. FIG.

In S601, the AF control unit 213 instructs the AF signal processing unit 204 to perform correlation calculation in order to perform focus detection calculation. An AF signal processing unit 204 performs a correlation calculation on a pair of phase difference image signals acquired from the image sensor 201 . Then, the AF control unit 213 calculates the defocus amount based on the image shift amount, which is the amount of shift at which the correlation amount obtained from the AF signal processing unit 204 becomes the minimum value. Correlation calculation is performed using a plurality of filters with different band characteristics. For example, four types of filters with different frequency bands are used, such as for low frequency, middle-low frequency, middle frequency, and high frequency. These filters are used to calculate four types of defocus amounts for low, middle and low, middle, and high frequencies. Since the configuration of each filter is publicly known, detailed description is omitted. The AF control unit 213 also instructs the AF signal processing unit 204 to calculate the correlation amount between the pair of phase difference image signals for each shift amount. The AF control unit 213 generates a correlation amount waveform for each shift amount acquired from the AF signal processing unit 204 .

In S602, processing for calculating the amount of correlation change between two images is performed. Calculation processing will be described with reference to FIG. FIG. 5 is a graph showing an example of correlation change amount (change amount of correlation amount) when driving the focus lens from a state where the degree of blur is large to near the in-focus position in imaging plane phase difference AF. The horizontal axis represents the degree of blurring of the subject. The blur increases as the distance from the in-focus position increases. The vertical axis represents the amount of correlation change (hereinafter referred to as MAXDER). The correlation change amount MAXDER can be calculated by the following formula (1).
MAXDER(k)=(COR[k-3]-COR[k-1])
-(COR[k-2]-COR[k]) (1)

k in the formula (1) is a variable for specifying the position and takes an integer value. COR[k] represents the amount of correlation between the two images at the position indicated by the variable k. Correlation change amounts MAXDER for low, medium and low, middle, and high frequencies corresponding to four types of filters with different frequency bands are calculated. It can be seen that in the imaging plane phase difference detection AF, the value of the correlation change amount MAXDER increases as the degree of blurring approaches the in-focus state (the state in which the subject is in focus).

In S603 of FIG. 4, the AF control unit 213 performs a process of calculating the standard deviation of the defocus amount based on the correlation change amount MAXDER. If the standard deviation of the defocus amount is expressed as Defocus3σ, it can be calculated by the following equation (2).
Defocus3σ=K×(A×MAXDER̂B) (2)

K in Expression (2) is a conversion coefficient for converting the image shift amount into the defocus amount. A and B are coefficients of a conversion formula for converting the amount of correlation change MAXDER into the standard deviation of the amount of image shift. "^" represents exponentiation. By substituting the correlation change amount MAXDER for the low-range, middle-low-range, mid-range, and high-range corresponding to the four types of filters with different frequency bands calculated in S603 into equation (2), A standard deviation Defocus3σ of the four types of defocus amounts is calculated.

In S604, in order to calculate a reliability evaluation value representing the reliability of the defocus amount, a process of calculating a threshold value of the standard deviation Defocus3σ of the defocus amount is performed. The Defocus3σ threshold is calculated based on the brightness condition and the contrast condition. A specific description will be given with reference to FIG.

FIG. 6 is a graph showing the relationship between the shooting sensitivity and the threshold value of the standard deviation of the defocus amount with respect to the luminance conditions. The horizontal axis represents the shooting sensitivity and indicates Sv1, Sv2, and Sv3, respectively. The vertical axis represents the threshold value of the standard deviation Defocus3σ of the defocus amount (denoted as Defocus3σLMN), indicating Th1, Th2, and Th3, respectively. Thresholds Th1 to Th3 correspond to Sv1 to Sv3, respectively, and Th1<Th2<Th3.

As shown in FIG. 6, the threshold Defocus3σLMN is calculated based on the shooting sensitivity in the shooting scene. When the shooting brightness is low, there is a correlation between the shooting brightness and the shooting sensitivity, and processing is performed to increase the shooting sensitivity and make the exposure appropriate. When the photographing sensitivity is high, noise increases and the standard deviation Defocus3σ of the defocus amount increases. Therefore, the threshold value Defocus3σLMN is set larger as the photographing sensitivity increases. In the present embodiment, the reliability evaluation value is increased to relax the focusing conditions, thereby making it easier to focus. Here, the calculation of the threshold Defocus3σLMN is performed with respect to the most reliable reliability evaluation value (Rel3 described later) for determining the focusing condition in order to relax the focusing condition. The details of the order of reliability evaluation values will be described later.

Thresholds Th2 and Th3 shown in FIG. 6 are determined based on the amplitude (denoted as PB), which is the difference between the maximum and minimum values of the image signal. A specific description will be given with reference to FIG. FIG. 7 is a graph showing the relationship between the amplitude PB on the horizontal axis and the threshold value Defocus3σLMN on the vertical axis. Amplitude PB is calculated by AF signal processing section 204 or AF control section 213 .

FIG. 7A shows a case where the imaging sensitivity is Sv2, and shows a threshold Th2_1 corresponding to the amplitude PB1 and a threshold Th2_0 corresponding to the amplitude PB2. PB1<PB2 and Th2_1>Th2_0. If the amplitude PB is less than PB1, the threshold is Th2_1. Also, when the amplitude PB is greater than PB2, the threshold is Th2_0. When the amplitude PB is greater than or equal to PB1 and less than or equal to PB2, the threshold linearly decreases within the range of Th2_1 to Th2_0 as the amplitude PB increases. FIG. 7B shows the case where the imaging sensitivity is Sv3, and shows the threshold Th3_1 corresponding to the amplitude PB3 and the threshold Th3_0 corresponding to the amplitude PB4. PB3<PB4 and Th3_1>Th3_0. If the amplitude PB is less than PB3, the threshold is Th3_1. Also, when the amplitude PB is greater than PB4, the threshold is Th3_0. When the amplitude PB is greater than or equal to PB3 and less than or equal to PB4, the threshold linearly decreases within the range of Th3_1 to Th3_0 as the amplitude PB increases.

As shown in FIG. 7, the threshold Defocus3σLMN is variably set according to the amplitude PB, which is the difference between the maximum and minimum values of the image signal. The amplitude PB is a value that changes according to the brightness of the subject. When the PB value is small (see PB1 and PB3), the brightness of the subject is low. processing is performed.

FIG. 7 shows an example in which the thresholds Th2 and Th3 have a linear relationship with the PB value in the range from PB1 to PB2 and in the range from PB3 to PB4. Without being limited to this, a method of setting the PB value and the threshold in a non-linear relationship or a method of stepwise correspondence between the two may be adopted. Further, the luminance evaluation value is not limited to the amplitude PB of the image signal, but may be the maximum or minimum value of the image signal or the luminance value of the object.

Contrast conditions will be described with reference to FIG. FIG. 8 is a graph for explaining a method of determining the Defocus3σ threshold based on the contrast evaluation value. The horizontal axis represents the ratio PB/Peak normalized by dividing the value of the amplitude PB by the maximum value Peak of the image signal, indicating PB_Peak1 and PB_Peak2. A contrast evaluation value expressed as a normalized ratio PB/Peak is calculated by the AF signal processing section 204 or the AF control section 213 . The vertical axis represents the Defocus3σ threshold value (Defocus3σCNT) corresponding to the contrast evaluation value, indicating Th4 and Th5. PB_Peak1<PB_Peak2. The relationship between the threshold Th5 corresponding to PB_Peak1 and the threshold Th4 corresponding to PB_Peak2 is Th5>Th4. If PB/Peak is less than PB_Peak1, the threshold is Th5. Also, when PB/Peak is greater than PB_Peak2, the threshold is Th4. When PB/Peak is greater than or equal to PB_Peak1 and less than or equal to PB_Peak2, the threshold linearly decreases within the range of Th5 to Th4 as PB/Peak increases.

When the value of PB/Peak is small, the contrast of the subject is low, so the threshold Defocus3σCNT is set to a large value. In the present embodiment, the reliability evaluation value is increased to relax the focusing conditions, thereby making it easier to focus. Calculation of the threshold Defocus3σCNT is performed for the most reliable contrast evaluation value that determines the focusing condition in order to relax the focusing condition. Note that a method of setting the PB/Peak and the threshold in a non-linear relationship or a method of stepwise correspondence between the two may be adopted without being limited to the example of FIG. 8 .

The AF signal processing unit 204 or the AF control unit 213 compares the value of the threshold Defocus3σLMN obtained from FIG. 7 and the value of the threshold Defocus3σCNT obtained from FIG. The larger of the two thresholds is determined as the threshold Defocus3σ of the standard deviation Defocus3σ.

In S605 of FIG. 4, a process of calculating a reliability evaluation value (denoted as Rel) representing the reliability of the defocus amount is performed. The reliability evaluation value Rel is determined in four stages, for example. The four types of reliability evaluation values are denoted as Rel3, Rel2, Rel1, and Rel0 in descending order of reliability. These can be calculated by the following formula (3).
Rel
= Rel3 if (Defocus3σ≤Def3σTH3)
Rel2 if (Def3σTH3<Defocus3σ≤Def3σTH2)
Rel1 if (Def3σTH2<Defocus3σ≤Def3σTH1)
Rel0 if (Def3σTH1≤Defocus3σ)
... (3)

Def3σTH3, Def3σTH2, and Def3σTH1 in Equation (3) are thresholds for the standard deviation Defocus3σ of the calculated defocus amounts. Def3σTH3 is the threshold calculated in S604 of FIG. An inequality in if( ) indicates a conditional expression.

When the reliability evaluation value Rel=Rel3, the reliability of the calculated defocus amount is high. This means that focusing is allowed, and corresponds to a condition where focusing is possible. Further, when the reliability evaluation value Rel=Rel2, it means that focusing may be performed only during frame addition processing of the defocus amount, which will be described later, and corresponds to a condition in which focusing is possible.

If the reliability evaluation value Rel=Rel1, it means that the direction of the calculated defocus amount is correct, and corresponds to a condition in which focusing is not possible. Further, when the reliability evaluation value Rel=Rel0, it means that the reliability of the defocus amount is the lowest, and corresponds to a condition in which focusing is impossible. Using the above equation (3), the reliability evaluation values for the defocus amounts for the low-range, medium-low range, medium-range, and high-range calculated in S601 are calculated.

In S606, a process of selecting one defocus amount is executed using the four types of defocus amounts calculated in S601 and the reliability evaluation value calculated in S605. For example, the defocus amount having the highest reliability evaluation value is selected from among the defocus amounts for the low range, middle-low range, middle range, and high range.

In S607, the AF control unit 213 performs frame addition determination calculation. The details of the frame addition determination calculation will be described later. In S<b>608 , the AF control unit 213 determines whether or not to perform frame addition with respect to the defocus amount calculated in the past frame based on the determination result obtained in S<b>607 . If it is determined to perform frame addition, the process advances to step S609, and if it is determined not to perform frame addition, the series of processes shown in FIG. 4 ends.

In S609, the AF control unit 213 performs averaging processing of the defocus amount. An addition average value of the defocus amount calculated for the current frame and the defocus amount calculated for the past frame is calculated. For example, the data calculated using the defocus amount for m frames acquired in the past and the defocus amount data calculated for the current frame are added and averaged to obtain the final defocus amount. After the amount is calculated, the process ends.

The frame addition determination calculation of S607 in FIG. 4 will be described with reference to the flowcharts shown in FIGS. In S901, the AF control unit 213 determines whether the defocus amount selected in S606 of FIG. Perform judgment processing. The accuracy of the defocus amount for the low range near the in-focus position is lower than the accuracy of the defocus amounts for the high, mid, and low-mid ranges, so the defocus amount for the low range is selected. If so, the process proceeds to S911. If the defocus amount for low frequencies has not been selected, the process proceeds to S902.

In S902, it is determined whether or not the reliability evaluation value Rel of the defocus amount selected in S606 of FIG. 4 is equal to Rel2 of the four types of reliability evaluation values. The reliability evaluation value Rel2 means that focusing is possible only during frame addition processing regarding the defocus amount. If Rel=Rel2 (determination result is true), the process proceeds to S903, and if Rel=Rel2 is not (determination result is false), the process proceeds to S911. In S903, it is determined whether or not the defocus amount selected in S606 of FIG. 4 is the defocus amount calculated using the high-frequency filter, that is, the high-frequency defocus amount. be. If the defocus amount for the high frequency band has a reliability evaluation value of Rel2, the accuracy of the defocus amount in the vicinity of the in-focus position is higher than that for the mid-range band, so frame addition processing is performed. By doing so, it is determined that focusing is possible. If the defocus amount for high frequencies has been selected, the process proceeds to S910. If the defocus amount for high frequencies has not been selected, the process proceeds to S904.

In S<b>904 , the AF control unit 213 performs a process of calculating an evaluation value DFD representing the degree of blurring of the image and the sharpness of the image. The evaluation value DFD can be calculated by the following formula (4).
DFD=Σ(A[k]+B[k]) ² /((Σ(A[k]) ² +Σ(B[k]) ² )/2)
... (4)

k in equation (4) is a variable for specifying the position and takes an integer value. A[k] and B[k] are the signal values of the A image and the B image at the position indicated by the variable k, respectively. Σ is the symbol for the summation operation with respect to k.

The sharpness of the image can be calculated by the following formula (5).
Sharpness=Σ(S[k+1]−S[k]) ² /Σ(S[k+1]−S[k])
... (5)

k in equation (5) is a variable for specifying the position and takes an integer value. S[k] is the signal value of the phase difference image signal at the position indicated by the variable k. Σ is the symbol for the summation operation with respect to k. The sharpness evaluation value sharpness/PB is calculated by dividing the sharpness calculated from the equation (5) by the amplitude PB of the image signal.

In S905, determination processing is executed to determine whether the evaluation value DFD calculated in S904 is equal to or greater than a predetermined threshold, or whether the sharpness evaluation value Sharpness/PB calculated in S904 is equal to or greater than a predetermined threshold. be done. A threshold for the evaluation value DFD (DFD threshold) is denoted as DFD_PARAM_TH, and a threshold for Sharpness/PB is denoted as SHARPNESS_TO_PB_TH. The first determination condition in S905 is as follows.
(I) DFD≧DFD_PARAM_TH or Sharpness/PB≧SHARPNESS_TO_PB_TH

If the determination condition (I) is satisfied, the AF control unit 213 determines that the focus lens position is in the vicinity of the in-focus position, that is, a condition that enables focusing, and advances the processing to S910. If the determination condition (I) is not satisfied, the process proceeds to S906.

In S<b>906 , the AF control unit 213 updates the DFD threshold based on the sharpness evaluation value Sharpness/PB in order to relax the focusing condition when shooting a low-luminance or low-contrast subject. A specific description will be given with reference to FIG. FIG. 11 is a graph showing the relationship between the sharpness evaluation value Sharpness/PB on the horizontal axis and the DFD threshold on the vertical axis. FIG. 11 illustrates graph lines for Sv1, Sv2, and Sv3 with different photographing sensitivities, where Sv1<Sv2<Sv3.

If Sharpness/PB is greater than SHARPNESS_TO_PB_TH, the DFD threshold is a constant value shown in DFD_PARAM_TH. When Sharpness/PB is smaller than SHARPNESS_TO_PB_RLX_TH, the DFD threshold differs for each shooting sensitivity. That is, for the shooting sensitivity Svj (j=1-3), the DFD threshold is DFD_PARAM_RLX_Svj_TH. DFD_PARAM_RLX_Sv1_TH>DFD_PARAM_RLX_Sv2_TH>DFD_PARAM_RLX_Sv3_TH.

If Sharpness/PB is greater than or equal to SHARPNESS_TO_PB_RLX_TH and less than or equal to SHARPNESS_TO_PB_TH, the DFD threshold increases as Sharpness/PB increases. For example, the DFD threshold linearly changes in the range from DFD_PARAM_RLX_Svj_TH to DFD_PARAM_TH according to the shooting sensitivity Svj (j=1-3).

As shown in FIG. 11, the DFD threshold is variably set according to the imaging sensitivity. The higher the imaging sensitivity, the smaller the DFD threshold. By doing so, when photographing an object with low contrast under low luminance conditions, it is possible to appropriately change the focusing condition according to changes in luminance conditions.

After S906 in FIG. 9, the process proceeds to Z907 in FIG. In S907, the AF control unit 213 executes determination processing regarding the second determination condition described below.
(II) DFD≧DFD_PARAM_Svj_TH and PB/Peak≦PB_TO_PEAK_TH
DFD_PARAM_Svj_TH is the DFD threshold updated in S906. Threshold PB_TO_PEAK_TH is a threshold for the contrast evaluation value PB/Peak for determining whether or not the subject has high contrast.

If the determination condition (II) is satisfied, that is, if the evaluation value DFD representing the degree of image blur is greater than or equal to the changed DFD threshold and PB/Peak is less than or equal to the threshold, the focus lens position is in focus. It is determined to be near the location. In this case, the process proceeds to S908. If the determination condition (II) is not satisfied, the process proceeds to S911.

In S908, determination processing is performed to determine whether the defocus amount selected in S606 of FIG. executed. If the defocus amount for middle-low range is selected, the process proceeds to S909. If the defocus amount for the mid-low range is not selected (that is, if the defocus amount for the mid-range is selected), it is determined that focusing is possible, and the process proceeds to S910.

In S909, the standard deviation Defocus3σ of the defocus amount for middle and low frequencies calculated in S603 of FIG. 4 is compared with a predetermined threshold. The threshold is represented as Def3σTH2LOW. This is a threshold for determining whether or not focusing is possible by performing frame addition processing with respect to the defocus amount calculated using the filter for middle and low frequencies. The threshold value Def3σTH2LOW is calculated according to the shooting sensitivity, the brightness of the subject, and the contrast of the subject, similarly to the threshold value Def3σTH3 calculated in S604. Alternatively, the threshold may be calculated only from the shooting sensitivity and the brightness of the subject, or the focusing condition may not be relaxed by changing the threshold based on the contrast of the subject. If it is determined in S909 that Defocus3σ is equal to or less than the threshold value Def3σTH2LOW, it is determined that focusing is possible, and the process advances to S910. If it is determined in S909 that Defocus3σ is greater than the threshold value Def3σTH2LOW, the process proceeds to S911.

In S910, the AF control unit 213 sets ON the flag for frame addition determination calculation. That is, it is determined that the frame addition related to the defocus amount is to be performed, and the processing ends after the flag indicating the determination result is set to ON. On the other hand, in S911, it is determined that the frame addition related to the defocus amount is not performed, and after the flag for frame addition determination calculation is set to OFF, the processing is terminated and the process proceeds to return processing.

In this embodiment, the focusing condition is adaptively relaxed based on the frequency band of the correlation calculation filter and the standard deviation of the defocus amount when photographing under low luminance conditions or when photographing a low-contrast subject. Reduces the probability of shooting scenes that may be judged to be out of focus, even though it should be judged to be in focus. At the same time, it is possible to increase the shooting scenes in which focusing can be achieved with high accuracy. According to the present embodiment, in AF control of a phase difference detection method (including an imaging plane phase difference detection method), more accurate focus detection and focus adjustment can be realized when shooting a low-brightness subject or the like.

103: Focus lens unit 105: Focus lens driving unit 106: Lens control unit 201: Image sensor 204: AF signal processing unit 212: Camera control unit 213: AF control unit

Claims (13)

Acquisition means for acquiring a plurality of image signals having phase differences from the imaging device;
calculating means for performing a correlation operation on the plurality of image signals using a filter having a plurality of frequency bands to calculate a plurality of focus detection information and a standard deviation of the focus detection information;
a control means for controlling focus adjustment using the plurality of focus detection information and standard deviation;
The control means is
A case where first focus detection information calculated by correlation calculation in a first frequency band is selected from among the plurality of frequency bands, and the first focus detection information calculated by correlation calculation in the first frequency band is selected. if the standard deviation of the first focus detection information is greater than a threshold , controlling the focus adjustment using the first focus detection information;
When second focus detection information calculated by correlation calculation in a second frequency band lower than the first frequency band is selected from among the plurality of frequency bands, the second focus detection information is selected . controlling the focus adjustment using the second focus detection information calculated by the correlation calculation in the frequency band of
A case where first focus detection information calculated by correlation calculation in a first frequency band is selected from among the plurality of frequency bands, and the standard deviation of the first focus detection information is the threshold value In the case of the following, the third focus detection information calculated from the first focus detection information and the focus detection information obtained by the correlation calculation in the first frequency band prior to the first focus detection information An imaging apparatus, wherein focus detection information is used to control the focus adjustment. The control means selects one of the plurality of focus detection information calculated by the calculation means by performing correlation calculations using the filters having the plurality of frequency bands, and selects a threshold for the standard deviation of the focus detection information. 2. The imaging apparatus according to claim 1, wherein control is performed to change the . When the first focus detection information calculated by the correlation calculation in the first frequency band is selected from among the plurality of frequency bands, the control means controls the first focus detection information When the standard deviation is equal to or less than the threshold , the first focus detection information and the focus detection information calculated by correlation calculation in the first frequency band prior to the first focus detection information. 3. The imaging apparatus according to claim 2, wherein the focus adjustment is controlled using the third focus detection information obtained by averaging the above. The calculation means calculates a reliability evaluation value of the focus detection information,
3. The method according to claim 2, wherein the control means selects one piece of focus detection information from the plurality of pieces of focus detection information using the reliability evaluation value when an image of a subject with low brightness is captured by the image sensor. The imaging device described. The control means determines whether or not to control the focus adjustment using the selected focus detection information and focus detection information calculated from focus detection information calculated prior to the focus detection information. 5. The imaging device according to claim 4, characterized in that: 3. The image pickup apparatus according to claim 2, wherein the control unit performs control to change a threshold value for the standard deviation of the focus detection information acquired for each of the plurality of frequency bands according to brightness conditions or contrast conditions. The control means sets the threshold when the luminance evaluation value calculated from the image signal is a first value to the threshold when the luminance evaluation value is a second value larger than the first value. 7. The imaging device according to claim 6, wherein the threshold is set larger than the threshold. 8. The imaging apparatus according to claim 7, wherein the calculation means or the control means calculates the luminance evaluation value from the maximum value or minimum value of the image signal, or the maximum value and minimum value of the image signal. The control means sets the threshold when the contrast evaluation value calculated from the image signal is a first value to the threshold when the contrast evaluation value is a second value larger than the first value. 7. The imaging device according to claim 6, wherein the threshold is set larger than the threshold. 10. The imaging apparatus according to claim 9, wherein the calculating means or the controlling means calculates the contrast evaluation value by normalizing the amplitude of the image signal by the maximum value of the image signal. 5. The image pickup apparatus according to claim 4, wherein the calculation means calculates the reliability evaluation value using a standard deviation of the focus detection information calculated from a change amount of the correlation amount of the image signal. The control means calculates a first evaluation value representing the degree of blurring of the image and a second evaluation value representing the contrast of the image associated with the image signal. and the second evaluation value is equal to or less than a second threshold, and the standard deviation of the first focus detection information is equal to or less than the threshold, the focus detection using the third focus detection information 12. The imaging device according to any one of claims 1 to 11, wherein adjustment control is performed. A control method executed in an imaging device that acquires a plurality of image signals having a phase difference from an imaging device and performs focus adjustment,
a calculation step of performing a correlation operation on the plurality of image signals using a filter having a plurality of frequency bands to calculate a plurality of focus detection information and a standard deviation of the focus detection information;
a control step of controlling focus adjustment using the plurality of focus detection information and standard deviation;
In the control step,
A case where first focus detection information calculated by correlation calculation in a first frequency band is selected from among the plurality of frequency bands, and the first focus detection information calculated by correlation calculation in the first frequency band is selected. when the standard deviation of the first focus detection information is greater than a threshold , the focus adjustment is controlled using the first focus detection information;
When second focus detection information calculated by correlation calculation in a second frequency band lower than the first frequency band is selected from among the plurality of frequency bands, the second focus detection information is selected . the focus adjustment is controlled using the second focus detection information calculated by the correlation calculation in the frequency band of
A case where first focus detection information calculated by correlation calculation in a first frequency band is selected from among the plurality of frequency bands, and the standard deviation of the first focus detection information is the threshold value In the case of the following, the third focus detection information calculated from the first focus detection information and the focus detection information obtained by the correlation calculation in the first frequency band prior to the first focus detection information A control method, wherein the focus adjustment is controlled using focus detection information.

Images