JP5600941B2 - Focus detection apparatus and imaging apparatus - Google Patents

Description

  The present invention relates to a focus detection apparatus and an imaging apparatus.

  In focus detection, a plurality of defocus amounts within a predetermined difference are extracted from a plurality of defocus amounts obtained at different aperture diameters of the optical system, and the final focus adjustment state is determined based on these defocus amounts. A focus detection apparatus that determines a defocus amount is known (for example, see Patent Document 1).

  For example, even if a vignetting caused by an imaging optical system occurs in one of a pair of focus detection light beams and a pair of signal data strings output from an image sensor cause relative distortion, this focus detection device is false. Focusing is prevented, the correlation between both signal data strings is accurately detected, and accurate focus detection is possible.

JP 2008-52009 A

  However, in the above-described conventional focus detection apparatus, the false focus prevention technology is disclosed when there are two shift amounts with which the degree of correlation is a predetermined value or more, but the case where there are three or more is disclosed. It wasn't.

A focus detection apparatus according to a first aspect of the present invention receives a pair of light beams passing through an aperture stop of an optical system and outputs a pair of focus detection signal data strings corresponding to a pair of images formed by the pair of light beams. Calculating a correlation amount between the pair of focus detection signal data sequences while relatively deviating the pair of focus detection signal data sequences when the aperture value of the aperture opening is the first aperture value; Correlation calculation means for calculating three or more image shift amounts related to the pair of images from the three or more minimum values when there are three or more minimum values of the correlation amount; and the three or more image shift amounts Each of which is multiplied by a first conversion coefficient corresponding to the first aperture value to calculate three or more first defocus amounts, and the three or more first defocuses. A first calculating three or more first differences between each of the quantities A minute calculation means calculates the three or more image shift amount difference between each of said three or more image shift amount, corresponding to the second aperture in each of said three or more image shift amount difference Second difference calculation means for multiplying the second conversion coefficient to calculate three or more second differences, and each of the three or more second differences is different from each of the three or more first differences. When the aperture value of the aperture stop is changed from the first aperture value to the second aperture value, the pair of focus detections after the change to the second aperture value is performed by the correlation calculation unit. A correlation calculation is performed on the signal data string to calculate three or more image shift amounts, and the defocus amount calculation unit multiplies each of the three or more image shift amounts by the second conversion coefficient to obtain 3 Control means for performing control for calculating at least two second defocus amounts; and at least three of the second defocus amounts. Determining means for determining a defocus amount that substantially matches any of the three or more second defocus amounts among the defocus amounts as a true defocus amount in a focus adjustment state of the optical system. Features.
An image pickup apparatus according to a seventh aspect of the present invention receives a photographing optical system and a photographing light beam that passes through a diaphragm aperture of the photographing optical system, and obtains imaging signal data corresponding to a subject image formed by the photographing light beam. A plurality of imaging pixels to be output and a pair of focus detection light beams passing through the aperture opening, and a plurality of focus detection signal data sequences corresponding to a pair of images formed by the pair of light beams are output. An image sensor in which focus detection pixels are two-dimensionally arranged and the pair of focus detection signal data strings when the aperture value of the aperture opening is the first aperture value are relatively displaced. When the correlation amount of the pair of focus detection signal data strings is calculated, and there are three or more minimum values of the correlation amount, three or more image shift amounts related to the pair of images from the three or more minimum values Correlation calculating means for calculating the above-mentioned three or more Defocus amount calculation means for calculating three or more first defocus amounts by multiplying each of the image shift amounts by a first conversion coefficient corresponding to the first aperture value, and the three or more first defocus amounts. a first difference calculation means for calculating at least three first differences between each of 1 defocus amount, and calculates the three or more image shift amount difference between each of said three or more image shift amount Second difference calculating means for calculating three or more second differences by multiplying each of the three or more image deviation amount differences by a second conversion coefficient corresponding to a second aperture value; and the three Changing means for changing the aperture value of the aperture opening from the first aperture value to the second aperture value when each of the second differences is different from each of the three or more first differences; Let the correlation calculation means perform a correlation calculation for the pair of focus detection signal data strings after being changed to the second aperture value. Three or more image shift amounts are calculated, and the defocus amount calculation means multiplies each of the three or more image shift amounts by the second conversion coefficient to obtain three or more second defocus amounts. And a control means for performing control for calculating a defocus amount that substantially coincides with any of the three or more second defocus amounts among the three or more first defocus amounts. Determining means for determining as a true defocus amount in the state.

  According to the present invention, in the pupil division type phase difference detection method, when there are three or more shift amounts at which the degree of correlation between a pair of signal data sequences output from the image sensor exceeds a predetermined value, false focusing is eliminated. Thus, an accurate defocus amount can be detected.

It is a cross-sectional view showing the configuration of the digital still camera of one embodiment. It is a figure which shows the focus detection position on the imaging | photography screen of an interchangeable lens. It is a figure explaining the principle of a pupil division type phase difference detection method. It is a front view which shows the detailed structure of an image pick-up element. It is a figure which shows the shape of the micro lens of an imaging pixel and a focus detection pixel. It is a front view of an imaging pixel. It is a front view of a focus detection pixel. It is a figure which shows the spectral sensitivity characteristic of a green pixel, a red pixel, and a blue pixel. It is a figure which shows the spectral sensitivity characteristic of a focus detection pixel. It is sectional drawing of an imaging pixel. It is sectional drawing of a focus detection pixel. It is a figure for demonstrating the mode of the imaging light beam which an imaging pixel receives. It is a figure for demonstrating the mode of the imaging light beam which a focus detection pixel receives. It is a flowchart which shows the imaging operation of a digital still camera. It is a flowchart which shows the imaging operation of a digital still camera. It is a figure which shows the relationship of the correlation amount C (k) with respect to the shift amount k of a pair of data. It is explanatory drawing in the case of converting an image shift amount into a defocus amount. It is a figure which shows the signal of the image in which false focusing generate | occur | produces. It is a figure which shows the signal of the image in which false focusing generate | occur | produces. It is a figure which shows the signal of the image in which false focusing generate | occur | produces. It is a figure which shows the fall (minimum value) of correlation amount C (k). 6 is a graph schematically illustrating linear conversion from an image shift amount to a defocus amount using a straight line with a conversion coefficient as an inclination. It is a figure which shows the signal of the image in which false focusing generate | occur | produces. It is a figure which shows the signal of the image in which false focusing generate | occur | produces. It is a figure which shows the signal of the image in which false focusing generate | occur | produces. It is a figure which shows the fall (minimum value) of correlation amount C (k). 6 is a graph schematically illustrating linear conversion from an image shift amount to a defocus amount using a straight line with a conversion coefficient as an inclination. It is a figure which shows the signal of the image in which false focusing generate | occur | produces. 6 is a graph schematically illustrating linear conversion from an image shift amount to a defocus amount using a straight line with a conversion coefficient as an inclination. It is a front view which shows the detailed structure of an image pick-up element. It is a front view of a focus detection pixel. It is sectional drawing of a focus detection pixel. It is a figure for demonstrating the focus detection operation | movement of a re-imaging pupil division system.

  As an imaging apparatus according to an embodiment, an interchangeable lens digital still camera will be described as an example. FIG. 1 is a cross-sectional view showing a configuration of a digital still camera according to an embodiment. A digital still camera 201 according to the present embodiment includes an interchangeable lens 202 and a camera body 203, and the interchangeable lens 202 is attached to the camera body 203 via a mount unit 204. An interchangeable lens 202 having various photographing optical systems can be attached to the camera body 203 via a mount unit 204.

  The interchangeable lens 202 includes a lens 209, a zooming lens 208, a focusing lens 210, a diaphragm 211, a lens drive control device 206, and the like.

  The lens drive control device 206 includes a microcomputer (not shown), a memory, a drive control circuit, and the like. The lens drive control unit 206 performs drive control for adjusting the focus of the focusing lens 210 and adjusting the aperture diameter of the aperture 211, and detecting the states of the zooming lens 208, the focusing lens 210, and the aperture 211. Further, transmission of lens information and reception of camera information (defocus amount, aperture value, etc.) are performed by communication with a body drive control device 214 described later.

  The aperture 211 forms an aperture having a variable aperture diameter at the center of the optical axis in order to adjust the amount of light and the amount of blur.

  The camera body 203 includes an image sensor 212, a body drive control device 214, a liquid crystal display element drive circuit 215, a liquid crystal display element 216, an eyepiece lens 217, a memory card 219, and the like. In the imaging element 212, imaging pixels are two-dimensionally arranged, and focus detection pixels are incorporated in portions corresponding to focus detection positions (focus detection areas). Details of the image sensor 212 will be described later.

  The body drive control device 214 includes a microcomputer, a memory, a drive control circuit, and the like. The body drive control device 214 repeatedly performs drive control of the image sensor 212, readout of an image signal and a focus detection signal, focus detection calculation based on the focus detection signal, and focus adjustment of the interchangeable lens 202, and the image signal Processing and recording, and operation control of the digital still camera 201 are performed.

  The body drive control device 214 communicates with the lens drive control device 206 through the electrical contact 213 to receive lens information and transmit camera information.

  The liquid crystal display element 216 functions as an electric viewfinder (EVF). The liquid crystal display element driving circuit 215 displays a through image on the liquid crystal display element 216 based on the image signal from the imaging element 212, and the photographer can observe the through image through the eyepiece 217. The memory card 219 is an image storage that stores image data of an image captured by the image sensor 212.

  A subject image is formed on the light receiving surface of the image sensor 212 by the light beam that has passed through the interchangeable lens 202. This subject image is photoelectrically converted by the image sensor 212, and an image signal and a focus detection signal are sent to the body drive control device 214.

  The body drive control device 214 calculates the defocus amount based on the focus detection signal from the focus detection pixel of the image sensor 212 and sends the defocus amount to the lens drive control device 206. The body drive control device 214 processes the image signal from the image sensor 212 to generate image data, stores the image data in the memory card 219, and transmits the through image signal from the image sensor 212 to the liquid crystal display element drive circuit 215. The through image is displayed on the liquid crystal display element 216. Further, the body drive control device 214 sends aperture control information to the lens drive control device 206 to control the aperture of the aperture 211.

  The lens drive controller 206 updates the lens information according to the focusing state, zooming state, aperture setting state, aperture opening F value, and the like. Specifically, the positions of the zooming lens 208 and the focusing lens 210 and the aperture value of the aperture 211 are detected, and lens information is calculated according to these lens positions and aperture values, or a lookup prepared in advance. Lens information corresponding to the lens position and aperture value is selected from the table.

  The lens drive control device 206 calculates a lens drive amount based on the received defocus amount, and drives the focusing lens 210 to the in-focus position according to the lens drive amount. Further, the lens drive control device 206 drives the diaphragm 211 in accordance with the received diaphragm value.

  FIG. 2 is a diagram showing a focus detection position (focus detection area) on the shooting screen of the interchangeable lens 202, and an image is sampled on the shooting screen when a focus detection pixel column on the image sensor 212 described later performs focus detection. An example of a region (focus detection area, focus detection position) to be performed is shown. In this example, focus detection areas 101 to 103 are arranged at the center and three left and right positions on the rectangular shooting screen 100. In the longitudinal direction of the focus detection area indicated by the rectangle, the focus detection pixels are linearly arranged in the vertical direction.

  Before describing the detailed configuration of the image sensor 212, the principle of the pupil division type phase difference detection method will be described with reference to FIG.

  A plurality of focus detection pixels 111 are arranged on the imaging surface 110. The focus detection pixel 111 includes a microlens 112 and a pair of photoelectric conversion units 113 and 114.

  The pair of photoelectric conversion units 113 and 114 are projected by the microlens 112 onto the distance measuring pupil plane 120 at a distance d (distance pupil distance) forward from the imaging surface 110 to form the distance measuring pupils 123 and 124. In other words, the luminous flux of the ranging pupil 123 out of the luminous flux passing through the distance measuring pupil plane 120 at a distance d ahead of the imaging surface 110 is received by the photoelectric conversion unit 113 of the focus detection pixel 111, and the distance measuring pupil plane. Among the light beams that pass on 120, the light beam of the distance measuring pupil 124 is received by the photoelectric conversion unit 114 of the focus detection pixel 111.

  The relative shift amount (phase difference, image shift amount) between the image signal of the photoelectric conversion unit 113 in the array of the focus detection pixels 111 and the image signal of the photoelectric conversion unit 114 is an image on the imaging surface. It changes according to the focus adjustment state of the optical system to be formed. Therefore, if the amount of deviation is obtained by calculating a pair of image signals generated by the focus detection pixels, the focus adjustment state of the optical system can be detected.

  By the way, the pair of distance measurement pupils 123 and 124 does not have a distribution obtained by simply projecting the pair of photoelectric conversion units 113 and 114, but diffracts light according to the aperture diameter (substantially coincides with the pixel size) of the microlens 111. Due to the effect, the distribution is blurred and subtracted. In FIG. 3, when the pair of distance measurement pupils 123 and 124 are scanned in the alignment direction using the slit in the direction perpendicular to the alignment direction of the pair of distance measurement pupils 123 and 124, a pair of distance measurement pupil distributions 133 and 134 are obtained. . Due to the diffraction effect, the pair of distance-measuring pupil distributions 133 and 134 have overlapping portions 135 at adjacent portions.

  As described above, in the pupil division type phase difference detection method, the image displacement amount on the imaging surface of the pair of images formed by the light flux passing through the pair of distance measurement pupils is detected and converted into the image displacement amount with a predetermined conversion. Multiply by a coefficient to convert to a defocus amount (focus adjustment state) in the optical axis direction.

  FIG. 4 is a front view showing a detailed configuration of the image sensor 212, and shows the details of the pixel arrangement by enlarging the vicinity of the focus detection area 101 in FIG. Imaging pixels 310 are densely arranged on the imaging element 212 in a two-dimensional square lattice pattern. The imaging pixel 310 includes a red pixel (R), a green pixel (G), and a blue pixel (B), and is arranged according to a Bayer arrangement rule. For focus detection, vertical focus detection focus detection pixels 313 and 314 having the same pixel size as the image pickup pixels are alternately arranged on the vertical straight line on which green elements and blue pixels should be continuously arranged. Are arranged in succession.

  The focus detection pixels 313 and 314 are arranged in a column where B and G of the imaging pixel 310 should be arranged. The focus detection pixels 313 and 314 are arranged in the column where the B and G of the imaging pixel 310 are to be arranged because an interpolation error occurs in the interpolation processing for obtaining an image signal for imaging at the position of the focus detection pixel. When this occurs, the interpolation error of the blue pixel is less noticeable than the interpolation error of the red pixel due to human visual characteristics.

  FIG. 5 is a diagram illustrating the shape of the microlens 10 of the imaging pixel 310 and the focus detection pixels 313 and 314. The shape of the microlens 10 of the imaging pixel 310 and the focus detection pixels 313 and 314 is originally a shape cut out from a circular microlens 9 larger than the pixel size in a square shape corresponding to the pixel size. A cross section in the direction of a diagonal line passing through the optical axis of the microlens 10 and a cross section in the direction of a horizontal line passing through the optical axis of the microlens 10 have shapes shown in FIG.

  As shown in FIG. 6, the imaging pixel 310 includes a rectangular microlens 10, a photoelectric conversion unit 11 whose light receiving area is limited by a light shielding mask described later, and a color filter (not shown). There are three types of color filters, red (R), green (G), and blue (B), and each spectral sensitivity has the characteristics shown in FIG. In the image pickup device 212, image pickup pixels 310 having respective color filters are arranged in a Bayer array.

  As shown in FIG. 7A, the focus detection pixel 313 includes a rectangular microlens 10, a photoelectric conversion unit 13 whose light receiving area is limited by a light shielding mask described later, and an ND filter (neutral density filter) (not shown). However, the shape of the photoelectric conversion unit 13 whose light receiving area is limited by the light shielding mask is rectangular. Further, as shown in FIG. 7B, the focus detection pixel 314 includes a rectangular microlens 10, a photoelectric conversion unit 14 whose light receiving area is limited by a light shielding mask described later, and an ND filter, and receives light by the light shielding mask. The shape of the photoelectric conversion unit 14 whose region is limited is a rectangle. When the focus detection pixel 313 and the focus detection pixel 314 are displayed so as to overlap each other with the microlens 10 as a reference, the photoelectric conversion units 13 and 14 whose light receiving areas are limited by the light shielding mask are arranged in the vertical direction.

  The focus detection pixels 313 and 314 are not provided with a specific color filter in order to perform focus detection for all colors, but instead are provided with the above-described ND filter for reducing the amount of incident light, and its spectral sensitivity. The characteristics are shown in FIG. That is, the spectral sensitivity characteristic is obtained by adding the spectral sensitivity characteristics of the green pixel, the red pixel, and the blue pixel shown in FIG. 8, and the light wavelength region in which the focus detection pixels 313 and 314 exhibit high spectral sensitivity is the green pixel, red The optical wavelength region in which the pixel and the blue pixel exhibit high spectral sensitivity is included. The density of the ND filter is determined so that, for example, the output level of the focus detection pixel is 3/4 or less of the output level of the green pixel when the image sensor is exposed to white light. This is because vignetting of the focus detection light flux occurs in a region where the image height on the screen is high (focus detection areas 102 and 103), the output balance of the pair of focus detection pixels 313 and 314 is lost, and the output level of one focus detection pixel Even when the value rises, the output level of the green pixel of the imaging pixels 310 is not exceeded.

  The imaging pixel 310 is designed in such a shape that the microlens 10 receives all the light flux that passes through the exit pupil diameter (for example, F1.0) of the brightest interchangeable lens. The focus detection pixels 313 and 314 are designed to have a shape that receives a pair of complementary light beams with respect to the light beam received by the imaging pixel 310 by the microlens 10.

  FIG. 10 is a cross-sectional view of the imaging pixel 310. In the imaging pixel 310, a light shielding mask 30 is formed in proximity to the photoelectric conversion unit 11 for imaging, and the photoelectric conversion unit 11 receives light that has passed through the opening 30 a of the light shielding mask 30. A planarizing layer 31 is formed on the light shielding mask 30, and a color filter 38 is formed thereon. A planarizing layer 32 is formed on the color filter 38, and the microlens 10 is formed thereon. The shape of the opening 30 a is projected forward by the microlens 10. The photoelectric conversion unit 11 is formed on the semiconductor circuit substrate 29.

  FIG. 11 is a cross-sectional view of the focus detection pixels 313 and 314. In the focus detection pixels 313 and 314, a light shielding mask 30 is formed in proximity to the photoelectric detection units 13 and 14 for focus detection. The photoelectric conversion units 13 and 14 receive light that has passed through the openings 30 b and 30 c of the light shielding mask 30. A planarizing layer 31 is formed on the light shielding mask 30, and an ND filter 34 is formed thereon. A planarizing layer 32 is formed on the ND filter 34, and the microlens 10 is formed thereon. The shapes of the openings 30b and 30c are projected forward by the microlens 10. The photoelectric conversion units 13 and 14 are formed on the semiconductor circuit substrate 29.

  FIG. 12 shows the configuration of a pupil division type phase difference detection type focus detection optical system using the microlens 10. Note that the focus detection pixels 313 and 314 are enlarged. In FIG. 12, the exit pupil 90 is disposed on the planned imaging plane of the interchangeable lens 202 and is set at a distance d in front of the microlens 10. The distance d is a distance determined according to the curvature and refractive index of the microlens 10, the distance between the microlens 10 and the photoelectric conversion units 13 and 14, and is referred to as a distance measuring pupil distance in this specification. FIG. 12 shows an optical axis 91 of the interchangeable lens, the microlens 10, the photoelectric conversion units 13 and 14, focus detection pixels 313 and 314, a photographing light beam 71, and focus detection light beams 73 and 74.

  The region 93 is a region where the opening 30b is projected by the microlens 10, and is referred to as a distance measuring pupil in this specification. Similarly, the region 94 is a distance measuring pupil in which the open portion 30 c is projected by the microlens 10. In FIG. 12, for the sake of easy understanding, the distance measuring pupils 93 and 94 are shown as elliptical areas. However, in actuality, the shape of the open portion 30b is enlarged and projected and becomes blurred due to diffraction.

  FIG. 12 schematically illustrates five focus detection pixels adjacent to the photographing optical axis. However, in the focus detection pixel array at the periphery of the screen, the photoelectric conversion units 13 and 14 each have a corresponding distance measurement. It is configured to receive a light beam coming from the pupils 93 and 94 to each microlens 10. The arrangement direction of the focus detection pixels is made to coincide with the arrangement direction of the pair of distance measuring pupils 93 and 94, that is, the arrangement direction of the pair of photoelectric conversion units 13 and 14.

  The microlens 10 is disposed in the vicinity of the planned imaging plane of the interchangeable lens 202. The shape of the openings 30b and 30c arranged close to the photoelectric conversion units 13 and 14 is projected by the microlens 10 onto the exit pupil 90 separated from the microlens 10 by the distance measurement pupil distance d, and the projected shape. Forms distance measuring pupils 93, 94.

  The photoelectric conversion unit 13 outputs a signal corresponding to the intensity of the image formed on the microlens 10 by the light beam 73 passing through the distance measuring pupil 93 and directed to the microlens 10 of the focus detection pixel 313. Further, the photoelectric conversion unit 14 outputs a signal corresponding to the intensity of the image formed on the microlens 10 by the light flux 74 that passes through the distance measuring pupil 94 and travels toward the microlens 10 of the focus detection pixel 314.

  A large number of the two types of focus detection pixels described above are arranged in a straight line, and the output of the photoelectric conversion unit of each pixel is grouped into a distance measurement pupil 93 and an output group corresponding to the distance measurement pupil 94, thereby measuring the distance measurement pupil 93 and the measurement pupil 93. Information on the intensity distribution of the pair of images formed on the pixel array by the focus detection light beams that respectively pass through the distance pupil 94 is obtained. By applying an image shift detection calculation process (correlation calculation process, phase difference detection process), which will be described later, to this information, an image shift amount of a pair of images is detected by a so-called pupil division type phase difference detection method. Furthermore, by performing a conversion calculation according to the proportional relationship between the distance between the center of gravity of the pair of distance measurement pupils and the distance measurement pupil distance to the image shift amount, the current image formation surface (the position of the microlens array) ( A deviation (defocus amount) of an actual imaging plane at a focus detection position determined on the photographing screen 100 is calculated.

  FIG. 13 is a diagram for explaining the state of the imaging light beam received by the imaging pixel 310 of the imaging device 212 shown in FIG. 6 in comparison with FIG. 12, and description of portions overlapping those in FIG. 12 is omitted.

  The imaging pixel 310 includes the microlens 10 and the photoelectric conversion unit 11 disposed behind the microlens 10. The shape of the opening 30 a arranged in the vicinity of the photoelectric conversion unit 51 is projected onto the exit pupil 90 that is separated from the microlens 10 by the distance measurement pupil distance d, and the projection shape is substantially the distance measurement pupils 93 and 94. A circumscribed region 95 is formed.

  The photoelectric conversion unit 11 outputs a signal corresponding to the intensity of the image formed on the microlens 11 by the imaging light flux 71 that passes through the region 95 and travels toward the microlens 10.

  14 and 15 are flowcharts showing the imaging operation of the digital still camera 201. When the power of the digital still camera 201 is turned on in step S100, the body drive control device 214 starts an imaging operation after step S110. In step S110, an aperture adjustment command is transmitted to the lens drive controller 206, and the aperture value of the interchangeable lens 202 is set to the open F value. In step S120, the data of the imaging pixel 310 is read out and displayed on the electronic viewfinder.

  In subsequent step S130, a pair of image data corresponding to the pair of images is read from the focus detection pixel array. The focus detection area is assumed to be selected in advance by the photographer using one of the focus detection areas 101 to 103 using a focus detection area selection member (not shown).

  In step S140, an image shift detection calculation process (correlation calculation process, phase difference detection process) described later is performed based on the read pair of image data, and the image shift amount is calculated and converted into a defocus amount. In step S150, it is determined whether false focusing has occurred. If not, the false focusing flag is set to 0 in step S151, and the process proceeds to step S200 in FIG. The false in-focus state is a state in which a plurality of highly reliable different defocus amounts are generated, and details of the process for determining whether or not false in-focus occurs will be described later. When the false focus flag is 0, it indicates that false focus has not occurred in the previous defocus detection cycle, and when it is 1, it indicates that false focus has occurred in the previous defocus detection cycle. Show.

  If it is determined in step S150 that false focusing has occurred, it is necessary to determine one true defocus amount from a plurality of different defocus amounts. For this purpose, first, in step S152, it is determined whether the false focus flag is 1 and there is no composition change. For example, the electronic viewfinder display image read out during the previous defocus detection cycle is stored and compared with the electronic viewfinder display image read out during the current defocus detection cycle. To do. When the motion vector between two images is equal to or greater than a predetermined threshold, it is determined that there is a composition change, and when it is equal to or less than the predetermined threshold, it is determined that there is no composition change. Also, for example, a posture sensor that detects the posture of the camera body is built in the camera body, and the posture information detected during the previous defocus detection cycle and the posture information detected during the current defocus detection cycle are Compare. If the difference between the two pieces of posture information is greater than or equal to a predetermined threshold, it may be determined that there is a composition change, and if it is less than the predetermined threshold, it may be determined that there is no composition change.

  If it is determined in step S152 that the false focus flag is 1 and there is no composition change, a predetermined condition is satisfied from among a plurality of defocus amounts generated in the false focus state in step S153. The defocus amount is set as a true defocus amount, and the process proceeds to step S200 in FIG. The predetermined condition is, for example, defocusing the movement amount of the lens driven between the previous defocus detection cycle and the current defocus detection cycle from the true defocus amount determined at the previous defocus detection cycle. The condition is that it is closest to the defocus amount obtained by subtracting the amount converted to the amount change. When such a condition is satisfied, an operation after step S154 described later (an operation for detecting the defocus amount again by changing the aperture diameter) can be omitted. Therefore, when false focusing continues without changing the composition, it is possible to realize a quick defocus detection operation without repeating the operation of changing the aperture diameter of the aperture.

  If it is determined in step S152 that the false focus flag is 0 or there is a composition change, the false focus flag is set to 1 in step S154, and the aperture diameter after step S155 is changed to detect the defocus amount again. Proceed to operation. Even when it is determined that there is a false focus when the first defocus is detected immediately after the power is turned on, the process proceeds to step S154 and subsequent steps.

  In step S155, the aperture diameter is determined based on a plurality of image shift amounts obtained in step S140 as will be described later.

  In step S160, an aperture adjustment command is transmitted to the lens drive control unit 206, and the aperture value of the interchangeable lens 202 is set to an F value corresponding to the aperture diameter determined in step S155. In subsequent step S170, a pair of image data corresponding to the pair of images is read from the focus detection pixel array. In step S180, an image shift detection calculation process (correlation calculation process, phase difference detection process) described later is performed based on the read pair of image data, and the image shift amount is calculated and converted into a defocus amount.

  In step S190, the defocus amount obtained by the full aperture F value is compared with the defocus amount obtained by the F value corresponding to the aperture diameter determined in step S155 by a process described later, and substantially matches. The defocus amount to be performed is set as a genuine defocus amount, and the process proceeds to step S200 in FIG. In step S190 of FIG. 14, the two defocus amounts that are substantially coincident with each other are not completely coincident with each other. Therefore, the average of the two defocus amounts is set as the true defocus amount. Any one of the focus amounts may be set as the true defocus amount.

  In step S200, it is checked whether or not the focus is close, that is, whether or not the calculated absolute value of the defocus amount is within a predetermined value. If it is determined that the lens is not in focus, the process proceeds to step S210, the defocus amount is transmitted to the lens drive control device 206, and the focusing lens 210 of the interchangeable lens 202 is driven to the focus position. Then, it returns to step S110 of FIG. 14 and repeats the operation | movement mentioned above.

  Even when focus detection is impossible, the process branches to this step, a scan drive command is transmitted to the lens drive control device 206, and the focusing lens 210 of the interchangeable lens 202 is driven to scan from infinity to the nearest. Then, it returns to step S110 and repeats the operation | movement mentioned above.

  If it is determined in step S200 that the focus is close to the in-focus state, the process proceeds to step S220, and it is determined whether or not a shutter release has been performed by operating a shutter button (not shown). If it is determined that the shutter release has not been performed, the process returns to step S110 and the above-described operation is repeated. On the other hand, if it is determined that the shutter release has been performed, the process proceeds to step S230, where an aperture adjustment command is transmitted to the lens drive control unit 206, and the aperture value of the interchangeable lens 202 is controlled by a control F value (F set by the photographer or automatically) Value). After completion of the aperture control, in step S250, the imaging device 212 performs an imaging operation, and image data is read from the imaging pixel 310 and all the focus detection pixels 313 and 314 of the imaging device 212.

  In step S250, pixel interpolation of pixel data at each pixel position in the focus detection pixel row is performed based on data of the imaging pixels 310 around the focus detection pixel. In subsequent step S260, image data composed of the data of the imaging pixel 310 and the interpolated data is stored in the memory card 219, and the process returns to step S110 of FIG. 14 to repeat the above-described operation.

  Details of the image shift detection calculation processing (correlation calculation processing, phase difference detection processing) in step S140 and step S180 in FIG. 14 will be described below.

Correlations disclosed in Japanese Patent Laid-Open No. 2007-333720 with respect to focus detection pixel data read from the image sensor 212, that is, a pair of data strings (A1 _{1 to} A1 _M , A2 _{1 to} A2 _M : M is the number of data) The calculation formula (1) is performed to calculate the correlation amount C (k).
C (k) = Σ | A1 _i × A2 _{i + s + k−} A2 _{i + k} × A1 _{i + s} | (1)

In Expression (1), the image shift amount k is an integer, and is a shift value representing a relative displacement amount in units of the data interval between the pair of image data strings A1 _i and A2 _i . C (k) is a correlation amount indicating the degree of correlation between the pair of image data strings A1 _i and A2 _{i at} the shift value k. The integration process (Σ) is performed over a predetermined section of data with respect to the data index i. The parameter s (s = 1, 2, 3...) Is a parameter for adjusting the frequency characteristic of the correlation calculation.

As shown in FIG. 16A, the calculation result of the expression (1) indicates that the correlation amount C (k) is obtained when the pair of data has a high correlation amount (k = k _j = 2 in FIG. 16A). Minimal (the smaller the value, the higher the degree of correlation). The shift amount x that gives the minimum value C (x) with respect to the continuous correlation amount is obtained using the three-point interpolation method according to the equations (2) to (5).
x = k _j + D / SLOP (2)
C (x) = C (k _j ) − | D | (3)
D = {C (k _j −1) −C (k _j +1)} / 2 (4)
SLOP = MAX {C ( _kj + 1) -C ( _kj ), C ( _kj- 1) -C ( _kj )}} (5)

  Whether or not the shift amount x calculated by Expression (2) is reliable is determined as follows.

  As shown in FIG. 16B, when the degree of correlation between the pair of data is low, the value of the minimum value C (x) of the interpolated correlation amount becomes large.

  Therefore, when C (x) is equal to or greater than a predetermined threshold value, it is determined that the calculated shift amount has low reliability, and the calculated shift amount x is canceled.

  Alternatively, in order to normalize C (x) with the contrast of data, when the value obtained by dividing C (x) by SLOP that is proportional to the contrast is equal to or greater than a predetermined value, the reliability of the calculated shift amount Is determined to be low, and the calculated shift amount x is canceled.

  Alternatively, when SLOP that is a value proportional to the contrast is equal to or less than a predetermined value, it is determined that the subject has low contrast and the reliability of the calculated shift amount is low, and the calculated shift amount x is canceled.

  As shown in FIG. 16C, when the correlation between the pair of data is low and there is no drop in the correlation amount C (k) between the shift ranges kmin to kmax, the minimum value C (x) is obtained. In such a case, it is determined that the focus cannot be detected.

When it is determined that the calculated shift amount x is reliable, the image shift amount shft is converted by the equation (6).
shft = PY · x (6)

  In Expression (6), PY is an image detection pitch by the focus detection pixels 313 and 314. Since two types of focus detection pixels are alternately arranged in the focus detection pixel array, the detection pitch PY is twice the pixel pitch of the imaging pixels 310.

The image shift amount calculated by Expression (6) is multiplied by a conversion coefficient Kd to convert it to a defocus amount def. The conversion coefficient Kd is a conversion coefficient corresponding to the F value of the aperture opening included in the lens information.
def = Kd · shft (7)

  FIG. 17 shows an image shift amount (relative deviation amount of a pair of images in a direction perpendicular to the optical axis of the optical system) as a defocus amount (actual surface relative to a reference plane in the optical axis direction of the optical system, that is, an imaging surface). It is an explanatory diagram in the case of conversion to (deviation amount of image plane).

In FIG. 17, the imaging surface 110, the image surface 140, the diaphragm surface 125, the optical axis 91, the defocus amount (distance from the imaging surface 110 to the image surface 140) def, the distance PO from the imaging surface 110 to the diaphragm surface 125, the diaphragm Positions G1 and G2 where the central light beams of the pair of focus detection light beams when the F value is F1 intersect the diaphragm surface, and the central light beams of the pair of focus detection light beams when the aperture F value is F2 (<F1) Are positions G3 and G4 at which the diaphragm surface intersects, a distance Q1 between the positions G1 and G2, a distance Q2 between the position G3 and the position G4, and an image shift amount S1 of the pair of images when the aperture F value is F1, An image shift amount S2 of a pair of images when the aperture F value is F2, a center beam opening angle θ1 of a pair of focus detection beams when the aperture F value is F1, and a pair of focus detections when the aperture F value is F2. The opening angle θ2 of the central light beam of the light beam is shown. Although the distance PO varies depending on the type of the interchangeable lens 202, the average distance is set substantially equal to the distance measuring pupil distance d in FIG. When the defocus amount def is small compared to the distance PO (that is, when it can be approximated as PO−def≈PO), the defocus amount def when the aperture F value is F1 is calculated by Expression (8).
def = S1 / (2 · Tan (θ1 / 2)) = S1 · PO / Q1 (8)

Therefore, the conversion coefficient Kd1 when the aperture F value is F1 is expressed by Equation 9.
Kd1 = 1 / (2 · Tan (θ1 / 2)) = PO / Q1 (9)

Similarly, the defocus amount def and the conversion coefficient Kd2 when the aperture F value is F2 (> F1) are calculated by Expressions (10) and (11).
def = S2 / (2 · Tan (θ2 / 2)) = S2 · PO / Q2 (10)
Kd2 = 1 / (2 · Tan (θ2 / 2)) = PO / Q2 (11)

  On the interchangeable lens side, a conversion coefficient Kd corresponding to the aperture F value is stored as lens information. When the image shift amount calculated with a certain aperture F value is converted into a defocus amount, a conversion coefficient Kd corresponding to the aperture F value is used.

  The above description is the description of the defocus amount calculation in a simple case where false focusing does not occur. That is, as shown in FIG. 16, there is a case where there is only one minimum value of the correlation amount. However, depending on the waveform of the formed image, there may be a case where the minimum value occurs at a plurality of locations. Is called false in-focus.

  First, a case where the minimum value of the correlation amount occurs at two locations due to false focusing will be described.

  FIG. 18 shows an image signal in which false focusing occurs as a function H0 (y), and a peak T0 similar to the peak R0 exists at a position away from the central peak R0 (y = 0). Here, y represents the distance from the optical axis on the imaging surface.

  FIGS. 19A and 19B show a pair of image signals generated by the focus detection pixel array when such an image is formed on a defocused surface with an aperture F value being an open F value. Functions H1 (y) and H2 (y) are shown. The functions H1 (y) and H2 (y) are functions obtained by shifting the original function H0 (y) in opposite directions based on y = 0 in accordance with the defocus amount. In the functions H1 (y) and H2 (y), the peak R1 and the peak R2 are relatively displaced by an image shift amount y1 corresponding to the defocus amount. Further, the peak R1 and the peak T2 similar to the peak R1 are relatively displaced by the image shift amount y2.

The section W1 including the peak R1 of the function H1 (y) is defined as one data string (A1 _{i in} Expression (1)), and the function H2 (y) is relatively shifted to the other data string (Expression (1) When the correlation calculation of Equation (1) is applied as A2 _i ) of), the correlation amount C (k) calculated with the horizontal axis as the shift amount k becomes the graph shown in FIG.

  In the graph shown in FIG. 21A, the shift amounts k1 and k2 corresponding to the drop (minimum values) V1 and V2 of the correlation amount C (k) are the shift amounts with high correlation between the two data.

  The shift amount k1 corresponding to the drop V1 is the shift amount obtained by matching the peak R1 of the function H1 (y) in FIG. 19A with the peak R2 of the function H2 (y) in FIG. This corresponds to the shift amount when the function H2 (y) is shifted to the right by y1 with reference to the function H1 (y) at W1.

  The shift amount k2 corresponding to the drop V2 is the shift amount obtained by matching the peak R1 of the function H1 (y) in FIG. 19A and the peak T2 of the function H2 (y) in FIG. 19B, that is, the section W1. This corresponds to the shift amount when the function H2 (y) is shifted to the left by y2 with reference to the function H1 (y).

  Since the correlation amounts E1 and E2 of the drops V1 and V2 (E1 and E2 are equal to or less than the predetermined threshold value E0) have almost the same value, which drop indicates the correlation amount indicating the correct match between the functions H1 (y) and H2 (y). do not know. When the shift amount k2 corresponding to the drop V2 is adopted as the shift amount of the two functions, a completely wrong focus detection result is calculated.

  The above is a description of the case where the minimum value of the correlation amount occurs in two places. Generally, the correlation calculation is highly reliable (the correlation value is equal to or less than a predetermined threshold value). If it is found (two or more locations), it is determined in step S150 that a false focus has occurred, and the subsequent processing is performed to determine the true defocus amount.

  Note that the reliability of the minimum value of the correlation amount may be determined by comparing the correlation value with a predetermined threshold value, or may be determined by the method described with reference to FIG.

  Next, the process from step S155 (determination of the aperture diameter) to step S190 in FIG. 14 when it is determined that there is a false focus will be described.

  First, the case where the minimum value of the correlation amount occurs at two locations will be described.

  When there are two minimum values, the aperture diameter (diaphragm aperture F value) may be any F value different from the open F value.

  Specifically, the aperture F value is set to FA (for example, when the F value is one step darker than the open F value, and when the open F value is F2.0, the aperture F value of the FA is F2.8). In this state, the aperture is controlled to the aperture F value FA, data is read from the focus detection pixel, and the correlation calculation process similar to the above is performed on the data. FIGS. 20A and 20B show a pair of image signals generated by the focus detection pixel array when the diaphragm aperture F value is FA and the image shown in FIG. 18 is formed on the defocused surface. The functions H3 (y) and H4 (y) are shown. The functions H3 (y) and H4 (y) are functions in which the original function H0 (y) is shifted in the opposite direction based on y = 0 in accordance with the defocus amount. In the functions H3 (y) and H4 (y), the peak R3 and the peak R4 are relatively displaced by an image shift amount y3 corresponding to the defocus amount, and the peak R3 and the peak T4 are The image displacement amount y4 is relatively displaced. The diaphragm aperture F value FA is an F value larger than the open F value, and therefore the relative image shift amount is a small value for the same defocus amount. Therefore, in the image shift amount, | y3 | <| y1 | and | y2-y1 | = | y4-y3 |.

The section W2 including the peak R3 of the function H3 (y) is defined as one data string (A1 _{i in} Expression (1)), and the function H4 (y) is relatively shifted to the other data string (Expression (1) When the correlation calculation of Equation (1) is applied as A2 _i ) of (), the correlation amount C (k) calculated with the horizontal axis as the shift amount k is shown in the graph of FIG.

  In the graph shown in FIG. 21B, the shift amounts k3 and k4 corresponding to the drop (minimum values) V3 and V4 of the correlation amount C (k) are the shift amounts with high correlation between the two data.

  The shift amount k3 corresponding to the drop V3 is the shift amount obtained by matching the peak R3 of the function H3 (y) in FIG. 20A and the peak R4 of the function H4 (y) in FIG. This corresponds to the shift amount when the function H4 (y) is shifted to the right by y3 with reference to the function H3 (y) at W2.

  The shift amount k4 corresponding to the drop V4 is the shift amount obtained by matching the peak R3 of the function H3 (y) in FIG. 20A and the peak T4 of the function H4 (y) in FIG. This corresponds to the shift amount when the function H4 (y) is shifted to the left by y4 with reference to the function H3 (y) at W2.

  Since the correlation amounts E3 and E4 of the drops V3 and V4 (E3 and E4 are equal to or less than the predetermined threshold value E0) have almost the same value, which drop indicates the correlation amount indicating the correct match between the functions H3 (y) and H4 (y). do not know.

  Here, after the shift amounts k1 and k2 obtained in the open F value are converted into the image shift amounts y1 and y2, the defocus amount def1 is expressed by the equations (6) and (7) using the conversion coefficient Kd0 in the open F value. , Def2. Further, after the shift amounts k3 and k4 obtained with the aperture value F value FA are converted into the image shift amounts y3 and y4, the conversion coefficient Kd3 at the aperture value F value FA is used, and the values are expressed by equations (6) and (7). The focus amounts are converted into def3 and def4.

  FIG. 22 is a graph schematically showing linear conversion from the image shift amount y to the defocus amount def using a straight line with the conversion coefficient K as an inclination, and shows a relationship of def = K × y. ing.

  In FIG. 22, the image shift amounts y1 and y2 obtained with the open F value are converted into defocus amounts def1 and def2 by the conversion coefficient Kd0, and the image shift amounts y3 and y4 obtained with the aperture F value FA are converted into the conversion coefficient Kd3. Is converted into defocus amounts def3 and def4.

  Since the image shift amounts y1 and y3 correspond to the true defocus amounts at the full aperture F value and the aperture F value FA, the defocus amounts def1 and def3 corresponding to the image shift amounts y1 and y3 are substantially the same values, The defocus amounts def2 and def4 and the true defocus amount def1 (≈def3) corresponding to the image shift amounts y2 and y4 are different values.

  In other words, a defocus amount that substantially matches among a plurality of defocus amounts obtained at different aperture F values (open F value and aperture F value FA) is a true defocus amount.

  Therefore, the defocus amounts def1 to def4 are compared, the one whose difference is within the predetermined threshold is extracted, and the true defocus amount is obtained based on the extracted defocus amounts (in this case, def1 and def3). For example, def1 and def3 may be averaged, or either def1 or def2 may be selected.

  The above is a description of the case where the minimum value of the correlation amount occurs at two locations. However, when there are three or more minimum values, the aperture is opened as in the case where the minimum value is two locations. If the defocus amount is detected with the aperture uniformly set as the aperture F value FA, there may be a plurality of coincidence of defocus amounts in the defocus amount at the open F value and the defocus amount at the aperture F value FA. In this case, the true defocus amount cannot be determined.

  The following is a description of a case where a plurality of defocus amounts coincide between the defocus amount at the open F value and the defocus amount at the opening F value FA.

  FIG. 23 shows an image signal as a function H10 (y). There are a peak R0 and a peak T0 at the same position as the function H0 (y) in FIG. 18, and a peak at a position between the peak R0 and the peak T0. There is a peak S0 similar to R0.

  24A and 24B show a pair of focus detection pixel arrays generated when such an image is formed on the same defocus plane as in FIG. 19 when the aperture F value is the open F value. The functions H11 (y) and H12 (y) of the image signal are shown. In the functions H11 (y) and H12 (y), the peak R1 and the peak R2 are relatively displaced by an image shift amount y1 corresponding to the defocus amount, and the peak R1 and the peak T2 are image shifted. The amount y2 is relatively displaced, and the peak R1 and the peak S2 are relatively displaced by an image shift amount y5.

  When the correlation amount C (k) is obtained by relatively shifting the function H12 (y) with the section W1 including the peak R1 of the function H11 (y) as a reference, the graph shown in FIG.

  In the graph shown in FIG. 26A, the shift amounts k1, k2, and k5 corresponding to the drop (minimum values) V1, V2, and V5 of the correlation amount C (k) are the shift values with high correlation between the two functions. .

  The shift amount k1 corresponding to the drop V1 is the shift amount obtained by matching the peak R1 of the function H11 (y) in FIG. 24A and the peak R2 of the function H12 (y) in FIG. 24B, that is, the section W1. This corresponds to the amount of shift when the function H12 (y) is shifted to the right by y1 with the function H11 (y) at.

  The shift amount k2 corresponding to the drop V2 is the shift amount obtained by matching the peak R1 of the function H11 (y) in FIG. 24A and the peak T2 of the function H12 (y) in FIG. This corresponds to a shift amount when the function H12 (y) is shifted to the left by y2 with reference to the function H11 (y) in W1.

  The shift amount k5 corresponding to the drop V5 is the shift amount obtained by matching the peak R1 of the function H11 (y) in FIG. 24A and the peak S2 of the function H12 (y) in FIG. 24B, that is, the section W1. This corresponds to the shift amount when the function H12 (y) is shifted to the left by y5 with reference to the function H11 (y).

  Since the correlation amounts E1, E2, and E5 of the drops V1, V2, and V5 (E1, E2, and E5 are equal to or less than the predetermined threshold value E0) have almost the same value, any drop is correct in the functions H11 (y) and H12 (y). I don't know the amount of correlation that shows a match.

  A case where the aperture value F is FA (the same F value as in the case of two minimum values) and the defocus amount is detected in the same manner will be described below. 25 (a) and 25 (b) show a pair of focus detection pixel arrays generated when the aperture F value is FA and the image shown in FIG. 23 is formed on the same defocus plane as FIG. The functions H13 (y) and H14 (y) of the image signal are shown.

  In the functions H13 (y) and H14 (y), the peak R3 and the peak R4 are relatively displaced by an image shift amount y3 corresponding to the defocus amount, and the peak R3 and the peak T4 are image shifted. The amount y4 is relatively displaced, and the peak R3 and the peak S4 are relatively displaced by an image shift amount y6.

  If the correlation amount C (k) is obtained by relatively shifting the function H14 (y) with the section W2 including the peak R3 of the function H13 (y) as a reference, a graph shown in FIG. 26B is obtained.

  In the graph shown in FIG. 26B, the shift amounts k3, k4, and k6 corresponding to the drop (minimum value) V3, V4, and V6 of the correlation amount C (k) are the shift amounts having high correlation between the two functions. Become.

  The shift amount k3 corresponding to the drop V3 is the shift amount obtained by matching the peak R3 of the function H13 (y) in FIG. 25A and the peak R4 of the function H14 (y) in FIG. This corresponds to the amount of shift when the function H14 (y) is shifted to the right by y3 with reference to the function H13 (y) at W2.

  The shift amount k4 corresponding to the drop V4 is the shift amount obtained by matching the peak R3 of the function H13 (y) in FIG. 25A with the peak T4 of the function H14 (y) in FIG. This corresponds to the shift amount when the function H14 (y) is shifted to the left by y4 with reference to the function H13 (y) at W2.

  The shift amount k6 corresponding to the drop V6 is the shift amount obtained by matching the peak R3 of the function H13 (y) in FIG. 25A with the peak S4 of the function H13 (y) in FIG. This corresponds to the shift amount when the function H14 (y) is shifted to the left by y6 with reference to the function H13 (y) at W2.

  Since the correlation amounts E3, E4, and E6 of the drops V3, V4, and V6 (E3, E4, and E6 are equal to or less than the predetermined threshold E0) have almost the same value, any drop is correct in the functions H13 (y) and H13 (y). I don't know the amount of correlation that shows a match.

  Here, the shift amounts k1, k2, and k5 obtained at the open F value are converted into image shift amounts y1, y2, and y5, and then converted into defocus amounts def1, def2, and def5 using the conversion coefficient Kd0 at the open F value. To do. Further, after the shift amounts k3, k4, and k6 obtained at the aperture value F value FA are converted into image displacement amounts y3, y4, and y6, the defocus amounts def3, def4, Convert to def6.

  FIG. 27 is a diagram similar to FIG. 22, and image shift amounts y1, y2, and y5 obtained with the open F value are converted into defocus amounts def1, def2, and def5 by the conversion coefficient Kd0, and the aperture F value FA. It is shown that the image shift amounts y3, y4, and y6 obtained in the above are converted to defocus amounts def3, def4, and def6 by the conversion coefficient Kd3.

  Since the image shift amounts y1 and y3 correspond to true defocus amounts at the full aperture F value and the aperture F value FA, the defocus amounts def1 and def3 corresponding to the image shift amounts y1 and y3 are substantially the same values. Become. The defocus amounts def5 and def4 corresponding to the image shift amounts y5 and y4 and the true defocus amount def1 (≈def3) are different values. However, the defocus amounts def2 and def6 corresponding to the image shift amounts y2 and y6 are substantially the same value. That is, a plurality of coincidence of defocus amounts occurs, and it becomes impossible to determine which defocus amount coincides with the true defocus amount.

  Therefore, when the minimum value of the correlation amount occurs at three or more locations, the F value of the aperture diameter that does not cause a plurality of coincidence of defocus amounts is set based on the image shift amount (shift value) of the open F value. The defocus amount is determined in advance, and the defocus amount is detected by the determined aperture opening F value, and the true defocus amount is determined.

  Hereinafter, for the sake of simplicity, the case where three minimum values of the correlation amount occur will be described. However, the case where there are three or more correlation values can be easily expanded.

  Consider a case where there are three image shift amounts y1, y2, and y5 as shown in FIG. 24B and the defocus amounts def1, def2, and def5 are calculated corresponding to the open F value.

  The defocus amounts corresponding to the image misalignment intervals | y2-y1 |, | y2-y5 |, | y5-y1 | are | def2-def1 |, | def2-def5 |, and | def5-def1 |.

  By the way, even if the aperture value F is changed, the interval between the plurality of peak images (that is, the difference in image shift amount) itself is universal. For example, in the open F value, as shown in FIG. 24B, the interval between the peak T2 and the peak R2 is the difference in image shift amount | y2-y1 |. However, when the aperture F value is FA, FIG. As shown in b), the distance between the peak T4 and the peak R4 is the image shift amount difference | y4-y3 |, which matches | y2-y1 |.

  Therefore, the defocus amount def21 obtained by converting the image shift amount intervals | y2-y1 |, | y2-y5 |, | y5-y1 | into the defocus amount by the conversion coefficient Kd3 of the aperture F value FA. (= Kd3 × | y2-y1 |), def22 (= Kd3 × | y2-y5 |), and def23 (= Kd3 × | y5-y1 |), the defocus amount at the open F value | def2-def1 |, | Def2-def5 |, and | def5-def1 | are substantially equal to each other, when the defocus amount at the first diaphragm aperture F value is compared with the defocus amount at the second aperture aperture F value. In addition, coincidence occurs at the defocus amounts at two locations. That is, since the true defocus amount is the same for the two F values (open F value and opening F value FA), if the defocus amount is different from the true defocus amount, the difference is The defocus amount which is taken also coincides between the two F values (open F value and opening F value FA).

  In FIG. 24B, since def23 (= Kd3 × | y5-y1 |) and | def2-def1 | coincide with each other, the defocus amount coincides with def2≈ in addition to the true defocus amount def1≈def3. def6 has occurred. The coincidence determination as described above does not need to detect the defocus amount in a state where the aperture aperture F value FA is actually changed, and if the conversion coefficient corresponding to the aperture aperture F value is known, the aperture aperture F value FA Can be done before changing. Therefore, when three minimum values of the correlation amount are detected in the correlation amount detected by the open F value, the diaphragm aperture F value that does not cause a plurality of defocus matches is determined in advance as follows. By controlling the aperture diameter to the determined aperture F value, detecting the defocus amount in that state, and extracting the defocus amount that substantially matches the defocus amount detected by the open F value, the true defocus is obtained. The amount can be determined.

  In the correlation amount detected by the open F value, the determination of the opening F value when three minimum values of the correlation amount occur is performed as follows.

(1) Prioritize opening F value candidates. For example, when the open F value is F2.0, the dark F value (in the order of F2.8, F4.0, F5.6...) Is selected step by step.

(2) Three defocus amounts def1a (= Ka × | y2−y1 |) and def2a (= Ka × | y2−y5) with the conversion coefficient Ka corresponding to the aperture F value in order of the aperture F value with the highest priority. |), Def3a (= Ka × | y5-y1 |) is calculated.

(3) The three defocus amounts def1a, def2a, def3a are compared with the three defocus amounts | def2-def1 |, | def2-def5 |, | def5-def1 | Find out if there is. If there is no substantial match, the aperture F value at that time is adopted, and if there is a substantial match, the same processing is performed on the aperture F value of the next priority.

  When the above processing is specifically applied to an image as shown in FIG.

  Data is read from the focus detection pixels at the open F value, and image functions as shown in FIGS. 24A and 24B are obtained.

  Correlation calculation processing is performed on the image function shown in FIG. 24, and the correlation amount graph shown in FIG. 26A is obtained. From the graph, the drop in the correlation amount (the drop in the correlation value is equal to or less than the threshold) is obtained. Three places are discovered.

  Defocus amounts (def1, def2, def5) corresponding to the image shift amounts (y1, y2, y5) with respect to the drop of the correlation amount at three locations are calculated.

  Three defocus amounts | def2-def1 |, | def2-def5 |, and | def5-def1 | are calculated.

  A conversion coefficient Kd3 corresponding to the first aperture aperture F value FA (= F2.8) is selected, and defocus amounts def21 (= Kd3 × | y2−y1 |), def22 (= Kd3 × | y2−y5). |), Def23 (= Kd3 × | y5-y1 |) is calculated.

  The defocus amounts def21, def22, and def23 and the three defocus amounts | def2-def1 |, | def2-def5 |, | def5-def1 | are compared, and the defocus amount def23 and the defocus amount | def2-def1 | It turns out that substantially matches.

  Therefore, even if the defocus amount is detected with the aperture F value FA, it can be seen that there are a plurality of matches as compared with the defocus amount with the open F value, so that the process proceeds to verification of the aperture F value FB of the next priority. .

  A conversion coefficient Kd4 corresponding to the aperture aperture F value FB (= F4.0) of the second priority is selected, and the defocus amounts def31 (= Kd4 × | y2-y1 |), def32 (= Kd4 × | y2-y5) |), Def33 (= Kd4 × | y5-y1 |) is calculated.

  The defocus amounts def31, def32, and def33 are compared with the three defocus amounts | def2-def1 |, | def2-def5 |, | def5-def1 |, and it is found that there is no matching defocus amount. The aperture stop F value FB at that time is determined as the stop aperture F value.

  The aperture aperture diameter is controlled to the aperture aperture F value FB, data is read from the focus detection pixels, and image functions as shown in FIGS. 28A and 28B are obtained.

  FIGS. 28A and 28B show a pair of image signals generated by the focus detection pixel array when the diaphragm aperture F value is FB and the image shown in FIG. 23 is formed on the defocused surface. The functions H15 (y) and H16 (y) are shown.

  In the functions H15 (y) and H16 (y), the peak R5 and the peak R6 are relatively displaced by an image shift amount y7 corresponding to the defocus amount, and the peak R5 and the peak T6 are image shifted. The amount y8 is relatively displaced, and the peak R5 and the peak S6 are relatively displaced by the image shift amount y9.

  If the correlation amount C (k) is obtained by relatively shifting the function H16 (y) with the section W3 including the peak R5 of the function H15 (y) as a reference, a graph shown in FIG. 26C is obtained.

  In the graph shown in FIG. 26C, the shift amounts k7, k8, and k9 corresponding to the drop (minimum value) V7, V8, and V9 of the correlation amount C (k) are the shift amounts having high correlation between the two functions. Become.

  The shift amount k7 corresponding to the drop V7 is the shift amount obtained by matching the peak R5 of the function H15 (y) in FIG. 28A and the peak R6 of the function H16 (y) in FIG. This corresponds to the shift amount when the function H16 (y) is shifted to the right by y7 with reference to the function H15 (y) at W3.

  The shift amount k8 corresponding to the drop V8 is the shift amount obtained by matching the peak R5 of the function H15 (y) in FIG. 28A and the peak T6 of the function H16 (y) in FIG. This corresponds to a shift amount when the function H16 (y) is shifted to the left by y8 with reference to the function H15 (y) in W3.

  The shift amount k9 corresponding to the drop V9 is the shift amount obtained by matching the peak R5 of the function H15 (y) in FIG. 28A and the peak S6 of the function H16 (y) in FIG. This corresponds to the shift amount when the function H16 (y) is shifted to the left by y9 with reference to the function H15 (y) at W3.

  Since the correlation amounts E7, E8, and E9 of the drop V7, V8, and V9 (E7, E8, and E9 are equal to or less than the predetermined threshold E0) have almost the same value, any drop is correct in the functions H15 (y) and H16 (y). I don't know the amount of correlation that shows a match.

  The shift amounts k7, k8, k9 obtained at the aperture value F value FB are converted into image shift amounts y7, y8, y9, and then the defocus amounts def7, def8, def9 are used using the conversion coefficient Kd4 at the aperture value F FB. Convert to

  FIG. 29 is a diagram similar to FIG. 22, and the image shift amounts y1, y2, and y5 obtained with the open F value are converted into defocus amounts def1, def2, and def5 by the conversion coefficient Kd0, and the aperture F value FB. It is shown that the image shift amounts y7, y8, and y9 obtained in the above are converted to defocus amounts def7, def8, and def9 by the conversion coefficient Kd4.

  Since the image shift amounts y1 and y7 correspond to the true defocus amounts at the full aperture F value and the aperture F value FB, the defocus amounts def1 and def7 corresponding to the image shift amounts y1 and y7 are substantially the same value. The other defocus amounts def2, def5, def8, and def9 have different values.

  The matched defocus amount def1 (≈def7) is adopted as the true defocus amount.

  As described above, in the present invention, when a false focus (a drop of the correlation amount is 3 or more) occurs in the first aperture F value (aperture full aperture F value), the first aperture F Based on a difference between a plurality of image shift amounts corresponding to a drop in a plurality of correlation amounts at a value (open aperture F value) and a conversion coefficient of a second aperture F value different from the first aperture F value The difference between the plurality of defocus amounts obtained in this way is compared with the difference between the plurality of defocus amounts corresponding to the drop of the plurality of correlation amounts at the first aperture F value (open aperture F value). If the comparison results do not match, the focus detection pixel data is actually read with the second aperture F value to calculate the defocus amount, and the defocus amount obtained with the first aperture F value is calculated as the true defocus amount. It is what.

  If the comparison results match, the difference between the plurality of image shift amounts corresponding to the drop of the plurality of correlation amounts at the first aperture F value (open aperture F value) and the second aperture F value A plurality of defocus amounts corresponding to a plurality of defocus amounts obtained based on conversion coefficients of different third aperture F values and a plurality of correlation amounts at the first aperture F value (open aperture F value). Compare the difference in quantity. The value of the aperture F value is changed according to the priority order until the comparison results do not match.

  As described above, according to the present invention, when false focusing (correlation amount drop is 3 or more) occurs in the first aperture F value (open aperture F value), the defocus amount is determined in advance. A diaphragm aperture F value is determined so that a plurality of matches does not occur. The defocus amount is detected based on the determined aperture opening F value, and the true defocus amount is determined based on the defocus amount that is substantially the same among the defocus amounts obtained from the two different aperture F values. By detecting the defocus amount with the two aperture F values, the true defocus amount can be determined with certainty, and the operation to detect the defocus amount by changing the aperture F value unnecessarily can be eliminated. And a rapid focus detection operation can be achieved.

---- Modified example ---
In the above-described embodiment of the present invention, the defocus amount detection at different diaphragm aperture F values is performed in accordance with the detection of the occurrence of false focusing, but the defocus amount detection activation / prohibition at different diaphragm aperture F values is performed. May be other factors.

  For example, by activating defocus amount detection at different aperture F values according to instructions from the camera user, it is possible to make a false focus determination based on the user's experience.

  In the embodiment of the present invention described above, the defocus amount is detected at different aperture F values during the occurrence of false focusing, and the true defocus amount is determined based on the substantially matched defocus amount. However, when the position of the image plane changes greatly due to the influence of spherical aberration by changing the aperture diameter, the spherical aberration between two different aperture aperture F values is taken into account, and the values at different aperture aperture F values are taken into consideration. Compare and judge the focus amount. Specifically, when the image plane of the second aperture opening F value is shifted by the defocus amount Z with respect to the image surface of the first aperture opening F value, the detection is made with the first aperture opening F value. A defocus amount that substantially matches the defocus amount and the defocus amount obtained by subtracting the defocus amount Z from the defocus amount detected by the second aperture F value is extracted.

  In the partial enlarged view of the image sensor 212 illustrated in FIG. 4, an example in which each pixel includes a pair of focus detection pixels 313 and 314 having one photoelectric conversion unit is illustrated. However, a pair of photoelectric conversions is included in one focus detection pixel. You may make it provide a part. FIG. 30 is a partially enlarged view of such an image sensor 212, and the focus detection pixel 311 includes a pair of photoelectric conversion units. The focus detection pixel 311 illustrated in FIG. 31 performs a function corresponding to the pair of the focus detection pixel 313 and the focus detection pixel 314 illustrated in FIGS. The focus detection pixel 311 includes a microlens 10 and a pair of photoelectric conversion units 13 and 14 as shown in FIG. The focus detection pixel 311 is not provided with a color filter in order to increase the amount of light. The spectral sensitivity characteristic of the focus detection pixel 311 is a spectral sensitivity characteristic (see FIG. 9) that combines the spectral sensitivity characteristic of a photodiode that performs photoelectric conversion and the spectral sensitivity characteristic of an infrared cut filter (not shown). That is, the spectral sensitivity characteristic is obtained by adding the spectral sensitivity characteristics of the green pixel, the red pixel, and the blue pixel shown in FIG. 8, and the light wavelength region in which the focus detection pixel 311 exhibits high spectral sensitivity is the green pixel, the red pixel. And the light wavelength region in which the blue pixel exhibits high spectral sensitivity.

  32 is a cross-sectional view of the focus detection pixel 311 shown in FIG. A light shielding mask 30 is formed in proximity to the photoelectric conversion units 13 and 14, and the photoelectric conversion units 13 and 14 receive light that has passed through the opening 30 d of the light shielding mask 30. A planarizing layer 31 is formed on the light shielding mask 30, and an ND filter 34 is formed thereon. A planarizing layer 32 is formed on the ND filter 34, and the microlens 10 is formed thereon. The shape of the photoelectric conversion units 13 and 14 limited to the opening 30d by the microlens 10 is projected forward to form a pair of distance measuring pupils. The photoelectric conversion units 13 and 14 are formed on the semiconductor circuit substrate 29.

  In the imaging device in the above-described embodiment, an example in which the imaging pixel includes a color filter with a Bayer array is shown, but the configuration and arrangement of the color filter are not limited to this, and a complementary color filter (green: G, yellow) : Ye, magenta: Mg, cyan: Cy) and other arrangements other than the Bayer arrangement.

  In the focus detection pixel in the above-described embodiment, an example in which the opening shape of the light shielding mask is rectangular has been described. However, the opening shape of the light shielding mask is not limited thereto, and may be other shapes. For example, a semicircle, an ellipse, or a polygon can be used.

  In the above-described embodiment, the focus detection operation by the pupil division method using the microlens has been described. However, the present invention is not limited to such a focus detection method, and is disclosed in Japanese Patent Laid-Open No. 2008-15157. The present invention can also be applied to a pupil detection type phase difference detection type focus detection apparatus using a polarizing element.

  In the above-described embodiment, the focus detection operation by the pupil division method using the microlens has been described. However, the present invention is not limited to such a focus detection method, but is used for focus detection by the re-imaging pupil division method. Is also applicable. The focus detection operation of the re-imaging pupil division method will be described with reference to FIG. In FIG. 33, the optical axis 491 of the interchangeable lens, the condenser lenses 410 and 420, the aperture masks 411 and 421, the aperture apertures 412, 413, 422, and 423, the re-imaging lenses 414, 415, 424, and 425, and for focus detection. Image sensors (CCD) 416, 426 are shown.

  In addition, focus detection light beams 432, 433, 442, and 443, and an exit pupil 490 that is set at a position of a distance d5 ahead of the scheduled imaging plane of the interchangeable lens are shown. Here, the distance d5 is a distance measurement pupil distance determined according to the focal length of the condenser lenses 410 and 420, the distance between the condenser lenses 410 and 420 and the aperture openings 412, 413, 422, and 423, and the like. A region 492 of the aperture openings 412 and 422 projected by the condenser lenses 410 and 420 is a distance measuring pupil. Similarly, a region 493 of the aperture openings 413 and 423 projected by the condenser lenses 410 and 420 is a distance measuring pupil. The condenser lens 410, the diaphragm mask 411, the diaphragm apertures 412, 413, the re-imaging lenses 414, 415, and the image sensor 416 constitute a focus detection unit for re-imaging pupil division method phase difference detection that performs focus detection at one position. Configure.

  In FIG. 33, a focus detection unit on the optical axis 491 and a focus detection unit outside the optical axis are schematically illustrated. By combining a plurality of focus detection units, it is possible to realize a focus detection apparatus that performs focus detection by re-imaging type pupil division phase difference detection at three focus detection positions on the screen.

  The focus detection unit having the condenser lens 410 is disposed between the condenser lens 410 disposed in the vicinity of the planned imaging surface of the interchangeable lens, the image sensor 416 disposed behind the condenser lens 410, and between the condenser lens 410 and the image sensor 416. A pair of re-imaging lenses 414 and 415 for re-imaging the primary image formed in the vicinity of the scheduled imaging plane on the image sensor 416, and arranged in the vicinity of the pair of re-imaging lenses (front side in FIG. 33). It further has a diaphragm mask 411 having a pair of diaphragm openings 412 and 413.

  The image sensor 416 is a line sensor in which a plurality of photoelectric conversion units are densely arranged along a straight line, and the arrangement direction of the photoelectric conversion units is a dividing direction of a pair of distance measuring pupils (that is, an arrangement direction of aperture openings). Match. Information corresponding to the intensity distribution of the pair of images re-imaged on the image sensor 416 is output from the image sensor 416. By applying an image shift detection calculation process (correlation process, phase difference detection process) to this information, an image shift amount of a pair of images by a so-called pupil division type phase difference detection method (re-imaging method) is detected. . Further, the deviation (defocus amount) of the current imaging plane with respect to the planned imaging plane is calculated by multiplying the image shift amount by a predetermined conversion coefficient.

  The image sensor 416 is projected on the planned imaging plane by the re-imaging lenses 414 and 415, and the detection accuracy of the defocus amount (image deviation amount) is determined by the detection pitch of the image deviation amount (in the case of the re-imaging method). Is determined by the arrangement pitch of the photoelectric conversion units projected on the planned imaging plane).

  The condenser lens 410 projects the aperture openings 412 and 413 of the aperture mask 411 as regions 492 and 493 on the exit pupil 490. Regions 492 and 493 are distance measuring pupils. That is, a pair of images re-imaged on the image sensor 416 is formed by light beams that pass through the pair of distance measuring pupils 492 and 493 on the exit pupil 490. Light beams 432 and 433 passing through the pair of distance measuring pupils 492 and 493 on the exit pupil 490 are focus detection light beams.

  Even in such a re-imaging pupil division method, the present invention can be applied if the distance measurement pupils 492 and 493 are equal in size to the distance measurement pupils 93 and 94 in FIG.

  Note that the imaging apparatus is not limited to a digital still camera or a film still camera in which an interchangeable lens is mounted on the camera body as described above. For example, the present invention can be applied to a lens-integrated digital still camera, film still camera, or video camera. Furthermore, the present invention can be applied to a small camera module built in a mobile phone, a surveillance camera, a visual recognition device for a robot, an in-vehicle camera, and the like.

9, 10 Microlens, 11, 13, 14 Photoelectric conversion part, 29 Semiconductor circuit board, 30 Shading mask, 30a, 30b, 30c Aperture, 31, 32 Planarization layer, 34 ND filter, 38 color filter, 71 73, 74 Focus detection luminous flux, 90 Exit pupil, 91 Optical axis of interchangeable lens, 93, 94 Distance pupil, 95 area, 100 Shooting screen, 101, 102, 103 Focus detection area, 110 Imaging surface, 111 Focus detection pixel , 112 Microlens, 113, 114 Photoelectric conversion unit, 120 Distance pupil plane, 123, 124 Distance pupil, 125 Diaphragm surface, 133, 134 Distance pupil distribution, 135 Superimposition unit, 140 Image plane, 201 Digital still camera, 202 Interchangeable lens, 203 Camera body, 204 Mount unit, 206 Lens drive control device 208 lens for zooming, 209 lens, 210 lens for focusing, 211 aperture, 212 imaging element, 213 electrical contact, 214 body drive control device, 215 liquid crystal display element drive circuit, 216 liquid crystal display element, 217 eyepiece lens, 219 memory card, 310 Imaging pixel, 311, 313, 314 Focus detection pixel, 410, 420 Condenser lens, 411, 421 Aperture mask, 412, 413, 422, 423 Aperture aperture, 414, 415, 424, 425 Re-imaging lens, 416, 426 Image sensor, 432, 433, 442, 443 Focus detection beam, 490 Exit pupil, 491 Optical axis of interchangeable lens, 492, 493 Distance pupil

Claims (8)

A light receiving means for receiving a pair of light beams passing through a diaphragm aperture of the optical system and outputting a pair of focus detection signal data strings corresponding to a pair of images formed by the pair of light beams;
The correlation amount of the pair of focus detection signal data sequences is calculated while relatively shifting the pair of focus detection signal data sequences when the aperture value of the aperture opening is the first aperture value, and the correlation amount Correlation calculating means for calculating three or more image shift amounts related to the pair of images from the three or more minimum values when there are three or more minimum values of
Defocus amount calculation means for calculating three or more first defocus amounts by multiplying each of the three or more image shift amounts by a first conversion coefficient corresponding to the first aperture value;
First difference calculation means for calculating three or more first differences between each of the three or more first defocus amounts;
Three or more image shift amount differences between each of the three or more image shift amounts are calculated, and a second conversion coefficient corresponding to a second aperture value is calculated for each of the three or more image shift amount differences. A second difference calculating means for calculating three or more second differences by multiplying
A change to change the aperture value of the aperture opening from the first aperture value to the second aperture value when each of the 3 or more second differences is different from each of the 3 or more first differences. Means,
The correlation calculation unit causes the pair of focus detection signal data strings after the change to the second aperture value to perform a correlation calculation to calculate three or more image shift amounts, and the defocus amount calculation unit Control means for performing control to calculate three or more second defocus amounts by multiplying each of three or more image deviation amounts by the second conversion coefficient;
Of the three or more first defocus amounts, a defocus amount that substantially matches any of the three or more second defocus amounts is determined as a true defocus amount in the focus adjustment state of the optical system. A focus detection apparatus comprising: a determination unit. The focus detection apparatus according to claim 1,
Drive control means for adjusting the focus by moving the optical system based on the defocus amount,
The correlation calculation means repeatedly calculates the image shift amount,
The determination unit corresponds to a movement amount of the optical system moved by the drive control unit based on the previous true defocus amount from the previous true defocus amount among the three or more first defocus amounts. A focus detection apparatus that determines a first defocus amount closest to a defocus amount calculated by subtracting a defocus amount as a true defocus amount. The focus detection apparatus according to claim 1 or 2,
The determining means includes at least three of the first defocus amounts obtained by subtracting a correction amount based on the spherical aberration of the optical system according to the second aperture value from each of the second defocus amounts. A focus detection device that determines a defocus amount that substantially matches any of the corrected defocus amounts as the true defocus amount. The focus detection apparatus according to any one of claims 1 to 3,
The focus detection apparatus, wherein the first aperture value is an open F value of the aperture opening. In the focus detection apparatus according to any one of claims 1 to 4,
The light receiving means includes a first focus detection pixel row that receives one of the pair of light beams, and a second focus detection pixel row that receives the other of the pair of light beams,
Each of the plurality of first focus detection pixels forming the first focus detection pixel array and each of the plurality of second focus detection pixels forming the second focus detection pixel array are alternately arranged. ,
Each of the plurality of first focus detection pixels and the plurality of second focus detection pixels includes a microlens and a photoelectric conversion unit. In the focus detection apparatus according to any one of claims 1 to 4,
The light receiving means includes a plurality of focus detection pixels,
Each of the plurality of focus detection pixels includes a microlens, a first photoelectric conversion unit that receives one of the pair of light beams, and a second photoelectric conversion unit that receives the other of the pair of light beams. And a focus detection device. Photographic optics,
A pair of imaging pixels that receive imaging light flux that passes through the aperture opening of the imaging optical system and output imaging signal data corresponding to a subject image formed by the imaging light flux, and a pair that passes through the aperture opening A plurality of focus detection pixels that receive a focus detection light beam and output a pair of focus detection signal data sequences corresponding to a pair of images formed by the pair of light beams are two-dimensionally mixed and arranged. An image sensor;
The correlation amount of the pair of focus detection signal data sequences is calculated while relatively shifting the pair of focus detection signal data sequences when the aperture value of the aperture opening is the first aperture value, and the correlation amount Correlation calculating means for calculating three or more image shift amounts related to the pair of images from the three or more minimum values when there are three or more minimum values of
Defocus amount calculation means for calculating three or more first defocus amounts by multiplying each of the three or more image shift amounts by a first conversion coefficient corresponding to the first aperture value;
First difference calculation means for calculating three or more first differences between each of the three or more first defocus amounts;
Three or more image shift amount differences between each of the three or more image shift amounts are calculated, and a second conversion coefficient corresponding to a second aperture value is calculated for each of the three or more image shift amount differences. A second difference calculating means for calculating three or more second differences by multiplying
A change to change the aperture value of the aperture opening from the first aperture value to the second aperture value when each of the 3 or more second differences is different from each of the 3 or more first differences. Means,
The correlation calculation unit causes the pair of focus detection signal data strings after the change to the second aperture value to perform a correlation calculation to calculate three or more image shift amounts, and the defocus amount calculation unit Control means for performing control to calculate three or more second defocus amounts by multiplying each of three or more image deviation amounts by the second conversion coefficient;
Of the three or more first defocus amounts, a defocus amount that substantially matches any of the three or more second defocus amounts is determined as a true defocus amount in the focus adjustment state of the optical system. An imaging apparatus comprising: a determination unit. The focus detection apparatus according to claim 1,
The changing means determines whether each of the three or more second differences is different from each of the three or more first differences, and when the difference is not different, Each of the three or more image shift amount differences between the above image shift amounts is multiplied by a third conversion coefficient corresponding to a third aperture value different from the second aperture value to be 3 or more. A focus detection device that calculates a second difference of

Images