JP5159205B2 - Focus detection device and control method thereof - Google Patents

Description

  The technical idea of the present application is suitable for a focus detection technique for controlling an imaging optical system based on a phase difference between images that have passed through different parts of the pupil region.

  Conventionally, a so-called image shift type focus detection apparatus is known as a focus detection method in a focus detection apparatus. The improvement in accuracy of this method requires that the two optical images to be compared are images of the same shape that are merely laterally shifted. This is because if the optical images to be compared have different shapes due to the vignetting of the light flux used for focus detection by the taking lens, the accuracy of the focus detection result is deteriorated. For this reason, it has been necessary to limit the aperture ratio of the photographic lens optical system and to limit the arrangement of the focus detection area so that the light flux used for focus detection does not cause vignetting by the photographic lens.

  In addition, it is known that the high-pass filter processing of the image signal is effective for the deterioration of the image signal due to the vignetting of the light beam used for focus detection. For example, it is patent document 1. FIG. However, this vignetting countermeasure has a problem that the focus detection capability is low for a subject having a low frequency component such as a human face. This vignetting countermeasure is to eliminate low frequency components of the electrical signal output from the focus detection device. Therefore, this method is effective in removing the adverse effects of minute vignetting of the light beam used for focus detection.

  On the other hand, two optical images are compared against image signal degradation due to vignetting of a light beam used for focus detection, and at least one of the two images is corrected based on the comparison result. It is effective to perform a correlation operation on the obtained image signal. For example, it is patent document 2. FIG.

As a method for correcting an optical image in which vignetting has occurred, a method for calculating a correction amount from the shape and position of the exit pupil of a photographing lens is known. For example, it is patent document 3. FIG.
JP 7-318793 A JP 2002-14277 A JP-A-3-214133

  However, the countermeasure against the vignetting of the light beam used for focus detection described in Patent Document 2 described above is accurate vignetting when the subject has a complicated luminance distribution or the optical image includes subjects at various distances. In some cases, the state cannot be detected, and a focus detection result with a large error is calculated. The countermeasure against the vignetting of the light beam used for focus detection described in Patent Document 2 is because the vignetting state is obtained from the output of the light receiving element that receives the optical image.

  Further, the optical image correction method described in Patent Document 3 cannot cope with a change in the amount of vignetting of an optical image due to a manufacturing error of a photographing lens or an imaging device or a secular change, and a variation in focus detection results increases. There was a problem. The optical image correction method described in Patent Document 3 is for determining a correction amount based on information of a photographing lens to be used.

The technical idea of the present application is a focus detection device that controls the imaging lens based on a phase difference of an optical image by a pair of light beams that have passed through different areas of an exit pupil of the imaging lens, the lens type information of the imaging lens and The correction means that may or may not correct vignetting according to the focal length information, and when the vignetting correction is performed by the correction means, the degree of coincidence of the image signals after the vignetting correction Whether or not to correct the vignetting is determined before performing the correlation calculation, and when the matching degree of the image is high, the imaging lens is controlled based on the phase difference obtained by the correlation calculation from the image signal. Control means for not controlling the photographing lens based on the image signal.

  According to the present invention, it is possible to reduce the loss of reliability of the focus detection result even if the image signal is corrected.

Example 1
The first embodiment is an example in which the focus detection apparatus of the present invention is applied to a single-lens reflex digital camera with interchangeable lenses. FIG. 1 is a central sectional view of a single-lens reflex camera incorporating a focus detection device of the present invention.

  In the figure, 200 is a single-lens reflex camera body, 201 is an imaging lens, and L is an optical axis of the imaging lens 201. An image sensor unit 204 including an optical low-pass filter, an infrared cut filter, and an image sensor is disposed near the planned imaging plane of the image lens 201. The imaging lens 201 and the imaging element unit 204 are provided with a main mirror 202 and a sub mirror 203 that are retracted out of the imaging light beam when imaging by a known quick return mechanism. The main mirror 202 is a half mirror, and the imaging light flux is separated into reflected light guided to the upper finder optical system and transmitted light incident on the sub mirror 203. The reflected light forms an image on the mat surface of the focusing plate 205 having the mat surface and the Fresnel surface, and is guided to the eyes of the observer through the pentaprism 206 and the eyepiece lens group 207. On the other hand, the transmitted light changes its optical path downward by the sub mirror 203 and is guided to the focus detection device 208 of the present invention.

  Next, the focus detection apparatus included in the camera described in FIG. 1 will be described. FIG. 2 is a perspective view of the focus detection apparatus 208 of the present invention, and uses phase difference focus detection as a distance measurement principle. Although the actual focus detection apparatus is made compact by folding the optical path with a reflecting mirror or the like, the figure is developed straight to eliminate the complexity of the figure.

  A field mask 210 has a central portion and cross-shaped openings 210a, 210b, and 210c on the left and right sides thereof. The field mask 210 is disposed in the vicinity of a position equivalent to the planned imaging plane of the imaging lens 201 and the imaging plane of the imaging element unit 204. A field lens 211 is disposed behind the field mask 210. The field lens 211 includes a plurality of lens portions 211a, 211b, and 211c having different optical actions, and each lens portion has a different lens optical axis. The lens portions 211a, 211b, and 211c correspond to the openings 210a, 210b, and 210c of the field mask 210, respectively.

  Reference numeral 212 denotes a stop having a plurality of openings, and reference numeral 213 denotes a secondary image forming lens unit (image forming lens unit) having lens portions corresponding to the plurality of openings of the stop 212. The secondary imaging lens unit 213 re-images the object image on the planned imaging surface imaged by the objective lens 201 on the light receiving element array of the light receiving elements 215 arranged behind the object image. An infrared cut filter that removes an infrared wavelength component unnecessary for focus detection is disposed in the vicinity of the stop 212, but is omitted.

  3 to 5 are plan views of members in the perspective view of FIG. Among these, FIG. 3 is a plan view of the diaphragm 212 as viewed from the field mask 210 side. FIG. 4A is a plan view of the incident surface side of the secondary imaging lens unit 213 as viewed. FIG. 4B is a plan view of the exit surface side of the secondary imaging lens unit 213. FIG. 5 is a plan view of the light receiving element 215 viewed from the field mask 210 side.

  In FIG. 3, the diaphragm 212 has two pairs of openings 212a-1 and 212a-2 and 212a-3 and 212a-4 at the center. Two pairs of openings 212b-1 and 212b-2, 212b-3 and 212b-4 are on the right side, and two pairs of openings 212c-1 and 212c-2, 212c-3 are on the left side. And 212c-4, respectively.

  4A, a prism portion corresponding to the aperture of FIG. 3 is formed on the incident surface side of the secondary imaging lens unit 213. In FIG. Therefore, like the diaphragm 212, two pairs of prism portions 213a-1 and 213a-2, 213a-3 and 213a-4 are formed in the central portion. Two pairs of openings 213b-1, 213b-2, 213b-3, and 213b-4 are on the right side, and two pairs of openings 213c-1, 213c-2, and 213c-3 are on the left side. And 213c-4 are formed respectively.

  In FIG. 4B, a spherical lens portion is formed on the exit surface side of the secondary imaging lens unit 213 corresponding to the prism portion of FIG. Similarly, two pairs of lens portions 214a-1 and 214a-2 and 214a-3 and 214a-4 are formed in the central portion. Also, two pairs of openings 214b-1 and 214b-2, 214b-3 and 214b-4 are on the right side, and two pairs of openings 214c-1, 214c-2 and 214c-3 are on the left side. And 214c-4 are formed respectively.

  In FIG. 5, the light receiving element 215 is formed with a light receiving element array corresponding to the lens portion of FIG. Similarly, two pairs of light receiving element arrays 215a-1 and 215a-2, 215a-3 and 215a-4 are formed in the central portion. Two pairs of openings 215b-1 and 215b-2, 215b-3 and 215b-4 are shown on the right side, and two pairs of openings 215c-1 and 215c-2 and 215c-3 are shown on the left side. And 215c-4 are formed.

  3 to 5 correspond to the opening portion of the field mask 210 and the lens portion of the field lens 211, respectively. For example, the light beam that has passed through the field mask opening 210b passes through the field lens 211b, and passes through two pairs of aperture openings 212b-1 and 212b-2, and 212b-3 and 212b-4, thereby forming four light beams. To be separated. The four light beams are incident on two pairs of prism portions 213b-1, 213b-2, 213b-3, and 213b-4 of the secondary imaging lens unit 213, respectively. And it inject | emits from two sets of lens parts 214b-1 and 214b-2, 214b-3 and 214b-4. Then, four optical images corresponding to the field mask opening 210b are formed on the two pairs of light receiving element arrays 215b-1, 215b-2, 215b-3, and 215b-4 of the light receiving element 215. Four crosses 216-1, 216-2, 216-3, and 216-4 indicated by dotted lines in FIG. 5 indicate respective optical images, and optical images 216-1 and 216-2, and optical images 216-3 and 216, respectively. -4 is a pair.

  In the focus detection apparatus having the above-described configuration, the optical images in the pair of optical images move toward or away from each other as the imaging lens 201 defocuses on the planned imaging plane.

  For example, the optical images inside the pair of optical images 216-1 and 216-2 move in the vertical direction, and this movement is detected by arranging the light receiving element rows in the vertical direction as shown in FIG. That is, the light quantity distribution relating to this optical image is detected based on the outputs of the light receiving element arrays 215b-1 and 215b-2. By calculating the light quantity distribution using a well-known correlation calculation means, the amount of change with respect to the distance between the pair of optical images associated with the imaging lens 201 is obtained.

  If the amount of change in the optical image interval of the pair can be known, the relationship between the amount of change and the defocus amount of the imaging lens 201 is approximated in advance by a polynomial or the like using the amount of change as a variable. The amount of focus can be predicted. Then, focus detection of the imaging lens 201 can be performed. In the case of the light receiving element arrays 215b-3 and 215b-4, the basic operation is the same except that the vertical direction described above is changed to the horizontal direction.

  The pair of light receiving element arrays 215-1 and 215b-2 are suitable for focus detection of a subject having a contrast component in the vertical direction because the light receiving elements are arranged in the vertical direction. On the other hand, the light receiving element arrays 215b-3 and 215b-4 are suitable for focus detection of a subject having a contrast component in the left and right direction because the light receiving elements are arranged in the left and right direction. When both are combined, so-called cross-type focus detection is performed that is not affected by the direction of the contrast component of the subject.

  In the above description, only the part relating to the subscript b has been described, but the same applies to the subscripts a and c.

  FIG. 6 is a diagram in which each light receiving element row of the light receiving element 215 is back projected onto the field mask 210. Since the field mask is disposed in the vicinity of the planned imaging plane of the imaging lens 201, there is no problem even if FIG. 6 is considered as the planned imaging plane. In the drawing, a rectangle indicated by a dotted line that is slightly larger than the field mask 210 is an imaging range by the imaging lens 201, and back projection images 218a, 218b, 218c, and 219a of light receiving element arrays indicated by crosses are formed in three field mask openings. 219b and 219c are formed. Since the pair of light receiving element arrays coincide on the planned imaging plane, they are displayed in an overlapping manner. Since the back-projected images 218a, 218b, 218c, 219a, 219b, and 219c are light receiving element arrays, the light amount distribution of the subject can be detected in this cross area. That is, these back-projected images become a so-called cross-type focus detection region. In this embodiment, the cross-type focus detection area is provided with a total of three places, that is, the center of the imaging range 217 and two off-axis positions as shown in FIG. 6, and the focus detection of the subject in this focus detection area can be performed. The cross-type focus detection area has two directions in which the spreading directions are orthogonal to each other, so that focus detection can be performed on most subjects regardless of the contrast component direction of the subject.

  Next, the relationship between the light beam that reaches each light receiving element array of the light receiving element 215 of the focus detection device 208 and the light beam that passes through the imaging lens 201 will be described with reference to FIGS.

  FIG. 7 is a top view in which the imaging lens 201 and the focus detection device 208 are straightly developed, and illustrates the light flux passing through the intersection of the field mask 210 and the optical axis L of the imaging lens 201. FIG. 7 shows only a part of the aperture 212 and the secondary imaging lens unit 213 for ease of explanation.

  Reference numerals 201 a, 201 b, and 201 c are lenses that constitute the imaging lens 201. Reference numeral 201d denotes an imaging lens diaphragm that adjusts the diameter of a light beam passing through the imaging lens 201. Reference numerals 201e and 201f denote a front frame member and a rear frame member that hold the imaging lens 201, respectively.

  In FIG. 7, the light flux passing through the intersection of the field mask 210 and the optical axis L of the imaging lens 201 is determined by the field lens stop 201d. Therefore, in the focus detection device 208 of the present invention, the light beam that passes through the region where the openings 212a-3 and 212a-4 of the stop 212 are back projected onto the imaging lens stop 201d reaches the light receiving element 215.

  FIG. 8 is a diagram in which projection images of the respective members viewed from the intersection of the field mask 210 and the optical axis L of the imaging lens 201 are superimposed on the imaging lens aperture 201d. Reference numerals 300a-3 and 300a-4 denote back-projected images of the openings 212a-3 and 212a-4 of the stop 212, respectively. Reference numeral 301 denotes a projected image by the imaging lenses 201a and 201b on the stop of the imaging lens front frame member 201e. Reference numeral 302 denotes an image projected by the imaging lens 201c on the stop of the imaging lens rear frame member 201f. As shown in FIG. 8, when the entire regions of the back-projected images 300a-3 and 300a-4 are included inside the projected images (301, 302) of the imaging lens members and the imaging lens stop 201d, the light flux used for focus detection There is no vignetting. The luminous flux that passes through the intersection of the field mask 210 and the optical axis L of the imaging lens 201 is determined by the diameter of the imaging lens aperture 201d, but is not affected by vignetting described later. Hateful.

  FIG. 9 is a top view in which the imaging lens 201 and the focus detection device 208 are straightly developed, and illustrates a light flux passing through an end point far from the imaging lens optical axis L of the backprojected image 219c of the light receiving element on the field mask 210. Is. H indicates the image height of the end point far from the imaging lens optical axis L of the back projection image 219c of the light receiving element on the field mask 210. In FIG. 9, the luminous flux passing through the end point of 219c is not determined by the field lens stop 201d, the lower luminous flux in the figure is taken by the imaging lens front frame member 201e, and the upper luminous flux is taken by the imaging lens rear frame member 201f. It has been decided. In other words, the luminous flux passing through the imaging lens stop 201d is blocked by the imaging lens front frame member 201e and the rear frame 201f member, and vignetting occurs.

  FIG. 10 shows an image obtained by superimposing projection images of the respective members viewed from the end point of 219c on the imaging lens stop 201d. The areas where the apertures 212c-3 and 212c-4 of the diaphragm 212 are back-projected onto the imaging lens diaphragm 201d by the field lens 211 and the imaging lens 201c are diaphragm back-projected images 300c-3 and 300c-4. Reference numeral 301 denotes a projected image of the imaging lens front frame member 201e viewed from the end point of 219c. Reference numeral 302 denotes a projected image of the imaging lens rear frame member 201f viewed from the end point 219c. Here, unlike FIG. 8, each member is projected onto the imaging lens aperture 201d from a position where the image height on the image plane is located (the end point of 219c), so that each projected image is the image of the imaging lens aperture 201d. Eccentric with respect to the center. As shown in FIG. 10, when the entire region of the back projection image 300c-3 is included inside the projection image (301, 302) of each member of the imaging lens and the imaging lens aperture 201d, the light receiving element array 215c-3 receives light. The luminous flux is not vignetted. However, a part of the back projection image 300c-4 exists outside the projection image 301 of the imaging lens front frame member (shaded portion in FIG. 10). In the region within the shaded area, vignetting occurs in the light beam used for focus detection, and the output of the light receiving element array 215c-4 decreases.

  The degree of vignetting of the light beam used for focus detection for a certain imaging optical system is determined by the amount of eccentricity of the projection image of each member on the imaging lens 201d, that is, the image height H of the focus detection area. In other words, the smaller the image height H of the focus detection area, the smaller the degree of vignetting of light used for focus detection by vignetting of the imaging lens.

  FIG. 11 shows the outputs of the light receiving element arrays 215c-3 and 215c-4 for a subject with uniform brightness when vignetting occurs as shown in FIG. The horizontal axis represents each pixel in ascending order of image height, and the vertical axis represents the output of each pixel. In general, the output of the optical element 215 is non-uniform even when the luminance of the subject is uniform due to the peripheral dimming of the imaging lens 201 and the focus detection optical system, variations in sensitivity of the optical element 215, and the like. . For this non-uniformity, the correction amount is calculated and stored in the camera in a state where there is no vignetting of the light beam used for focus detection by the imaging lens, so that output correction of the photoelectric element, so-called shading correction, is performed. FIG. 11 shows the output after the shading correction.

  FIG. 11a shows the output of the light receiving element array 215c-3. In FIG. 10, since vignetting does not occur in the back-projected image 300c-3 of the stop 212c-3, a constant output can be obtained by correcting the decrease in the amount of light around the imaging lens. FIG. 11b shows the output of the light receiving element array 215c-4. In FIG. 10, even if the decrease in the peripheral light amount of the imaging lens is corrected, the output decreases as the image height in the focus detection area increases. This is because the back projection image 300c-4 of the aperture stop 212c-4 has vignetting caused by the projection image 301 of the imaging lens front frame member 201e.

  When designing a focus detection device of an imaging lens exchangeable camera so that vignetting does not occur in the light beam used for focus detection, the back projection images of the focus detection device's aperture are all on the imaging lens aperture for various imaging lenses. It is necessary to arrange it inside the projected image of the member. This can be solved by reducing the image height of the focus detection area, shortening the interval between the back projection images of the aperture of the focus detection device, and reducing the area. However, this solution leads to a narrow focus detection range, a deterioration in focus detection accuracy due to a shortened baseline length, and a deterioration in accuracy for low-luminance subjects in focus detection. Therefore, a means for performing focus detection in a state where vignetting has occurred in the light beam used for focus detection is necessary. If this focus detection can be performed with high accuracy, the accuracy of focus detection can be improved and the focus detection range can be expanded. Hereinafter, means for performing focus detection in a state where vignetting has occurred in the light beam used for focus detection will be described.

  FIG. 12 is a block diagram showing an example of a camera focus detection apparatus suitable for carrying out the present invention.

  In the figure, PRS is a central processing circuit of the camera, which is a one-chip microcomputer in which, for example, a CPU, RAM, ROM, ADC (A / D converter), input / output ports and the like are arranged. In the ROM, a series of camera control software including AF control and parameters are stored. Information for determining the presence or absence of vignetting of a light beam used for focus detection by an imaging lens described later is also stored in the ROM. The PRS constitutes a part of the calculation means.

  DBUS is a data bus, and SHT receives data input via the data bus DBUS while the control signal CSHT is input from the central processing circuit PRS. A shutter control circuit (APR) for controlling the travel of the shutter front curtain and rear curtain (not shown) based on the data receives data input via the data bus DBUS while the control signal CAPR is input. The lens communication circuit that performs serial communication with the lens control circuit LNSU transmits lens driving data DCL to the lens control circuit in synchronization with the clock signal LCK. This is based on data that receives data input via the data bus DBUS while the control signal CLCOM is input. At the same time, lens information DLC is serially input. The lens driving data DCL includes information on the type of the mounted camera, the type of the focus detection device of the camera, information on the lens driving amount at the time of imaging, and the like.

  BSY is a signal for informing the camera side that a focus adjusting lens (not shown) is moving. When this signal is generated, the serial communication is not performed. SPC is a photometric circuit, and when it receives a control signal CSPC from the control circuit, it sends a photometric output SSPC to the central processing circuit PRS. The photometric output SSPC is A / D converted by the ADC in the central processing circuit PRS and used as data for controlling the shutter control circuit SHT and the aperture control circuit APR.

  Reference numeral 400 denotes a light projection circuit for projecting auxiliary light for focus detection, which drives the LED to emit light by the control signal ACT and the synchronous clock CK from the central processing circuit PRS.

  SNS is a ranging light receiving circuit having a plurality of pairs of light receiving element arrays 215a-1 / 215a-2, 215a-3 / 215a-4,... 215c-3 / 215c-4. The SNS is configured such that each light receiving circuit for distance measurement receives an image at a position corresponding to the focus detection areas 218 and 219 on the screen. SDR is a sensor driving circuit that controls the light receiving circuits 1 to 5 in accordance with signals STR and CK input from the central processing circuit PRS. The SDR controls the light receiving circuits 1 to 5 by the control signals φ1, φ2, CL, and SH, selects one of the light receiving circuits by the selection signals SEL1 to SEL6, and an image signal obtained from the selected light receiving circuit. SSNS is transmitted to the central processing circuit PRS.

  FIG. 13 shows how the lens control circuit LNSU obtains the focal length information and distance ring information of the lens. The memory means 401 stores the position of the distance ring, the position of the zoom ring, the coefficient of the defocus amount versus the focus adjustment lens extension amount, and the like. The stored information is converted into a signal, calculated by the CPU 403 in the lens control circuit LNSU, and input as lens information DLC to the central processing circuit PRS via the lens communication circuit LCOM. The focal length information of an imaging lens having a plurality of focal lengths, that is, a so-called zoom lens, is a representative value of each range obtained by dividing a continuously changing focal length into a plurality of ranges. Further, since the position information of the distance ring in this case is not directly used for the focus calculation, so much accuracy is not required.

  The focus detection operation of the camera focus detection apparatus having the above configuration will be described with reference to the flowchart of FIG.

  In the camera imaging flow, when a first stroke operation (so-called half-press operation) of the release switch of the camera or a dedicated focus detection operation start switch is detected, the focus detection operation is started.

  In step 1 (abbreviated as “S” in the figure), a focus detection area is designated from the focus detection areas 218 and 219 provided in the screen in response to a manual operation of the photographer or according to a predetermined algorithm. To do. The designated focus detection areas are numbered with focus detection area numbers in order from 1 in the order of focus detection processing. In step 2, the light receiving element is exposed with respect to the designated focus detection area. The exposure amount adjusting means provided in the light receiving element 215 is controlled so as not to saturate the amount of light received, and the exposure ends. Thereafter, one of the light receiving circuits is selected according to the order of focus detection processing, and the image signal SSNS obtained from the selected light receiving circuit is transmitted to the central processing circuit PRS.

  In step 3, by inputting 1 to the focus detection area number N which is a parameter for managing the area for performing the focus detection process, the area for performing the focus detection process is set, and the calculation process is started.

  In step 4, it is determined whether or not there is a possibility that vignetting may occur with respect to the focus detection region where the arithmetic processing is performed. As described above, this determination is determined by the combination of the imaging lens and the focus detection device. Therefore, it is possible to determine whether or not there is a possibility of vignetting by storing in the camera a combination of the imaging lens type (lens type information), focal length information, and focus detection area in which vignetting occurs. .

  FIG. 15 shows an example of a table stored in the camera. The imaging lens type (lens type information) and the focal length information of the lens information DLC that are always input from the memory means 400 in the imaging lens to the central processing circuit PRS via the lens communication circuit LCOM are stored in the camera. Search from the table. For example, when the mounted imaging lens is “lens 3” and the focal length used is “focal length 3-3”, the focus detection areas where vignetting occurs are 218b and 219b. When the focus detection area where the focus detection process is to be performed is 218b or 219b, YES is determined in step 4. On the other hand, otherwise, it is determined as NO. In this embodiment, as an example of the table stored in the camera, the focus detection areas 218b, 218c, 219b, and 219c of the imaging lens 201 that are symmetrical with respect to the optical axis are stored as one piece of data. In this respect, the focus detection areas 218b and 219b are symmetrical with respect to the optical axes of the focus detection areas 218c and 219c and the imaging lens 201, respectively, and are theoretically the same, so only the focus detection areas 218b and 219b are stored. Also good.

  If YES in step 4, a correction amount calculation formula and an adjustment coefficient are simultaneously selected. In the example described above (in the case of “lens 3” and “focal length 3-3”), the correction amount calculation formula is selected as “correction formula C” and the adjustment coefficient “× 0.8”. Details of the correction method will be described later.

  Returning to FIG. 14, in step 5, when NO is determined in step 4, focus detection processing is performed when there is no possibility of vignetting. Here, based on the image signal from the light receiving element array 215, the focus detection result of the designated focus detection region, that is, the defocus amount is calculated. Here, the phase difference is detected by using a known method using correlation calculation, for example, a method disclosed in Japanese Patent Publication No. 5-88445, using image signals obtained from the pair of light receiving element arrays 215. . Thereby, the defocus amount including the defocus direction can be calculated. Also, it is determined whether or not each focus can be detected by a known method based on the contrast of the image signal. If YES is selected in step 4, the process proceeds to step 6 to perform focus detection processing when there is a possibility of vignetting. First, in step 6, vignetting correction of the image signal is performed. Details will be described later.

  Next, in step 7, the suitability of focus detection is determined based on the corrected image signal, and the magnitude of the correction error is determined with respect to a predetermined threshold value. A method for determining the correction error will be described later. When it is determined that the focus detection is appropriate, that is, when it is determined that the correction error is small with respect to the threshold value (in the case of YES at step 7), the process proceeds to step 5 as in the case where the image signal is not corrected, Focus detection processing is performed. If it is determined that the focus detection is inappropriate, that is, if it is determined that the correction error is large with respect to the threshold value (NO in step 7), the focus detection in this focus detection area becomes impossible, and the process proceeds to step 8. This is because even if focus detection is performed as it is, there is a high possibility that an erroneous focus detection result will be obtained. This technical idea is that the reliability of the correction performed can be determined by determining the correction error of the corrected image signal after correcting the output of the light receiving element. Thereby, even if the image signal is corrected, the reliability of the focus detection result is not impaired. Further, even when the vignetting condition varies due to manufacturing errors and changes over time, a focus detection result with high reliability is obtained by not performing focus detection when the correction error is large.

  In step 8, it is determined whether or not the processing for all focus detection areas that need to be performed the focus detection process specified in step 1 has been completed. If NO in step 8, the focus detection area is set again. Specifically, the focus detection area number N plus 1 is input (step 9). Then, Step 4 and subsequent steps are repeated for the newly set focus detection area.

  If YES in step 8, that is, if the focus detection process for the entire field of view is completed, the process proceeds to step 10, where the focus detection area is determined from the focus detection result, and the in-focus state is determined from the calculated defocus amount of the area. Determine if there is. The method for determining the focus detection area can be performed using a known technique for selecting a focus detection area with higher reliability, for example, a technique disclosed in Japanese Patent Application Laid-Open No. 2007-052072. If NO in step 10, that is, if it is determined that it is not in focus, the imaging lens starts to be driven according to the defocus amount calculated from the focus detection processing result, and the focus detection operation is performed again from step 2 ( Step 11). If YES in step 10, that is, if it is determined that the camera is in focus, the focus detection process is terminated by waiting for a second stroke operation (so-called full press operation) of the release switch for starting the imaging operation. If the standby state continues for a certain period of time, the focus detection situation may have changed, so the focus detection operation is started again.

Next, the vignetting correction of the image signal performed in step 6 of FIG. 14 will be described. The vignetting correction of the image signal is performed using the correction amount calculation formula and the adjustment coefficient selected in Step 4 of FIG. In general, the F number of a so-called zoom lens having a plurality of focal lengths increases as the focal length increases. In such an imaging lens, when the focal length is the longest, the amount of vignetting of the light beam used for focus detection becomes large. Therefore, in this embodiment, the correction when the amount of vignetting is the greatest is emphasized, and the correction is performed by the following equation.
OUT (g) = IN (g) × (K (g) × T) (Equation 1)
In Equation 1, g is the number of a pixel constituting the light receiving element array, and IN (g) indicates an output before correction of the pixel g. K (g) is a coefficient to be applied to the output of each pixel constituting the light receiving element array when the focal length is the longest. K (g) may be a set of numbers corresponding to each pixel or may be a polynomial.

  Next, the adjustment coefficient T will be described. FIG. 18 shows an output before correction of each pixel of a certain light receiving element. The horizontal axis indicates the pixel number, and the vertical axis indicates the output of each pixel, which corresponds to “g” and “IN (g)” in Equation 1, respectively. In the figure, vignetting in three focal length states is shown. IN (g) -1 indicates a state where the vignetting is the largest in the state where the focal length is the longest. IN (g) -2 and IN (g) -3 indicate that the focal length is gradually shorter than that of IN (g) -1, and the vignetting state is improved. In this embodiment, the vignetting state of IN (g) -1 is corrected, and K (g) is determined so that the output of each pixel becomes constant. That is, in that case, the adjustment coefficient T is 1. Therefore, if IN (g) -2 and IN (g) -3 are corrected by K (g), so-called overcorrection occurs, and the output of the pixel in which vignetting has occurred becomes larger than necessary. Therefore, in this embodiment, the adjustment amount T is adjusted using the adjustment coefficient T. For example, by setting the adjustment coefficient of IN (g) -2 to 0.6, 60% of the correction amount determined by K (g), and the adjustment coefficient of IN (g) -3 to 0.8 As a result, 80% of the correction amount determined by K (g) can be used as the correction amount, and the output can be corrected. By adopting such a method, it is possible to reduce the amount of information stored and processed in the camera as compared with the case where K (g) is stored according to the vignetting state. Further, since the vignetting state changes almost proportionally as the focal length becomes longer, there is no problem with the accuracy of correction.

  By multiplying these three coefficients, OUT (g), that is, the output of the corrected pixel g can be calculated. By adopting such a correction mode, it is possible to accurately perform correction in a state where the degree of vignetting is the largest while suppressing the amount of data stored in the camera. The correction amount other than the case where the focal length is the longest is a constant multiple of that when the focal length is the longest, but when the correction amount is not so large, the correction amount can be accurately corrected.

  Next, the correction error determination method performed in step 7 of FIG. 14 will be described. In this embodiment, a correction error determination method is performed using 1) the degree of coincidence of a pair of image signals, and 2) an H level that is an evaluation value using a predicted value of a focus detection shift amount due to the correction error as a parameter. Whether or not the target focus detection area can be used is determined based on the magnitude of the H level value calculated from the pair of image signals.

First, 1) the degree of coincidence between a pair of image signals will be described. When performing focus detection processing, it is known that the result of correlation calculation deviates from a true value if the degree of coincidence of a pair of image signals is poor due to ghost light or the like. Similarly, when the luminous flux used for focus detection is vignetted, the degree of coincidence of the image signals is deteriorated and affects the result of the correlation calculation. Therefore, at the H level, a function using the output difference s of a pair of pixels constituting a pair of light receiving elements (for example, 215b-1 and 215b-2) as a parameter is defined as a matching degree U (Equation 2).
Matching degree U = U (s) (Equation 2)
By determining the correction error using the degree of coincidence U defined in this way, the focus detection process can be performed when the degree of coincidence of the images is good and the reliability is high.

Next, 2) the predicted value of the focus detection shift amount due to the correction error will be described. If correlation calculation is performed after performing known filter processing, the focus detection error may not increase even if the degree of coincidence between the two images is not so good. For example, the spatial frequency component of the subject image in the portion where the degree of coincidence of the two image signals is deteriorated due to vignetting may be removed by the filter. Therefore, by calculating a predicted value for how much the focus detection result is shifted from the image signal due to the correction error, it is possible to make the focus detection process impossible when there is a possibility that the focus detection shift is greatly generated. it can. In this embodiment, the predicted value of the focus detection deviation amount due to the correction error is defined as a function E using parameters of the first contrast information c1 and the second contrast information c2.
Predicted value of focus detection deviation amount due to correction error E = E (c1, c2) (Equation 3)
The parameter c1 in Expression 3 is a sum of absolute values of output differences between adjacent pixels in the image signal, and c2 is a square sum of output differences between adjacent pixels in the image signal. In general, c1 and c2 have a correlation with a predicted value of a focus detection deviation amount due to a correction error.

  The correlation will be described with reference to FIGS. FIG. 16a shows a pair of image signals (215c-3, 215c-4) when the value of c1 is large, and FIG. 16b shows a pair of image signals (215c-3, 215c-4) when the value of c1 is small. ). Since c1 is the sum of the absolute values of the output differences between adjacent pixels, when c1 increases, there is a tendency for the height difference to increase in the output of the image signal. Therefore, in FIGS. 16a and 16b, a state where the height difference of the image output is large and a state where it is small are representatively shown. In both figures, correction is not sufficiently performed due to the correction error, and the left side of the output of each pixel of the light receiving element array of 215c-3 is affected by vignetting, and the output of 215c-3 is more than the output of 215c-4. It shows a reduced state. J1 and J2 indicate the barycentric positions of the image outputs of 215c-3 and 215c-4. The barycentric position is an average of the integrated values of the pixel outputs in the pixel array direction.

  In FIG. 16A and FIG. 16B, the output difference of the image due to vignetting is indicated by the hatched portion. In FIG. 16a, since the height difference of the output of the image signal is larger than in the state of FIG. 16b, the shaded area is larger, that is, the output of 215c-3 with respect to 215c-4 is smaller. This is because the output decreases at a constant rate due to vignetting, and thus the absolute value of the amount of change in output due to vignetting increases as the absolute value of the output increases.

  When there is no vignetting or when the correction error is 0, the paired optical images (215c-3 and 215c-4) are substantially equal, so that the center of gravity, J1, and J2 are substantially equal in the focused state. . However, when vignetting occurs, the similarity between the pair of optical images is lost, and the center of gravity is displaced. In FIG. 16, the image signal of 215c-3 is shaped such that the left side is cut compared to 215c-4, and the gravity center position J2 moves to the right in the figure with respect to the gravity center position J1.

  Comparing the states of FIGS. 16a and 16b, since the hatched area occupied in the output sum (area of the entire image signal) of each pixel in the pair of light receiving element rows is larger in FIG. The difference of J2 becomes large. That is, the deviation of the center of gravity position increases as the difference in height between the output of the image signals in the area where the vignetting occurs is larger. This indicates that the larger the value of c1, the higher the possibility that the shift amount of the center of gravity position becomes large.

  Since the correlation calculation at the time of focus detection is almost the same as calculating the shift of the centroid position of the pair of images, the focus detection result is shifted when the centroid position is shifted. From this, it can be seen that there is a correlation between the shift amount between c1 and focus detection, and if c1 is large, there is a high possibility of calculating an erroneous focus detection result due to a correction error.

  Next, a correlation between c2 and the predicted value of the focus detection deviation amount due to the correction error will be described with reference to FIG. 17a shows a pair of image signals (215c-3, 215c-4) when the value of c2 is small, and FIG. 17b shows a pair of image signals (215c-3, 215c-4) when the value of c2 is large. ). Since c2 is the square sum of the output differences of adjacent pixels in the image signal, when the value of c2 is small, the number of irregularities in the shape of the image signal tends to be small and the amplitude tends to be small. Therefore, FIGS. 17a and 17b representatively show a state where the number of irregularities in the shape of the image signal is small and a large number. In both figures, correction is not sufficiently performed due to the correction error, and the effect of vignetting remains on the left side of the output of each pixel of the light receiving element array, and the output of 215c-3 is smaller than 215c-4. . J1 and J2 indicate the barycentric positions of the image outputs 215c-3 and 215c-4, as in FIG.

  In FIGS. 17a and 17b, an output difference of an image due to vignetting is indicated by a hatched portion. The shape of the image signal output in FIG. 17a has one convex shape, whereas the shape of the image signal output in FIG. 17b has two convex shapes, so that the area of the entire image signal is as shown in FIG. Smaller than FIG. 17b.

  Also in FIG. 17, as in FIG. 16, the image signal of 215 c-3 is shaped such that the left side is cut compared to 215 c-4, and the gravity center position J <b> 2 is in the drawing with respect to the gravity center position J <b> 1. Move to the right. Comparing the states of FIGS. 17a and 17b, since the hatched area in the area of the entire image signal is larger in FIG. 17a, the difference between the gravity center positions J1 and J2 of the image signal becomes larger. That is, when the area of the area where the vignetting occurs (the shaded area) is constant, the displacement of the center of gravity position becomes larger as the area of the entire image signal is smaller. This indicates that the smaller the c2 is, the higher the possibility that the shift amount of the center of gravity position becomes large.

  As described above, since the focus detection result is shifted when the position of the center of gravity is shifted, there is a correlation between the shift amount between c2 and focus detection, and when c2 is small, there is a high possibility of calculating an incorrect focus detection result due to a correction error. I understand that.

  As described above, by using c1 and c2, the predicted value of the focus detection deviation amount due to the correction error can be calculated.

Next, calculation of the H level will be described. The H level is calculated from the coincidence degree U determined from the paired image signals and the predicted value E of the focus detection deviation amount due to the correction error determined from the characteristics of the image signals as in the following equation.
H = U (s) × E (c1, c2) (Equation 4)
The H level is calculated from the deviation of the degree of coincidence of elephants caused by the correction error and the predicted value of the focus detection deviation, thereby determining the correction error. Therefore, even when the degree of coincidence of the paired images is good, avoid performing focus detection processing when it is expected that there is a high possibility of calculating an erroneous focus detection result due to a correction error, and vice versa. be able to. Thereby, even if the image signal is corrected, the reliability of the focus detection result is not impaired. Further, the correction error amount of the image signal correction varies due to manufacturing errors and changes with time. By not performing focus detection when the correction error is large, only a focus detection result with high reliability can be obtained.

  In this embodiment, the focus detection method has been described using a focus detection apparatus having an optical system that performs re-imaging. In this regard, the focus detection method can be applied not only to this but also to various other focus detection methods. For example, the phase difference type focus detection may be performed using a pair of pixels arranged on the scheduled imaging surface 210 of the imaging lens, using light beams passing through different portions of the imaging lens 210. In Steps 5 and 7, when the correction error is large, the focus detection process is not performed for the focus detection field of view. In this regard, when the correction error is large, the focus detection process may be performed with priority given to the focus detection field of view with a smaller correction error. In this case, the imaging lens is driven in step 11 to focus on the focus detection field of view with a smaller correction error than the focus detection field of view with a large correction error.

  In the present embodiment, since information on vignetting is stored in the camera, the amount of information to be stored in the imaging lens can be suppressed. In addition, for an imaging lens designed and developed before manufacturing the camera, vignetting information of the camera cannot be stored in the imaging lens. In this regard, according to the present embodiment, the focus detection process is not performed when the correction error amount is large. This increases the accuracy of focus detection.

  In this embodiment, information on vignetting is stored in the camera, but it may be stored in the lens. In this case, the amount of information to be stored in the camera can be suppressed, and vignetting information of the imaging lens cannot be stored in the camera for an imaging lens designed and developed after the camera is manufactured. However, according to the present embodiment, the focus detection process of the present invention can be performed.

(Example 2)
In the first embodiment, the determination of the presence or absence of vignetting of the light beam used for focus detection by the imaging lens and the correction of the image signal are performed using the imaging lens type (lens type information) and focal length information passed from the lens to the camera. This was done by collating with a table stored in the table. In the present embodiment, a description will be given of a case where a focus detection area in which vignetting occurs and information on the amount of vignetting are calculated and corrected from information communicated from each imaging lens to the camera.

  The configuration of the focus detection apparatus of this embodiment is the same as that of the first embodiment, and is as shown in the block diagrams shown in FIGS. The difference from the first embodiment is the content of the focus detection device information stored in the camera and the content of information communicated between the imaging lens and the camera.

  In this embodiment, in FIG. 12, a representative value of the position of the focus detection area of the focus detection device from the optical axis L of the imaging lens is stored in the ROM in the central processing circuit PRS of the camera. Further, the position of the aperture opening 212 of the focus detection device projected by the optical system of the focus detection device and the information on the shape of the projection image are also stored in the ROM.

  In FIG. 13, the memory unit 400 stores exit pupil information corresponding to the focal length of the imaging lens. The exit pupil information is a member that restricts the effective F number of the light beam passing through the imaging lens 201, for example, on the imaging lens optical axis L such as the lens diaphragm 201d in FIG. 7, the front frame, the rear frame members 201e, and 201f. Information on position and diameter. The exit pupil information is input as lens information DLC to the central processing circuit PRS via the lens communication circuit LCOM according to the current focal length.

  The flowchart of the focus detection operation of the present embodiment is almost the same as that shown in FIG.

  The only change is the method of determining whether or not vignetting in step 4 occurs.

  In step 4 of the present embodiment, a known method using exit pupil information in the imaging lens, position information of the focus detection area, and information related to a projection image by the optical system of the focus detection device of the aperture opening 212 of the focus detection device. Is used to determine whether or not vignetting may occur. For example, this is a technique disclosed in Japanese Patent Publication No. 3-214133.

  In the case of YES at step 4, the light amount reduction information generated by vignetting is calculated by the central processing circuit PRS of the camera. Also in this case, exit pupil information in the imaging lens, position information of the focus detection area, and information related to a projection image by the optical system of the focus detection device of the aperture opening 212 of the focus detection device are used. Details of the method are as disclosed in Japanese Patent Publication No. 3-214133. A necessary correction amount is calculated from the calculated light amount reduction information, and correction is performed by the method described in the first embodiment.

  In this embodiment, the exit pupil information is stored in the imaging lens, and the position information of the focus detection area and the information related to the projection image by the optical system of the focus detection device of the aperture opening of the focus detection device are stored in the camera. Thus, the presence / absence of vignetting and the correction amount are determined. Therefore, vignetting correction can be performed for any combination of cameras and lenses as long as these pieces of information are available. Generally, since exit pupil information is stored in the imaging lens, vignetting correction can be performed by storing information on the focus detection device in the camera regardless of the manufacturing time of the lens. .

  Further, in this embodiment, since information related to the exit pupil is stored in the imaging lens, the amount of information to be stored in the camera can be suppressed.

(Example 3)
In Embodiments 1 and 2, the imaging lens interchangeable camera has been described. In this embodiment, an imaging lens fixed camera will be described.

  The configuration of the focus detection apparatus of the present embodiment is almost the same as that of the first and second embodiments and is as shown in the block diagram of FIG. The difference from the first and second embodiments is that a part of the information shown in FIG. 15 stored in the camera or the imaging lens is stored in the camera. Further, since the camera and the lens are integrated, in the first and second embodiments, the memory means, CPU, and the like provided in the imaging lens are omitted.

  The flowchart of the focus detection operation of the present embodiment is almost the same as that shown in FIG.

  The only change is the method of determining whether or not the combination of the imaging lens and the focus detection device in which the vignetting in step 4 occurs.

  In step 4 of the present embodiment, it is possible to determine whether or not there is a possibility of vignetting by storing a combination of focal length information and a focus detection area in which vignetting occurs in the camera. FIG. 18 shows a table stored in the camera.

  The table shown in FIG. 19 stored in the central processing circuit PRS is compared with the current focal length information and the focus detection area, and if there is a match, a correction amount calculation formula and an adjustment coefficient are obtained, 4 to select YES. If there is no corresponding item, NO is selected in Step 4. If YES in step 4, correction is performed by the method described in the first embodiment using the correction amount calculation formula and the adjustment coefficient. In this embodiment, since the correspondence between the camera and the imaging lens is one-to-one, there is relatively little information regarding vignetting to be stored. For this reason, there is no problem even if the correction formulas are stored one by one in accordance with the focal length and the focus detection region, and the correction can be performed with higher accuracy. .

Central sectional view of a single-lens reflex camera incorporating the focus detection device of the present invention The perspective view of the focus detection apparatus of this invention A plan view of the diaphragm 212 viewed from the field mask 210 side The top view which looked at the entrance plane side of the secondary image formation lens unit 213 The top view which looked at the light receiving element 215 from the visual field mask 210 side The back projection of each light receiving element row of the light receiving element 215 on the field mask 210 A top view in which the imaging lens 201 and the focus detection device 208 are developed straight. One form of projection of each member on imaging lens stop 201d The figure which shows one form of the light beam which passes an imaging lens One form of projection of each member on imaging lens stop 201d A diagram showing vignetting on a subject with uniform brightness 1 is a block diagram showing an example of a camera focus detection apparatus suitable for carrying out the present invention. Explanatory diagram of the lens control circuit LNSU The figure which shows the focus detection operation | movement of the focus detection apparatus of a camera An example of an information table used for image signal vignetting correction according to the first exemplary embodiment Parameter c1 explanatory diagram of the predicted value E of the focus detection deviation amount due to the correction error Parameter c2 explanatory diagram of the predicted value E of the focus detection deviation amount due to the correction error The adjustment coefficient T will be described An example of an information table used for image signal vignetting correction in FIG. Block diagram for explaining the configuration of the third embodiment

Explanation of symbols

DESCRIPTION OF SYMBOLS 201 Imaging lens 201 d Imaging lens aperture 208 Focus detection apparatus 210 Field mask 211 Field lens 212 Diaphragm 215 Light receiving element 216 Optical image

Claims (4)

A focus detection device that controls the imaging lens based on a phase difference of an optical image by a pair of light beams that have passed through different regions of an exit pupil of the imaging lens,
Correction means that may or may not correct vignetting according to lens type information and focal length information of the imaging lens;
When said correction vignetting is performed by the correction means determines before performing the correlation operation the appropriateness of the correction of the vignetting from the matching of the image of the image signal after the correction of the vignetting, when the degree of coincidence of the image is Control means for controlling the imaging lens based on a phase difference obtained by correlation calculation from the image signal, and for not controlling the photographing lens based on the image signal when the degree of coincidence of the images is low. Focus detection device. A focus detection device that controls the imaging lens based on a phase difference of an optical image by a pair of light beams that have passed through different regions of an exit pupil of the imaging lens,
Correction means that may or may not correct vignetting according to exit pupil information of the imaging lens and information of the focus detection device;
When said correction vignetting is performed by the correction means determines before performing the correlation operation the appropriateness of the correction of the vignetting from the matching of the image of the image signal after the correction of the vignetting, when the degree of coincidence of the image is A focus detection apparatus that controls the imaging lens based on a phase difference obtained by correlation calculation from the image signal, but does not control the photographing lens based on the image signal when the degree of coincidence of the images is low.   The focus detection apparatus according to claim 1, wherein the control unit determines whether or not to perform focus detection processing based on an image signal corrected by the correction unit. Wherein the control means, the focus detection apparatus according to claim 1 or 2, characterized in that to determine the field of view focused on the basis of the image signal corrected by said correction means.

Images