JP4558781B2 - Camera focus detection device - Google Patents

Description

  The present invention relates to a camera focus detection apparatus used for a digital camera or a video camera, for example.

  Conventionally, in an imaging apparatus such as a digital camera or a video camera, focus detection by a so-called hill-climbing method, in which a photographing lens is driven to a position where the high-frequency component of an output signal of an imaging element is maximized to focus on a subject, has been widely adopted. ing. However, since this method focuses on the subject while driving the photographic lens, there is a problem that it takes a relatively long time to reach the in-focus state. On the other hand, the focus detection by the triangulation method or the phase difference method can obtain the distance to the subject and the amount of focus shift, so that it is possible to detect the focus quickly, and particularly high speed performance is required. This method is suitable for still cameras.

Further, in Patent Document 1 , a pair of solid-state imaging devices that receive reflected light from the same subject incident from two lens systems that are separated from each other by a base line length are provided, and a pair that forms an image on the imaging surface of the pair of solid-state imaging devices. A technique related to a digital electronic still camera is disclosed in which an arbitrary position in an image of the subject is designated and the distance to the position designated point of the subject can be obtained by an algorithm using a triangulation method.
Japanese Patent Laid-Open No. 5-18748

In the technique described in Patent Document 1 , since the baseline direction of the pair of lens systems is arranged substantially parallel to the bottom surface of the camera, there is a possibility that the photographing lens touches the finger when the camera is held. In order to solve this problem, one lens system (used for both imaging and focus detection) is arranged at a position near the center of the front of the camera in the same way as a camera constituted by a conventional single lens, and the other lens It is assumed that the system (dedicated to focus detection) is placed at an upper position, for example, obliquely above the finger when touching the camera. In this case, the image sensor arranged at the imaging position of the lens system arranged obliquely above also has the long side direction aligned with the horizontal direction of the camera, like the image sensor arranged near the center of the front surface of the camera. In this case, the pixel arrangement direction and the baseline direction of the two lens systems do not coincide with each other, so that the signal processing for focus detection becomes complicated and the processing takes time .

SUMMARY OF THE INVENTION An object of the present invention is to provide a camera focus detection device that is simple and capable of high-speed processing .

In order to achieve the above object, according to the present invention, the first optical system that is used for both photographing and focus detection is disposed at the imaging position of the first optical system, and the long side direction or the short side direction is the horizontal direction of the camera. match the arranged first imaging element to a second optical system for focus detection, which is spaced above the first optical system and a predetermined base line length, the imaging position of said second optical system Arranged such that the long side direction or the short side direction has a predetermined angle with the long side direction and the short side direction of the first image sensor and coincides with the baseline direction of the second optical system and the first optical system. the group and placed second imaging device, a first image data representing a first image captured by the first image sensor, a second image data representing a second image captured by the second image sensor the a reference image that has been set in the partial region of the first image, consistent with the reference image of the second image That between the reference image is an image, the focus detection device for a camera, characterized in that it and a correlation amount calculating means for calculating a relative deviation amount of the base line Direction is provided.

According to the present invention, the second imaging device for focus detection, since group line Direction is arranged with a long side direction and the short side direction at a predetermined angle of the first image pickup device, when gripping the camera The possibility that the focus detection optical system is blocked by the hand is reduced. The second image pickup element for focus detection by Rukoto the long side direction are arranged so as to coincide with the base line direction of the first optical system and the second optical system, easy operation by the correlation amount calculating means become.

According to the present invention, it is possible to provide a camera focus detection device that is simple and capable of high-speed processing .

  Hereinafter, embodiments of the present invention and techniques related to the present invention will be described with reference to the drawings.

[First technology]
FIG. 1 shows and describes a configuration example of an imaging apparatus that employs a focus detection apparatus according to a first technique related to the present invention. As shown in FIG. 1, the imaging apparatus 1 includes a zoom lens 2 and a focus lens 3 as a photographing optical system. The light beam passes through the lenses 2 and 3 and the diaphragm 4 and forms an image on the light receiving surface of the image sensor 5 such as a CCD. A signal photoelectrically converted by the image pickup device 5 is input to the image pickup circuit 6, and a video signal is generated by the image pickup circuit 6. The video signal is converted into a digital video signal (image data) by the A / D converter 7 and temporarily stored in the memory 8. The image data stored in the memory 8 is read out at a predetermined screen rate (for example, 1/30 second) and converted into an analog video signal by the D / A converter 9. A liquid crystal display element (hereinafter abbreviated as LCD) 10 displays a subject image based on the analog video signal.

  When a recording operation is performed by operating a release switch included in an operation switch (hereinafter abbreviated as operation SW) 24, the image data in the memory 8 is compressed by the compression circuit of the compression / expansion circuit 11; Stored in the recording memory 12. When a reproduction operation is performed, the data compressed and stored in the recording memory 12 is decompressed by the decompression circuit of the compression / expansion circuit 11 and then temporarily stored in the memory 8. / A converter 9 converts it to an analog video signal. On the LCD 10, a reproduced image is displayed based on the analog video signal.

  Here, the LCD 10 displays an index 201 indicating a visual field capable of triangulation and an index 202 indicating a distance measurement area on the display screen (see FIG. 2). That is, the LCD 10 can display the index 201 indicating the field of view that can be measured by the triangulation that changes according to the zoom value of the zoom lens 2 of the photographing optical system, so as to overlap the image captured by the CCD 5. By 201, it is possible to easily confirm an area where triangulation is possible. If the distance measurement area is instructed by the operation of the operation SW 24 or the like outside the index 201 indicating the field of view capable of triangulation (area where distance measurement is impossible), AF is performed based on the so-called hill-climbing method. Made.

  Image data A / D converted by the A / D converter 7 is input to an auto exposure processing circuit (hereinafter abbreviated as AE processing circuit) 13 and an autofocus processing circuit (hereinafter abbreviated as AF processing circuit) 14. Is done. The AE processing circuit 13 calculates an AE evaluation value corresponding to the brightness of the subject by calculating a luminance value of image data for one frame (one screen), and outputs it to the CPU 15. Further, the AF processing circuit 14 extracts a high frequency component from a luminance component of image data for one frame (one screen) with a high-pass filter or the like, calculates a cumulative addition value, etc., and converts it into a high frequency side contour component or the like. A corresponding AF evaluation value is calculated and output to the CPU 15.

  On the other hand, a light beam that has passed through a focus lens 102 that is an optical system dedicated to autofocus (hereinafter referred to as an AF optical system) that is disposed apart from the optical axis of the imaging optical system by a base line length L passes through a diaphragm 103, A subject image is connected to an image sensor 104 such as a CCD. A signal photoelectrically converted by the imaging element 104 is input to the imaging circuit 105, and a video signal is generated by the imaging circuit 105. The video signal is converted into a digital video signal (image data) by the A / D converter 106 and temporarily stored in the memory 107.

  The AF processing circuit 108 uses a so-called triangle using image data obtained based on the photographing optical system and stored in the memory 8 and image data obtained based on the AF optical system and stored in the memory 107. Based on the principle of distance measurement, the distance to the subject selected by the distance measurement area selection circuit 113 is obtained. The ranging area selection circuit 113 may be manually selected by a camera operator or selected by a known visual line detection device. After driving the focus lens 102 of the photographic optical system to the in-focus position based on the object distance information obtained based on the principle of triangulation, the focus is adjusted as necessary or every time in order to increase the accuracy of focusing on the object. Every time the lens 102 is driven to focus, the final focus position adjustment is performed by a so-called hill-climbing method. For this purpose, the AF processing circuit 108 extracts a high-frequency component in the luminance component of image data for one frame (one screen) with a high-pass filter or the like, calculates a cumulative added value, and the like, thereby calculating a high-frequency side contour component. An AF evaluation value corresponding to the amount or the like is calculated and output to the CPU 100.

  A predetermined timing signal synchronized with the screen rate is input to the CPU 100 from a timing generator (hereinafter abbreviated as TG circuit) 109. The CPU 100 performs various control operations in synchronization with this timing signal. The timing signal of the TG circuit 109 is also input to the imaging circuit 105. Further, the TG circuit 109 controls the CCD driver 110 so as to drive the CCD 104 at a predetermined timing.

  A predetermined timing signal synchronized with the screen rate is input from the TG circuit 16 to the CPU 15. The CPU 15 performs various control operations in synchronization with this timing signal. The timing signal of the TG circuit 16 is also input to the imaging circuit 6. The imaging circuit 6 performs various processes such as color signal separation in synchronization with the timing signal. The TG circuit 16 controls the CCD driver 17 so as to drive the CCD 5 at a predetermined timing.

  The CPU 15 controls the first to third motor drive circuits 18 to 20 to drive the diaphragm 4, the focus lens 3, and the zoom lens 2 via the first to third motors 21 to 23. Control. That is, the CPU 15 controls the first motor drive circuit 18 on the basis of the AE evaluation value and rotationally drives the first motor 21 to adjust the aperture amount of the aperture 4 to an appropriate value, that is, auto Perform exposure control.

  The CPU 15 is connected to the CPU 100, receives distance data based on the principle of triangulation from the CPU 100, and controls the second motor drive circuit 19 to control the second motor 22 based on the distance data. The focus lens 3 is driven by rotation to set the focus state. However, in general, the accuracy is improved when the control is performed based on the hill-climbing method in which the focus lens 3 is driven to the position where the contrast of the image sensor 5 is maximized. Therefore, after driving the focus lens based on the principle of triangulation, the second motor drive circuit 19 is controlled to rotate and drive the second motor 22 based on the AF evaluation value as necessary. The AF evaluation value from the AF processing circuit 14 is obtained. Then, based on the AF evaluation value obtained in this way, the CPU 15 drives the focus lens 3 to the lens position where the AF evaluation value is maximized to set the focus state, that is, performs autofocus.

  However, since it takes a relatively long time to control the hill-climbing method, it goes without saying that the driving of the focus lens 3 based on the hill-climbing method may be omitted.

  The CPU 15 is connected with, for example, an electrically rewritable and nonvolatile read-only memory EEPROM 25, and the EEPROM 25 performs various controls and various operations via the CPU 15. Data used for image processing, information for knowing the relationship between the imaging positions of the image sensor 5 and the image sensor 104 described later, information for correcting the brightness of the imaging surface of the image sensor 5 and the image sensor 104, etc. Is stored. These pieces of information are read and used when the power of the imaging device 1 is turned on. Note that when the CPU 15 detects the voltage of the battery 26 and detects that the voltage is lower than a predetermined voltage value, the CPU 15 displays a display indicating that the remaining amount of the battery 26 is low and a display prompting charging or replacement of the battery. To do.

  Note that the image pickup surface size of the image pickup element 104 may be smaller than the image pickup surface size of the image pickup element 5, and the number of pixels of the image pickup element 104 may be smaller than the number of pixels of the image pickup element 5. According to this, since an inexpensive image sensor can be used for AF, the cost is reduced.

  Further, since the photographing optical system is also used for triangulation, parallax does not occur, and a subject in an arbitrary region can be accurately focused.

  Here, the principle of distance measurement related to the focus detection apparatus according to the first technology will be described with reference to FIG. This is common to other technologies and embodiments related to the present invention described later.

In FIG. 3, the photographing optical system M1 and the AF optical system M2 are disposed apart from each other by the base line length L. The focal length of the imaging optical system M1 with respect to the imaging device 5 is f1, and the focal length of the AF optical system M2 with respect to the imaging device 104 is f2. Further, the deviation of the image of the subject O from the optical axis of the photographing optical system M1 is ΔL1, the deviation from the optical axis of the AF optical system is ΔL2, and the distance to the subject O is R. In such a case, the following equation (1) is established.

R / L1 = f1 / ΔL1
R / L2 = f2 / ΔL2
L = L1 + L2
L / R = (1 / f1) · (ΔL1 + (f1 / f2) · ΔL2) (1)
That is, from this equation (1), the deviation ΔL1 of the image O from the optical axis of the photographing optical system M1 and the deviation ΔL2 from the optical axis of the AF optical system M2, the focal lengths f1, f2 of the two optical systems M1, M2, If the base line length L is determined, the distance R to the subject can be found.

  Hereinafter, the operation of the imaging apparatus employing the focus detection apparatus according to the first technique will be described in detail with reference to the flowchart of FIG.

  When the power switch included in the operation SW 24 is turned on and the power is turned on, the CPU 15 and the CPU 100 are reset to power on, and this operation is started.

  First, the AE processing circuit 13 calculates an AE evaluation value corresponding to the brightness of the subject by calculating a luminance value of image data for one frame (one screen), and outputs the AE evaluation value to the CPU 15. (Step S1). Subsequently, the CPU 15 determines whether or not the first release (hereinafter referred to as 1st release) of the release switch included in the operation SW 24 has been performed, and enters a standby state waiting for the release (step 1). S2).

  If the first release is performed in step S2, the CPU 100 reads the position of the distance measurement area selected by the distance measurement area selection circuit 113 (step S3).

  Next, the CPU 100 calls a subroutine “triangulation ranging AF” and obtains the distance to the subject based on the principle of triangulation (step S4). The operation of triangulation by this subroutine “triangulation AF” will be described later with reference to FIG.

  When the distance to the subject is obtained in step S4, the CPU 15 drives the focus lens 3 to the in-focus position based on the distance to the subject (step S5).

  Subsequently, the CPU 15 determines whether or not the first release has been released, and enters a standby state for waiting for the release (step S6). If the first release has been released, the CPU 15 determines whether or not the second release (hereinafter referred to as the second release) of the release switch included in the operation SW 24 has been performed (step S7). If not, branch to J1. On the other hand, if the second release has been performed in step S7, the CPU 15 next performs auto exposure control based on the AE evaluation value (step S8), thus ending a series of photographing operations.

  Next, the subroutine “triangulation AF” executed in step S4 of FIG. 4 will be described in detail with reference to the flowchart of FIG.

  If the position of the focus lens 3 of the photographic optical system is greatly deviated from the in-focus position, the larger the aperture opening, the more blurred the image, making it difficult to accurately determine the amount of deviation.

  For this purpose, the CPU 15 narrows down the diaphragm 4 of the photographing optical system to a predetermined value prior to the calculation of the shift amount (step S11). Even if the photographic lens is greatly deviated from the in-focus position, a relatively clear image is formed, which is effective in accurately obtaining the amount of deviation.

  Further, it is desirable that the aperture value of the photographing optical system is equal to that of the aperture 103 of the AF optical system.

  This is because the blur and illuminance distribution of the image formed by the photographing optical system and the image formed by the AF optical system are made equal to obtain the amount of deviation accurately. That is, if the aperture value of the photographic optical system is set in this way, the blur and illuminance distribution of the image formed by the photographic optical system and the image formed by the AF optical system are approximated. It is effective in seeking.

  Next, the CPUs 15 and 100 perform exposure of the image sensor 5 and the image sensor 104 based on the AE evaluation value obtained by the AE processing circuit 13, and acquire image data (step S12).

  Then, the CPU 100 forms an image on the AF optical system by ΔL1, which is a deviation from the optical axis center of the imaging optical system of the image of the distance measuring area imaged by the imaging optical system, by the subroutine “ΔL1, ΔL2 calculation”. The AF processing circuit 108 is controlled so as to calculate ΔL2 which is a deviation from the center of the optical axis of the AF optical system of the image corresponding to the image in the distance measuring area (step S13). This subroutine “ΔL1, ΔL2 calculation” will be described in detail later with reference to FIG. Thus, the CPU 100 calculates the distance R to the subject based on the above equation (1) (see FIG. 3) (step S14), and returns from this subroutine “triangulation ranging AF”.

  Here, with reference to FIGS. 6 and 7, the principle of the correlation calculation for obtaining the image shift amount will be described.

First, FIG. 6A shows an array of pixels of the image sensor 5 in the first distance measurement area selected by the distance measurement area selection circuit 113. The size of one pixel is H1 × W1. An image in a predetermined range P1 in the first distance measurement area is set as a reference image. The distance from the position of the center of gravity of the reference image to the optical axis of the photographing optical system is denoted by ΔL1. Then, the output of the i-th pixel to S (i) based on line Direction from the optical axis of the imaging optical system of the reference image.

On the other hand, FIG. 6B shows an arrangement of pixels in the second distance measuring area of an image formed by the AF optical system corresponding to the first distance measuring area. The size of one pixel is H2 × W2. Set a reference image P2 having the same number of pixels as the reference image in the second ranging area, the j-th output from the center of the AF optical system of the reference image based on line Direction and C (j) To do.

And the sum of the absolute value of the difference of the output of a corresponding pixel is taken, this is set to A (m), and it is set as a correlation amount. In this case, the following expression (2) is established.

  Here, i is the number of the first pixel in the reference portion, and j is the number of the first pixel in the reference portion. Further, k means the kth pixel counted from the first pixel of the standard part and the reference part.

  The reference portion expects a margin of about ± m on the left and right in j + n.

  Here, m attached to the data of the reference part is attached to the CPU 15 for the operation of fixing the data number of the reference part and sending the data number of the reference part one by one.

  Then, the correlation amount A (m) is obtained while sequentially changing m from 1 to p, and when the correlation amount A (m) takes the minimum value, the images of the reference portion and the reference portion coincide with each other. The relative distance of the image of the reference part is obtained. Since the correlation amount A (m) is a value for each minimum pixel, the accuracy of the relative distance between the image of the reference portion and the reference portion is limited to one pixel.

  Therefore, as shown in FIG. 7, a relative distance with high accuracy of one pixel pitch or less is obtained by interpolation from the geometric relationship of the correlation amount before and after the correlation amount A (m) takes the minimum value. According to this interpolation, it is possible to empirically determine the relative distance between the image of the reference portion and the reference portion with a high accuracy of 1/10 pixel or less. This interpolation calculation itself is known, but will be briefly described below. In FIG. 7, the third smallest value A (1) and the minimum value A (2) of the correlation amount A (m) are represented by a straight line. When the angle θ between the straight line and the horizontal line is made equal to the angle θ between the straight line passing through the second smallest value A (3) of the correlation amount A (m) and the horizontal line, the above two The amount of deviation m of the points where the straight lines intersect is obtained.

  By the way, in the first technique, it is necessary to satisfy the following “first to third conditions” in order to perform correlation calculation between the standard image and the reference image and obtain a highly accurate deviation amount.

  This is the same concept for other technologies and embodiments related to the present invention.

More specifically, the "first condition", and line the center line baseline direction of the center line of the base line Direction reference image and the projected line on the object of the reference image is projected onto the object matches It is to be. The “second condition” is that the area and shape on the subject surface projected onto the photoelectric conversion unit of the unit pixel are the same. Furthermore, the “third condition” is that the levels of image data corresponding to the same subject are the same. Hereinafter, specific means for satisfying these first to third conditions will be described in more detail.

(First condition)
Specific means for satisfying the first condition will be described.

  Spot light (light source) is arranged at an infinite position on the optical axis of the photographing optical system. The focal length and aperture value of the photographic optical system are set to f1 and F1, respectively, the pixel data of the image sensor 5 at this time is read, and the coordinate position P1 (the image position of the spot light on the imaging surface of the image sensor 5 ( Px1, Py1) are detected. Similarly, the pixel data of the image sensor 104 is read out to detect the coordinate position P2 (Px2, Py2) at which the spot light forms an image on the imaging surface of the image sensor 104, and the coordinate positions of P1 and P2 are set as a set of data. Is recorded in the EEPROM 25.

  In the above, a set of data corresponding to spot light at an infinite position on the optical axis of the photographing optical system was recorded. However, spot light is arranged at a plurality of infinite positions around the optical axis of the photographing optical system. Then, the coordinate positions on the image pickup surfaces of the image pickup device 5 and the image pickup device 104 on which the plurality of spot lights are imaged may be recorded in the EEPROM 25. In this case, since the imaging position does not change even if the imaging magnification of the imaging optical system and the AF optical system changes, the subject that forms an image on the imaging surface of the image sensor 5 is on the imaging surface of the image sensor 104. The position where the image is formed can be easily obtained, and the correlation calculation between the reference image and the reference image can be easily performed.

  Further, the spot light (light source) is set at a finite distance outside the optical axis of the photographing optical system, and the coordinate positions on the image pickup surfaces of the image pickup device 5 and the image pickup device 104 on which the spot light is imaged are stored in the EEPROM 25. Of course, it may be recorded.

  The position of the imaging surface of the imaging device 5 on which the subject located at infinity on the optical axis from the coordinate positions P1 (Px1, Py2), P2 (Px2, Py2) recorded in the EEPROM 25 in this way, and The position of the imaging surface of the image sensor 104 is known.

  Therefore, the coordinate positions on the imaging surfaces of the imaging element 5 and the imaging element 104 on which an object outside the optical axis of the imaging optical system forms an image can be easily obtained by calculation.

  For example, the focal length of the imaging optical system is f1, the focal length of the AF optical system is f2, and the coordinate (x1) with the coordinate position P1 (Px1, Py1) recorded on the EEPROM 25 on the imaging surface of the imaging device 5 as the origin is used. , Y1), an image is formed.

  In this case, the position (x2, y2) on the image sensor 104 where an image corresponding to the image at the position (x1, y1) is formed is based on the coordinate position P2 (Px2, Py2) recorded in the EEPROM 25 as the origin. (X2, y1 · f2 / f1). The value x2 corresponding to x1 is obtained by correlation calculation. If x2 = x1 · f2 / f1, the subject distance is infinite. The closer the subject is, the larger the difference between x2 and x1 · f2 / f1 is. Become.

  The correlation calculation is to obtain the difference between the two.

  Here, FIG. 8 shows a configuration example of a system for writing the data into the EEPROM 25 when the imaging apparatus 1 is manufactured. In FIG. 8, when a predetermined signal is sent from the writing device 801 to the CPU 15 of the imaging device 1, the CPU 15 executes a command for writing data relating to the coordinate position on the imaging surface to the EEPROM 25.

  Hereinafter, with reference to the flowchart of FIG. 9, the operation of writing data relating to the coordinate position to the EEPROM 25 of the imaging apparatus 1 by this system will be described in more detail.

  First, the imaging device 1 is set to the writing mode of the EEPROM 25 by the writing device 801 (step S21), the focal length of the photographing optical system is set to a predetermined focal length (step S22), and the aperture of the photographing optical system is determined in advance. The aperture value is narrowed down (step S23).

  Then, image data of the image sensor 5 and the image sensor 104 are read (step S24), and based on the read image data, the coordinate positions P1 and P2 on the image pickup surface are written in the EEPROM 25 (step S25), and this operation is performed. finish.

(Second condition)
A specific means for solving the second condition will be described.

  Since the photographing magnification of the photographing optical system and the AF optical system and the size of one pixel of the image pickup element 5 and the image pickup element 104 are different, it is necessary to consider these in order to satisfy the second condition.

  Therefore, the ratio α1 / α2 of the image magnification between the photographing optical system and the AF optical system is calculated.

  Note that α1 / α2 = f1 / f2. However, f1 and f2 are the focal length of the photographing optical system and the focal length of the AF optical system, respectively, as described above with reference to FIG.

Next, FIG. 6 (a), the as shown in FIG. 10 (a), a plurality of pixels corresponding to the group line Direction of the first ranging region of the imaging element 5 for receiving the image formed by the photographing optical system, based on line rectangular aliquoted into each length W1 countercurrent. The length in the direction perpendicular to the base line length is H1, and W1 × H1 is a new pixel. Then, a signal obtained by adding the plurality of original pixels included in W1 × H1 or an average value thereof is obtained and used as an output of a new pixel (hereinafter referred to as “virtual pixel”) of the imaging device 5. The virtual pixels may be added by thinning out a specific line included in the W1 × H1.

Similarly, FIG. 6 (b), the as shown in FIG. 10 (b), a plurality of pixels corresponding to the second ranging area groups line Direction of the imaging device 104 which receives the image formed by the AF optical system, aliquoted into each length W2 based on line direction. The length in the direction perpendicular to the base line length is set as H2, and W2 × H2 is set as one new pixel. Then, a signal obtained by adding the signals of the plurality of original pixels included in W2 × H2 or an average value thereof is set as an output of the virtual pixel of the image sensor 104.

Here, W1, W2, H1, and H2 are
α1 / α2 = W1 / W2 = H1 / H2
Set to satisfy the relationship.
As described above, if the ratio between the size of the unit pixel of the image sensor 5 and the size of the unit pixel of the image sensor 104 is made equal to the ratio of the magnification of the photographing optical system and the magnification of the AF optical system, the optical for imaging is used. The system (photographing optical system) and the image sensor 5 can be used together to quickly focus on a wide area of the subject with a simple configuration. The size of the unit pixel means that a plurality of pixels are collectively regarded as one virtual pixel (that is, the virtual pixel).

  Next, the virtual pixel size is specifically designed based on the basic concept described above.

From the above equation (1), the relationship between the detection error dΔL2 of the AF optical system and the distance measurement error dR is
dR = −dΔL ² · R ² / (L · f ² ) (3)
It becomes.

On the other hand, the depth of field of the taking lens is
Depth of field = −F · δ · R ² / f1 ² (4)
F: Aperture value of the photographing optical system, δ: Allowable circle of confusion
It becomes.

Here, if the distance measurement error is within the depth of field of the photographic lens, it can be said that the camera is in focus. For this,
| Ranging error | <| Depth of field |
From the above formulas (3) and (4)
dΔL ² · R ² / (L · f ² ) <F · δ · R ² / f 1 ²
Is established.

Therefore,
dΔL2 <F · δ · L · f2 / f1 ² (5) is established.

Now, as design values, F = 1, δ = 1/30 mm, L = 40 mm (the distance measurement error at the center of the optical axis of the photographing optical system is obtained, L2 = L), f2 = 10 mm, f1 = 30 mm Then,
dΔL2 <15 (μm)
It becomes.

In this example, the AF optical system is assumed to have been manufactured as designed, as long as it is a group line side pitch pixel pitch less 15μm in direction of the imaging device 104, necessary by correlation operation does not depend on the aforementioned interpolation Accuracy will be obtained.

However, assuming that the final shift amount is calculated by the above-described interpolation method, and the accuracy of 1/10 of one pixel pitch can be secured by this, the required accuracy can be obtained if the pixel pitch is 150 μm or less. It will be. When the group line direction pixel pitch of the direction of the image sensor 104 and 10 [mu] m, so that sufficient accuracy can be obtained.

Therefore, on the condition that the above interpolation calculation is performed, the above equation (5) is corrected,
dΔL2 <10 · F · δ · L · f2 / f1 ² (6)
And

  When the photographic optical system is a zoom lens and the zoom magnification becomes large, as is clear from the above equation (6), the AF accuracy and the range that can be measured are limited.

  FIG. 11 shows such a relationship. That is, in FIG. 11B, it is assumed that the field of view 1101b of the photographing optical system and the field of view 1102b of the AF optical system coincide with each other, and the AF accuracy is ensured by the above-described specific numerical values. However, when the focal length of the photographing optical system is shortened and the magnification of the image is increased (wide angle side), the field of view 1102a of the AF optical system becomes narrower than the field of view 1101a of the photographing optical system (see FIG. 11A). In this case, as is apparent from the above equation (5), the tolerance value of the deviation amount of the AF optical system is larger than that in FIG. 11B, and sufficient accuracy can be obtained. On the other hand, when the focal length of the photographic optical system is increased and the magnification of the image is reduced (on the telephoto side), the visual field 1102c of the AF optical system becomes wider than the visual field 1101c of the photographic optical system (see FIG. 11C). In this case, the allowable value of the deviation error of the AF optical system is smaller than that in FIG. 11B, and sufficient accuracy may not be obtained.

  Here, the effective size of the imaging surface of the imaging element 5 (for imaging and AF) is 18.5 mm (H) × 13.5 mm (v), the horizontal pixel pitch is 10 μm, and the imaging element 104 (for AF only) is imaged. The effective size of the surface is 7.4 mm (H) × 5.94 mm (V), the horizontal pixel pitch is 10 μm, the focal length of the AF optical system is f2 = 10 mm, and the focal length f1 of the photographing optical system is from 10 mm to 100 mm. Ratio of the horizontal field of view of the AF optical system to the horizontal field of view of the photographic optical system when changed in steps of 10 mm (f1 · H2 / (f2 · H1)) (where H1 is the image pickup device 5 (for image pickup and AF)) The horizontal effective size (18.5 mm) and H2 represents the horizontal effective size (7.4 mm) of the image sensor 104 (for AF)). . Note that the size of the image sensor 104 is smaller than that of the image sensor 5. In general, the image sensor 104 has a large number of pixels and is large and expensive, but the image sensor 104 is cheaper than that. This is because a relatively small number of pixels is used. Further, Table 1 shows the result of obtaining the allowable distance measurement error of the AF optical system in units of the number of pixels of the image sensor 104 using the above equation (6).

Then, to be equal to or less than the allowable range finding errors, next to the virtual pixel of the image sensor 104, that is, a group line direction size of direction W2 selected as appropriate Table 1 the number of pixels of the image pickup element 104 as a unit, imaging As described above, the horizontal size W1 of the virtual pixel of the element 5 is automatically determined from the relationship of f1 / f2 = W1 / W2, and is as shown in Table 1 with the number of pixels as a unit.

In Table 1, when the focal length of the photographic optical system is 10 mm, W2 is 10 pixels compared to the allowable distance measurement error of 135 pixels, which is slightly more accurate than the other photographic optical systems. This is for unifying W2 as much as possible to simplify the algorithm. Theoretically, W2 may be 135 pixels or less.

  As can be seen from Table 1, the range that can be measured varies depending on the focal length of the photographing optical system. Therefore, in the present invention, as shown in FIG. 11, a visual field capable of distance measurement based on triangulation is displayed according to the focal length of the photographing optical system. It also displays the ranging area.

  On the other hand, the longer the focal length of the photographic optical system, the smaller the allowable distance measurement error, and there is a possibility that accuracy cannot be ensured on the telephoto side. Therefore, in the second technique related to the present invention described later, the zoom value of the AF optical system is changed in accordance with the zoom value of the photographing optical system (details will be described later).

  The size H1 and H2 in the direction perpendicular to the base line length of the virtual pixel may be appropriately determined so as to satisfy the relationship of f1 / f2 = H1 / H2.

(Third condition)
Specific means for satisfying the third condition will be described.

  The image data value of the image sensor 5 and the image data value of the image sensor 104 are different because the configurations of the photographing optical system and the AF optical system and the sensitivity of the image sensor 5 and the image sensor 104 are different.

  Therefore, as described below, correction data is recorded in the EEPROM 25 at the time of manufacturing the camera, and the image data of the image sensor 5 and the image data of the image sensor 104 are normalized before the correlation calculation is performed. The output level of the corresponding image data is made substantially the same.

  That is, the signal level or the ratio of the same image area of the first image formed by the photographing optical system and the second image formed by the AF optical system is stored in the EEPROM 25, and this information is stored. Based on this, the output level is corrected. According to this, the level of the image signal of the reference image corresponding to the reference image can be matched. Details will be described below.

  Now, it is assumed that an image Im1 of the subject area A1 is formed on the imaging surface of the imaging device 5 in a state where the aperture of the imaging optical system is reduced to a predetermined value. At the same time, it is assumed that an image Im2 of the subject area A1 is formed on the imaging surface of the imaging element 104.

The ratio of the average value Vm1 of the pixel data of the image sensor 5 corresponding to the Im1 at this time and the average value Vm2 of the pixel data of the image sensor 104,
K = Vm1 / Vm2
Is recorded in the EEPROM 25.

  Note that the subject may be a wide diffusion plate, and the ratio of the average value of the output of the plurality of pixels of the image sensor 5 and the average value of the output of the plurality of pixels of the image sensor 104 at this time may be K. The Vm1 and Vm2 may be recorded in the EEPROM 25, and the ratio between the two may be calculated later. Alternatively, the aperture of the imaging optical system may be set to a predetermined aperture value corresponding to the focal length of the imaging optical system, and the coefficient K may be obtained for each focal length and stored in the EEPROM 25.

  By multiplying the image data of the image sensor 104 by K before the correlation calculation, the brightness of the standard image and the reference image can be made substantially the same.

  Here, FIG. 12 shows a configuration example of a system for writing the data in the EEPROM 25 when the imaging apparatus 1 is manufactured. Here, the same components as those in FIG. When a predetermined signal is sent from the writing device 1201 to the CPU 15 of the imaging device 1, the CPU 15 executes a command for writing data to the EEPROM 25.

  Hereinafter, the flow of the data writing operation to the EEPROM 25 will be described in more detail with reference to the flowchart of FIG. The imaging device 1 is set to the writing mode of the EEPROM 25 by the writing device 1201 (step S31). Next, the focal length of the photographic optical system is set to a predetermined focal length (step S32), and the aperture of the photographic optical system is narrowed down to a predetermined aperture value (step S33). Then, the data of the image sensor 5 and the image sensor 104 are read (step S34), the coefficient K is written in the EEPROM 25 based on the read image data (step S35), and this operation is finished.

  The above is the description of specific means for satisfying the first to third conditions.

  Next, the subroutine “ΔL1, ΔL2 calculation” executed in step S13 of FIG. 5 will be described in detail with reference to the flowchart of FIG. In this subroutine, deviation amounts ΔL1 and ΔL2 from the optical axis center of the image in the distance measurement area are calculated.

  First, the CPU 100 reads the ranging area selected by the ranging area selection circuit 113 (step S41). Next, the CPU 100 appropriately selects the reference image from the distance measurement area in order to calculate the barycentric position (ΔL1) of the reference image. Of course, the position of this reference image may be equal to the distance measurement area, or a small area having a predetermined contrast suitable for correlation calculation may be selected from the distance measurement areas. Then, the CPU 100 obtains the barycentric position of the reference image, sets the shift amount from the optical axis of the photographing optical system to ΔL1 (step S42), and obtains the barycentric position of the reference image corresponding to the barycentric position of the reference image (step S42). S43).

  Next, the CPU 100 stores the focal length f1 of the photographing optical system obtained from Table 1 and the virtual pixel sizes (W1, W2, H1, H2) of the image pickup device 5 and the image pickup device 104 in the EEPROM 25 as a table. Referring to this table, the output of each virtual pixel of the reference image and the reference image is calculated. Note that the output value of the virtual pixel is the average value of the outputs of the original plurality of pixels (step S44). Next, the level of the virtual pixel output corresponding to each of the standard image and the reference image is normalized based on the above-described coefficient K stored in the EEPROM 25 (step S45).

  Then, under the control of the CPU 100, the AF processing circuit 108 performs the correlation calculation between the reference image and the reference image based on the above-described correlation calculation expression, that is, the above expression (2). Since this correlation calculation is to obtain the deviation ΔL2 of the reference image from the optical axis of the AF optical system, the correlation calculation is performed assuming that the center of gravity position of the reference image is on the optical axis of the AF optical system, and the above-mentioned deviation calculation is performed. The distance to the subject is calculated based on the amounts ΔL1 and ΔL2 (step S46), and the process returns to the process of FIG.

  As described above, according to the first technique related to the present invention, the imaging optical system (imaging optical system) and the imaging element 5 are used together, and the focus is quickly focused on a wide area of the subject with a simple configuration. Can be combined. In addition, even if the magnification of the photographing optical system and that of the AF optical system are different, it is possible to accurately perform distance measurement on a subject in an arbitrary region.

[Second technology]
FIG. 15 shows and describes the configuration of an imaging apparatus that employs a focus detection apparatus according to a second technique related to the present invention. Here, for simplification of description, the same components as those in FIG. 1 described above are denoted by the same reference numerals, overlapping description is omitted, and differences between the two will be mainly described.

  If the zoom value of the zoom lens 2 of the photographing optical system and the zoom value of the zoom lens 101 of the AF optical system are greatly different from each other, the field of view of the image by the photographing optical system and the field of view of the image by the AF optical system are greatly different. The subject distance calculation error increases, and the subject distance may not be obtained for a wide field of view of the subject.

In view of such a point, in the imaging apparatus employing the focus detection apparatus according to the second technique , the zoom value of the zoom lens 2 of the photographing optical system is read, and the zoom of the AF optical system is read according to the zoom value. The CPU 200 controls the fifth motor drive 111 to drive the motor 112 so that the zoom value of the lens 101 is set to a predetermined value. Of course, the zoom value of the zoom lens 101 may be set to a predetermined value step by step.

  In this way, the zoom value of the zoom lens 101 is set to a predetermined value step by step so that the zoom value of the zoom lens 101 of the AF optical system is continuously changed in the same manner as the zoom lens 2 of the photographing optical system. This is because the apparatus becomes large and the cost increases.

  As described above, by setting the zoom value of the zoom lens 101 to a predetermined value in a stepwise manner, it is not necessary to ensure the accuracy of the intermediate zoom value, so a simple configuration is possible, and a stepwise zoom value is also possible. It is possible to improve the accuracy of the.

  Here, the basic principle and main operation of the imaging apparatus employing the focus detection apparatus according to the second technique are substantially the same as the contents described in the first technique related to the present invention. The subroutine “triangulation ranging AF” executed in step S4 of FIG.

  Accordingly, the flow of processing of the subroutine “triangulation AF” according to the second technique will be described in detail below with reference to the flowchart of FIG.

  First, the CPU 200 reads the zoom value of the zoom lens 2 of the photographing optical system via the CPU 15 (step S51). Then, the motor 112 is driven based on the fifth motor drive 111 so as to set the zoom value of the zoom lens 101 of the AF optical system to a predetermined value according to the zoom value (step S52). The reason why the zoom value of the zoom lens 101 of the AF optical system is set to a predetermined value in accordance with the zoom value of the zoom lens 2 is as follows. That is, as described above, if the zoom value of the zoom lens 2 of the photographing optical system and the zoom value of the zoom lens 101 of the AF optical system are greatly different from each other, the field of view of the image by the photographing optical system and the image of the image by the AF optical system are different. This is because the field of view differs greatly and the subject distance calculation error increases, making it impossible to determine the subject distance for a wide field of view of the subject.

  Here, it goes without saying that the zoom value of the zoom lens 101 may be set to a predetermined value step by step. The zoom value of the zoom lens 101 is set to a predetermined value step by step for the following reason, that is, the zoom value of the zoom lens 101 of the AF optical system is set in the same manner as the zoom lens 2 of the photographing optical system. This is because if the change is made continuously, the apparatus becomes large and the cost increases. If the zoom value of the zoom lens 101 is set to a predetermined value step by step in this way, it is not necessary to ensure the accuracy of the intermediate zoom value, so a simple configuration is sufficient, and the accuracy of the stepwise zoom value is improved. It becomes possible to raise.

  Subsequently, the CPU 15 narrows down the diaphragm 4 of the photographing optical system to a predetermined value (step S53), and then acquires image data by the AE processing circuit 13 (step S54). The details of steps S53 and S54 are the same as steps S11 and S12 of FIG. 5, and thus further description thereof is omitted. Next, the CPU 200 controls the AF processing circuit 108, and ΔL1 which is a deviation from the optical axis center of the image of the first distance measuring area imaged by the photographing optical system, and the above-mentioned first image imaged by the AF optical system. ΔL2 that is a deviation from the center of the optical axis of the image corresponding to the image of the two distance measurement areas is calculated (step S55). The subroutine “ΔL1, ΔL2 calculation” is as described above with reference to FIG.

  Thus, the CPU 200 calculates the distance R to the subject based on the above expression (1) (see FIG. 3) by the AF processing circuit 108 (step S56), and returns.

  As described above, according to the second technique, it is possible to quickly focus on a wide area of a subject with a simple configuration using both the imaging optical system (imaging optical system) and the imaging element 5. It becomes.

[Third technology]
This is positioned as an improvement example of the first or second technique related to the present invention.

  Since the configuration of the imaging apparatus that employs the focus detection apparatus according to the third technique related to the present invention is the same as that of FIG. 1 or 15, here, the differences will be mainly described with reference to the drawings.

  The above-described triangulation may cause a mechanical error of the AF optical system or an error accompanying a correlation calculation. On the other hand, the so-called hill-climbing method, in which the lens is driven to a position where the high-frequency component of the image signal imaged on the image sensor becomes the maximum while driving the photographing lens, is ideal because the image signal itself is used for focusing. However, if the photographing lens is far away from the in-focus position, sufficient contrast cannot be obtained, and there is a problem that it takes time until in-focus.

  In view of this point, in the third technique, first, the photographic lens is driven based on the principle of triangulation, and finally, the photographic lens is driven to the in-focus position based on the hill-climbing method.

  Hereinafter, with reference to the flowchart of FIG. 17, the main operation by the imaging apparatus employing the focus detection apparatus according to the third technique will be described in detail.

  Here, the description will be made with reference to the configuration of FIG. 1 as appropriate.

  Steps S61 and S62 are the same as steps S1 and S2 in FIG. Next, the CPU 100 reads the distance measurement area selected by the distance measurement area selection circuit 113 (step S63), and determines whether or not the selected visual field area can be measured by the triangulation (see FIG. 3). (Step S64), and if it is an area where triangulation is possible, triangulation is performed (step S65). The subroutine “triangulation AF” is the same as that described with reference to FIG. 5 or FIG.

  Next, the CPU 15 compares the subject distance obtained by the triangulation with the focus position of the focus lens 3 of the current photographing optical system (step S66).

  The focus position of the focus lens of the current photographing optical system can be known by a known method of detecting the drive position of the focus lens 3 by providing an encoder on the drive member of the focus lens 3. Description of this is omitted.

  When the CPU 15 determines that the focus lens 3 is in the vicinity of the focal point (step S67), the CPU 15 immediately drives the focus lens 3 by a so-called hill-climbing method (step S69). On the other hand, when determining that it is not in the vicinity of the in-focus point, the CPU 15 drives the focus lens 3 based on the distance information obtained by the triangulation (step S68), and then drives the focus lens 3 by a so-called hill-climbing method (step S69). . Steps S70 to S72 are the same as steps S6 to S8 in FIG.

  Here, FIG. 18 shows the drive position of the focus lens 3 on the horizontal axis and the time on the vertical axis. As shown in the figure, in the third technique, the photographing lens is driven based on the triangulation until time t1, and is driven by a so-called hill-climbing method between t1 and t2.

  As described above, according to the third technique, it is possible to quickly focus on a wide area of a subject with a simple configuration using both the imaging optical system (imaging optical system) and the imaging element 5. It becomes. Further, for example, even when the photographing lens of the photographing optical system is far away from the in-focus position, it is possible to focus in a short time.

[Fourth technology]
This is positioned as an improvement of the first technique related to the present invention.

  Since the configuration of the imaging apparatus that employs the focus detection apparatus according to the fourth technique related to the present invention is the same as that in FIG. 1, the differences will be mainly described with reference to the drawings.

  In an imaging apparatus that employs the focus detection apparatus according to the fourth technique, an AF method based on the principle of triangulation and an AF method based on a so-called hill-climbing method are used in combination, and one of the methods is given priority depending on the situation. It is characterized by adopting.

  More specifically, when it is determined that the brightness of the subject light passing through the AF optical system is brighter than a predetermined value, the triangulation AF is preferentially performed, and the brightness of the subject light passing through the AF optical system is determined. When it is determined that the image is darker than the predetermined value, AF based on the so-called hill-climbing method is preferentially performed.

  However, the situation to be determined is not limited to brightness.

  Hereinafter, with reference to the flowchart of FIG. 19, the main operation by the imaging apparatus employing the focus detection apparatus according to the fourth technique will be described in detail.

  Steps S81 and S82 are the same as steps S1 and S2 in FIG. Next, the CPU 100 reads the distance measurement area selected by the distance measurement area selection circuit 113 (step S83). Then, the CPU 100 detects the brightness B1 of the image that has passed through the AF optical system (step S84). More specifically, an output signal from the image sensor 104 is read, and an average brightness of a relatively wide area of the subject is obtained. Next, it is determined whether or not the average brightness B1 of the subject is smaller than a predetermined value α (step S85).

  If the average brightness B1 of the subject is smaller than the predetermined value α in step S805, it is determined that triangulation cannot be performed, and the process proceeds to so-called hill-climbing AF processing in step S87. As such a case, for example, the photographer may be blocking the AF optical system by hand at the time of photographing. On the other hand, if the average brightness B1 of the subject is greater than the predetermined value α in step S85, the CPU 100 next performs AF by triangulation using the subroutine “triangulation AF” (step S86). ). The subroutine “triangulation AF” is the same as that described with reference to FIG.

  Subsequently, the CPU 15 controls the second motor drive 19 to drive the focus lens 3 by a so-called hill climbing method (step S87). The subsequent operations from step S88 to S90 are the same as the operations from step S6 to step S8 in FIG. Thus, a series of processing is completed.

  As described above, according to the fourth technique, when the photographing lens is in the vicinity of the in-focus position by detecting whether the photographing lens is in the vicinity of the in-focus position by a simple method regardless of the triangulation AF, Can eliminate the triangulation AF and focus at high speed with high accuracy.

[First Embodiment]
This is positioned as an improvement example of the first or second technique related to the present invention.

  Since the configuration of the imaging apparatus adopting the focus detection apparatus according to the first embodiment of the present invention is the same as that of FIG. 1 or FIG. 15, here, the differences will be mainly described with reference to FIG. .

  FIG. 20 shows an external configuration of a camera as an imaging apparatus that employs the focus detection apparatus according to the first embodiment of the present invention, and will be described in detail.

As shown in this FIG. 20, connecting the optical axes lines (group line Direction) of the photographing optical system (photographing lens 204) AF-only optical system 203 is inclined relative to the horizontal direction of the camera body.

Distance to the object, as described above in FIG. 3, it means obtained based on the relative shift amount of the base line Direction of 2 images you focused on the imaging plane of the two imaging devices, conventional to determine the distance to the subject using techniques, better-matched reading direction of each of the pixels of the imaging device 5 and the image pickup device 104 (the long side direction or the short side direction) based on line direction, described above It is easily guessed that the process of calculating the deviation amount as described above becomes simple.

However, such an arrangement cannot be adopted because the image sensor 5 used for photographing is inclined with respect to the camera body. To solve such a problem, when the group line Direction in the horizontal direction of the camera body, the design is not preferable because there is a high possibility that AF optical system is blocked by the hand when the photographer grips the camera. From this point of view, even in a camera according to the first embodiment, as shown in the external view of FIG. 20, configured as described above based on line Direction is arranged with a horizontal and a predetermined angle of the camera body ing.

In this case, if the image pickup device 5 arranged at the image forming position of the photographing optical system and the image pickup device 104 arranged at the image forming position of the AF dedicated optical system 203 are configured as shown in FIG. When the position of the reference image 2202 corresponding to the reference image on the image plane received by the image sensor 104 corresponding to the reference image 2201 set on the image plane is calculated, the reference image 2202 is relative to the reference image. since shifts based line direction, the reference image 2202 and calculates a correlation amount based on the above-described equation (2) while shifting by a predetermined pitch in direction said group line, the position of the reference image matching the reference image Ask for. As a result, the pixel readout direction of the image sensor 104 is different from the normal image sensor readout direction, and is inclined with respect to the normal readout direction. Therefore, a relatively complicated calculation is required for setting the reference image. And it takes time to calculate.

  Even in this case, since the reference image does not need to be shifted, the configuration of FIG. 22 may be used.

However, in the first embodiment, and further by inclining only the arrangement of the imaging device 104 based on line Direction As shown in FIG. 21, to match the reading direction of the pixel signals of the image sensor 104 based on line Direction It was. In the first embodiment, such an arrangement simplifies the calculation of the correlation amount and enables high-speed correlation calculation.

Next, a method for generating virtual pixels of the image sensor 5 and the image sensor 104 when the image sensor is arranged as shown in FIG. 21 will be described in detail. First, the address conversion of pixels of the image sensor 5 as the arrangement direction and group lines Direction of virtual pixels of the image sensor 5 coincides.

  The address conversion will be described below with reference to FIG.

Figure 23 shows the original a, theta (horizontal and base line direction angle of direction of the camera) by a rotation to address the positional relationship in obtaining the thick line image due to the oblique scanning indicated by the thin line. In the figure, white circles indicate actual pixels stored in the memory 8, and black circles indicate virtual pixels read from the memory 8. Corresponding to each address position P (00), P (10), P (20), P (01), P (11), P (21), P (02), P (12), P (22) Pixel data is written in the memory 8, and the corresponding address positions Q (10) and Q (20) indicated by bold lines after rotating by θ around the position P (00) using the pixel data at these address positions. ), Q (01), Q (11), Q (21),... Are obtained and sent to the memory 8 as an address signal Add.

  For example, the virtual pixel addresses at the address positions Q (10), Q (20), Q (01), and Q (11) in FIG. 23 are obtained as follows from the illustrated relationship.

Q (10) x ... Px (00) + cos θ
y ... Py (00) + sin θ
Q (20) x ... Px (00) +2 cos θ
= Px (10) +2 cos θ−1
y ... Py (00) + 2sinθ
= Py (10) + 2sin θ
Q (01) x ... Px (00) -sinθ
y ... Py (00) + cosθ
Q (11) x... Px (00) −sin θ + cos θ
= Px (01) −sin θ + cos θ
y ... Py (00) + cos θ + sin θ
= Py (01) + cos θ + sin θ−1
However, x and y are values representing Q (XY) in coordinates before conversion, and Px (XY) and Py (XY) are values of the X coordinate and Y coordinate of the address position P (XY), respectively. It shall represent.
FIG. 24 shows an address conversion diagram when the address conversion principle diagram shown in FIG. 23 is rotated by θ = 30 degrees. Let (XST, YST) be the address of the origin.

In FIG.
XST = 0 XW = 0.866 X0 = −2.4 × 0.5
YST = 0 YW = 0.5 / 2.4 Y0 = 0.866
As is clear from FIG. 24, Xmn and Y address in which the X address in the number of columns m and the number of lines n (coordinates after conversion (m, n)) is represented by the coordinates before conversion are the coordinates before conversion. The general formula representing Ymn is as follows.

Xmn = XST + m · XW + n · X0
Ymn = YST + m · YW + n · Y0
For example, the addresses (coordinates) representing the coordinates (m, n) after the conversion of the 0th line (n = 0) with the coordinates before the conversion are (XY) = (0, 0), (0.866, 0 .208), (1.732, 0.417),... (XY) = (− 1.2, 0.866), (−0.334, 1.074) in the first line (n = 1). ), (0.532, 1.28),. Here, it is clear from the figure that the integer part of each address indicates the address Add and the decimal part indicates the interpolation coefficient K.

  For the interpolation process, it is preferable to use a four-point weighting method as shown in FIG.

As shown in FIG. 25, pixel data X1 and pixel data X2 are defined as shown in FIG. 25, and the surrounding four points P (11), P (21), P (12) are defined as Q. the pixel data of the P (22), using the same reference numerals, P (11), respectively, P (21), is defined as P (12), P (22 ), obtaining the pixel data Q by the following equation .

Q = (1-Ky) X1 + Ky · X2
X1 = (1-Kx) P (11) + KxP (21)
X2 = (1-Kx) P (12) + KxP (22)
Therefore,
Q = (1-Kx) (1-Ky) P (11) + Kx (1-Ky) P (21)
+ Ky (1−Kx) P (12) + Kx · Ky · P (22) (7) The calculation of the equation (7) is P (11), P which is data of 4 pixel addresses in one cycle. This can be realized by reading (21), P (12), and P (22).

As described above, in the first embodiment, the value of pixel data at each address after address conversion is calculated as described above, and a plurality of pixel data at these new addresses are synthesized, as described above. Virtual pixel data is generated. As described above, in order to obtain pixel data at a new address when coordinates are rotated, a relatively complicated operation is required. However, in the first embodiment, the imaging device 104 is the slope is zero with respect to the base line Direction, so arranged to be inclined with respect to the base line Direction only the imaging device 5, the amount of computation of the coordinate transformation Is less.

Moreover, the reference image so setting the reference image on the image received at the image pickup device 5, since it is sufficient to address translation of only the reference image, which needs to shift an image based on line Direction for correlating the amount of Since there is no need to perform complicated address conversion, there is also an advantage that the amount of calculation is extremely small. In the first embodiment, has been described only rotated the address conversion can be easily realized by the same 4-point weighted scheme applies address translation based line Direction and vertical reference image as a reference image . Thus, by correcting the deviation of the group line Direction and vertical pixels of the reference image and the reference image can be performed with high correlation calculation accuracy.

[Summary]
The first to fourth techniques and the first embodiment related to the present invention have been described above. However, the present invention is not limited to this, and various improvements and modifications can be made without departing from the spirit of the present invention. Of course, it is possible. For example, FIGS. 1 and 15 show an example in which the CPU that controls the triangulation AF and the CPU that controls the so-called hill-climbing AF are separated. However, the present invention is not limited to this, and one CPU controls both. However, a further CPU may be used.
The first to fourth techniques and the first embodiment related to the present invention are not only digital cameras and digital video cameras but also so-called mobile phones with camera functions, PDAs, notebook personal computers, and the like. Of course, it can be widely applied to mobile devices and the like.

  Of course, some or all of the technical matters of the first to fourth techniques and the first embodiment related to the present invention can be combined as appropriate.

1 is a block diagram illustrating a configuration example of an imaging apparatus that employs a focus detection apparatus according to a first technique related to the present invention. The figure which shows the example which displayed on the display screen of LCD10 of FIG. 1 the parameter | index 201 which shows the visual field which can perform a triangular distance measurement, and the parameter | index 202 which shows a distance measurement area | region together. The conceptual diagram for demonstrating the ranging principle relevant to the focus detection apparatus which concerns on the 1st technique relevant to this invention. 6 is a flowchart for explaining in detail the operation of the imaging apparatus employing the focus detection apparatus according to the first technique related to the present invention. 5 is a flowchart for explaining in detail a subroutine “triangulation AF” executed in step S4 of FIG. (A) is a figure which shows the arrangement | sequence of the pixel of the image pick-up element 5 of a 1st ranging area, (b) is a figure which shows the arrangement | sequence of the pixel of the image pick-up element 104 of a 2nd ranging area. The figure which shows the relationship between the data number m of a reference part, and correlation amount A (m). 1 is a configuration diagram of a system that writes data to an EEPROM 25 when manufacturing an imaging apparatus 1 that employs a focus detection apparatus according to a first technique related to the present invention. 9 is a flowchart for explaining a data writing operation related to a coordinate position in the EEPROM 25 of the imaging apparatus 1 by the system of FIG. (A) is a figure which shows the pixel size of the image pick-up element 5, (b) is a figure which shows the pixel size of the image pick-up element 104. (A) thru | or (c) is a figure which shows the relationship between the visual field of an imaging optical system, and the visual field of AF optical system. 1 is a configuration diagram of a system that writes data to an EEPROM 25 when manufacturing an imaging apparatus 1 that employs a focus detection apparatus according to a first technique related to the present invention. 13 is a flowchart for explaining a flow of data writing operation to the EEPROM 25 of the imaging apparatus 1 by the system of FIG. 6 is a flowchart for explaining in detail a subroutine “ΔL1, ΔL2 calculation” executed in step S13 of FIG. The block diagram which shows the structural example of the imaging device which employ | adopted the focus detection apparatus which concerns on the 2nd technique relevant to this invention. 10 is a flowchart for explaining in detail the flow of processing of a subroutine “triangulation AF” according to a second technique related to the present invention; 10 is a flowchart for explaining in detail the main operation of an imaging apparatus employing a focus detection apparatus according to a third technique related to the present invention. The horizontal axis represents the drive position of the focus lens 3 and the vertical axis represents time. The flowchart explaining in detail the main operation | movement by the imaging device which employ | adopted the focus detection apparatus which concerns on the 4th technique relevant to this invention. 1 is an external configuration diagram of a camera as an imaging apparatus that employs a focus detection apparatus according to a first embodiment of the present invention. Tilt the only arrangement of the image sensor 104 based on line Direction, diagram showing a state in which to match the reading direction of the pixel signals of the image sensor 104 based on line Direction. The figure which shows the positional relationship of the image pick-up element 5 arrange | positioned in the image formation position of an imaging optical system, and the image pick-up element 104 arrange | positioned in the image formation position of the AF exclusive use optical system 203. FIG. Address conversion principle diagram. FIG. 23 is a principle diagram of address conversion when FIG. 22 is rotated by θ = 30 degrees. The conceptual diagram for demonstrating a 4-point weighting system.

  DESCRIPTION OF SYMBOLS 1 ... Imaging device, 2 ... Zoom lens, 3 ... Focus lens, 4 ... Diaphragm, 5 ... Imaging element, 6 ... Imaging circuit, 7 ... A / D converter 8 ... Memory, 9 ... D / A converter, 10 ... LCD, 11 ... Compression / decompression circuit, 12 ... Recording memory, 13 ... AE processing circuit, 14. ..AF processing circuit, 15 ... CPU, 16 ... TG circuit, 17 ... CCD driver, 18-20 ... first to third motor drivers, 21-23 ... motor, 24 ... Operation switch, 25 ... EEPROM, 26 ... Battery, 100 ... CPU, 101 ... Zoom lens, 102 ... Focus lens, 103 ... Aperture, 104 ... Image sensor 105 ... Imaging circuit 106 ... A / D converter 107 ... Mori, 108 ... AF processing circuit, 109 ... TG circuit, 110 ... CCD driver, 111 ... motor driver, 112 ... motor, 113 ... ranging area selection circuit, 200 ... CPU.

Claims (2)

A first optical system that is used for both photographing and focus detection;
Disposed at an imaging position of the first optical system, a first imaging element the long side direction or the short side direction are arranged to coincide with the horizontal direction of the camera,
A second optical system for focus detection arranged at a predetermined baseline length from the first optical system;
The second optical system is disposed at the imaging position, and the long side direction or the short side direction has a predetermined angle with the long side direction and the short side direction of the first image sensor, and the second optical system and the first optical system. A second image sensor arranged to coincide with the baseline direction of the optical system ;
A first image data representing a first image captured by the first image pickup device, based on the second image data representing a second image captured by the second image sensor, the portion of the first image a reference image that has been set in the area, and the between the reference image is an image that matches the reference image of the second image, the correlation amount calculating means for calculating a relative deviation amount of the base line direction ,
Focus detecting apparatus of a camera, characterized in that it comprises a. The reference image is what is represented by the data of the virtual pixels arranged in the base line Direction of the first image data, the reference image is to the baseline direction of the second image data The camera focus detection apparatus according to claim 1, wherein the focus detection apparatus is represented by data of pixels to be arranged .

Images