JP4032877B2 - Auto focus system - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an autofocus system, and more particularly to an autofocus system that controls the focus of a photographing lens by a contrast method.
[0002]
[Prior art]
In general, autofocus for a video camera or the like is based on a contrast method. This contrast method integrates the high frequency components of a video signal within a certain range (focus area) of the video signal obtained from the image sensor to obtain a focus evaluation value, and the focus evaluation value is maximized (maximum). Thus, the focus adjustment is automatically performed. As a result, it is possible to obtain the best focus (focusing) that maximizes the sharpness (contrast) of the image captured by the image sensor.
[0003]
As a method for setting the focus at the in-focus position (the maximum point of the focus evaluation value), a so-called hill-climbing method is widely known. This method detects the direction in which the focus evaluation value increases by comparing the focus evaluation values at two different points when the focus is moved, moves the focus in that direction, and changes the focus evaluation value from increase to decrease. The focus is returned to the position before the focus evaluation value decreases, and the focus is set to the maximum point of the focus evaluation value.
[0004]
In addition, in the case of the above-described hill-climbing method, there is a drawback in that it is not possible to determine the increase direction or focus of the focus evaluation value unless the focus is actually moved. In other words, Japanese Patent Application Nos. 2001-168246 and 2001-168247 have proposed methods for detecting the focus state (front focus, rear focus, in-focus) of the taking lens without moving the focus. No., Japanese Patent Application No. 2001-168248, Japanese Patent Application No. 2001-168249, etc.). According to this focus state detection method, it is possible to immediately know the current focus state from the magnitude relationship between the current focus evaluation values obtained from the respective image sensors, so that the focus movement direction and focus can be adjusted without moving the focus. Judgment can be made. Therefore, the autofocus using this method has an advantage that the focus can be quickly set to the in-focus position.
[0005]
[Problems to be solved by the invention]
By the way, when performing autofocus using a plurality of image sensors as described above, in order to increase the accuracy of autofocus, the sensitivity of the focus evaluation value obtained from each image sensor (the subject light incident on each image sensor and Therefore, it is necessary to match the magnitude of the focus evaluation value obtained with respect to it). That is, autofocus control is performed on the assumption that the characteristics of each image sensor and the processing circuit for processing a video signal from each image sensor to obtain a focus evaluation value are the same. Since there are variations in the characteristics of the image sensor and the processing circuit, it is important to calibrate the sensitivity of the focus evaluation value obtained from each image sensor in advance in order to increase the accuracy of autofocus.
[0006]
The present invention has been made in view of such circumstances. In the case where autofocus control is performed based on focus evaluation values obtained from a plurality of image sensors, the sensitivity of the focus evaluation value is appropriately calibrated and the accuracy is high. An object of the present invention is to provide an autofocus system that can perform autofocus.
[0007]
[Means for Solving the Problems]
In order to achieve the above object, the invention according to claim 1, the subject light incident on the photographing lens is imaged by a plurality of imaging means arranged at positions having different optical path lengths, and is imaged by the plurality of imaging means. The focus of the photographing lens is moved to the in-focus position based on the focus evaluation value indicating the sharpness of each image. In addition, for each focus evaluation value obtained from each imaging means, a value obtained by multiplying the amount of change of the focus evaluation value from a predetermined reference value by a predetermined gain is used for controlling the focus of the photographing lens. Focus evaluation value In the autofocus system, By setting the reference value and the gain to appropriate values The sensitivity is calibrated so that the sensitivity of the focus evaluation values obtained from each of the imaging means is the same. A calibration unit, and the calibration unit moves a focus of the photographing lens, and a reference value setting unit that sets a focus evaluation value obtained from each imaging unit as the reference value when an image without contrast is captured. Gain setting means for setting the gain so that the maximum values of the focus evaluation values obtained from the respective imaging means coincide with each other, and when setting the gain by the gain setting means, Moving the focus with the zoom of the taking lens set to a predetermined position. It is a feature.
[0008]
The invention described in claim 2 The subject lens incident on the photographic lens is imaged by a plurality of imaging units arranged at positions having different optical path lengths, and the photographic lens is based on a focus evaluation value indicating the sharpness of each image captured by the plurality of imaging units. Move the focus to the in-focus position, For each focus evaluation value obtained from each of the imaging means, a value obtained by multiplying the amount of change of the focus evaluation value from a predetermined reference value by a predetermined gain is used for controlling the focus of the photographing lens. Evaluation value In the autofocus system, By setting the reference value and the gain to appropriate values The sensitivity of the focus evaluation values obtained from the respective imaging means is matched. Calibrate the sensitivity Proofreading means The calibration means comprises Reference value setting means for setting the focus evaluation value obtained from each imaging means as the reference value when an image having no contrast is taken, and obtained from each imaging means when the focus of the photographing lens is moved. Gain setting means for setting the gain so that the maximum values of the focus evaluation values obtained coincide with each other Then, after the focus of the photographing lens is moved at a high speed to confirm the presence of the maximum value of the focus evaluation value obtained from each image sensor, the focus is moved at a low speed and obtained from each image sensor. A gain setting means for accurately detecting the maximum focus evaluation value; It is characterized by having.
[0009]
The invention according to claim 3 is the invention according to claim 1 or claim 2, wherein the calibration means is configured to perform initial setting when a predetermined switch is turned on, when power is turned on, or before shipment. Sometimes the sensitivity calibration is automatically performed.
[0010]
The invention according to claim 4 Claim 1 or According to a second aspect of the present invention, the calibration means stores the reference value and gain set by the reference value setting means and the gain setting means in a memory as calibration data.
[0011]
The invention according to claim 5 Claim 1 or The invention according to claim 2, wherein the reference value setting means captures an image without the contrast by each of the imaging elements by closing an iris of the photographing lens. 1 or 2 autofocus system.
[0013]
Claims 6 The invention described in claim 1 The gain setting means moves the focus at a low speed after confirming the existence of the maximum value of the focus evaluation value obtained from each image sensor by moving the focus of the photographing lens at a high speed. The maximum value of the focus evaluation value obtained from each image sensor is detected with high accuracy.
[0014]
Claims 7 The invention according to claim 1 is characterized in that, in the invention according to claim 1 or 2, there is provided display means for displaying that the calibration of the sensitivity by the calibration means is being executed.
[0015]
According to the present invention, since the calibration means for calibrating the sensitivity of the focus evaluation value obtained from each image sensor is provided, the sensitivity of the focus evaluation value can be appropriately calibrated, and highly accurate autofocus can be performed. In addition, it is possible to automatically calibrate the sensitivity of the focus evaluation value, so that troublesome work can be eliminated.
[0016]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, preferred embodiments of an autofocus system according to the present invention will be described in detail with reference to the accompanying drawings.
[0017]
FIG. 1 is a block diagram of a television camera system to which an autofocus system according to the present invention is applied. As shown in the figure, this TV camera system is composed of a lens device 10 and a camera body 12, and the camera body 12 shoots a video for broadcasting and outputs or records a video signal of a predetermined format. An image sensor for recording on a medium (hereinafter referred to as an image sensor for video), a required circuit, and the like are mounted.
[0018]
The lens device 10 is detachably attached to the camera body 12 and mainly includes an optical system (photographing lens) and a control system. First, the configuration of the photographing lens will be described. The photographing lens includes a focus lens (group) 16, a zoom lens (group) 18, an iris 20, and a relay lens (relay optical) including a front relay lens 22A and a rear relay lens 22B. System) etc. are arranged. A half mirror 24 is arranged between the front relay lens 22A and the rear relay lens 22B of the relay optical system for branching the subject light for focus state detection from the subject light incident on the photographing lens. .
[0019]
The half mirror 24 is installed with an inclination of approximately 45 degrees with respect to the optical axis O of the photographing lens, and a part of the subject light (for example, 1/3 light amount) that has passed through the front relay lens 22A is used for focus state detection. Is reflected at right angles as subject light.
[0020]
The subject light transmitted through the half mirror 24 is emitted from the rear end side of the photographing lens as subject light for an image and then enters the imaging unit 14 of the camera body 12. Although the configuration of the imaging unit 14 is omitted, subject light incident on the imaging unit 14 is separated into, for example, three colors of red light, green light, and blue light by a color separation optical system, and an image pickup device for video for each color. Is incident on the imaging surface. As a result, a color image for broadcasting is taken. In addition, the focus surface P in the figure indicates the optically equivalent position on the optical axis O of the imaging lens with respect to the imaging surface of each image pickup device.
[0021]
On the other hand, the subject light reflected by the half mirror 24 travels along the optical axis O ′ perpendicular to the optical axis O as subject light for focus state detection, and enters the relay lens 26. Then, the light is condensed by the relay lens 26 and enters the focus state detection unit 28.
[0022]
The focus state detection unit 28 includes two prisms 30A and 30B constituting a light splitting optical system and a pair of image pickup devices 32A and 32B for focus state detection (hereinafter referred to as focus state detection image pickup devices 32A and 32B). It is configured.
[0023]
As described above, the subject light reflected by the half mirror 24 travels along the optical axis O ′ and first enters the first prism 30A. Then, the light is equally divided into reflected light and transmitted light by the half mirror surface M of the first prism 30A. Of these, the reflected light is incident on the imaging surface of one focus state detection image sensor 32A, and the transmitted light is incident on the other focus state detection image sensor 32B. Note that, for example, 1/6 of the total object light incident on the photographing lens is incident on the respective imaging surfaces of the focus state detection imaging elements 32A and 32B.
[0024]
FIG. 2 shows the optical axis of the subject light incident on the video image sensor of the camera body 12 and the optical axis of the subject light incident on the pair of focus state detection image sensors 32A and 32B on the same straight line. . As shown in the figure, the optical path length of the subject light incident on one focus state detection image pickup device 32A is set shorter than the optical path length of the subject light incident on the other focus state detection image pickup device 32B. The optical path length of the subject light incident on the image pickup surface (focus plane P) of the image pickup device is set to be an intermediate length. That is, the pair of focus state detection image pickup devices 32A and 32B (the image pickup surfaces thereof) are respectively disposed at positions equidistant (d) in the front-rear direction from the image pickup surface (focus surface P) of the image pickup device.
[0025]
Accordingly, the subject light for focus form detection branched by the half mirror 24 is equidistant from the image pickup surface (focus surface P) of the image pickup device for video by the pair of focus state detection image pickup devices 32A and 32B. Imaged at position. The focus state detection imaging elements 32A and 32B acquire a video signal for focus state detection (autofocus control) as will be described later, and do not need to capture a color image. In this embodiment, it is assumed that the CCD is for capturing a black and white image.
[0026]
Next, the control system of the lens apparatus 10 will be described. The focus lens 16, the zoom lens 18, and the iris 20 are respectively connected to the focus motor 42, the zoom motor 46, and the iris motor 50 shown in FIG. When the focus motor 42 is driven, the focus lens 16 is moved in the optical axis direction to change the focal position (shooting distance) of the photographing lens. When the zoom motor 46 is driven, the zoom lens 18 is moved to the optical axis. When the zoom magnification of the photographing lens is changed by moving in the direction and the iris motor 50 is driven, the aperture blades of the iris 20 are opened and closed to change the aperture diameter (aperture value).
[0027]
Each of the motors 42, 46, 50 is supplied with a driving voltage from a focus motor driving circuit 44, a zoom motor driving circuit 48, and an iris motor driving circuit 52, and each of the driving circuits 44, 48, 52 has a D / A conversion. A control signal is given from the CPU 40 mounted on the lens apparatus 10 via the device 54.
[0028]
The control signal output from the CPU 40 indicates, for example, the voltage value corresponding to the rotational speed of the motor to be driven, that is, the operating speed of the target to be driven (focus lens 16, zoom lens 18, iris 20). When the voltage value is converted into an analog signal by the D / A converter 54 and applied to the corresponding drive circuit 44, 48, 52, the voltage is amplified by each drive circuit 44, 48, 52, and the amplified voltage is driven. A voltage is applied to the corresponding motor 42, 46, 50. As a result, the rotational speeds of the motors 42, 46 and 50 are controlled by the CPU 40.
[0029]
The current positions of the focus lens 16, the zoom lens 18, and the iris 20 are detected by a focus lens position detector 56 such as a potentiometer, a zoom lens position detector 58, and an iris position detector 60, respectively. Detection signals output from 56, 58 and 60 are supplied to the CPU 40 via the A / D converter 68.
[0030]
Therefore, in the processing of the CPU 40, the operation speeds of the focus lens 16, the zoom lens 18, and the iris 20 can be controlled to desired speeds by controlling the rotation speeds of the motors 42, 46, and 50 as described above. At the same time, the rotational speeds of the motors 42, 46, 50 are controlled by reading the current positions of the focus lens 16, the zoom lens 18, and the iris 20 based on detection signals from the position detectors 56, 58, 60. The setting positions of the lens 16, the zoom lens 18, and the iris 20 can be controlled to desired positions.
[0031]
In general, the focus and zoom of the photographing lens can be manually operated by an operator by connecting a controller such as a focus demand 62 and a zoom demand 64 to the lens device 10. Then, a focus command signal (focus demand data) having a voltage corresponding to the rotation position of the manual operation member (focus ring) is output and supplied to the CPU 40 via the A / D converter 68. For example, the CPU 40 sets the value of the focus demand data as a value indicating the movement target position of the focus lens 16, and the difference between the movement target position and the current position (focus position data) acquired from the focus lens position detector 56. As described above, the control signal for instructing the movement at the movement speed corresponding to is output to the focus motor drive circuit 44 via the D / A converter 54. As a result, the focus lens 16 moves to the movement target position commanded by the focus demand 62 and stops.
[0032]
In general, a voltage corresponding to the rotation position of the operation member (thumbing) is supplied from the zoom demand 64 to the CPU 40 as a value indicating the movement target speed of the zoom lens 18, and the CPU 40 moves at the movement target speed. Is output to the zoom motor drive circuit 48, and the zoom lens 18 is moved at the movement target speed commanded by the zoom demand 64. For the iris 20, generally, a command signal for instructing the operation target position of the iris 20 is given from the camera body 12 to the CPU 40, and the position of the iris 20 is controlled by the CPU 40 so as to be the operation target position.
[0033]
In addition, the focus control of the photographing lens includes manual focus (MF) control using the focus demand 62 and auto focus (AF) control based on the video signals from the focus state detection imaging elements 32A and 32B. There is an AF switch 66 for switching between MF control and AF control in the lens device 10 or the focus demand 62. The on / off state of the AF switch 66 is detected by the CPU 40. When the AF switch 66 is off, MF control is performed, and based on the focus command signal (focus demand data) from the focus demand 62 as described above. The focus lens 16 is controlled.
[0034]
On the other hand, when the AF switch 66 is turned on, AF control is performed. That is, the image (video) imaged by the pair of focus state detection imaging elements 32A and 32B is a video signal for sequentially transmitting the pixel values along a plurality of scanning lines (horizontal lines) constituting one screen. And is input to the focus evaluation value generation unit 70. The configuration and processing of the focus evaluation value generation unit 70 will be described later. In the focus evaluation value generation unit 70, the contrast of each image captured by the focus state detection imaging elements 32A and 32B from the input video signal. Focus evaluation values indicating the level of (sharpness) are generated, and the generated focus evaluation values are given to the CPU 40. The focus evaluation value generated based on the video signal from the focus state detection image sensor 32A is referred to as a focus evaluation value of channel A (chA), and is based on the video signal from the focus state detection image sensor 32B. The generated focus evaluation value is referred to as a focus evaluation value of channel B (chB). As will be described in detail later, the CPU 40 acquires the focus evaluation values of chA and chB acquired from the focus evaluation value generation unit 70, and based on the acquired focus evaluation values, the focus state of the photographing lens (front pin, rear pin) The focus lens 16 is controlled so that the photographing lens is in focus.
[0035]
In the figure, a memory 72 is a non-volatile memory in which, for example, calibration data described later for calibrating the sensitivity of the focus evaluation value is stored, and a calibration start switch 74 is a switch for instructing the start of the calibration. Yes, the display 76 is a display means such as an LED for indicating that the calibration is being executed.
[0036]
The AF control in the camera system configured as described above will be described in detail below. First, the configuration and processing of the focus evaluation value generation unit 70 will be described. As shown in FIG. 3, the video signals output from the focus state detection imaging elements 32 </ b> A and 32 </ b> B are input to high-pass filters (HPF) 80 </ b> A and 80 </ b> B of the focus evaluation value generation unit 70. Here, since the focus state detection imaging elements 32A and 32B are both CCDs that capture black and white images, the video signals output from the focus state detection imaging elements 32A and 32B constitute respective screens. It is a luminance signal indicating the luminance value of each pixel
The high frequency components of the video signals input to the HPFs 80A and 80B are extracted by the HPFs 80A and 80B, respectively, and the signals of the high frequency components are subsequently converted into digital signals by the A / D converters 82A and 82B. The Only digital signals corresponding to pixels within a predetermined focus area (for example, the center portion of the screen) among the digital signals for one screen (one field) of the images captured by the focus state detection imaging elements 32A and 32B. Are extracted by the gate circuits 84A and 84B, and the values of the digital signals in the extracted range are added by the adders 86A and 86B. Thereby, the sum total of the values of the high frequency components of the video signal in the focus area is obtained. The values obtained by the adders 86A and 86B are focus evaluation values indicating the sharpness of the image in the focus area, and the focus evaluation values obtained by the adder 86A are the focus evaluation values of the channel A (chA). As described above, the focus evaluation value obtained by the adder 86B is given to the CPU 40 as the focus evaluation value of the channel B (chB).
[0037]
It should be noted that various synchronization signals are given to the respective circuits such as the focus state detection imaging devices 32A and 32B and the gate circuits 84A and 84B from the synchronization signal generation circuit 88 shown in FIG. Is planned. Further, a vertical synchronizing signal (V signal) for each field of the video signal is supplied from the synchronizing signal generation circuit 88 to the CPU 40.
[0038]
Next, focus state detection and focus (focus lens 16) control based on the focus evaluation value will be described. As described above, the current focus state of the photographic lens with respect to the imaging surface (focus surface P) of the imaging device for video can be detected based on the focus evaluation values of chA and chB acquired from the focus evaluation value generation unit 70.
[0039]
FIG. 4 is a diagram illustrating the state of the focus evaluation value with respect to the focus position when a certain subject is imaged, with the horizontal axis representing the position of the focus lens 16 (focus position) and the vertical axis representing the focus evaluation value. . A curve C indicated by a dotted line in the figure indicates a focus evaluation value when it is assumed that a focus evaluation value is obtained from a video signal from a video image sensor (or an image sensor arranged at a conjugate position with the video image sensor). The curves A and B indicated by the solid line in the figure indicate the focus evaluation values of chA and chB obtained from the focus state detection imaging elements 32A and 32B, respectively, with respect to the focus position. Is. In the figure, the position F3 at which the focus evaluation value of the curve C is maximum (maximum) is the in-focus position.
[0040]
When the focus position of the photographic lens is set to F1 in the figure, the chA focus evaluation value V _A1 Is a value corresponding to the position F1 of the curve A, and the focus evaluation value V of chB _B1 Is a value corresponding to the position F1 of the curve B. In this case, the chA focus evaluation value V _A1 Is the focus evaluation value V of chB _B1 From this, it can be seen that the focus position is set closer to the in-focus position (F3), that is, the state of the front pin.
[0041]
On the other hand, when the focus position of the photographic lens is set to F2 in FIG. _A2 Is a value corresponding to the position F2 of the curve A, and the focus evaluation value V of chB _B2 Is a value corresponding to the position F2 of the curve B. In this case, the chA focus evaluation value V _A2 Is the focus evaluation value V of chB _B2 From this, it can be seen that the focus position is set to the infinity side from the in-focus position (F3), that is, the state of the rear pin.
[0042]
On the other hand, when the focus position of the photographing lens is set to F3, that is, the in-focus position, the chA focus evaluation value V _A3 Is a value corresponding to the position F3 of the curve A, and the focus evaluation value V of chB _B3 Is a value corresponding to the position F3 of the curve B. At this time, the focus evaluation value V of chA _A3 And chB focus evaluation value V _B3 From this, it can be seen that the focus position is set to the in-focus position (F3).
[0043]
As described above, it is possible to detect whether the current focus state of the photographing lens is the front pin, the rear pin, or the in-focus state based on the chA and chB focus evaluation values obtained from the focus evaluation value generation unit 70. it can.
[0044]
Therefore, the focus lens 16 can be moved to the in-focus position by controlling the position of the focus lens 16 based on the chA and chB focus evaluation values obtained from the focus evaluation value generation unit 70. That is, when the focus evaluation values of chA and chB are determined to be the front pins, the focus lens 16 is moved in the infinity direction, and when the focus evaluation values are determined to be the rear pins, When the focus lens 16 is moved in the closest direction and the focus lens 16 is determined to be in focus, the focus lens 16 can be moved to the focus position by stopping the focus lens 16 at that position. it can.
[0045]
The processing of the CPU 40 corresponding to the above description will be specifically described. When the focus evaluation value of chA acquired from the focus evaluation value generation unit 70 is AFV_A and the focus evaluation value of chB is AFV_B, AFV_A> AFV_B. Since it is in the state of the front pin, the CPU 40 changes the currently set movement target position of the focus lens 16 to the infinity side by a movement amount (positive value) described later, and focuses on the new movement target position. A control signal for moving the lens 16 is output to the focus motor drive circuit 44 via the D / A converter 54. On the other hand, when AFV_A <AFV_B, the rear pin is in the state, so that the currently set movement target position of the focus lens 16 is changed to the closest side by a movement amount (negative value) described later, and the new A control signal for moving the focus lens 16 to a desired movement target position is output to the focus motor drive circuit 44 via the D / A converter 54. Such processing is repeated, and when AFV_A = AFV_B, the movement of the focus lens 16 is stopped. As a result, the focus lens 16 moves to the in-focus position.
[0046]
Here, the value of the detection signal (focus position data) indicating the current position of the focus lens 16 acquired from the focus lens position detector 56 is F_POSI, and the movement target position of the focus lens 16 set as described above is AF_CTRL. Then, the CPU 40 sets a value obtained by subtracting the current position F_POSI from the movement target position AF_CTRL, that is, AF_CTRL−F_POSI as a control signal value F_SPEED to be output to the focus motor drive circuit 44. The control signal output to the focus motor drive circuit 44 is a value corresponding to the rotational speed of the focus motor 42 (moving speed of the focus lens 16) commanded to the focus motor drive circuit 44, and the control signal set as described above. By outputting the value F_SPEED to the focus motor drive circuit 44, the focus lens 16 moves at a speed corresponding to the difference (AF_CTRL−F_POSI) between the movement target position AF_CTRL and the current position F_POSI.
[0047]
Next, the amount of movement added to the current movement target position when setting a new movement target position of the focus lens 16 as described above will be described. As described above, the difference between the current position F_POSI of the focus lens 16 and the movement target position AF_CTRL corresponds to the movement speed of the focus lens 16, and the current movement target position is set when a new movement target position AF_CTRL is set. The larger the amount of movement to be added is, the faster the moving speed of the focus lens 16 is. The smaller the amount of movement is, the smaller the moving speed of the focus lens 16 is.
[0048]
On the other hand, when the focus lens 16 is moved to the in-focus position, in order to surely stop the focus lens 16 with the stable operation at the in-focus position, the above-mentioned movement is performed as the focus lens 16 approaches the in-focus position. It is necessary to reduce the moving amount of the focus lens 16 by reducing the amount, and to move the moving amount of the focus lens 16 to 0 when the focus position is reached.
[0049]
Therefore, the CPU 40 calculates the difference ΔAFV (= AFV_A−AFV_B) between the focus evaluation values of chA and chB, and moves the difference ΔAFV * AFG obtained by multiplying the difference ΔAFV (= AFV_A−AFV_B) by a predetermined AF gain AFG. Set as a quantity. As a result, when the focus lens 16 reaches the in-focus position, that is, when the focus evaluation value difference ΔAFV becomes 0 (AFV_A = AFV_B), the movement amount ΔAFV * AFG becomes 0, and the focus lens 16 Stops at the in-focus position. As can be seen from FIG. 4, when the focus lens 16 approaches the in-focus position from the focus position, the focus evaluation value difference ΔAFV decreases and the movement amount ΔAFV * AFG gradually approaches zero. The moving speed of the focus lens 16 is also gradually reduced.
[0050]
Further, as described above, instead of using the value ΔAFV * AFG obtained by multiplying the focus evaluation value difference ΔAFV between chA and chB by a predetermined AF gain AFG as the movement amount, the movement amount is set as follows. You can also. That is, the CPU 40 first obtains a ratio ΔAFV = AFV_A / AFV_B between the focus evaluation value AFV_A of chA and the focus evaluation value AFV_B of chB. In the case of AFV_A> AFV_B (ΔAFV> 1), that is, in the state of the front pin (see FIG. 4), the movement amount is set to (ΔAFV−1) * AFG. AFG represents a predetermined AF gain value. On the other hand, in the case of AFV_A ≦ AFV_B (ΔAFV ≦ 1), that is, in the rear pin state (or in-focus state), the movement amount is set to − (1 / ΔAFV−1) * AFG.
As a result, when the focus lens 16 reaches the in-focus position, ΔAFV = 1, so that the movement amount becomes 0, and the focus lens 16 stops at the in-focus position. Further, when the focus lens 16 approaches the in-focus position from near the in-focus position, (ΔAFV-1) or (1 / ΔAFV-1) decreases and the movement amount gradually approaches 0, and the focus The moving speed of the lens 16 is also gradually reduced. Further, when the focus evaluation value ratio ΔAFV = AFV_A / AFV_B is used as an element for determining the movement amount in this way, the movement amount (movement speed) due to the size of the focus evaluation value itself is little and more stable. The focus operation can be realized.
[0051]
Next, calibration of the sensitivity of the chA and chB focus evaluation values acquired from the focus evaluation value generation unit 70 will be described. In the above description, it is assumed that the sensitivity of the chA and chB focus evaluation values obtained from the focus evaluation value generation unit 70 (hereinafter referred to as focus evaluation value sensitivity) is the same. That is, the focus state detection imaging device 32A for generating the chA focus evaluation value and the characteristics of various circuits related to chA in the focus evaluation value generation unit 70, and the focus state for generating the chB focus evaluation value It is assumed that the characteristics of various circuits related to chB in the detection image sensor 32B and the focus evaluation value generation unit 70 match. However, in reality, the focus evaluation value sensitivities may not match, and in this embodiment, the sensitivity of the chA and chB focus evaluation values obtained from the focus evaluation value generation unit 70 can be calibrated. The case will be described.
[0052]
First, in the CPU 40, the focus evaluation value for chA acquired from the focus evaluation value generation unit 70 is AFV_A0, and the focus evaluation value for chB is AFV_B0. Then, the focus evaluation values of chA and chB with the calibrated sensitivity, that is, the focus evaluation values used for the AF control (referred to as corrected focus evaluation values) are respectively AFV_A and AFV_B as described above. At this time,
[0053]
[Expression 1]
AFV_A = (AFV_A0−AFV_A_OFFSET) * AFG_A… ▲ 1 ▼
[0054]
[Expression 2]
AFV_B = (AFV_B0−AFV_B_OFFSET) * AFG_B… ▲ 2 ▼
Thus, the corrected focus evaluation value is calculated. The focus evaluation values AFV_A0 and AFV_B0 acquired from the focus evaluation value generation unit 70 before correction by the above formulas (1) and (2) are referred to as focus evaluation values before correction.
[0055]
Therefore, the focus evaluation value sensitivity can be calibrated by setting the values of AFV_A_OFFSET, AFV_B_OFFSET, AFG_A, and AFG_B in the above formulas (1) and (2) to appropriate values, and the focus evaluation value sensitivity of chA and chB Can be matched.
[0056]
Although specific calibration processing will be described later, the AFV_A_OFFSET and AFV_B_OFFSET (reference values) obtained by the calibration processing are focus evaluations of the black levels of chA and chB obtained in a state where the focus state detection imaging elements 32A and 32B are shielded from light, respectively. The value (focus evaluation value before correction) is set. For example, the focus evaluation values of chA and chB acquired when the iris 20 is completely closed are set. However, the present invention is not limited to this, and for example, the focus evaluation values of chA and chB obtained when a plain subject (an image without contrast) is captured may be set.
[0057]
On the other hand, the above AFG_A and AFG_B (gain) are chA and chB obtained when the focus lens 16 is moved while photographing a predetermined subject such as a chart (however, it may not be a specially determined subject). Based on the maximum value of the respective focus evaluation values, the values are set so that the corrected focus evaluation values match.
[0058]
Using the AFV_A_OFFSET, AFV_B_OFFSET, AFG_A, and AFG_B set by the calibration process in this way, the focus evaluation values AFV_A0 and chB of chA acquired from the focus evaluation value generation unit 70 by the above formulas (1) and (2) By correcting the evaluation value AFV_B0, the chA and chB focus evaluation values AFV_A and AFV_B in a state where the focus evaluation value sensitivities coincide with each other can be acquired.
[0059]
Further, the AFV_A_OFFSET, AFV_B_OFFSET, AFG_A, and AFG_B set by the calibration process are stored as calibration data in the memory 72 shown in FIG. 1, and when AF control is performed, the calibration data is read and the focus evaluation value is read. The focus evaluation value acquired from the generation unit 70 is used for correction. The calibration process is executed by turning on the calibration start switch 74 shown in FIG. 1. During the execution of the calibration process, the indicator 76 (for example, LED) shown in FIG. Then, the display 76 is turned off. The calibration start switch 74 and the display 76 may be installed in the lens device 10 or may be installed in a controller such as the focus demand 62. In addition, the calibration process is not executed by turning on the calibration start switch 74, but is executed when the power of the lens apparatus 10 is turned on or during initial adjustment before product shipment, and the set calibration data is stored in the memory 72. It may be.
[0060]
Next, an AF control processing procedure in the CPU 40 will be described. First, the flow of the entire processing in the CPU 40 will be described with reference to the flowchart of FIG. 5. After the CPU 40 performs the necessary initial settings (step S10), it is determined whether or not to start calibration of focus evaluation value sensitivity (generation of calibration data). Is determined (step S12). Whether or not the focus evaluation value sensitivity is calibrated is determined by whether or not the calibration start switch 74 is turned on as described above.
[0061]
If YES is determined in step S12, a focus evaluation value sensitivity calibration process described later is executed (step S14). If NO is determined, the calibration data already generated and stored in the memory 72 is stored. Read from the memory 72 (step S16).
[0062]
Next, the CPU 40 performs iris control based on the iris command signal given from the camera body 12 (step S18). Next, zoom control is performed based on the zoom command signal from the zoom demand 64 (step S20).
[0063]
Subsequently, the CPU 40 determines whether or not the AF switch 66 is ON (step S22). If YES is determined, the AF start flag is turned ON (step S24), and then the focus control process is performed. Execute (step S26). On the other hand, if NO is determined in step S22, the focus control process is executed without turning on the AF start flag (step S26). When the focus control process in step S26 ends, the process returns to step S18, and the processes from step S18 to step S26 are repeatedly executed.
[0064]
6 to 9 are flowcharts showing the procedure of the focus evaluation value sensitivity calibration process in step S14. When performing the focus evaluation value sensitivity calibration, it is desirable for the operator to set the photographing lens in a state of photographing a chart having a high contrast as a subject. First, in FIG. 6, the CPU 40 outputs a control signal to the iris motor drive circuit 52 to drive the iris motor 50 and completely closes the opening of the iris 20 (step S30). As a result, the focus state detection imaging elements 32A and 32B are shielded from light. Subsequently, the chA focus evaluation value AFV_A and the chB focus evaluation value AFV_B are acquired from the focus evaluation value generation unit 70 (steps S32 and S34). Subsequently, the count value of the focus evaluation value sampling counter is incremented by 1 (step S36), and it is determined based on the count value of the sampling counter whether or not the specified number of samplings have been completed (step S38). When it determines with NO, the process from said step S32 is repeated.
[0065]
On the other hand, if YES is determined in step S38, the maximum values AFV_A_MAX and AFV_B_MAX are detected from the focus evaluation values sampled for chA and chB, and the detected maximum values AFV_A_MAX and AFV_B_MAX are set to chA. The chB offset values (focus evaluation values of the black level) are AFV_A_OFFSET and AFV_B_OFFSET. That is,
[0066]
[Equation 3]
AFV_A_OFFSET = AFV_A_MAX
[0067]
[Expression 4]
AFV_B_OFFSET = AFV_B_MAX
(Step S40).
[0068]
Next, the CPU 40 outputs a control signal to the iris motor drive circuit 52 to drive the iris motor 50 and open the opening of the iris 20 (step S42). As a result, the subject light enters the focus state detection imaging elements 32A and 32B. Subsequently, the CPU 40 outputs a control signal to the zoom motor driving circuit 48 to drive the zoom motor 46, and moves the zoom lens 18 to an appropriate position (predetermined position) (step S44). Further, a control signal is output to the focus motor drive circuit 44 to drive the focus motor 42 and move the focus lens 16 to the closest end (step S46).
[0069]
Note that the position of the zoom lens 18 set in step S44 is a position suitable for calibration of the focus evaluation value sensitivity. For example, if it is set to the wide side, the focus lens 16 is moved at high speed in the following processing. On the other hand, if the telephoto side is set, there is an advantage that an accurate peak (maximum value) of the focus evaluation value can be detected in the following processing.
[0070]
Next, the CPU 40 acquires the focus evaluation value AFV_A of chA from the focus evaluation value generation unit 70, and sets the acquired focus evaluation value AFV_A to AFV_A_MIN (step S48). Further, as shown in the flowchart of FIG. 7, the CPU 40 acquires the focus evaluation value AFV_B of chB from the focus evaluation value generation unit 70, and sets the acquired focus evaluation value AFV_B to AFV_B_MIN (step S50). . In step S48 and step S50 and the following processing, the chA and chB focus evaluation values represented by AFV_A and AFV_B are the chA and chB set in step S40 from the focus evaluation values acquired from the focus evaluation value generation unit 70, respectively. The offset AFV_A_OFFSET and AFV_B_OFFSET may be subtracted.
[0071]
Next, the CPU 40 sets the moving speed F_SPEED_1ST of the focus lens 16 for detecting the subject (step S52). Then, the moving speed F_SPEED_1ST is output as a control signal to the focus motor drive circuit 44 via the D / A converter 54, and the focus lens 16 is moved in the infinity direction (step S54).
[0072]
While moving the focus lens 16 in the infinity direction in this way, the CPU 40 acquires the focus evaluation value AFV_A for chA and the focus evaluation value AFV_B for chB from the focus evaluation value generation unit 70 (steps S56 and S58). It is determined whether or not peaks (maximum values) of the chA focus evaluation value and the chB focus evaluation value have been detected (step S60). When it determines with NO, the process from said step S56 is repeated.
[0073]
On the other hand, if YES is determined in step S60, the CPU 40 subsequently acquires the focus evaluation value AFV_A for chA and the focus evaluation value AFV_B for chB from the focus evaluation value generation unit 70 (steps S62 and S64). For AFV_A_MIN and AFV_B_MIN set in step S50,
[0074]
[Equation 5]
AFV_A ≦ AFV_A_MIN
Or
[0075]
[Formula 6]
AFV_B_MIN ≦ AFV_B_MIN
It is determined whether or not holds (step S66). When it determines with NO, the process from said step S62 is repeated. On the other hand, if YES is determined, the focus lens 16 is stopped as shown in the flowchart of FIG. 8 (step S68).
[0076]
Next, the CPU 40 sets a moving speed F_SPEED_MIN (slower than the moving speed F_SPEED_1ST) of the focus lens 16 that can accurately detect the maximum focus evaluation value (step S70). Then, the movement speed F_SPEED_MIN is output as a control signal to the focus motor drive circuit 44 via the D / A converter 54, and the focus lens 16 is moved in the closest direction (step S72).
[0077]
While the focus lens 16 is thus moved in the closest direction, the CPU 40 acquires the chA focus evaluation value AFV_A and the chB focus evaluation value AFV_B from the focus evaluation value generation unit 70 (steps S74 and S76), and chA. The maximum values of the focus evaluation value AFV_A and the focus evaluation value AFV_B of chB are searched, and the maximum values are set to AFV_A_MAX and AFV_B_MAX (step S78). Then, it is determined whether a maximum value is found for both chA and chB (step S80). When it determines with NO, the process from step S74 is repeated. On the other hand, if YES is determined, the focus lens 16 is stopped as shown in the flowchart of FIG. 9 (step S82).
[0078]
Next, the CPU 40 determines that AFV_A_MAX and AFV_B_MAX set in step S78 are as follows:
[0079]
[Expression 7]
AFV_A_MAX> AFV_B_MAX
It is determined whether or not the condition is satisfied (step S84). If YES is determined, the calibration value AFG_B in the above formulas (1) and (2) is set to 1, and the calibration value AFG_A is set to AFV_B_MAX / AFV_A_MAX (step S86). On the other hand, if NO is determined, the calibration value AFG_A in the above formulas (1) and (2) is set to 1 and the calibration value AFG_B is set to AFV_A_MAX / AFV_B_MAX (step S88).
[0080]
Then, the calibration values AFG_A and AFG_B set in step S86 or step S88 are written in the memory 72 (nonvolatile memory) as the focus evaluation value calibration data (step S90). The offset AFV_A_OFFSET and AFV_B_OFFSET set in step S40 are also written in the memory 72 as calibration data. The focus evaluation value sensitivity calibration processing is completed by the above processing.
[0081]
FIG. 10 is a flowchart showing the focus control process in step S26 of FIG. When executing the focus control process, first, the CPU 40 obtains the focus evaluation value AFV_A of the focus state detection imaging element 32A (chA) from the focus evaluation value generation unit 70 (step S100) and at the same time, the focus state detection imaging. The focus evaluation value AFV_B of the element 32B (chB) is acquired (step S102).
[0082]
Here, the focus evaluation values AFV_A and AFV_B acquired in step S100 and step S102 indicate the focus evaluation values after correction, and the correction procedure is shown in FIG. First, the CPU 40 reads the chA and chB focus evaluation values (focus evaluation values before correction) from the focus evaluation value generation unit 70, and sets them to AFV_A0 and AFV_B0, respectively (step S120). Then, as indicated by the above formulas (1) and (2), the corrected focus evaluation values AFV_A and AFV_B using the calibration data AFV_A_OFFSET, AFV_B_OFFSET, AFG_A, and AFG_B read from the memory 72 in step S16 of FIG. The following formula,
[0083]
[Equation 8]
AFV_A = (AFV_A0−AFV_A_OFFSET) * AFG_A… ▲ 1 ▼
[0084]
[Equation 9]
AFV_B = (AFV_B0−AFV_B_OFFSET) * AFG_B… ▲ 2 ▼
Calculated by
[0085]
Returning to FIG. 10, after acquiring the focus evaluation values AFV_A and AFV_B for chA and chB calculated as described above, the CPU 40 next determines whether or not the AF start flag is set to ON. (Step S104). If NO is determined, MF processing is executed.
[0086]
In the case of the MF process, the CPU 40 acquires focus position data F_POSI indicating the current position of the focus lens 16 from the focus lens position detector 56 (step S106), and indicates the movement target position of the focus lens 16 from the focus demand 62. Focus demand data F_CTRL is acquired (step S108). Then, a difference F_POSI−F_CTRL between the acquired focus position data F_POSI and focus demand data F_CTRL is obtained, and the value is set as a movement speed F_SPEED for moving the focus lens 16 to the movement target position commanded by the focus demand 62. (Step S110). Then, the movement speed F_SPEED is output as a control signal to the focus motor drive circuit 44 via the D / A converter 54. (Step S114).
[0087]
On the other hand, if YES in step S104, that is, if it is determined that the AF start flag is ON, the CPU 40 executes AF processing (step S112).
[0088]
FIG. 12 is a flowchart showing the AF processing procedure in step S12. First, the CPU 40 determines whether or not the parameter F_MEMO_FLG is set to 1 (step S130). In the first process that shifts from the MF control to the AF control, it is determined as NO. In this case, the focus demand data indicating the current movement target position is acquired from the focus demand 62, and the data value is the initial (current) movement target. The position F_CTRL is set (step S132). Next, the parameter F_MEMO_FLG is set to 1 (step S134). If YES is determined in step S130, the processes in steps S132 and S134 are not performed.
[0089]
Next, the CPU 40 obtains the difference ΔAFV = AFV_A−AFV_B between the corrected chA focus evaluation value AFV_A and the chB focus evaluation value AFV_B acquired in step S100 and step S102 of FIG. 10 (step S136).
[0090]
Then, a value (movement amount) ΔAFV * AFG obtained by multiplying the ΔAFV by a predetermined AF gain AFG is added to the current movement target position AF_CTRL, and the value is set as a new movement target position AF_CTRL (step S138). That is, AF_CTRL = AF_CTRL + ΔAFV * AFG.
[0091]
Next, the CPU 40 reads focus position data F_POSI indicating the current position of the focus lens 16 from the focus lens position detector 56 (step S140), and the difference between the focus position data F_POSI and the movement target position AF_CTRL set in step S138. AF_CTRL-F_POSI is set as a moving speed F_SPEED for moving the focus lens 16 (step S142). Then, returning to the flowchart of FIG. 10, the movement speed F_SPEED is output as a control signal to the focus motor drive circuit 44 via the D / A converter 54 (step S114).
[0092]
With the above processing, the focus lens 16 moves to the in-focus position at a moving speed corresponding to the focal length and F value of the photographing lens.
[0093]
Next, instead of using the difference ΔAFV = AFV_A−AFV_B between the focus evaluation value AFV_A of chA and the focus evaluation value AFV_B of chB as in the AF process shown in FIG. An AF process when the ratio ΔAFV = AFV_A / AFV_B between the focus evaluation value AFV_A of chA and the focus evaluation value AFV_B of chB is used as an element for setting the movement amount as described above will be described with reference to the flowchart of FIG. Note that the processing from step S150 to step S154 in the flowchart in FIG. 13 is exactly the same as the processing from step S130 to step S134 in FIG. 12, and therefore will be described from the processing in step S156 in FIG.
[0094]
The CPU 40 performs the processing from step S150 to step S154, and then the ratio ΔAFV between the corrected chA focus evaluation value AFV_A and chB focus evaluation value AFV_B acquired in step S100 and step S102 of FIG. = AFV_A / AFV_B is obtained (step S156).
[0095]
Then, the CPU 40 determines whether or not the focus evaluation value ratio ΔAFV is greater than 1.0 (step S158). If YES is determined, ΔAFV = (ΔAFV−1.0) * AFG is set (step S160). On the other hand, if NO is determined, -ΔAFV = (1 / ΔAFV−1.0) * AFG is set (step S162). AFG represents a predetermined AF gain value. Then, the CPU 40 adds the obtained value (movement amount) ΔAFV to the current movement target position AF_CTRL, and sets the value as a new movement target position AF_CTRL (step S164). That is, AF_CTRL = AF_CTRL + ΔAFV.
[0096]
Next, the CPU 40 reads focus position data F_POSI indicating the current position of the focus lens 16 from the focus lens position detector 56 (step S166), and the difference between the focus position data F_POSI and the movement target position AF_CTRL set in step S164. AF_CTRL-F_POSI is set as a moving speed F_SPEED for moving the focus lens 16 (step S168). Then, returning to the flowchart of FIG. 10, the movement speed F_SPEED is output as a control signal to the focus motor drive circuit 44 via the D / A converter 54 (step S114).
[0097]
In the above embodiment, the case where two focus evaluation values are acquired from the two focus detection imaging elements 32A and 32B and AF control is performed has been described. However, the present invention is not limited thereto, and the optical path lengths are arranged at different positions. Even when AF control is performed based on three or more focus evaluation values obtained from three or more image pickup devices, the process of matching the focus evaluation value sensitivities can be performed in the same manner as in the above embodiment.
[0098]
In the above embodiment, in AF control, the movement target position of the focus lens 16 is set by the difference or ratio between the focus evaluation values of chA and chB, and the movement speed corresponding to the difference between the movement target position and the current position. However, the present invention is not limited to this, and the moving speed may be set directly based on the difference or ratio between the chA and chB focus evaluation values, and the focusing lens 16 may be moved at the moving speed. Good.
[0099]
Further, although cases have been described with the above embodiment as examples where the present invention is applied to a television camera system, the present invention is not limited thereto, and can also be applied to a video camera and a still camera for capturing a still image. .
[0100]
【The invention's effect】
As described above, according to the autofocus system of the present invention, since the calibration means for calibrating the sensitivity of the focus evaluation value obtained from each image sensor is provided, the sensitivity of the focus evaluation value is appropriately calibrated, and the accuracy is improved. High autofocus can be performed. In addition, it is possible to automatically calibrate the sensitivity of the focus evaluation value, so that troublesome work can be eliminated.
[Brief description of the drawings]
FIG. 1 is a block diagram of a television camera system to which an autofocus system according to the present invention is applied.
FIG. 2 is a diagram in which an optical axis of subject light incident on a video image sensor and an optical axis of subject light incident on a pair of focus state detection image sensors are displayed on the same line.
FIG. 3 is a block diagram illustrating a configuration of a focus evaluation value generation unit.
FIG. 4 is a diagram illustrating a state of a focus evaluation value with respect to a focus position when a certain subject is imaged, with a horizontal axis indicating a focus position of the photographing lens and a vertical axis indicating a focus evaluation value.
FIG. 5 is a flowchart showing the overall flow of processing in a CPU.
FIG. 6 is a flowchart illustrating a procedure of focus evaluation value sensitivity calibration processing in a CPU.
FIG. 7 is a flowchart illustrating a procedure of focus evaluation value sensitivity calibration processing in a CPU.
FIG. 8 is a flowchart illustrating a procedure of focus evaluation value sensitivity calibration processing in a CPU.
FIG. 9 is a flowchart illustrating a procedure of focus evaluation value sensitivity calibration processing in the CPU.
FIG. 10 is a flowchart showing a processing procedure of focus control in FIG. 5;
FIG. 11 is a flowchart illustrating a processing procedure for calculating a focus evaluation value in a CPU.
FIG. 12 is a flowchart showing a processing procedure of AF processing in FIG. 10;
FIG. 13 is a flowchart illustrating a processing procedure in another form of AF processing;
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10 ... Lens apparatus, 12 ... Camera body, 14 ... Imaging part, 16 ... Focus lens (group), 18 ... Zoom lens (group), 20 ... Iris, 22A ... Front side relay lens, 22B ... Rear side relay lens, 24 ... Half mirror, 26 ... relay lens, 28 ... focus state detection unit, 30A ... first prism, 30B ... second prism, 32A, 32B ... focus state detection image sensor, 40 ... CPU, 42 ... focus motor, 44 ... focus Motor drive circuit 56 ... focus lens position detector 62 ... focus demand 66 ... AF switch 70 ... focus evaluation value generator 72 ... memory 74 ... calibration start switch 76 ... display 80A, 80B ... high pass Filter (HPF), 82A, 82B ... A / D converter, 84A, 84B ... Gate circuit, 86A, 86B Adder

Claims (7)

The subject lens incident on the photographic lens is imaged by a plurality of imaging units arranged at positions having different optical path lengths, and the photographic lens is based on a focus evaluation value indicating the sharpness of each image captured by the plurality of imaging units. And a value obtained by multiplying the amount of change from a predetermined reference value of the focus evaluation value by a predetermined gain for each focus evaluation value obtained from each imaging means, In an autofocus system that is used as a focus evaluation value for controlling the focus of the photographing lens ,
A calibration unit that calibrates the sensitivity so that the sensitivity of the focus evaluation value obtained from each imaging unit matches by setting the reference value and the gain to appropriate values ;
The calibration means includes
Reference value setting means for setting a focus evaluation value obtained from each of the image pickup means as the reference value when an image without contrast is picked up;
Gain setting means for setting the gain so that the maximum value of the focus evaluation values obtained from the respective imaging means when the focus of the photographing lens is moved;
And an autofocus system that moves the focus while setting the zoom of the photographing lens to a predetermined position when the gain is set by the gain setting means . The subject lens incident on the photographic lens is imaged by a plurality of imaging units arranged at positions having different optical path lengths, and the photographic lens is based on a focus evaluation value indicating the sharpness of each image captured by the plurality of imaging units. And a value obtained by multiplying the amount of change from a predetermined reference value of the focus evaluation value by a predetermined gain for each focus evaluation value obtained from each imaging means, In an autofocus system that is used as a focus evaluation value for controlling the focus of the photographing lens ,
Comprising a calibration means for calibrating the sensitivity as the sensitivity of the focus evaluation values obtained from the respective imaging means coincides by setting the reference value and the gain to an appropriate value,
The calibration means includes
Reference value setting means for setting a focus evaluation value obtained from each of the image pickup means as the reference value when an image without contrast is picked up;
Gain setting means for setting the gain so that the maximum focus evaluation values obtained from the respective imaging means coincide with each other when the focus of the taking lens is moved ; A gain for accurately detecting the maximum value of the focus evaluation value obtained from each image sensor by moving the focus at a low speed after confirming the existence of the maximum value of the focus evaluation value obtained from each image sensor by moving Setting means;
Features and to Luo over autofocus system further comprising a.   3. The autofocus system according to claim 1, wherein the calibration unit automatically executes calibration of the sensitivity when a predetermined switch is turned on, when the power is turned on, or when initial setting is performed before shipment. . The calibration means according to claim 1 or 2 autofocus system, characterized in that to the memory to store the reference value and the gain set by the reference value setting means and said gain setting means as the calibration data. It said reference value setting means, by closing the iris of the photographing lens, according to claim 1 or 2 autofocus system characterized by capturing an image with no the contrast by the respective imaging element. The gain setting means moves the focus of the photographing lens at a high speed and confirms the presence of the maximum value of the focus evaluation value obtained from each of the image sensors, and then moves the focus at a low speed to move the focus from each of the image sensors. 2. The autofocus system according to claim 1 , wherein a maximum value of the obtained focus evaluation values is detected with high accuracy.   3. The autofocus system according to claim 1, further comprising display means for displaying that the calibration of the sensitivity by the calibration means is being executed.

Images