JPH0727108B2 - Automatic focus adjustment device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic focus adjustment device that detects the focus state of an image pickup lens by receiving subject light that has passed through the image pickup lens of a camera to perform focus adjustment.

Prior art and its problems First of photographic lens which is symmetrical with respect to the optical axis
The object light fluxes that have passed through the first and second areas are re-imaged to form two images, the mutual positional relationship between these two images is determined, and the amount of deviation of the imaging position from the planned focus position and its Direction (whether the image formation position is in front of the planned focus position,
A focus detecting device for obtaining the rear side, that is, the front focus or the rear focus has already been proposed. The optical system of such a focus detection device has a structure as shown in FIG. 7, and this optical system has a planned focal plane (4) behind the taking lens (2) or a position further rearward from this plane. Has a condenser lens (6), and further has re-imaging lenses (8) and (10) behind them, and each re-imaging lens (8), (10)
Line sensors (12) and (14) each having a CCD as a light receiving element are disposed on the image forming plane of. As shown in FIG. 8, the images on the line sensors (12) and (14) have an optical axis (in the case of a so-called front focus) in which an image of an object to be focused is formed in front of the planned focal plane. In the case of rear pins, on the contrary, they are far from the optical axis (18). When the two images are in focus, the distance between two corresponding points of the two images becomes a specific distance defined by the configuration of the optical system of the focus detection device. Therefore,
In principle, the focus state can be determined by detecting the distance between two corresponding points of two images. FIG. 7 shows that the above interval is detected by the correlator 16 upon receiving the signals from the line sensors (12) and (14).

By the way, even in a camera equipped with the focus detection device as described above, it is necessary to perform focus adjustment by measuring the distance to a moving object at a high speed like a sports photograph as well as a stationary object. is there. Therefore, the focus detection device needs to have a distance measurement time as short as possible. On the other hand, there is a strict demand for distance measurement accuracy required for single-lens reflex cameras, and the focus detection device must be able to perform highly accurate distance measurement in order to meet the demands of such single-lens reflex cameras.

However, when the focus detection device repeats the focus detection operation a plurality of times for the same subject at a certain distance, the focus detection device changes the position of the subject in a plane perpendicular to the distance detection direction, or when the camera is held by camera shake due to camera shake. If there is a slight vibration, the detection results do not always match, and they are distributed with a slight variation around a certain value. That is, there is a drawback that accurate focus detection information cannot be obtained by only one detection operation.

This variation is caused by the nonuniformity of the characteristics and arrangement of each element of the focus detection element array consisting of many light receiving elements, the instability of the processing circuit, and the illuminance distribution on the detection element surface (corresponding to the subject brightness distribution). The spatial frequency characteristics of the focus detection elements are determined by the arrangement pitch of the detection element groups arranged in rows, and correct measurement is not possible for frequency components higher than the spatial frequency determined by the Nyquist sampling theorem. This is because the luminance distribution of the subject is discontinuously measured due to the inability to perform it, or the dead zone existing between the elements, and even if there is a luminance change in the dead zone, it cannot be detected.

Therefore, if the position of the subject changes in the plane perpendicular to the distance detection direction, the pattern of the subject image projected on the focus detection element surface changes, and the sampling position of the subject image in the focus detection element row is relatively changed. Also changes, the detection results may not match due to the above factors of the detection element,
When the same measurement is repeated many times, the detection results will be distributed with some variation around a certain value. Therefore, even if the focus of the projection lens is adjusted based on the defocus amount obtained by one detection operation, the adjustment accuracy is not guaranteed. Therefore, a device adapted to obtain data of a plurality of defocus amounts by detecting the in-focus state a plurality of times and obtain an average value thereof, and perform focus adjustment of the photographing lens based on the average value. Is proposed in Japanese Patent Laid-Open No. 56-78811. However, in this apparatus, the photographing lens is stopped, the detection operation is repeated, and the driving of the photographing lens is started after the average value is obtained, so that the focus cannot be adjusted quickly.

By the way, when considering the focus detection process, the photographing lens is usually out of focus in the initial state. In such a state, in order to shorten the time required for focus detection, it is prioritized that the approximate value of the lens extension amount or the lens extension amount is calculated and the photographic lens is moved immediately, rather than the precise calculation. Therefore, when the defocus amount is large, high sampling accuracy is not necessary for image shift detection.

As a device for shortening the focus adjustment time based on such an idea, a device has been proposed in which the sampling width (number) and / or the filter characteristic of the image is switched according to the defocus amount (for example, , JP-A-59-
107312, JP-A-59-142506 and JP-A-60
-4913 gazette). The reason why the filter characteristic is switched according to the defocus amount as described above is as follows. That is, projecting a subject image on the focusing element array and computing the photoelectric output thereof to detect the focusing position means computing the frequency component relating to a certain spatial frequency band in the subject image. When the spatial frequency components to be calculated are only relatively high-order ones, they are not easily affected by electrical and optical error factors, and the defocus amount can be accurately obtained near the in-focus position. it can.
However, when the defocus amount becomes large, the high-order spatial frequency components in the subject are drastically reduced, and the focus detection accuracy is significantly reduced. Therefore, the filters may be switched so that the high-order spatial frequency information is handled in the vicinity of the focus position and the low-order spatial frequency information is handled in the position away from the focus position.

By the way, in the case of handling the high-order spatial frequency information and the low-order spatial frequency information as described above, the photographing lens has to rotate twice before reaching the in-focus position with high probability, and the movement of the photographing lens is smooth. In addition, the time required for focusing cannot be shortened too much.

It is an object of the present invention to provide an automatic focus adjustment device in which the movement of a taking lens is smooth and it takes a short time to reach a focused state.

Means for Solving the Problems To achieve the above object, the present invention provides an optical system that divides an exit pupil of an objective lens into two parts, and a pair of photoelectric conversion element arrays that receive light images from the respective exit pupils. And a first focus detection operation that operates in response to the start of the focus adjustment operation and that detects the focus adjustment state of the objective lens based on the higher spatial frequency component in a relatively narrow region based on the output of the photoelectric conversion element array. A focus detecting means, a determining means for determining whether or not the first focus detecting means is in a state in which a reliable focus adjustment state is detected, and an operation when the determining means determines that there is no reliability, Second focus detection means for detecting the focus adjustment state of the objective lens based on the low spatiotemporal frequency component in a relatively wide area based on the output of the photoelectric conversion element array, and the determination means are determined to be reliable. If you do The objective lens is driven on the basis of the output of the first focus detecting means, and the objective lens is driven on the basis of the output of the second focus detecting means when it is determined that there is no reliability. And

Operation According to the present invention, after the focus adjustment operation is started, the focus detection by the high frequency component is first performed in the narrow region by the above configuration. If the determination means determines that the focus detection result is reliable,
The means for driving the objective lens drives the objective lens based on the output of the first focus detection means. Then, as a result of performing focus detection by the high frequency component in the narrow area, when the determination means determines that the focus detection result is not reliable, the second focus detection means operates for the first time, and the focus by the low frequency component in the wide area. The detection is performed, and the means for driving the objective lens drives the lens based on the detection result.

Embodiments Embodiments of the present invention will be described below with reference to the accompanying drawings.

A block diagram of a focus detection device according to the present invention is shown in FIG.

In FIG. 1, the control circuit (31) including a microcomputer starts the focus detection operation by pressing the shutter release button (not shown) one step when the focus detection mode switch AFSW is on.

First, the CCD as the first and second photoelectric conversion element arrays provided in the photoelectric conversion circuit (30) from the control circuit (31).
The pulse-shaped integral clear signal ICGS is output to the image sensor (see FIG. 2), which causes the photoelectric conversion circuit (30)
Each pixel of the CCD image sensor is reset to the initial state, and the output AGCOS of the brightness monitor circuit (not shown) built in the CCD image sensor is set to the power supply voltage level. At the same time, the control circuit (31) is
The gh "level shift pulse generation enable signal SHEN is output. Then, at the same time as the integration clear signal ICGS disappears, the integration of the photocurrent is started in each pixel in the CCD image sensor of the photoelectric conversion circuit (30), and at the same time the photoelectric conversion is started. The output AGCOS of the brightness monitor circuit of the conversion circuit (30) begins to drop at a speed according to the subject brightness, but the reference signal output DOS from the reference signal generation circuit built into the photoelectric conversion circuit (30) is at a constant reference level. The gain control circuit (25) compares AGCOS with DOS, and within a predetermined time (100msec.
The gain of the variable gain operational amplifier (22) is controlled depending on how much S decreases with respect to DOS. In addition, the gain control circuit (25) is within a predetermined time after the integration clear signal ICGS disappears.
When it is detected that AGCOS has dropped below a predetermined level with respect to DOS, at that time, a "High" level TINT signal is output.
The TINT signal is input to the shift pulse generating circuit (26) through the AND circuit (AN) and the OR circuit (OR), and in response thereto, the shift pulse SH is output from the circuit (26). When this shift pulse SH is input to the CCD image sensor of the photoelectric conversion circuit (31), the photocurrent integration by each pixel ends, and the charge corresponding to this integration value is paralleled from the CCD image sensor to the corresponding cell of the shift register. Be transferred. On the other hand, based on the clock pulse CL from the control circuit (31), the phase is 18 from the transfer pulse generation circuit (27).
Two sensor drive pulses φ1 and φ2 which are deviated by 0 ° are output and input to the photoelectric conversion circuit (30). The CCD image sensor of the photoelectric conversion circuit (30) discharges the charge of each pixel of the CCD shift register one by one in series in synchronism with the rising of φ1 of these sensor drive pulses φ1 and φ2. The OS signals forming the are sequentially output. this
The OS signal has a higher voltage as the incident intensity on the corresponding pixel is lower, and the subtraction circuit (24) uses this as the reference signal DO.
Subtract from S and output (DOS-OS) as a pixel signal. If the TINT signal is not output after the integration clear signal ICG disappears and a predetermined time elapses, the control circuit (31) outputs “Hig
The h "level shift pulse generation command signal SHM is output.
Therefore, even if the predetermined time has elapsed after the integration clear signal ICG disappeared, the AGC controller (25) outputs the "High" level TINT.
When no signal is output, the shift pulse generation circuit (26) generates the shift pulse SH in response to the shift pulse generation command signal SHM.

On the other hand, in the above operation, when the control circuit (31) outputs pixel signals corresponding to the seventh to tenth pixels of the CCD image sensor of the photoelectric conversion circuit (30),
Outputs sample hold signal S / H. This part of the CCD image sensor is covered with an aluminum mask for the purpose of removing the dark output component, and the light receiving pixel of the CCD image sensor is in a light-shielded state. On the other hand, the sample hold signal causes the peak hold circuit (21) to hold the difference between the output OS and DOS corresponding to the aluminum mask part of the CCD image sensor of the photoelectric conversion circuit (30). Is input to the variable gain amplifier (22). The variable gain amplifier (22) amplifies the difference between the pixel signal and its difference output with a gain controlled by the AGC controller (25), and the amplified output is A / D converted by the A / D converter (23).
After being converted, it is taken in by the control circuit (31) as pixel signal data.

The A / D conversion of the A / D conversion circuit (23) is performed with 8 bits, but the upper and lower 4 bits are transferred to the control circuit (31).

After that, the control circuit (31) sequentially stores the pixel signal data in the internal memory. When the storage of the data corresponding to all the pixels of the image sensor is completed, the data is given to the operation determination circuit (32). . The calculation discriminating circuit (32) calculates the defocus amount and its direction according to a predetermined program and causes the display circuit (33) to display them, while the lens drive device (28) is used to defocus amount and its direction. Drive the lens to automatically focus the taking lens (29).

FIG. 2 is a front view of the sensor according to the embodiment of the present invention. In this embodiment, the two line sensors, namely the first line sensor
And the second photoelectric conversion element array is substituted by using two different regions of one line sensor.

In FIG. 2, (X) indicates the position where the optical axis of the taking lens passes. The output of the pixel near the optical axis passage position (X) is not used. (L ₁ ) to (l ₃₂ ) represent pixels in the reference portion (L) corresponding to one line sensor, and the reference portion (L) is the first block (I 1) of the pixels (l ₁ ) to (l ₃₂ ). ), Pixel (l ₅ )
It is divided into two blocks of the second block (II) of (l ₂₈ ). A monitor light receiving element for monitoring the illuminance on the pixel is provided above the reference portion (L) in the drawing.

(R ₁ ) to (r ₄₀ ) indicate pixels in the reference portion (R) corresponding to the other line sensor. The number of pixels in the reference part (R) is 40
The number is larger than the number of pixels (32) in the reference portion (L). Then, when the taking lens is in focus on the planned focal plane, the images on the pixels (l ₁ ) to (l ₃₂ ) of the first block (I) of the reference portion (L) are the images of the reference portion (R). Pixel (r ₅ )
~ (R ₃₈ ) Suppose it matches the image above. The images (r ₅ ) to (r ₃₆ ) are used as the focusing block (F) in the reference portion (R), and the boundary between the pixels (l ₁₆ ) and (l ₁₇ ) in the center of the reference portion (L) is referred to. Focus block (F) of section (R)
Let L ₂ be the distance between the boundary between the pixel (r ₂₀ ) and the pixel (r ₂₁ ) at the center of, that is, the image interval at the time of focusing detection.

FIG. 3 shows the control circuit (31) and the operation discrimination circuit (3
3 is a flowchart of an embodiment of the present invention showing a flow of operation when (2) is set as one microcomputer.

When a power switch (not shown) is turned on, power is supplied to the camera. Then, the flow of Fig. 3 starts and # 1
Waiting for the AF switch to be turned on in the AF switch determination step in step # 3, when the AF switch is turned on, the control circuit (31) causes the CCD to accumulate charge in step # 2, and when this is complete, # In the Data Dump step of 3, the CCD output is sequentially output from the CCD as a video signal (OS). This video signal (OS) is subtracted by the subtraction circuit (24) and becomes a pixel signal. This pixel signal is amplified by a gain according to the brightness of the subject, and then A / D conversion circuit (23) / D converted to digital value. The control circuit (31) receives this data and the gain data. Next, every third difference data is created from the step CCD pixel signals of # 4 and # 5. This purpose is a process for removing a low frequency error factor as a spatial frequency of the illuminance distribution on the standard portion and the reference portion caused by a deviation from the design value of the distance measuring optical system. −
Since it is described in Japanese Patent No. 4914, its description is omitted. In steps # 6 and # 7, the correlation calculation is performed using the second block (II) to indicate the position of the reference area having the highest degree of correlation LM ₂
To calculate. In step # 8, it is determined whether or not detection is possible using the second block (II). If it is determined that detection is possible, after step interpolation calculation in steps # 9 and # 10, a highly accurate image interval shift is performed. The quantity P is calculated. Regarding the correlation calculation, the interpolation calculation, and the undetectability determination, the applicant of the present application filed
59-100069, Japanese Patent Application No. 59-104212 and Japanese Patent Application No. 59-1042.
No. 13, which is described in detail. Then # 11,
In step # 12, the image interval shift amount P is the defocus amount D.
F, further converted to the lens drive amount. In step # 13, it is determined whether or not the obtained defocus amount (DF) is within the focusing range. If it is within the range, focusing display is performed in step # 15. If it is out of the range, the lens is driven in step # 14, and the process returns to the CCD integration step # 2. If it is determined in step # 8 that the detection by the second block (II) is impossible, then the correlation calculation using the entire reference portion is started. In steps # 16 and # 17, summation data lswk and rswk are created by the difference data of the pixels of the standard part and the reference part. This processing is a measure for improving the ranging accuracy and the ranging probability by increasing the contrast between pixels used for the correlation calculation even in a low-frequency subject. Next, in steps # 18 and # 19, the correlation calculation using the entire standard portion is performed, and lM ₁ indicating the position of the area of the reference portion having the highest degree of correlation is calculated. In step # 20, it is determined whether or not detection by the entire reference portion is possible. If possible, in steps # 21 and # 22, after the interpolation calculation, a highly accurate image interval deviation amount P is obtained. It is calculated and the process goes to step # 11. When it is determined that the LO-CON SCAN has not been completed, it is determined that the LO-CON SCAN has already been completed in step # 23. LO-CON SCAN was conceived as a countermeasure when the amount of focus shift is too large to make distance measurement possible.It is required when the distance is measured while moving the lens and the amount of focus shift is within the range This is a scan in which the amount of extension of the lens is changed in order to control the lens to the in-focus state according to the distance measurement value, that is, the amount of image distance deviation. If the LO-CON SCAN is completed, the # 25 step LO-CON display is performed and the # 2 C
Return to CD integration step. LO-CON SCA in step # 23
N If not finished, LO-CON SCAN in step # 24.
Starts and returns to the CCD integration step of # 2 again.

FIG. 4 is a flowchart showing a part of another embodiment of the present invention. In this embodiment, when the second block (II) cannot be detected, the contrast sum value C ₂ is calculated,
If the value is less than the fixed value A, it is determined as a low-frequency object, and the correlation calculation is performed by the entire reference part with a wide area and pixel pitch. However, if the value is more than the fixed value A, the amount of out-of-focus is too large to measure the distance. Therefore, in order to reduce the time, the correlation calculation is not performed by the whole reference part, and it is put in the LO-CON SCAN.

In FIG. 4, in step # 8, the second block (II)
If it is determined that it is undetectable in step # 26, the total value of the contrast based on the difference data of the second block (II) in step # 26.
C ₂ is calculated. Next, in step # 27, C ₂ is a constant value A
It is determined whether or not the above. If it is less than A, the correlation calculation by the whole reference part after step # 16 is started. If it is A or above, it is judged to be out of focus and LO of # 23
-Enter the step for determining the end of CON SCAN.

5 and 6 are flow charts showing a part of still another embodiment of the present invention.

In this embodiment, the reference part is divided into three blocks and the correlation calculation is performed for each block, and the correlation calculation is performed by the entire reference part only when all blocks cannot be detected. Furthermore, only when the correlation calculation of the entire reference part becomes undetectable, it is judged as undetectable as a whole, and the LO-CON SCA
It is the one entering N.

That is, in FIG. 5, # 6, # 7, # 8, # 28,
In steps # 29, # 30, # 33, # 36, # 37, # 44 and # 45, the correlation calculation is performed for the three blocks of the reference part, and only when all the blocks become undetectable, the reference is set. Correlation calculation is performed by the entire section. That is, steps # 16 to # 20 and # 23 of FIG. 6 are executed. Then, only when the correlation calculation by the entire reference part becomes undetectable, it is determined that the correlation cannot be detected as a whole, and the LO-
CON SCAN is performed.

EFFECTS OF THE INVENTION According to the present invention, focus detection is first performed by a high frequency component in a narrow region, and focus detection by a low frequency component is performed in a wide region only after it is determined that the detection result is unreliable. In the case of a subject having high focus detection reliability such as a subject having high brightness and high contrast, a highly accurate focus detection operation is immediately performed by a high frequency component, and an accurate focus adjustment operation is performed in a short time.

[Brief description of drawings]

FIG. 1 is a block diagram of an embodiment of an automatic focusing apparatus according to the present invention, FIG. 2 is a front view of a CCD image sensor used in the automatic focusing apparatus of FIG. 1, and FIG. 3 is FIG. FIG. 4 is a flowchart showing a flow of the operation of the automatic focusing apparatus of FIG. 4, FIG. 4 is a flowchart of a modified portion of the flowchart of FIG. 3, and FIGS. 5 and 6 are flowcharts showing modified examples of the flowchart of FIG. FIG. 8 and FIG. 8 are schematic configuration diagrams of the optical system of the focus detection device according to the present invention. 29 …… Shooting lens, 30 …… Photoelectric conversion circuit, 31 …… Control circuit, 32 …… Calculation discrimination circuit.

Claims (1)

[Claims] 1. An optical system that divides an exit pupil of an objective lens into two, a pair of photoelectric conversion element arrays that receive light images from the respective exit pupils, and an optical system that operates in response to the start of a focus adjustment operation. The first focus detecting means for detecting the focus adjustment state of the objective lens based on the higher spatial frequency component in a relatively narrow area based on the output of the photoelectric conversion element array, and the first focus detecting means are reliable. Discriminating means for discriminating whether or not a proper focus adjustment state is detected, and it operates when the discriminating means judges that there is no reliability, and a relatively wide area based on the output of the photoelectric conversion element array. Second focus detection means for detecting the focus adjustment state of the objective lens based on the low-order spatial frequency component, and based on the output of the first focus detection means when the determination means determines that there is reliability. The objective lens Dynamic and, when it is determined that no reliable automatic focusing apparatus characterized by comprising, a means for driving the objective lens based on the output of the second focus detection unit.