JPH01185506A - Focus detecting device for camera - Google Patents

Abstract

PURPOSE:To smoothly detect a focus without being influenced by a fluctuation, etc., of an output signal of an image sensor by changing a focusing range in accordance with contrast and brightness of an object to be photographed, a moving state of a zoom lens and its lens position, history of the defocus quantity or a focus detection mode. CONSTITUTION:An object luminous flux which has transmitted through a photographic lens is photodetected by an image sensor 100, and based on an output signal from this image sensor 100, the defocus quantity of the photographic lens is operated by a defocus quantity arithmetic means 200. In this case, if this defocus quantity is within a focusing range, it is decided to be focused by a focusing deciding means 300, and also, by a focusing range varying means 400, the focusing range is varied in accordance with contrast or brightness of an object to be photographed, a moving state of a zoom lens or its lens position, history of the defocus quantity or a focus detection mode. In such a way, a focus can be detected smoothly without being influenced by a fluctuation, etc., of the output signal of the image sensor 100.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an automatic focus detection device for a video camera or a camera.

(Prior Art) In recent video cameras and cameras, autofocus (AF=automatic focus adjustment), in which a photographic lens is automatically moved to a focusing position according to the subject distance, has been widely adopted. There are various methods for autofocus, part 1
As a first step, an image sensor receives the subject light beam that has passed through the photographic lens or a dedicated external lens linked to the photographic lens, and calculates the amount of defocus of the photographic lens based on the output signal of the image sensor. A phase difference method is known.

In the phase-difference method, two light beams from the subject that have passed through different areas of the photographic lens are received by different light-receiving elements, and when the focus is on, the output of the light-receiving element that received one light beam and the other light beam are different. The outputs of the light-receiving elements that received the light coincide with each other, but when the light is out of focus, a phase shift occurs in the outputs of the respective light-receiving elements. The amount of phase shift can be detected and the amount of defocus of the photographic lens can be determined based on the amount of shift using a predetermined arithmetic expression. Furthermore, since the upper and lower rays of the photographic lens are switched before and after the focal point, this property can be used to detect the direction of defocus (front focus or rear focus).

Now, there are several known phase difference methods, and one of them, shown in Figure 6, is the T
In figure 0 showing the CL (Through Camera Lens) method, l is a photographing lens, 2 is an infrared cut filter, 3 is a correction lens, 4 is a light shielding plate, 5 is a microlens array, and 6 is behind the microlens array 5 ChargeCoupled
The exit pupil of the photographing lens l and the CCD
It has a conjugate relationship with the light-receiving surface of CCD 6, and the upper half of the exit pupil is viewed by the A-row light-receiving elements of the CCD 6, and the lower half is viewed by the B-row light-receiving elements. The output signals from the row light receiving elements and the output signals from the B row light receiving elements match, but when they are out of focus, a shift occurs between the two output signals as shown in FIG. This amount of deviation S is determined by a calculation called HF logic shown below using a microcomputer (Japanese Patent Laid-Open No. 5
7-45510).

is+1*j2ljl where a,,b 1 is the i-th A of CCD6, respectively
These are the outputs of the row light receiving elements and the B row light receiving elements.

Addition value vJ of equation (1) is calculated as 8th for j=-12 to 12.
Plot as shown in the figure and calculate V k > O, V k++ < O
Find the difference D0 between Vk and ■, or □, or find the maximum k, and interpolate between Vk and Vk, 1 using the following formula to find the amount of deviation S.

The difference ■@-Vk*x at this time has a correlation with the contrast of the detected image, and is also used for contrast discrimination during AF heat processing.

The amount of deviation S obtained as above is converted into the amount of deviation (defocus amount) between the image forming position of the photographing lens and the imaging element surface of the video camera, and this amount of defocus is determined to be within a predetermined focusing range. If the amount of defocus is outside the 0 focusing range, the focus motor moves the photographing lens l until the detected amount of defocus falls within the focusing range.

By the way, the focusing range is generally determined by the depth of focus ε of the photographing system. , the focus shift detection accuracy es of the focus detection element, and the focus shift accuracy @1 of the drive control precision of the focusing lens 9. It is desirable that the focal depth εD=F−d determined from the value and the minimum confusion circle d of the image sensor be approximately equal to D>ε. In this case, the larger the difference, the more noticeable the image blurring will be on the image sensor due to the amount of residual defocus during automatic focus control. In addition, @5 is preferably ε,<D, and if ε5≧D, even though there is no change in the subject distance due to fluctuations in the output signal of the focus detection element,
The detected value of the defocus amount frequently deviates outside the in-focus range, and the focusing lens is driven each time to correct the pseudo-focal shift, resulting in a very unstable system. Furthermore, it is desirable that the control accuracy of the focusing lens, ε, is ε,D.If q,≧D, it will be difficult to accurately stop the focusing lens within the focusing range, and as mentioned above, If 0 or more results in an unstable system, the following relationship becomes a desirable condition for achieving a good autofocus system.

rLs <D, e, <D, eD#D Now, in the conventional focus detection device, since this focusing range is fixed, the following problem occurs.

(1) When the detected amount of defocus is on the boundary between in-focus and out-of-focus, if the output signal of the image sensor fluctuates even slightly, it may become in-focus or out-of-focus, causing the in-focus lens to This causes unnatural movements such as hunting, and when manually operating the focusing lens, the in-focus and out-of-focus indications repeat, resulting in an unstable state.

(2) Depth of focus ε. As described above, is expressed by εD=F−d, and changes depending on the brightness of the subject (aperture opening). When the focusing range is fixed, D is determined based on the F value of the lens, so when the subject is bright, focus detection is performed with a precision that is actually unnecessary. In addition, if εS<a is not guaranteed at that time, that is, if there is fluctuation in the output signal of the image sensor, ε,
>D, and focus detection cannot be performed smoothly.

(3) It reacts sensitively to external noise and objects momentarily passing in front of the lens, and when using the focusing lens or manual operation, the focus display means such as an LED may operate unnecessarily.

(4) In the case of a subject with low contrast, the reliability of the image sensor's output signal itself is low, so it is easily affected by fluctuations in the output signal, and the probability of ε,>D increases, making focus detection difficult. It disappears.

(Objective and Structure of the Invention) The present invention has been made in view of the above points, and an object of the present invention is to enable smooth focus detection without being affected by fluctuations in the output signal of an image sensor. do.

In order to achieve this object, in the present invention, as shown in the overall configuration in FIG. The defocus amount of the photographing lens is calculated by the defocus amount calculation means 200, and if this defocus amount is within the focusing range, the focus determination means 300 determines that the focus is in focus, and the focusing range variable means 400 determines that the focus is in focus. The focusing range is configured to be changed according to the contrast or brightness of the subject, the moving state of the zoom lens or its lens position, the history of the amount of defocus, or the focus detection mode.

(Example) The following describes an example in which the present invention is applied to a TCL system.

FIG. 2 is a block diagram of a video camera equipped with an automatic focus adjustment device according to the present invention. In the figure, 7 is a focusing lens for focusing, and 8 is a focusing lens for moving the focusing lens 7 in the optical axis direction. A focus motor, 9 is a drive circuit that drives the focus motor 8, 10 is a 6x zoom lens that changes the focal length f of the focusing lens 7 in the range of 12 to 72 mm, and 11 is a zoom ring (not shown) that is rotated. A zoom motor that moves the zoom lens 10, 12 a drive circuit for the zoom motor 11, 13 a position sensor that detects the position of the zoom lens 10, and 14 an image sensor 15 that captures the subject image.
This is the master lens that forms the image. The subject light flux is guided to the focus detection optical system by a half mirror 16, and is then guided to the CCD 1 by an AF lens 17 and a microlens 18.
9 is imaged. Furthermore, 20 is a CCD driver that drives the CCD 19, 21 is an A/D converter that converts the analog output from CCD 19 into a digital value, and 22 is a CO
It is an interface between the D driver 20 and the A/D converter 21 and the microcomputer 23.

The microcomputer 23 uses the HF logic mentioned above to
is determined, and the defocus amount is calculated from this shift amount S.

By comparing this defocus spear with a preset focusing range, it is determined whether the object is in front focus, in focus, or in back focus. When the front focus is on, a signal is output from the microcomputer 23 to the drive circuit 9, and the focus motor 8 is driven for a time corresponding to the previously calculated defocus amount, so that the focusing lens 7 is moved in the focusing direction. Conversely, when the rear focus is on, a signal is output from the microcomputer 23 to the drive circuit 9, and the focusing lens 7 is moved in the extending direction.

24 is a mode changeover switch for changing the focus detection mode; when the mode changeover switch 24 is off, the focusing lens 7 is driven by the focus motor 8 in auto mode, and when it is on, the focusing lens 7 is manually operated by the user. This is a manual mode in which movement is performed by This mode is switched according to the user's selection, and in the manual mode, a focus display unit 25 provided in the viewfinder displays one of front focus, in-focus, and back focus.

FIG. 3 and FIG. 4 are diagrams for explaining how the focusing range changes in the example, in which the horizontal axis is the focusing range D, and the vertical axis is the focal length f of the focusing lens 17. In this embodiment, the focusing range is changed as follows.

(1) The focusing range is changed according to the history of the defocus amount.

In order to bring the focusing lens 7 as close as possible to the best focus position, the focusing range is set to be narrow to D0=εo/2. If it comes into focus twice in a row within this narrow focusing range,
It is determined that the current position is near the best focus, and the focusing range is expanded five times (D=5DO') as shown in FIG. By setting a wide focusing range in this way, the system becomes extremely resistant to fluctuations in the image sensor output. However, if this continues, it will not respond to slight changes in the subject distance, so double the focusing range (D = 2D,) at intervals of a few seconds (for example, 3 seconds) so that the reaction speed does not slow down too much. Perform focus judgment, and if it becomes out of focus again, increase the focus range to 1x (D=DQ)
Make it.

(2) The focusing range is changed depending on the zoom position of the zoom lens and the brightness of the subject.

When the zoom position is on the wide side (f = 12 to 25 mm), the amount of focus shift corresponding to close subject distance is small, so if you increase the focus range magnification too much, the response will appear to be poor. It ends up.

Therefore, in (1), when the focus range is uniformly expanded 5 times during continuous focusing, when the zoom position is on the wide side, the depth of field is also taken into consideration, and the accumulation time of the CCD 19 is 101sec.
c or more, that is, when the subject is dark, the focusing range is tripled (D = 3D,), and when the accumulation time is less than 10m5ec, that is, when the subject is bright, the focusing range is increased five times (D = 5D,
).

By the way, when the zoom lens 10 is moved, the CCD
The output signal of No. 19 tends to fluctuate, and in practice, the focus may shift in a strange way. Therefore, in order to make focus detection smoother,
While operating the zoom lens lO, the focusing range is increased by 5 times (D=5D
Expand to o).

(4) Changing the focusing range according to the focus detection mode.

In other words, in the manual mode in which the focusing lens 7 is moved manually by the user, the focusing range is first set to DQ=ε8/2 as shown in FIG. When the focus is on, set the focus range to D”
3D.

expand to. - Once out of focus, D =
Do. As a result, even in the manual mode, the in-focus display will not change unstablely due to fluctuations in the output signal of the CCD 19.

Now, in the example, the focusing range was changed in the manners (1) to (4) above, but any one of (1) to (4)
The focusing range may be changed in one or a combination of the two.

Next, the operation of the embodiment will be explained using the flowchart shown in FIG.

First, the CCD try 8 2 is activated by a command from the microcomputer 23.
0 is driven, charge accumulation in the CCD 19 starts (F-1), and the output signal from the CCD 19 is converted into a digital value by the A/D converter 21 and input to the microcomputer 23 (F-2). ), the microcomputer 23 calculates the amount of deviation S using the above-mentioned HF logic, and calculates the amount of defocus from this amount of deviation S (F-3). The defocus amount calculated for focus detection is stored in the memory within the microcomputer 23. Therefore, the memory that stores the current defocus amount is rewritten by erasing the previously stored information (F-4).

Next, judge the contrast of the subject from the value of VK VK +1 (see page 4 of the specification) mentioned above.F-5)
, if the contrast is low contrast below the reference value, the process proceeds to step (F-28) and the focus motor 8 is stopped. Next, although it is not low contrast, it is necessary to determine whether the value is close to low contrast (F-6).
If the contrast is near low contrast, the process advances to step (F-26) and the focusing range is set to 5 times (D=5D).

This makes it possible to perform smooth focus detection without any fluctuation in focusing even in the vicinity of low contrast.

Next, determine whether the focus detection mode is manual mode or auto based on the state of the mode changeover switch 24 (F-7),
If it is manual mode, proceed to step (F-12), if it is auto mode, determine whether there is a change in the zoom position of the zoom lens 10 (F-8), and if there is a change in the zoom position, focus It is determined whether the range has already been expanded or not based on the expansion flag in the microcomputer 23 (F-9). If it is enlarged, step (F-1
Proceed to step 4), and if it is not enlarged, check whether the previous and previous shots were in focus (F-1O2F-11), and if any of them are in focus, go to step (F-16) and set the enlargement flag to L. Make it. If the previous time and the previous time are in focus,
Set the counter in the microcomputer 23 and start measuring the time (3 seconds) to expand the focusing range (F-12), then set the expansion flag to H (F-13) and decrement the counter ( F-13) Next, determine whether the counter is O (that is, whether 3 seconds have passed), and if O, set the enlargement flag to L (F-16) and set the focus range to D.
o=ε. /2, and if the counter is not O, set the focus range to Do=(D/2) with the enlargement flag=H (F-17).

Next, determine whether the previous focus detection result is in focus (F
-18) If the focus is not in focus, use the focus range D-D set in step (F-17) to determine whether the defocus amount is within the focus range. If it is within the focus range, stop the focus motor 8 (F-28)
, if it is not within the focusing range, the focus motor 8 is driven based on the defocus amount (F-29).

If the previous focus detection result is in focus (F-18), the amount of out-of-focus is averaged (F-19) to eliminate fluctuations due to noise, etc., and smooth focusing. After that, double the focus range (F-20), check whether the enlargement flag is H (F-21), and if not, proceed to step (F-27) to double the focus range. Focusing is determined based on D=2D.

Next, if the enlargement flag is H (F-21), the focusing range is expanded three times (F-22), and then it is determined whether the focus detection mode or manual mode is selected (F-23), and the If the mode is the focus range D=3D set in step (F-22), focus determination is performed (F-27). When not in manual mode, that is, in auto mode (F-23), next, if the focal pressure 111f of the focusing lens 7 determined from the output of the position sensor 13 is less than 25 mm (if the zoom lens 10 is on the Te1e side) (e.g.), expand the focusing range by 5 times (F-26), and perform focus judgment (F-27). When the focal length f is 25 mm or less (that is, when the zoom position is on the Wide side), Next, the CCD
Check the accumulation time of 19 (F-25). CCD
The accumulation time of 19 is controlled by the microcomputer 23 according to the magnitude of the output signal of CCD l 9, and this accumulation time is less than 10 m5ec (when the subject is bright or the subject is bright).
Expands the focusing range by 5 times (F-26), 10m5ec
Step (F-
Focus determination is performed using the three times the focus range set in step 22) (F-27).

By setting the focusing range by changing it as described above, the conventional problems (1) to (4) when setting the focusing range to a fixed value can be solved (see pages 6 to 7 of the specification). I can do it. In other words, (1) Even when the amount of defocus is on the border between in-focus and out-of-focus, the focusing range is expanded (in the example, 5 times in auto mode and 3 times in manual mode), so that the device (2) Accurate focus judgment can be made depending on the brightness of the subject, and (3) The focus range is expanded, so it is possible to avoid unnatural movements due to external noise or objects that suddenly cross the subject. Focusing is not disturbed (
4) Even for objects with low contrast, by widening the focusing range, smooth focusing can be performed without being affected by fluctuations in the output signal of the image sensor.

In the above embodiments, a video camera was used as an example, but the present invention is not limited thereto, and can also be applied to a general still camera. The focus detection method is also applicable not only to the TCL method but also to other known phase difference methods and contrast methods.

Furthermore, in the embodiments, the TTL method is described in which the focal point is detected by the subject light flux transmitted through the photographic lens, but the present invention is also applicable to an external method in which an AF optical system separate from the photographic lens is provided.

(Effects of the Invention) As explained above, in the present invention, the focusing range can be adjusted according to the contrast and brightness of the subject, the moving state and position of the zoom lens, the history of the amount of defocus, or the focus detection mode. Since the focus is changed, focus detection can be performed smoothly without being affected by fluctuations in the output signal of the image sensor.

[Brief explanation of the drawing]

FIG. 1 is an overall configuration diagram of a camera focus detection device according to the present invention, FIG. 2 is a block diagram of an embodiment of the present invention, and FIGS. 3 and 4 are modes of changes in the focusing range of the embodiment. FIGS. 5A and 5B are flowcharts explaining the operation of the embodiment, and FIGS. 6 to 8 are diagrams explaining the principle of the TCL system.

Claims (1)

[Claims] an image sensor that receives a subject light beam that has passed through a photographic lens or a dedicated external lens linked to the photographic lens; a calculation means that calculates the amount of defocus of the photographic lens based on an output signal from the image sensor; A focus determination means determines that the focus is in focus when the focus amount is within the focus range, and the focus range is determined by the contrast or brightness of the subject, the moving state of the zoom lens or its lens position, and the history of the defocus amount. , or a focus range variable means for changing the focus range according to the focus detection mode.