JPS5928886B2 - automatic focus detection device - Google Patents

Description

[Detailed description of the invention] The present invention relates to a pupil-splitting automatic focus detection device.

In such devices, the imaging effect rays emanating from the exit pupil of the objective are generally split through a group of lenslets;
Each split light beam is made incident on a self-scanning optical sensor consisting of a pair of continuous photoelectric elements, and the pit of the objective lens is detected based on the phase difference between the output signals from each optical sensor.

In such devices, the pit detection accuracy is usually fixed, in which case the detection accuracy is set to the highest condition, and therefore no small movement of the objective lens or subject However, the focus position will shift. In addition, when focusing is performed by manually moving the objective lens based on the detection results and looking at the focus display in the viewfinder, etc., if the detection accuracy is high, the changes in the display will be subtle, so it will take very fine attention to detail. is required. Furthermore, in the case where the objective lens is automatically moved by a motor or the like based on the detection result, the motor or the like is constantly rotating, which is extremely troublesome. Furthermore, since the F value of the objective lens during photographing is often narrowed down, if the detection accuracy is fixed, focusing will be performed with higher precision than necessary. An object of the present invention is to provide a sliding focus detection device that is easy to use by making pit detection accuracy, that is, focusing accuracy variable.

Hereinafter, the present invention will be described with reference to the attached drawings.

First, referring to FIG. 1, the outline of the pupil-splitting automatic focus detection device will be explained. The exit pupil 12 of the objective lens 11 is divided by a pupil division optical system 10 consisting of a group of small lenses such as a pinhole, slit, lenticular or fly's eye lens, and the exit pupil 12 passes through parts 12A and 12B of the exit pupil 12. The image-forming action ray of the objective lens 11 is the optical sensor SA of group A.
o,... SAi... SAn and group B optical sensor SBo... SBi... -S
The incident light corresponds to Bn. These optical sensors are SAo
and SBo, ...SAi and SBi, ...
SAn and SBn are each paired, and the imaging action light beams that pass through portions 12A and 12B of the light beams incident on the same location on the pit detection surface correspond to the two optical sensors in the pair. incident. As shown in FIG. 2, when the pit of the photographic lens is shifted from the subject, a phase difference occurs in the outputs of the paired optical sensors. Furthermore, the position of the output phase shift of the A group and B group of the optical sensor is reversed due to the front focus and the rear focus. For example, the arrow ← indicates the direction in which the output phases of the A group and the B group of the optical sensor are shifted when focusing from the front focus position and the rear focus position to the in-focus position. FIG. 3 shows the illuminance of the subject image on the pit detection surface, where A is when the subject is in focus, the mouth is in focus from the front, and C is when the mouth is in focus from the rear. The arrangement of the above-mentioned optical sensors is as shown in FIG.
Element SO having SBn...Sn-1 is Y
Alternatively, a pair of optical sensors SAO, SBO, . . . SAn, S may be arranged in the direction shown in FIG.
Element SO having Bn...Sn in the 4D direction 2
They may be arranged in columns. Optical sensor SAO-SA
n, SBO-SBn is a set of CCDs (ChargeCO
(upledDevlee), and its output data is processed by the circuit shown in FIG.

This set of CCDs 13 is driven by a CCD drive circuit 14, and the output signals of the CCDs 13 are converted into digital signals by an A/D converter 16 via a video amplifier 15 and stored in memories 17 and 18. In this case, the memory 7 has a signal AO-A from the optical sensor SAO-SAn of group A.
n is memorized, and at the same time, the memory 18 stores the optical sensor S of group B.
Output signals BO-Bn of BO-SBn are stored. The output signal of this memory 18 is shifted by a shift circuit 19, and the difference between it and the opposing signal from the memory 17 is calculated by a differential circuit 20. The output signal of the differential circuit 20 is converted into an absolute value circuit 21
The absolute value is taken at , and the sum is taken at the integrating circuit 22 . The control circuit 23 changes the shift amount j of the shift circuit 19 and resets the integration circuit 22 to repeat the above calculation, and uses the memory 24 to calculate the value k of j when X is the minimum.
The focus state of the photographic lens and camera shake are detected. When detecting the focus state of the photographic lens, first select C in steps 1 to 5 as shown in the flowchart in Figure 6.
Data AO-An, BO-Bn from CDl3 to memory 1
Continuing to 7, 18.

Next, the calculations in steps 6 to 11 are performed as described above.

However,.

This calculation result X becomes a function X(j) of the shift amount j, and is as shown in FIG. 7: 1 for front focus, 2 for in-focus, and 3 for rear focus. Therefore, the relationship between the value k of j that minimizes X(j) and the focus state of the photographic lens is as follows.

Therefore, in steps 12 to 18, the first (at the time of Jb-n)
treats X as the minimum value XMIN and stores it in the memory 24, stores j-bn as k in the memory 24, increments the shift amount j, and returns to step 7, and from the second time onwards (when j\Bn ) is XMIN,k in the memory 24 if the operation result X is smaller than the minimum value XMIN in the memory 24.
is changed to that X, j, j is incremented, and the process returns to step 7. If XMIN>x, the process directly returns to step 7. Here, data AO-An in memories 17 and 18,
The relationship between BO-Bn, the shift amount j, and the range in which equation (1) is calculated is as shown in FIG. When it reaches j-a, the process proceeds to steps 19 to 23, and the control circuit 2
3 detects the focus state of the photographing lens by making a determination using equation (2) from k in the memory 24, and the output thereof drives the photographing lens with the lens drive motor to perform a focusing operation, or the detection output thereof. The focus state is displayed on the display device and the focusing operation is performed manually. It goes without saying that the objective lens 11 constitutes a photographic lens, or is separate from and interlocks with the photographic lens. Next, pit detection accuracy will be explained with reference to FIG.

Now, when the pupil diameter is DFl and the allowable blur amount is δ, the depth of focus is RFl. Also, when the pupil diameter is DF2 smaller than DFl and the allowable blur amount is the same δ, the depth of focus is R
It will be F2K. Therefore, the smaller the pupil diameter, the F
The larger the number, the deeper the depth of focus, the wider the focusing range, and the less precise the detection accuracy. In addition, even if there is a phase difference k in the output of the optical sensor, the allowable blur amount δ
If it is within the range, it is allowed. Allowable blur amount δ
If the amount of blur becomes larger than , this indicates that the front focus or the back focus has been reached, and the phase difference k is also shifted by 5 in either direction. Therefore, by determining the allowable shift amount Kx corresponding to the focusing range RF for each F number (Kx≧O) and comparing this with the phase difference k, it is possible to obtain the accuracy corresponding to each F number. pits can be detected. this is,
This can be carried out using a flowchart as shown in FIG. That is, the output k from step 18 shown in FIG. 6 is compared with 1k1 and Kx in step 19, and if the absolute value of the phase difference k is equal to or smaller than the allowable shift amount Kx, the phase difference k is the sum. Since it is within the focus range, a focus signal is issued. Otherwise, in step 20, if the phase difference k is smaller than the negative allowable shift amount - Kx, a signal for the rear pin is output; otherwise, if k>-Kx, a signal for the front pin is output. . This corresponds to the above-described rear pin k<0 and front pin k>O. The allowable shift amount is input from the exposure control circuit 25 connected to the control circuit 23 in FIG. 5 in response to the F-number signal of the objective lens during photographing. In the above case, the pupil division angle θF is fixed, but
As shown in Fig. 9, the pupil division angles θF1 and θF2 are different depending on the pupil diameters DF and DF2, and the focusing ranges RFl and RF2 are also different. Pit can be detected with detection accuracy that corresponds to the pupil division angle, that is, corresponds to different F numbers.

This will be explained in more detail with reference to FIG. In order from the focal plane G of the objective lens 11, a first optical sensor array A is located at a distance d1, a second optical sensor array A2 is located at a distance D2, and a third optical sensor array A3 is located at a distance D3. As shown in FIG. 12, they are arranged perpendicularly to the optical axis S of the lens and in close proximity to each other in parallel to each other so as not to overlap with each other, with the optical axis S at the center. Each array consists of a pair of photoelectric element rows, and the pitch P between each element is constant. Moreover, each array can also be arranged radially around the optical axis S.
The pupil division angle θF is determined from the F number FNO of the objective lens by the following equation.

Further, the relationship between the pupil division angle θF, the distance d from the focal plane G of the lens to the light receiving surface of each array, and the pitch P between photoelectric elements in each array is as follows.

Therefore, by determining d and P in accordance with each θF, each split light beam can be made to enter the corresponding photoelectric element correctly.

A large value d means that the focusing range is large. Therefore, by selectively using the array arranged at d corresponding to each pupil division angle, pit detection including pit detection accuracy can be performed. Selection of each array is performed automatically or manually based on the pupil diameter of the objective lens when it is open or the F-number signal from the exposure control circuit during photographing.
Further, the change in d can also be obtained by using a small lens group arranged on the focal plane G of the objective lens as a zoom lens consisting of a large number of small lens elements. In this case, only one optical sensor array is required and its position is fixed.
Also, from equation (4) above, instead of increasing d, d
It is also possible to keep P constant and change P in each optical sensor as shown in FIG. 13 according to changes in θF. In this case, a plurality of optical sensor arrays with different values of P are similarly prepared, but since d is constant, the same small lens group can be used at the same position. Further, these changes in P and d can also be used in combination.
Furthermore, as shown in FIG. 14, the pit detection accuracy can be changed by arranging a plurality of optical sensor arrays with the same pitch P between the photoelectric elements in each array and different pitches Q between the pairs of photoelectric elements. be able to.

Q is determined corresponding to the amount of blur at a specific F number. This method uses the fact that the larger the F number is, the more relaxed the pit detection accuracy is, and makes the standard for the pit detection accuracy itself looser. That is, an array with a large Q has less detection accuracy than an array with a small Q. This can be done by carefully measuring the distance measurement light from the exit pupil, or
The difference is whether it is partially extracted and measured. Therefore, by arranging a plurality of arrays with different Q values corresponding to a plurality of specific F-numbers and selectively using each array in accordance with the F-number at the time of imaging, the pit detection accuracy can be improved. It can be made to correspond to the F number. The calculation in this case is shown in FIG. 10, for example, Kxl
This is done by substituting . Such a method corresponds to, for example, a sampling inspection in the field of quality control.

This is because if accuracy is strictly required, a 100% inspection is necessary, but if the accuracy is not strict, a sampling inspection is sufficient. Therefore, another method is to use a single row array in which the photoelectric elements are arranged as densely as possible, without using multiple arrays with different Qs, and to The outputs from the elements can be used, for example from every other or every second element. This can be achieved by performing an interlaced operation in FIGS. 6 and 10 with j←j+l as j←j+Kx. However, bn=NbXKx (however, Nb is an integer smaller than zero), and a-NaXKx (however, Na is an integer larger than zero). This is exactly the same as a sampling inspection, and Kx corresponds to the number of samplings and is determined corresponding to the F number. These methods are effective when moving the objective lens in one direction, that is, from infinity to close range or from close range to infinity. Also, the 14th
In the figure, for each array, instead of making the small lens elements of the corresponding small lens groups the same diameter, Q can be changed by using small lens elements with different diameters, and P can also be changed accordingly. .

As described above, according to the present invention, the pit detection accuracy can be changed depending on the F number, so it is possible to realize a focus detection device that is easy to use.

Moreover, if the pit detection accuracy can be arbitrarily changed according to the photographer's will, regardless of the F number, a great degree of freedom can be obtained in terms of drawing, such as soft focus.

[Brief explanation of drawings]

Fig. 1A shows an outline of a pupil-splitting automatic focus detection device to which the present invention is applied, and Figs. 2 and 3 show in-focus states, front focus states, and back focus states in the above device. Figure 4A is a diagram showing the arrangement of the optical sensors in the above device, Figure 5 is a diagram showing the block diagram of the electronic circuit in the above device, and Figure 6 7 is a diagram showing the output of the arithmetic processing in the above circuit, FIG. 8 is a diagram showing the range of the arithmetic processing, and FIG. 9 is a diagram showing the principle of the present invention. FIG. 10 is a diagram showing a part of a flowchart of arithmetic processing in one embodiment of the invention, FIG. 11 is a diagram showing another embodiment of the invention, and FIG. FIG. 11 is a schematic side view, FIG. 13 is a diagram showing a modification of the embodiment shown in FIG. 11, and FIG. 14 is a diagram showing still another embodiment of the present invention. 10...Pupil division optical system, 11...Objective lens, SA, SB...Photo sensor pair.

Claims (1)

[Claims] 1 Split the light from the outer circumferential portion of the exit pupil of the objective lens into two, and provide each divided light through a small lens group arranged on the focal plane of the objective lens, corresponding to each lens element of the small lens group. In an automatic focus detection device that detects focus based on the phase difference between the output signals from these photoelectric element arrays, the automatic focus detection device detects focus based on the phase difference between the output signals from these photoelectric element arrays. An automatic focus detection device comprising means for changing the allowable phase difference of the output signal or for selectively selecting the output signal in order to change the focus detection accuracy.