JPS6017414A - Focus detecting device - Google Patents

Abstract

PURPOSE:To simplify an arithmetic processing circuit by specifying the shape of a cell for forming a photodetector array, and making the photodetector array itself have weighting and integral function. CONSTITUTION:Photodetector arrays 6', 7' are divided into weighted cells 6a- 6d and 7a-7d, respectively. As for these cells, the shape for generating a weighted photocurrent when the light from an object to be photographed is received is given. Also, an arithmetic processing circuit 11 to which the output of the arrays 6', 7' is provided, consists of buffers 34-41, subtracting circuits 43-46, dividing circuits 23, 31, a comparing circuit 32 and a polarity deciding circuit 33. Since the photodetector array is formed in this way, a weighting circuit and an integration circuit in the arithmetic processing circuit can be omitted. Accordingly, the arithmetic processing circuit can be simplified.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a focus detection device for a camera or the like using a light receiving element array.

2. Description of the Related Art In recent years, there has been a demand for systems that automatically measure (or answer) the distance (or focus) of objects (such as subjects) in various industrial and consumer devices. For example, autofocus systems expected for cameras are typical systems that automatically focus (or measure the distance to a subject).

As a representative system for position detection using conventional light receiving elements (CCD, SPD, etc.), a triangulation type (one-object matching type) has been put into practical use. However, some of these systems have problems with focusing accuracy and insufficient functions to determine whether front focus is back focus or the like. Therefore, the inventor of the present invention has already developed a focus detection device that takes measures to address the problems of the conventional device, and this focus detection device will be described below.

In the focus detection device already developed by the inventor of the present invention, incident light from a subject passes through first and second imaging optical systems through first and second passages that are spatially separated. However, images are formed on a pair of first and second light receiving element arrays located at a fixed focal plane of the imaging optical system.

At this time, each light-receiving element array has N pieces (light-receiving elements are divided into N equal parts in the length direction, and the number of divisions N is N-4, N
=8. Can be arbitrarily selected such as N=16. ), and when the incident light reaches the first and second light receiving element arrays and forms an image, an illuminance distribution signal that depends on the intensity of the incident light The above-mentioned signals obtained from the first and second light receiving element arrays are processed by a weighting circuit, an integrating circuit, a dividing circuit, a comparing circuit, etc., which will be explained below, to detect and determine focus, front focus, and rear focus. is possible.

The above device will be explained in more detail.

In FIG. 1, 1 is a subject S serving as a light source, and light from this subject S passes through a first passage 2 and a second passage 3 that are spatially separated, and then passes through a first and second lens. (
(imaging optical system) 4, 5, and the first image is formed on its fixed focal plane.
Since the second light-receiving element array (CCD or the like) 6, 7 is located, the image of the subject S is formed on the light-receiving element array 6.7.

Also, 1st. An objective lens 9 is placed near the second lenses 4 and 5.
is located, and the light from the subject S passes through the objective lens 9 via the optical path 8, and the film plane 10 is located at the fixed focal plane.
is located. Here, the first .phi. The imaging optical systems 4 to 10 are set in advance so that when the second light-receiving element array 6.7 is in focus, the film plane 10 is in focus.

Also, 1st. The second lens 4.5 is exactly the same lens (focal length, glass material, aperture, etc.) and is located at the same position from the subject S in an almost optically equivalent relationship, and both lenses move exactly the same (two (including cases in which the lenses are integrally formed on the same plate surface and move synchronously), and similarly, the first and second light receiving element arrays 6.7 are also almost optically equivalent from the subject S. (the two light receiving element arrays are integrally formed on the same plate surface, and the first
．． (including the case where it is placed at the fixed focal point position of the second lens).

Next - above 1. The illuminance distribution signal obtained from the second light-receiving element array 6°7 is weighted.

The signal enters a signal processing circuit (arithmetic processing circuit) 11 having an integral function, a division function, a comparison function, a polarity judgment function, etc., is subjected to arithmetic processing, and generates a focus signal, front focus, and rear bin signals, and these signals are output. A control circuit 121 is inserted. This output control circuit 12 has a motor drive output section, and when the front focus is on, a positive signal is output. When the object is in focus, the motor drive output becomes O, the motor 13 stops, and the object is in focus.

Also, when the rear focus is achieved, a negative signal is output to reverse the motor M (negative feedback), and similarly, when the focus is achieved, the motor 13 is stopped and a single focus detection function is obtained.

FIG. 2 is a schematic diagram showing the object (light source S) and the operating state of the imaging optical system when in focus and out of focus (front focus, back focus).

Figure 3 shows each operating state (focus, front focus, etc.) in Figure 2.

The same 1st . The image of the light beam from the object is shown on the second light-receiving element array 6.7.

Next, using FIGS. 2 and 3, the same 1. The process in which a light beam from an object forms an image on the second light receiving element array 6.7 and the movement of the image on the light receiving element array will be explained.

In FIG. 2, when the first light-receiving element array 6 is arranged in the order from top to bottom from block 1 to block 16, the second light-receiving element array 7 is arranged in the reverse order, that is, downward. From above, they are arranged in the order of -1 to m to -16.

In such a configuration, when in focus as shown in 2nd e71(a), the 1st. Second light receiving element array 6.7
The light beam from the object is imaged in the upper center part, as shown in Figure 3 (a).
).

Next, as shown in FIG. 2(b), when the object gradually retreats from the in-focus state (front focus), the first. Second light receiving element array 6.
7, the image is moved downward, and as shown in FIG. 3(b), the image is IIkl-9 on the first light receiving array 6.
From I! Similarly, the image formed on the second light receiving element array 7 moves in the direction from level -9 to level -1.

Next, as shown in FIG. 2(C), when the object gradually moves forward from the focused state (rear focus), the first. On the second photodetector array 6.7, the image is moved upward, as shown in FIG.
), the image formation on the first photodetector array 6 is NQ.
Similarly, the image formed on the second light-receiving element array 7 moves from position -9 to position -16.

From the above explanation, it can be seen that the movement of the object back and forth, and the first. The relationship between the movement and the image formed on the second light receiving element array 6.7 will be explained by taking the first light receiving element array 6 as an example.

When an object moves from the front to the back, its image is formed on the light-receiving array 6 from the cell at floor-1 to l! When the object moves in the direction of the cell 1L-16 and conversely moves from the back to the front, it moves on the light receiving element array 6 from the cell 1k-16 to the cell 1K-1. With the same operating principle as described above, with respect to the movement of the object on the first and second light-receiving element arrays 7, the movement of the image moves in the completely opposite direction to the movement of the image on the first light-receiving element array 6, When in focus, the image is focused on the first. The same cell address (f
fi-1 to N11-16).

To summarize the above explanation, the simplest object is a single object (a point light source or a line ray), and when the images on the first and second photodetector arrays 6 and 7 are each thinner than in each cell. (corresponding to the images shown in FIGS. 3(a), (b), and (C)) as an example, if the object is the first one. When an image is formed on each of the second light-receiving element arrays 6.7, the relationship between the cell address number of each imaged cell and the in-focus, front focus, and rear focus is as follows: ■When in focus...the image is formed 1st. Second light receiving element array 6
．． Each cell number of 7 is equal.

, °, Cell number of the first light receiving element = Cell number of the second light receiving element ■When focusing on the front...Cell number of the first light receiving element>Cell number of the second light receiving element ■When focusing on the back... - Cell number of the first light receiving element <cell number of the second light receiving element By comparing the cell numbers of each light receiving element that forms an image, the in-focus, front focus, and rear focus states can be detected.

However, if there are more than one light source as a subject, for example, there are two objects (point light sources) that correspond to a single object (point light source) as described above, separated by a little distance, and those objects (subjects) are the first. When two images are formed on each of the second photodetector arrays 6 and 7, it is not possible to determine which imaged cell number should be compared, and the front focus and rear focus states are detected when out of focus. I won't be able to do it. Therefore, the following processing is performed so that the state before and after in-focus (degree of out-of-focus) can be determined.

That is, FIG. 4 shows an embodiment of a signal processing circuit for performing the above processing, and this circuit is similar to the signal processing circuit 11 in FIG.
This corresponds to The outline of the arithmetic operations for performing the above processing will be explained using this Fig. 4, and then Fig. 4,
The operation of detecting the in-focus, front-focus, and rear-focus states will be explained in detail with reference to FIG.

In FIG. 4, the ccD drive clock (here, 16 pulse trains and correction pulses are repeatedly generated) created by the scanning circuit 14 is transmitted via the output line 15 to @X,
The light flux from the object is supplied to the O3 optical element array (here, divided into 16 equal parts by a -dimensional CCD and composed of 16 cells) 6 and 7, and then converged on the first light receiving element array 6. The illuminance distribution signal is sent out in time series from the output terminal 16, and is applied to the first and second weighting circuits 19 and 20 via two output lines 17 and 18, respectively. (Details will be described later), A(L), Bit) (
Here, the time series signal is (=mouth to mouth 6).

Each signal A(t), B(t) is the first . are respectively integrated by the second integrating circuits 21 and 22, and the output Vl
.

■2 is v1 = f A(t) dt - (1)V 2 = f
B(tl d t ---(21 (however, 1 to 16 correspond to cell addresses 1 and -1 to 16 of the light receiving element array). Each of these outputs VI and V2 is output from the first division circuit. 23, and its output C becomes.

Similarly, the light beam from the object forms an image on the first and second light-receiving element arrays 7, and the illuminance distribution signals thereof are sent out in time series from the output terminal 24 and transmitted to the third light receiving element array 7 through two output lines 25 and 26. ．．
The fourth weighting is applied to circuits 27 and 28, where the weighting is calculated, resulting in a time series signal output of A(t), B'(1), and a third . are respectively applied to the fourth integration circuit 29 and 30, and their outputs ■8+v4 are VB-f, A(t) dt --14)v4-ha B
(t) dt - ri (51. Each of these outputs V8
．． %'4 is added to the second division circuit 31, and the output C' is as follows. The above +31, (Formula 61 is based on the spatial position information tone described later, the first and second light receiving element arrays 6.7 (7)
) * Ru address 1! This corresponds to any one of floors 1-1 to 16.

1st. The outputs c, c, and ' of the second division circuit 23.31 are added to the comparison circuit 32. When in focus...C=C...(7) When out of focus...c>d... ...(8) When out of focus...c<c'...(9) An output is obtained, and when in focus, it is taken out as the output of equation (7). Also, when out of focus, to determine front focus and back focus, (1), +21, +41, +51
, +81, each output of equation (9) is connected to the polarity determination circuit 33
added to. When the following formula 011~α holds true, it is determined that the front pin state is present, and ■Vs>0, v4)0 and VI <o + ■2>0−- (10)■va <
0, v 4 < 0 and V1> OI V2 <
O-,..., Circle■c>c' (12) When the following U□□□~(15+ formula holds true), it is determined that the rear pin state is established.

■ ■l＞0IV2＞0 One eye v8〈0, v4〉
0 ・ ・ (I3) ■■l〈0. v2〈0andtsuvB
>o, V4 <0 -= (141 ■c <c'
...(151 (However, as described later, v 1 = = A, v 2 - B
, v B == A .

v4=B) In the above, the determination of the front pin state is 10), (111
Set the expression to take priority, and at other times (QO), 09
(when the formula does not hold), the determination is made based on a set of conditions (details will be described later). Similarly, the determination of the rear pin state is α31, (
Priority is given to formula 141, and in other cases (when the formula (13, α) does not hold), the condition of formula 051 is used (details will be described later).

Next, using FIG. 5, as a simple example of a general subject, a single object (a point light source or a line light source) is imaged on the first and second light receiving element arrays 6 and 7, respectively. , when the width is thinner than each cell width, as shown in Figs. 3(a) and (b).

(corresponding to (C)) are located a little apart, and the first. Second
Assuming various cases in which two sets of images are formed on each of the light receiving element arrays 6 and 7, focusing and front focus are performed.

The operation for determining the rear pin state will be explained in detail.

First, 1. Of the second light receiving element array 6.7, the first
This will be explained by taking the light receiving element 6 as an example. Figure (a) shows a CCD drive clock signal for driving the light receiving element array 6, and the horizontal axis shows time t ([1 to [16]) and spatial position addresses (cell addresses -1 to 16). I'm taking it. This clock signal outputs the photoelectric conversion output of the cell address positive-1, which is the spatial address of the CCD, at time t1, and the output of cell address floor-2, which is the spatial address of the CCD, at time t2, and so on. Cell address positive-1, which is the spatial location address at time 116
The outputs of the first light receiving element array 6 are allocated in such a way that the outputs of the first light receiving element array 6 can be extracted in a corresponding (synchronized manner) from the spatial position addresses (cells lhl to 16) in the order of time t1 to 116 in a time series manner. The next first weighting will be taken out as a signal and sent to the circuit 19. The same figure (b
) shows an example of a profile in which the time-series output signal from the output terminal 16 is weighted by the first weighting circuit 19, and the photoelectric conversion output of -1 (first) at cell address 1 is 100% ( The obtained output (as is) is weighted, and the photoelectric conversion output of cell address 1-5 (fifth) is set to 0%, and the envelope of the first cell space between them is a straight line. Weighted by a function (generally a function that is differentiable between them and whose derivative or negative value is a function), and cell address 11kl-9
Let the photoelectric conversion output of (9th) be -100%, and the cell address is -5 and cell address l.
! Similarly, the second cell space ■, which is between 1L and 9, is weighted with a function whose envelope is a straight line (generally, it is differentiable between them and its derivative is negative), and the cell address floor is −
13 (13th) photoelectric conversion output is 0%, and the third cell space ■ between cell address N1-9 and cell address group-13 is similarly expressed by a function whose envelope is a straight line (general is differentiable between them and its derivative is positive), and the photoelectric conversion output of cell address floors ~16 (16th) is
5%, and the fourth cell space ■ between cell address 1 -13 and cell address floor -16 is similarly a function whose envelope is a straight line (in general, the function is differentiable between them, is positive).

FIG. 5(C) shows an example of a profile in which the time-series output signal from the output terminal 16 is weighted by the second weighting circuit 20, and the photoelectric conversion output of cell address floor -1 is set to 096. The photoelectric conversion outputs of cell addresses Ni1-5 are weighted to 100%, and the first cell space I between them is defined as a function whose envelope is a straight line (generally, it is differentiable between them, and its derivative The photoelectric conversion output of cell address 11kL-9 is weighted by 096, and the envelope between cell address floor -5 and cell address floor -9, which is the second cell space ■, is is a straight line (generally, it is weighted to is a function that is differentiable between them and whose derivative is negative), and the photoelectric conversion output of cell address floor -16 is weighted at -25%, and the cell address floor that is the fourth cell space This figure shows weighting between -13 and cell address floor -16 using a function whose envelope is a straight line (generally, a function that is differentiable between -13 and cell address floor -16 and whose derivative is positive).

FIG. 5 (dl is each cell address (floor -
1 to ff1-. 16). The second weighting is the circuit 19. The weighting of 20 is the quantity Ai (i=1 to 16)
, indicates the value obtained by dividing Bi (i = 1 to 16) - if this is Ci - Ci - 2...00 i (however, i = 1 to 16, and the value including polarity) ) is defined as The value of Ci is a spatial position conversion value set corresponding to the spatial position (location S) of each cell number of the light receiving element array 6, and is specifically as shown in the table below. Among the values of Ci in the lower right table, the earth bell is clamped at a value determined by the upper and lower limits of the dynamic range.

The curve in FIG. 5(d) is a plot of the values of Ci in the table, and the horizontal axis represents the spatial position addresses (cell addresses &-1 to 16 )
is taking. The spatial position information value (value of the above equation (3)) C, which is the output of the division circuit 23 processed by the signal processing circuit αD shown in FIG. As a result, the spatial position address is determined, and as will be described later, the function of detecting focus, front focus, and rear focus can be obtained.

The weighting shown in FIG. 5 and the table above is based on the calculation
The weighting of circuit 19.20 and the first . The results obtained using the second integration circuits 21 and 22 are the same as (1) above.

When calculated according to formula (2), the following (17), (1
It will be type 81.

v, =: A-c A(t)dt 3 2 1 = (Kl+ - to 2+- to 8+- to 4+ 0・K5-!
-Ks4 4 4 4 2 3 3 - KV -- to 8- to 9--KIO-"Kl+-2-
Kl.

44 444 1 2 3 +0- to 18+7 to +4+HK+5+7 to +6)-...
-(lηv2=B =f B(t) dL 1 2 3 = b(OK++- to 2+-K11+- to 4+ to 5+3-
7444 to 6+ 44 -1-- to 8+ 0- to 9-'-KIO-4++ Once to Ky+2- This is to standardize the photoelectric conversion output, and +a (Ki) to Kl- indicates that the light flux from the object (light source S) reaches the cell address floors 1 to 16 of the light receiving element array 6, which is the spatial position address. It is a coefficient that is modulated in proportion to the illuminance distribution when an image is formed, and the Ki is - -Z, <Ki Z... (+9).The spatial position information C that is the output of the division circuit 23 is
From the formula (3), 11η, and α, the following formula C!■ is obtained.

The above value of C is a value having polarity, and the above αη.

Q8) If we condition with the values A and B of the formula (the above (10)
~(logic of formula 151), takes the value of any one point in the spatial position conversion value Ci in FIG. 5(d).

In this way, in the signal processing circuit 11, from the above equation It can be converted into spatial position information C and used and expressed.

Regarding the second light receiving element array 7, the third. The fourth weighting is performed by setting the weighting quantities of circuits 27 and 28 to profiles as shown in FIG. By setting , it is possible to obtain spatial position information d using a similar logic by treating an arbitrary object consisting of a collection of a plurality of single objects as a physical quantity equivalent to a single object.

From the above explanation, the two sets of subjects as simple examples of the general subjects mentioned above are 1. Second light receiving element array 6.
7, the signals can be processed and treated in the same manner as when a single object (point light source) is imaged. Second
The cell number of the photodetector array 6゜7 (l!lL1~16
) or the above-mentioned spatial position information C9C, in-focus, front focus, and rear focus can be detected by comparing the respective spatial position information C9C.

Furthermore, in the case of 3 sets of objects (black light source), there are 12 sets of objects and 1 set of objects, of which 2 sets are
As mentioned above, it can be replaced equivalently with one object, so
In effect, it can be replaced with two sets of objects, and these two
The set of objects can ultimately be replaced with a single set of objects (point light sources), and if this is expanded inductively, it can be generally replaced with -N sets of objects (N is a positive number larger than l). Even if
Focus, front focus, and back focus detection is possible.

Next, as a simple example of a collection of a plurality of single objects, the above-mentioned two sets of objects (point light sources) are formed as two images on the first and +32 photodetector arrays 6.7, respectively. When the two images are generated, the two images are combined (the photoelectric conversion output is one above 07) ~
(20) signal processing), one spatial position information c is obtained as a physical quantity equivalent to a single image corresponding to a single object.
, c' will be explained using FIGS. 6 to 18.

6 and 7 are diagrams when two objects are formed as two images on the first light-receiving element array 6, and FIGS. ．． Second. Third. An example is shown in which the above-mentioned two images are formed in the fourth cell spaces I to II, respectively, and FIGS. Second. Third. An example is shown in which the above two images are formed across each of the fourth cell spaces - (2).

8 to 12 are diagrams for explaining the operation in the states of FIGS. 6(a) to (d), and FIGS. 13 to 18 are diagrams for explaining the operations in the states of FIGS. 7(a) to (e). It is a drawing for explaining the operation.

FIG. 8 shows the light receiving element array 6 as shown in FIG. 6(a).
This is an explanatory diagram corresponding to when two images are formed in the first cell space 1 of , and the spatial position information, size, and brightness of the images are indicated by solid lines A2. A4, B2, and B4 are shown, and if the output when each photoelectric conversion output is subjected to signal processing with the first weighting circuit 19 and the first integrating circuit 21 is AOt, then AOt is the above-mentioned 09 According to the formula 2. Since everything other than 4 is O, the following at+ formula is obtained, and AOt- and 2+ - are 4 (2114 When the brightness of the image is the same and 22 and 4: KO - An
+=Ko-=C2'lJ. Similarly, if the second weighting is the circuit 20 and the output when signal processing is performed by the first second integrating circuit 22 is Bat, then the BOI is the 18
From equation 1, we get the following equation, and Bor: − to 2
+ - to 4...Shi14 When 4 to 2=to 4==Ko, Bo + = Ko -= (241).At this time, the spatial position information Go+ is the above (20).
From the formula -C01-w1=1...C5).

From the above equation (support), the spatial position of -Ci=l on the spatial position address axis is a position equivalent to cell number -3 since AOt > 0 and Bat > O, and at this time AOt, Bat
The position information and size including polarity are shown in Figure 8(b).
(C) is shown by a broken line, and similarly, Cot is shown in Fig. 8 (
d) is indicated by a dashed line.

From this, after two images form two photoelectric conversion outputs,
It can be seen that by performing the signal processing of formulas 07) to ① above, it is possible to convert the signal into physical quantities AOt, Bat, and Cot equivalent to the signal obtained when imaging a single object (point light source).

Figure 9 shows the case where the image is formed at exactly the same position as in Figure 8, and the brightness of the image is twice as bright as A'4 than A'2 (
Other conditions are exactly the same as in Figure 8), in this case 2 = 2 = = 4 = = Kg, and this is (21) above.

(c) Substituting into the formula, A'ol=Ko... (e) BO1=-Z-
KO...(4). This spatio-temporal position information C
'OI becomes.

Similar to Figure 8, Figure 9 also shows the above (26+, (2)
, A'or, B'o+, C determined by formula (28)
It shows a value of 'o+, and even if the brightness of two objects (image formation) is different, the same object (point light source)
A physical quantity equivalent to the signal obtained during imaging, AOt,
It can be seen that it can be converted into B'o+ and C'o+.

At this time, the above (support), (24+, (251 formula) and
Comparing formulas (A), @, and (C), C721, (2
4+ , (from equation 25+, the brightness and size of the images of the two objects) are equal at 4 o'clock, the spatial location address corresponding to the spatial location information Go+ shown by equation (25) is also obtained from FIG. , two images A2. You can see that it is located in the center of A4.

In addition, (support)), (4), from the formula □□□, when the brightness power S of the two images is different, the spatial position information C'o + 4 shown by the formula □□□The corresponding spatial position address is From Figure 9,
It can be seen that approximately in proportion to the relative brightness ratio, the image formation is closer to the bright image (here, cell C, cell l!1L4) (cell address value approximately 3.3).

Furthermore, the brightness of the two images is K2:4=1:10
Even when there is an extreme difference from 0, -100K2 = 4 =K
By substituting this into the above equation 211, n, (25), Go + = 2.92 01, and the spatial location address corresponding to the spatial location information C"01 at this time is almost - cell address 4. (, °, spatial position information C' on the nacelle ground 4)
o4=3.0, which is a value very close to the above 2.92. ) Also, when the brightness of the image on cell addresses 2 and 4 is opposite to the above explanation, that is, when 2:2 L4=Ko or 2"100L4=Ko, the cell address Spatial position information that is approximately proportional to the relative value of the brightness of the images 2 and 4 and closer to the brighter image formation side (here, cell address 2) and the corresponding cell address value are obtained.

From the above explanation, the above 0η, (distance) and the relative value of the brightness of the two images.For this reason, the above (2), (
Spatial position information Co+, C'o1 represented by the formula
(and the corresponding spatial position address) can be used as a determination signal for focus, front bin, and rear bin, including information on the position of the two images and their size and brightness.

Figures 10, 11, and 12 are shown in Figure 6 (b), respectively.
), (C), and (d), the second . third and fourth cell spaces n, 10.
, IV are explanatory diagrams corresponding to when two images are formed. First, FIG. 1O is similar to the explanation of FIG. 8, K6. The brightness of the two images is the same and the brightness of the two images is 6:
When 8=Ko(y), AO2, B'02, CO2
From the above formula (17+, (181, (20+), the following equation is obtained:

4 4...29) Bo 2 = 3! -Ks +-! -Ka = K.

４　４　・・・・・・O■ 02 CO2: ~=-1・-Gll O2 As shown in FIG.

Next, in FIG. 11, as in the explanation of FIG. 8, +o is added to −,
' All values other than K12 are 0, and when the brightness of the two images is the same TeKlO:KI2=Ko (7), Aoa,
Bog, Cos are 0η, α811 (a) formula, Aos = − + o −−! −Kl 2 =
-KO44°"°°゛C21 Bo3= -1 to +o-'!-to +2= -K.

1... (Foundation) 4 4 This is as shown in FIG.

Next, in FIG. 12, KI4. K
The values other than I6 are O, and the brightness of the two images is the same.
When 4 = KI6 = Ko, An4, BO4, C
O4 is the above αη, (181, (from formula 201, AO4 knee!−Kl 4 +−!− + s = K.

４　４　・・・・・・(351 Bo 4=-1 + 4-! -Kl　6=-K.

４　４　・・・・・・(to) This is as shown in FIG.

That is, as in the explanation of FIG. 8, two images are equivalent to the physical quantity of the signal obtained when imaging a single object, as shown in FIGS. 1θ, 11, and 12. AO2, BO
It can be seen that it can be converted into 2, CG2, , Aoa, Bog, Can and AO4°BO4, CO4. Furthermore, in FIGS. 10, 11, and 12, it can be seen that even when two objects (or images) have different brightnesses, conversion can be performed in the same manner as explained in FIG. 9 for FIG. 8.

FIG. 13 corresponds to the case where a total of two images are formed across the first and second cell spaces (■, ■) of the light-receiving element array 6, as shown in FIG. 7(a). In the explanatory diagram, the position information, size, and brightness of the image formation are indicated by the solid line A3.
, As, Bg, and Bg, and the respective photoelectric conversion outputs are shown as 1st . The second weighting is the circuit 19.20-first
．． Letting Aos and Bos be the outputs when the signals are processed by the second integrating circuits 21 and 22, the outputs AO5 and BO5 are expressed by the following expression -+ and (39) from the αη and α expressions.

However, K8. The values other than 8 are 0, and the values are shown when the brightness of the two images is equal and Ka=Ks=Ko.

3 Ao 5 Knee to 8-- to 8: --! -KO...
□□□) 4 4 4 B05=S Kg +-! -Ka=-LKo...
G9) 4 4 4 At this time, the spatial position information CO5 becomes the following equation (40) from the above equation (Shiro).

From the above formula 038), +40), Aos <
Q, Bos > O, C knee 3's spatial position is equivalent to cell floor -6, and the position information of AO5, Boa at this time and the size including polarity are shown by broken lines, and similarly. Cos is shown by a dashed line.

From this, after two images form two photoelectric conversion outputs,
It can be seen that by performing the signal processing of equations 0η to (a) above, it is possible to convert into physical quantities Aos, Bos, and CO5 equivalent to the signal obtained when imaging a single object (point light source).

Figure 14 is an explanatory diagram when the image is formed at exactly the same position as in Figure 13, and the brightness of the image is twice as bright as A'8, and the other conditions are exactly the same as in Figure 13. When 2K
Assuming that a = Ka = Ko, this is the above (support).

(To) Substitute into the expression, A'os-zK8-'Ka =-! -0K.

4 4 8 ......(41) B'05=tan ni 8 +-! - to 8 = Dan KO1,...
・(47J4 4 8, and this spatio-temporal position information C'05 is '""'
,,1. , - (43)CO3--end--=5 O6.

Similar to FIG. 13, FIG. 14 also shows the above (41),
Ta.

A'o6) O, B'05) 0 determined by formula (43)
, shows a value of '06, and two objects (image formation)
Even when the brightness of
, B'05, and C'05.

In this case as well, using the same logic as explained in FIGS. 8 and 9, aη, QB+, (from the 20+ formula,
The spatial position information obtained by signal processing (calculation) and the corresponding spatial position address are the position information of the two images (2
(distance between the two images) and the relative value of the brightness of the two images. For this reason, the above +40+
, Spatial position information CQ5 represented by +421 formula, c
'os (and its corresponding spatial location address) is 2
It is possible to use the information (position information, size, and brightness) of the two images as a judgment signal for front focus and rear focus.

Figures 15, 16, 17, and 18 are similar to Figure 7 (
As shown in b), (C), (d), and (e), the first and third cell spaces I, I[, the first and fourth cell spaces I, IV, the second and 3 cell space ■
, (2) - is an explanatory diagram corresponding to the case where a total of two images are formed, one each, spanning the third and fourth cell spaces (2) and (2).

First, in Fig. 15, K8°KI is similar to the explanation of Fig. 13.
2 is 0, the brightness of the two images is the same, and Ks
=+2=Ko), Ao6. 'Bo6. Co
s is the above Q? ), O8), From formula (20),
AO6-7Kg --! −Kl 2 =−! -KO44
4... (4(a) 3 Bo 6=-to 8--Kl 2 =-!-K.

4 4 4 ......(451, Aos) Q, Bog (Q, C06
=-1, as shown in FIG.

Next, in FIG. 16, as in the explanation of FIG. 13, K4°K'
Anything other than 6 is O, and if the two images have the same brightness, then 4=
KI6=KO(7) Toki-AO? , BO7, CO7
Ha, previous mun + (181, (from the 20+ formula, Ao 7--!- to 4 + -!-Kl 6: Ko
...(4714 BO7 4-!-Kl' a = S1(0...
・(4 mothers 4 4 4 07 CO7= -= 0.5 ... (49) O7, Aoy') O, BO7>0, CO7:0.5
This is as shown in FIG.

Next, in FIG. 17, as in the explanation of FIG. 13, K6°K1
Anything other than 1 is 0, the brightness power of the two images is the same 5, and K
When a = + + = Ko, Aos, Bog,
Co8I, the above α7+, (1B+, +201
From the formula Aos=-Kei ni 6-su Ku=-'-Ko...
...Gloss 4 4 4 BQll=- to 6-Ei ++=-! -Ko...
・(51) 4 4 4 Boa1 ...(521 Cos = −= −− O083, Aos <O, Bo8 >0, Co8=−
', as shown in FIG. 17.

Next, in FIG. 18, similar to the explanation of FIG. 13, K+o,
The values other than KI4 are 0, and when the brightness of the two images is the same and KIO==K14=KO, AO9, BO9, C
O9 is from the above 0η, α mark, and formula ■, Ao9=-, +o-)-7 +4==-rKo Mountain°
°Eel Bo9= −−KIO−1+4 =−KO=−(
5414, AO9<O, Bo9(0, C09=2),
As shown in FIG.

As shown in FIGS. 15 to 18, the 13th
As in the explanation of the figure, physical quantities Ao6. BO5, GO
It turns out that it can be converted to 5~Aos, BO9, CO9, and also in Figures 15, 16, and 17.

In FIG. 18, it can be seen that even when two objects (or images) have different clarity, conversion can be performed in the same way as in the explanation of FIG. 14 for FIG. 13.

Using the drawings from Figure 6 to Figure 18 above, two images caused by two sets of objects (light sources) are converted into physical quantities equivalent to a single image corresponding to a single object. We have explained that it can be handled with respect to all combinations (including cases spanning two spatial cells). From this, generally speaking, for any two objects (point light sources), each signal information associated as a function of the spatial position information of the two images and the relative value of the brightness of the two images is It is obvious that the transformation is possible, so the above transformation converts one object (point light source) into
An output equivalent to the spatial position information (and the corresponding spatial position address) obtained from the above is obtained. Further - generally N
The above conversion is also possible for a set of objects in the same way as in the above explanation. - Spatial position information (and corresponding spatial position address) obtained from one object (point light source) from N sets of images. You will get an output equivalent to .

By performing electrical conversion and signal processing (conversion of any N images into a signal output equivalent to one image formation), it is possible to combine any object (an aggregate of N sets of single objects). Detection of focus, front focus, and rear focus can be easily realized.

In addition, in the above device, the weighting of the circuit is as shown in FIG. 5 as an example of the profile. The second weighting starts from the timing date of circuit 19.20 as 100% and 0%, respectively, but cells l'kl~1
6, if the relative relationship between the first weighting is the circuit (incoming tatami thread) and the second weighting is the circuit CB multiplied signal) is maintained, then
As shown in Figure 19, Cell! ! It is not a particular problem that the weighting coefficients shift between 11 and 16.
The same result (output) as the calculation in the figure can be obtained.

Similarly, the first. The second weighting circuit 19.20 uses the coefficients as shown in FIG. 20, even if the weighting of the A signal system and the weighting of the B signal system are reversed compared to FIG. The same effect as the calculation in FIG. 5 can be obtained. However, the spatial position conversion value Ci has a reciprocal relationship between FIG. 20 and FIG. 5, as shown.

Furthermore, in the above device, the weighting is performed in the fifth circuit as an example of the circuit.
As shown in the figure, the first. The second weighting is circuit 19.20
timing (between 1 and [5 (cell floors l to 5), [5 to
Between [9 (cells NIIL5 to 9) - (9 to +g (cells 9 to 13) and 113 to [1 et al. (cell floors 13 to 1)
G) is weighted by an envelope obtained by a straight line function, but even if one sine wave function and one polygonal line function are used as shown in Figures 21 and 22, the difference in Figure 5 is The same result (output) as the calculation is obtained.

In addition, in the above device, weighting is performed by dividing the cell areas 1 to 16 of the light-receiving array into four sets of cell spaces (with inflection points) as shown in FIG. 5 as an example of a circuit. Even if the cell space is set to be divided into 8 groups (with inflection points) as shown in Fig. 23, the same result as the calculation in Fig. 5 will be obtained ( output) is obtained.

As shown in Figure 1, the optical system of the focus detection device separates the light beam from the object S into two, passes through two imaging lenses, and then passes through the first and second light receiving element arrays. 6.7
As shown in FIGS. 24 and 25, the transmitted light of one imaging lens 9 is focused on the mirror M+.
, M2 or through the imaging lenses Ll, L2.
Even with a configuration in which an image is formed on the second light-receiving array 6.7, the focus detection function described above can be obtained, similar to the optical system shown in FIG.

Also, as an example of an overview of the operation of the optical system and signal processing, Section 1. We have explained the arithmetic method using serial signal processing using a CCD as the second photodetector array, but as shown in the optical system in Fig. It is also possible to take out 16 photoelectric conversion outputs at once in parallel and replace the calculations in the block diagram shown in Figure 4 in parallel to perform has signal processing. In this case, Although it has advantages such as being able to improve calculation speed, it also has problems such as an increase in the number of elements. You can choose the processing.

The focus detection device developed by the inventor of the present invention is configured as described above. Since each cell constituting the second photodetector array 6.7 has a shape (rectangle, parallelogram, square, etc.) as shown in Figure 4, weighting requires a circuit or an integrating circuit. The circuit configuration becomes complicated.

The present invention has been made in view of the above points, and the present invention has been made in view of the above points.
By giving the cells that make up the ray a shape that generates a weighted photoelectric current when receiving light from the subject, and by providing the light-receiving element array itself with weighting and integration functions, signal processing is possible. The purpose of the weighting in the circuit is to provide a focus detection device in which the circuit and the integration circuit can be omitted, and the configuration of the arithmetic processing circuit can be simplified.

Hereinafter, one embodiment of the present invention will be described with reference to the drawings.

FIG. 27 shows a signal processing circuit of an uninspected IJ3 device according to an embodiment of the present invention. 5a, 5b, 5C, 5d, 6e- and 73°7
b+ 7 C+ 7' + 7 e). 34 to 37 are buffer stages for amplifying the photocurrent from the light receiving element array 6'; 38 to 41 are buffer stages for amplifying the photocurrent from the light receiving element array;
3.44.45.46 are subtraction circuits S1-S4 for subtracting the outputs from the buffers 34-41.

Next, the operation of this embodiment will be explained using FIG. 27 and FIGS. 5 to 18. Regarding the light receiving element array 6' in FIG. 27, the light receiving element array 6' is divided into five weighted cells 63 to 6e, and the cells 6b, 6d→-
The photocurrent of cells 6c and 6e corresponds to the second weighting (signal system B), and the photocurrent of cells 6c and 6e corresponds to the second weighting (signal system B) in FIG. are equivalent to each other, and the buffer 3 to which the light current is input
4 to 37- and the outputs VBl to V of the buffers 34 to 37
The signal vI + v2 obtained from the subtraction circuit 43.44 into which B4 is input is -v, = vB, -vB2・-'t5B)v2
"vB8 'B4 """"η.Here, the light-receiving strand array 6' of FIG.

Since the weighting of D is equivalent to each of the weightings of FIG. , (561, (571 formulas) are equivalent to the above-mentioned 0η and 081 formulas, respectively.The gloss obtained in this way,
69 formula can be similarly substituted into (support) to thinning, and the focus detection function can be achieved using the same principle as the conventional method.

28 to 31 show a light-receiving array 6'.

We show another method of weighting the game.

Figures 28 and 29 show that the weighted pattern is divided into N layers (N:
Mj is set to an integer greater than or equal to 2 (in the figure -4), and even if the subject is close to a point light source and becomes small, the first weighting (incoming tatami thread) - the second weighting (B signal system) is It becomes easier to hit both of the given light-receiving element cells equally, and the calculation accuracy improves. Furthermore, according to the light receiving element array shown in FIG.
Only one /<sofa for the B signal system is required, and the subtraction circuit can be omitted.

In addition, FIG. 31 shows that the weighting shown in FIG. 30 is a method in which four periods of the pattern are layered, and the patterns are overlapped on concentric circles.
Calculation accuracy is improved, such as by eliminating the directional characteristics of the subject.

In this way, if the weighting of the light-receiving element array is characterized by its shape, the advantages of the focus detection device according to the present invention can be further brought out.

The focus detection device according to the present invention can be used for consumer purposes such as still cameras, video cameras, microscopes, etc., and can also be used for industrial purposes such as distance measuring function application devices and collision prevention devices.

It can be widely applied to search devices and the like.

As described above, according to the present invention, the shape of the cells constituting one light-receiving lattice array is such that a weighted photocurrent is generated when the array receives light from a subject. Since the element array has weighting and integration functions, the configuration of the arithmetic processing circuit into which the photocurrent from the array is input can be greatly simplified, and a highly accurate focus detection device can be obtained. .

[Brief explanation of drawings]

Fig. 1 is a block diagram showing an example of an optical system in a focus detection device already developed by the inventor of the present invention, and Fig. 2 shows a focus detection device.
Schematic diagram showing the object and the operating state of the imaging optical system when out of focus,
Figure 3 shows the 1. An explanatory diagram showing the state of image formation on the second light receiving element array, FIG. 4 is a block diagram of the signal processing circuit of the above device, and FIG.
1 in the diagram. The second weighting is a diagram for explaining the process from circuit coefficients and spatial position conversion values to detection of spatial position information. In Figures 6 and 7, two objects (point light sources) are connected to the first light receiving element array. 8 to 12 are explanatory diagrams of the operation when the above images are formed as two images, and FIGS. 8 to 12 are explanatory diagrams of the operation in the states of (a) to (d) in FIG. FIG. 19 is an explanatory diagram of the operation in states (2) to (e) of FIG. The second weighting is an explanatory diagram of the operation when the coefficients are shifted (shifted), and FIG. 20 is the first weighting in FIG. The second weighting is an explanatory diagram of the operation when the coefficients are reversed (switched). FIGS. 21 and 22 are the first in FIG. The second weighting is an explanatory diagram of the operation when the coefficients of the circuit are weighted by an envelope obtained by a trigonometric (sine wave) function or a polygonal line function. 24 and 25 are explanatory diagrams of the operation when 11 to 16 are divided into 8 sets of cell spaces, and the transmitted light of one imaging lens is imaged on the first light receiving element array and the second light receiving element array. Optical system block diagram; FIG. 26 is a block diagram of the device shown in FIG. 1 using a parallel signal processing method; and FIG. 27 is a block diagram of an arithmetic processing circuit of a focus detection device according to an embodiment of the present invention. Block diagram showing a configuration example, Fig. 28~
FIG. 31 is a diagram showing an example of a light-receiving element array used in each of the devices. 1...Subject, 4,5...1st. second imaging optical system, 6', Te...1st. Second light receiving element array-11...
-Signal processing circuit (arithmetic processing circuit). Note that the same reference numerals in the figures indicate the same or corresponding parts. Agent Masuo Oiwa Figure 1 n Figure 2 Figure 31 Figure 6 Figure n Figure 7 Figure 28 Figure 29 Figure 30

Claims (1)

[Claims] (1) A first lens is provided for light incident from the subject through two different optical paths. a second imaging optical system, and a second imaging optical system, respectively. A first cell comprising a plurality of cells disposed at a fixed focal plane of the second imaging optical system and shaped so as to generate a weighted photocurrent across the array when receiving light from the subject. ．． a second light receiving element array; A focus detection device comprising: an arithmetic processing circuit that performs arithmetic processing on the photocurrent from each cell of the second light-receiving element array and determines an in-focus state or an out-of-focus state.