JPH0690357B2 - Deviation amount detector - Google Patents

Description

Description: TECHNICAL FIELD OF THE INVENTION The present invention relates to a deviation amount detection device for detecting a relative deviation amount of an optical image projected on a pair of photoelectric element arrays, and particularly to a focus amount based on the deviation amount. The present invention relates to a shift amount detection device that detects an adjustment state or a distance to a target object.

(Background of the Invention) A conventional shift amount detecting device of this type detects a relative shift amount of two optical images by using a photoelectric output pattern of one photoelectric element array as a photoelectric output pattern of the other photoelectric element array. On the other hand, the correlation amount of both photoelectric output patterns at the time of electrically shifting by a predetermined amount is calculated, and the pattern relative shift amount is obtained from the shift amount when the correlation becomes maximum.

Specifically, assuming that each photoelectric element array is composed of photoelectric elements Ph arranged at a pitch P as shown in FIG. 1 (a), a photoelectric output pattern composed of photoelectric output of each photoelectric element is obtained. While relatively shifting in units of the pitch P, the correlation between the two photoelectric output patterns is obtained, and the correlation amount is calculated for each pattern shift amount. The correlation amount giving the maximum correlation is obtained from the plurality of correlation amounts thus obtained, and the shift amount giving the correlation amount is obtained. However, since the shift amount is in pitch units, the shift amount that gives the maximum correlation is naturally in pitch units, and in the end, the above-mentioned optical image shift amount detection device limits the optical image shift amount detection accuracy to the pitch units. There is a drawback that. This detection device also has the following drawbacks.

As shown in FIGS. 1 (b), (c), and (d), respectively, the optical image A on one photoelectric element array and the optical image B on the other photoelectric element array relatively match each other. 2 (a), 2 (b), and 2 (c), the correlation amount C (L) with respect to each shift amount L in the cases of 0.5 P shift, 4.5 P shift, and 4.5 P shift, respectively, is shown. When the relative shift amount between the two optical images A and B is an integral multiple of the pitch P as shown in FIG. 1 (b), the shift amount (white outline) corresponding to the shift amount as shown in FIG. 2 (a). (Indicated by an arrow), the correlation amount C (L) is significantly smaller than the correlation amounts of other shift amounts, so that the shift amount that gives the maximum correlation can be easily obtained. However, the shift amount of the light image is not an integral multiple of the pitch P, and especially as shown in FIGS. 1 (c) and (d).
In the case of an odd multiple of 0.5P, as shown in FIGS. 2 (b) and 2 (c), the correlation amount is relatively smaller than the others, but almost equal to each other (indicated by a large black circle in the figure). Are only obtained for a plurality of shift amounts (indicated by arrows). Therefore,
In such a case, it is not possible to obtain the shift amount that gives the maximum correlation, or even the shift amount that is close to it, and it is impossible to detect the shift amount.

Further, as shown in FIG. 2, the number of shift amounts is, for example, L =-
A large number (21) of 10, -9, -8, ... 8, 9, 10 causes a dramatic increase in circuit scale or arithmetic scale. Therefore, the number of shifts is reduced, and for example, L = −3, −2, −
Fig. 2 (c) when the total of 1, 0, 1, 2, 3 is 7
Then, the smallest value in this shift region l is the shift amount L.
Since it appears only at one point of = 0, the relative shift amount of the optical image is
Despite the 4.5P, there is a drawback that the shift amount L = 0 is erroneously detected as the shift amount at this time.

An optical image shift amount detection device that solves this drawback is disclosed in Japanese Patent Laid-Open No. 57-45.
It is disclosed in 510. In principle, this device detects the optical image shift amount as follows. That is, the correlation amount is obtained while mutually shifting the photoelectric output patterns of the pair of photoelectric element arrays in units of the photoelectric element pitch P, and the function V (L) = C (L-1) corresponding to the difference between these correlation amounts. -C (L +
Find 1). 3 (a), (b), and (c) are respectively FIG. 1 (b),
The function V (L) for the optical images of (c) and (d) is plotted for each unit shift amount. The slope of each straight line obtained by connecting adjacent points of the plotted points with a straight line is detected, the maximum of this slope is detected, and the point at which this straight line having the maximum slope crosses the horizontal axis L is obtained. This point is the shift amount that gives the maximum correlation. As described above, in this detection apparatus, the correlation amount obtained from the shift amount in units of pitch P is interpolated by calculating the function V (L), the maximum slope straight line, and the like. ), (B), (c), as indicated by the white arrow, it is possible to accurately detect the deviation amount of the optical image to a unit smaller than the pitch P. Further, as shown in FIG. 3, the straight line having the maximum inclination does not appear except for the one relating to the maximum correlation. Therefore, if at least the shift region is selected in a wide area, there will be no erroneous detection or non-detection.

However, even this detection device has a drawback that erroneous detection occurs if the number of shifts is small, like the above detection device. For example, the number of shifts is, for example, L = −3, −
If a total of 7 of 2, -1, 0, 1, 2, 3 is set, the third
In FIG. 6 (c), the straight line with the maximum inclination in the shift region 1 is a straight line that passes L = 0, so the misregistration amount is erroneously detected as zero even though the misregistration amount of the optical image is 4.5P. Will end up. Furthermore, in order to perform the above-mentioned interpolation, that is, to obtain the above-mentioned straight line, four different correlation amounts are required, which causes a drawback that the circuit scale and the calculation scale are increased.

(Object of the Invention) An object of the present invention is to detect a deviation amount of an optical image to a unit amount sufficiently smaller than a shift amount and to prevent the erroneous detection without unduly increasing a circuit scale or a calculation scale. An object is to provide a quantity detection device.

It is an object of a second invention of the present application to provide a deviation amount detecting device capable of executing interpolation from three correlation amounts without causing a significant increase in circuit scale or calculation scale.

An object of a third invention of the present application is to provide a deviation amount detection device that is not affected by the contrast (change in pattern) of a target object and that can prevent erroneous detection without significantly increasing the circuit scale or the calculation scale. Is.

The configuration of the first invention will be described with reference to FIG.

The optical means 100 creates a pair of optical images of almost the same object. The relative positions of the pair of optical images change depending on the focus adjustment state of the objective lens or the distance to the target object. The photoelectric conversion means 101 includes first and second photoelectric element arrays 102 and 103. The first photoelectric element array 102 receives one of the both optical images and the second photoelectric element array 103 receives the other optical image. Then photoelectric conversion is performed. The correlation calculating means 104 converts the photoelectric output of the first photoelectric element array 102 or the output pattern related to the photoelectric output obtained by appropriately processing it into the photoelectric output of the second photoelectric element array or the output pattern obtained by appropriately processing the photoelectric output. On the other hand, a predetermined amount ..., L-1,
The correlation amount of both output patterns at each time when shifted by L, L + 1, ... (where L is an integer) is obtained. For example,
Correlation amount C (L-1) when only a predetermined amount L-1 shifts,
Correlation amount C (L) when shifted by a predetermined amount L, predetermined amount L
The correlation amount C (L-1) when shifted by -1 is calculated. The interpolating means 105 uses the correlation amount ... C from the correlation calculating means 104.
From (L-1), C (L), C (L + 1) ..., The extreme value correlation amount that takes an extreme value and the shift amount that gives it are interpolated. The discriminating means 106 compares whether or not the extreme value correlation amount exceeds a predetermined value. When this exceeds, the interpolation shift amount that gives this extreme value is used as a signal representing the focus adjustment state or the target object distance.

The configuration of the second invention will be described with reference to FIG.

Optical means 100, photoelectric element arrays 102 and 103, correlation calculation means 10
4 is the same as that of the first invention. The interpolating means 107 is
From the three correlation amounts C (L-1), C (L) and C (L + 1) for the three shift amounts L-1, L and L + 1, the shift amount corresponding to the extreme value therebetween is calculated. A third aspect of the invention is to remove the influence of the contrast of the target object.
The shift amount is calculated by the following equation using the value E representing contrast; Interpolate by. Here, E = MAX {C (L-1) -C
(L), C (L + 1) -C (L)}.

Embodiments of the Invention One embodiment of the present invention will be described below with reference to the drawings.

FIG. 6 shows an optical system of the focus detection device according to the embodiment.
In the figure, a field lens 2 is arranged in the vicinity of a planned image formation (primary image plane) of an image forming optical system 1 such as a photographing lens, and the field lens 2 has a rectangular light transmission area 2a in the center thereof. The area other than the area 2a is a light shielding area. The substantially rectangular parallelepiped transparent block 3 is made of a high refractive index material such as glass or plastic, and the field lens 2 is attached to one end face 3a thereof. On the other end surface 3b facing the one end surface 3a, a pair of concave mirrors 4 and 5 slightly inclined in opposite directions are provided. Both end surfaces 3a, 3b
In the block 3 between them, a pair of mirrors 6 and 7 are obliquely installed at an angle of about 45 ° with a predetermined gap. Below the transparent block 3, photoelectric conversion devices 8 are arranged, respectively. The photoelectric conversion device 8 has photoelectric device arrays 9 and 10 formed below the mirrors 6 and 7, respectively. Each photoelectric element array is composed of a plurality of linearly arranged photoelectric elements.

The light flux that has passed through the imaging optical system 1 passes through the light transmission region 2a of the field lens 2 and enters the block 3 where the mirrors 6, 7
The light enters the pair of concave mirrors 4 and 5 through the gap between them. One concave mirror 4 reflects the incident light toward the mirror 6, and the other concave mirror 5 reflects the incident light toward the mirror 7, and the respective reflected lights are reflected to the photoelectric element arrays 9 and 10 via the mirrors 6 and 7, respectively. To reach. In this way, a pair of subject images for almost the same subject are formed on the arrays 9 and 10. These object images are
The relative position on each of the arrays 9 and 10 changes according to the focus adjustment state of the imaging optical system 1. Specifically, when in focus,
The two images are relatively matched and are displaced in one direction when the front pin is used, and are displaced in the opposite direction when the rear pin is used. Such an optical system is not limited to the above-mentioned example, and even if it is a re-imaging optical system disclosed in JP-A-54-104859, the lenslet array optical system disclosed in JP-A-54-159259 is disclosed. System, but also JP-A-57
It may be the distance measuring optical system disclosed in Japanese Unexamined Patent Publication No. 40212.

The photoelectric element array 9 includes photoelectric outputs a ₁ , a ₂ , and a _{3 of} each photoelectric element.
... a _i ... are generated in time series as shown in Fig. 7 (a),
Similarly photoelectric output b ₁ also the photoelectric element array _{10, b 2, b 3 ...} b i ...
Occurs in time series as shown in FIG. 7 (b). Fig. 7
Photoelectric output pattern 9A of the photoelectric element array 9 shown in (a)
Corresponds to the intensity distribution pattern of the light image projected on the array, and the same applies to the photoelectric output pattern 10A of the photoelectric element array 10 shown in FIG. 7 (b). Therefore, the shift amount Δx between the two photoelectric output patterns 9A and 10A corresponds to the relative shift amount of the optical images on the arrays 9 and 10.

In FIG. 8, photoelectric output data a ₁ , a ₂ , ... A _i ... And b _{1 of} the photoelectric element arrays 9 and 10 of the photoelectric conversion device 8 are shown in FIG.
b ₂ , ... B _i ... Are input to the processing circuit 11. The processing circuit 11 subjects the input data to non-linear conversion such as logarithm or filtering. The photoelectric output data subjected to such processing is converted into a digital value by the A / D converter 12. MC is a microcomputer,
The functions of the solid line blocks 13 to 21 shown in the dotted line block MC are fulfilled. The memory unit 13 stores the output of the converter 12. The correlation calculation unit 14 is a series of data regarding the photoelectric element array 9.
The correlation between a ₁ , a ₂ , ... A _i ... And a series of data b ₁ , b ₂ , ... B _i ... The value is sequentially calculated while shifting by L. Specifically, the correlation amount C (L) is calculated by the following formula.

Here, L is an integer corresponding to the shift amount of the data string as described above, and the first term q and the last term r may be changed depending on the shift amount L. Of course, the correlation amount C (L) is not limited to the above equation, and Σ | a _i −b _j | ^r or Σ (−a _i ) · b _j can be used.

The memory unit 15 has three correlation amounts C when the data shift amounts are increased by one data by one amount, L-1, L, and L + 1.
_{(L-1) = C -1} , C (L) = C 0, C (L + 1) = a C ₁ respectively to memory. Extremum existence discriminating unit 16, the correlation amount C _-1, which is a memory, for C _0, C _1, compares the C _-1 and C _0, C ₀ and the magnitude relationship of each of C _1, condition C _-1 ≧ It is determined whether C ₀ and C ₁ > C ₀ are satisfied. The discriminator 16 determines a discrete correlation amount C
When (L) is plotted as shown in FIG. 9 (a), the correlation function F (shown by the dotted line) created by interpolating these correlation amounts has a minimum value between C _-1 and C ₁ . Whether or not it has is determined, and there is a possibility that a minimum value exists when the above conditions are satisfied.

The first interpolation unit 17 stores the memory unit 15 when the above conditions are satisfied.
The above-mentioned minimum value C _min is interpolated from the correlation amounts C _-1 , C ₀ and C ₁ in the following equation.

D _L = 0.5 × (C ₋₁ −C ₁ ) ………… (1) E = MAX {C ₁ −C ₀ , C ₋₁ −C ₀ } ……… (2) C _ext ＝ C ₀ − | D _L | ... (3) Here, MAX {Ca, Cb} means selecting the larger one of Ca and Cb.

This interpolation method will be described in detail with reference to FIG. 9 (b). In the figure, it is assumed that C ₀ <C _-1 <C ₁ . Conclusion the maximum value C ₁ and the minimum value C ₀ of these three correlation amount with a straight line l _1, line l is the slope and the absolute value of the straight line l ₁ is equal with the slope sign is reversed _{- 1,} pulled so as to pass through the intermediate value C _-1. The intersection of these straight lines l ₁ and l _-1 becomes the minimum value C _ext of the correlation function F. D is the distance between C ₀ and C _ext in the coordinate axis C (L) direction
_{Assuming L} , this distance D _L is obtained by the above equation (1). Therefore, the minimum value C _ext becomes C ₀ − | D _L |. The correlation amount C
Since the size of (L) changes largely depending on the pattern of the subject image, it is standardized so that the minimum value does not depend on the subject image pattern. Therefore, as a normalization factor,
Using E in equation (2), the value obtained by dividing C ₀ − | D _L | in equation (3) by E And In this way, if the minimum value is standardized so as not to depend on the subject image, the minimum value of the correlation function F, that is, the correlation amount F _m that gives the maximum correlation, as shown in FIG. It becomes a value close to zero, and the other minimum values F _e become considerably larger than that.

The memory unit 18 was calculated by the interpolation unit 17. To temporarily store.

The maximum correlation discriminator 19 is stored in this memory 18. And the reference value C _ref are compared, and when the former is smaller than the latter,
Generate comparison output. The reference value C _ref is selected as an intermediate value between the correlation amount F _m that gives the maximum correlation and the other minimum value F _e . Therefore, this discrimination unit 19 is
It is determined whether or not the interpolated minimum value F _e corresponds to the correlation amount F _m that gives the maximum correlation.

The second interpolation unit 20 responds to the comparison output of the maximum correlation determination unit 19, that is, when the minimum value F _e corresponds to the correlation amount F _m that gives the maximum correlation, the shift amount L corresponding to the correlation amount F _m. formula _m Interpolate by.

k is a coefficient for converting this shift amount L _m into a shift amount Z in the optical axis direction between the subject image and the predetermined image plane.

The memory unit 21 stores the shift amount L _m that gives the maximum correlation from the second interpolation unit 20, and the drive display unit 22 stores the front pin and the rear pin according to the shift amount L _m stored in the memory 20. Alternatively, the indication of focus is displayed, and the imaging optical system 1 of FIG. 6 is driven in the direction of the focus position. Further, in the second interpolation unit 20, the shift amount L _m is converted into a shift amount Z in the optical axis direction between the subject image and the predetermined image plane by the formula Z = k · L _m , and this image plane shift is performed. The amount Z is stored in the memory unit 21, and the drive display unit 22
May respond to this amount Z.

The processing circuit 11 may perform sampling processing of input data instead of or in addition to the above processing.

Next, this operation will be described with reference to FIG. 8 and its flowchart.
This will be described with reference to FIG.

The outputs a ₁ , ... A _{i of the} photoelectric element array 9 and the outputs b ₁ , ... B _{j of the} photoelectric element array 10 are appropriately processed by the processing circuit 11, and are A / D converted by the A / D converter 12. After that, it is stored in the memory unit 13. The microcomputer MC sets the shift amount L to zero (step in FIG. 10). The correlation calculator 14 calculates the correlation amount C (L- when the shift amounts are L-1, L, and L + 1.
1), C (L), C (L + 1) are calculated, and these correlation amounts C (L-1), C (L), C (L + 1) are stored in the memory unit 14
Are stored in the respective memories C _-1 , C ₀ , C ₁ (step). The extreme value existence determining unit 16 determines the contents of the memory C ₁ and the memory C _0.
And the contents of memories C ₁ and C ₀ respectively,
It is determined whether or not the conditions C ₁ > C ₀ and C ₋₁ ≧ C ₀ are satisfied, and when they are satisfied, it is determined that an extreme value exists between the contents of the memory C _{−1 and} the contents of the memory C ₁ (step ). First interpolator 17
Is the value from the contents of memories C _-1 , C ₀ , and C ₁ when the above conditions are met. And calculate these values Is stored in the memory unit 18 (step). Determined by this interpolation And the reference value C _ref are compared by the maximum correlation discriminator 19, It is determined whether or not is satisfied (step). When this condition is satisfied, the minimum value C _ext is the correlation function F of FIG.
Is determined to be the minimum value of Then, the value L _m is stored in the memory unit 21 (step). The drive display means 22 drives the imaging optical system toward the in-focus position according to this value L _m , and also displays the focus adjustment state.

On the other hand, when the step condition is not satisfied, it is determined whether or not the shift amount L at this time satisfies the condition L> 0 (step). Since the shift amount L is set to zero now,
The above condition L> 0 is not satisfied. At this time, the shift amount is −
L + 1, that is, updated to 1 (step). The newly set shift amount L and the predetermined number l _f are compared (step). The predetermined amount l _f is an integer value that determines the maximum shift amount. When the condition of this step is not satisfied, the process returns to the above step. When returning to this step after returning to this step, the condition L> 0 in the step is satisfied because L = 1, and the shift amount is -L,
That is, it is updated to -1 (step), and the process returns to the step again. In this way, the shift amount L is sequentially changed from 0 to 1,-
The amount of shift that gives the maximum correlation is obtained by gradually increasing the numbers of 1, 2, -2, 3, -3, .... Without getting the maximum correlation,
When the shift amount L reaches a predetermined value l _{f, in order} to show this, in steps And end.

As the imaging optical system approaches the in-focus position, the shift amount that gives the maximum correlation naturally becomes smaller.
Gradually increasing the shift amount from zero in order to obtain the maximum correlation is generally advantageous in that a quick focus adjustment state detection or distance measurement can be achieved.

In the above correlation calculation, for example, the correlation amount C when L = 0
(L) and C (L + 1) are the correlation amounts C (L
−1), C (L), and the correlation amount C (L
−1) and C (L) are the correlation amount C when L = −1, respectively.
(L), equal to C (L + 1). Similarly, C (L) and C (L + 1) when L = 1 are C (L−) when L = 2, respectively.
1), equal to C (L), and C (L- when L = -1
1) and C (L) are C (L) and C (L + when L = -2.
Equal to 1). Despite such a relationship, in the flow chart of FIG. 10, when L changes, three correlation amounts C (L-
1), C (L), and C (L + 1) were all newly calculated.
Since the calculation of C (L) takes a considerable amount of time, this means an increase in the calculation time.

Therefore, an improved example that prevents such an increase in calculation time is given below.
Shown in Figure 11.

In FIG. 11, In the correlation amount C (L-1), C (L), after calculating the C (L + 1), these respectively memory C _-1, stores the C _0, C _1, the memory C _0, C ₁ Content of C (L), C (L +
1) respectively memory _{^{▲ C + -1 ▼, ▲ C}} + 0 in ▼, similarly memory C _-1, respectively the contents of the C ₀ 's memory ^{_{▲ C - 0 ▼, C -}} 1 stored in ▼. After the shift amount is increased by 1 in step, the correlation amount C (L + 1) is newly calculated in step and the content of the memory C ₁ is updated with this value. The contents of the memories C _-1 and C ₀ are respectively updated with the contents of the memories ▲ C ⁺ _-1 ▼ and ▲ C ⁺ ₀ ▼ stored in
The memories ▲ C ⁺¹ _-1 ▼ and ▲ C ⁺¹ ₀ ▼ are updated with the contents of C ₀ and C ₁ , respectively. Memory updated in this step C _-1 , C ₀ , C
The contents of ₁ are processed in steps. Similarly, after the sign of the shift amount is inverted in step, the correlation amount C (L-1) is newly calculated in step, and the value C
(L-1), and a memory _{^{▲ C - 0 ▼, ▲ C}} - 1 respectively at contents of ▼ memory C _-1, and updates the contents of C _0, C _1, the memory C _-1, the content of C ₀ To update the contents of memory ▲ C ^- ₀ ▼, ▲ C ^- ₁ ▼. Memory updated in this step C _-1 , C
The contents of ₀ and C ₁ are also processed in steps. Others are 10th
It is the same as the figure.

In the above description, the shift amount is sequentially increased from zero to the positive side and the negative side alternately. Next, an example of increasing the shift amount in one direction will be described with reference to the flowchart of FIG.

In FIG. 12, the shift amount L is set to a negative initial value l _i in step. At step, the correlation amount C (L) is calculated and stored in the memory C _-1 . In the step, L is increased by 1, and in the step, the correlation amounts C (L) and C (L +
1) is calculated and stored in the memories C ₀ and C ₁ , respectively.
In step, compare contents of memories C ₁ and C ₀ , and condition C ₁
When> C ₀ is satisfied, the contents of memories C ₋₁ and C ₀ are compared in a step, and when the condition C ₋₁ ≧ C ₀ is satisfied, these C ₋₁ and C ₁
Since there is a minimum value between and, D _L , E
From this value D _L , E and C ₀ To interpolate. In step this Is smaller than the reference value C _ref , this is determined to be the correlation amount that gives the maximum correlation, the shift amount L _m for this maximum correlation is interpolated in step, and the sequence ends.

On the other hand, when the condition C ₁ > C ₀ is not satisfied in the step, the shift amount is further increased by 1 in the step. If the shift amount is smaller than the positive final value l _{f in} the step, the correlation amount C (L + 1) is newly calculated in the step, the contents of the memory C ₁ are updated with this, and the memories C _-1 , C
The contents of ₀ to update the contents of the memory C _0, C _1. After this, return to the step. Further, when the step condition is not satisfied, even if L is increased by 1, the shift amounts L-1 and L at that time are increased.
It is clear that we cannot get a local minimum between +1 and so we increase L by 2 in steps. Also, when the condition of the step is not satisfied, the step is moved to the step for the same reason. In the step, when L after this increase is smaller than the final value l _f , the contents of the memory C _-1 are updated to the contents of the memory C _{1 in} the step, and then the process returns to the step.

In step or, when L reaches the final value l _f , in order to indicate that in step The reference value C _ref is stored in and the sequence ends.

In addition, the above-mentioned embodiment, Since the value E used for the interpolation calculation of is also used as a normalization factor, there is an advantage that it is useful for shortening the calculation time. However, if such an advantage is ignored, any value related to the contrast of the subject light image can be selected as the normalization factor, and for example, the value disclosed in Japanese Patent Laid-Open No. 55-98709, that is, the difference between adjacent pixels, can be selected. An autocorrelation function given as the sum of absolute values can also be used, and Max {C (L) -C
(L-1)}, Max {C (L + 1) -C (L-1)}, Max {C (L)}, or the like can also be used.

Next, an example in which the minimum value is not standardized will be described with reference to FIG.

In the shift regions l _{i to} l _f , which of the plurality of local minimum values corresponds to the minimum value cannot be determined from the individual local minimum value values unless the local minimum value is normalized. Can not. Therefore, as described below, it is necessary to calculate all the minimum values in the above-mentioned region and obtain the minimum value from them.

In FIG. 13, in step, an arbitrary initial value A _0, which is sufficiently larger than a normal minimum value, is stored in the memory C _min , and a memory L _{0 is} stored.
Store arbitrary value A ₀₀ in each. Step ~ is the 12th
It is the same as the steps 1 to 3 in the figure. In the step, 0.5 × (C ₋₁ −C ₁ ) operation is performed and the result is stored in the memory D
And the calculation result of C ₀ − | D | is stored in the memory C _ext . The minimum value of the correlation amount thus interpolated is stored in the memory C _ext . In the step, the contents of the memories C _ext and C _min are compared, and the condition C _ext <C
When _min is satisfied, in step, the contents of memory C _min the contents of the memory C _ext of this time, the contents of the memory L ₀ with the contents of the memory L representing the shift amount at this time,
Further, the contents of the memory D _L are updated with the contents of the memory D, respectively. Further, the result of the operation Max {C ₁ −C ₀ , C ₋₁ −C ₀ } is stored in the memory E.
To store. In this way, the memory C _min is updated with C _ext that satisfies the step condition, so that the memory C _min is updated.
_The minimum value that has been calculated so far is always stored in min. After the step or when the condition of the step is not satisfied, the step is performed. Steps ~ are the same as steps ~ in Fig. 12. If the condition of the step is not satisfied, move to the step.
Steps ~ are the same as steps ~ in Fig. 12. The shift amount L is the final value l _f in step or
Is reached, the shift amount L _m corresponding to the minimum value C _min of the correlation amount in the shift regions l _{i to} l _f is calculated in step. Interpolated by

(Effect of the Invention) According to the first invention of the present application, since the control means for repeatedly controlling the correlation operation, the interpolation operation, and the discriminating operation are provided until an appropriate judgment is made, a minimum shift amount is provided. The true extreme value can be discriminated and an accurate shift amount can be calculated, an erroneous extreme value can be excluded, and the circuit scale and operation time do not increase.

Further, according to the second invention of the present application, the interpolation means for obtaining the interpolated extreme value using the three representative values and the correlation calculation, the interpolated extreme value calculation and the discriminating operation are repeated until an appropriate decision is made. Since the control means for controlling is provided, it is possible to determine whether or not the deviation can be detected with the minimum shift amount, and the circuit scale and the calculation time are not increased.

Further, according to the third invention of the present application, an interpolating means for normalizing the interpolated pole value obtained by interpolating using the value E representing the contrast of the target object and obtaining the shift amount corresponding thereto,
By providing a discriminating means for discriminating whether or not the normalized interpolated extremum value exceeds a predetermined value, it is possible to eliminate the influence of the change in the pattern of the target object, to distinguish the true extremum value, and to accurately shift the value. The amount can be calculated, and the circuit scale and the calculation time are not increased.

[Brief description of drawings]

FIG. 1 (a) is a diagram showing an arrangement of photoelectric element arrays, FIG.
1 (b) to 1 (d) are diagrams showing the relative positional relationship between two optical images, and FIGS. 2 (a) to 2 (c) are correlations with the shift amount by the conventional shift amount detecting device. FIG. 3 (a) to FIG. 3 (c) are diagrams showing the relationship between the shift amount and the correlation amount by another conventional shift amount detecting device, and FIG. 4 is a configuration of the present invention. 5 is a block diagram showing the configuration of the second invention, FIG. 6 is a perspective view showing the optical system of one embodiment of the present invention,
7 (a) and 7 (b) are diagrams showing photoelectric output waveforms of the photoelectric element array, FIG. 8 is a block diagram showing the above embodiment, and FIG. 9 (a) is a correlation amount according to the above embodiment. FIG. 9 (b) is an explanatory diagram for explaining the interpolation method, and FIGS. 10 to 13 are flowcharts showing the operation of the present embodiment. 100 ... Optical means, 102, 103 ... Photoelectric element array 104 ... Correlation calculation means, 105 ... Interpolation means 106 ... Discrimination means, 107 ... Interpolation means

Claims (7)

[Claims] 1. Optical means for producing first and second optical images of a target object whose relative position changes according to the focus adjustment state of the objective lens or according to the distance to the target object, respectively. Photoelectric conversion means including first and second photoelectric element arrays for photoelectrically converting the first optical image and the second optical image, respectively, an output pattern related to photoelectric output of the first photoelectric element array, and the second photoelectric element. The amount of shift relative to the output pattern related to the photoelectric output of the element array ..., L-1, L,
L + 1, ... (where L is an integer), the correlation calculation means for obtaining the correlation amount of both output patterns at the time of shifting, and a plurality of representative values smaller than the correlation amount are obtained from the plurality of correlation amounts. An interpolating means for obtaining a predetermined function from the representative value of, and an interpolating pole value obtained by interpolating from the function, and an interpolating shift amount giving the interpolating pole value, and whether or not the interpolating pole value exceeds a predetermined value. If the interpolating pole value does not exceed the predetermined value, the shift amount is further increased by the correlation calculating means to control the correlation by determining the determining means, the correlation calculating means, the interpolation means, and the determining means. When the interpolated extreme value exceeds the predetermined value, the interpolated extreme value is obtained by the interpolated means based on the plurality of correlation amounts and the interpolated extreme value is obtained by the interpolated means. Set the shift amount to the above focus adjustment state. Shift amount detection apparatus characterized by comprising a control means for outputting a signal representative of the distance. 2. The interpolation means calculates the interpolation pole value by obtaining two straight lines l1 and l2 which are the predetermined functions, using the three correlation values as the representative value, and also calculates the interpolation pole value. A value E related to contrast is obtained, and the determining means determines whether or not the interpolation pole value exceeds the predetermined value by using the value E in a form that does not depend on the contrast of the target object. The deviation amount detecting device according to claim (1). 3. The correlation means obtained by the correlation calculating means by the interpolation means starting from the vicinity of zero and gradually increasing the shift quantity L + 1, L-1, L + 2, L-2 ,. Claim 3 characterized in that the operation for obtaining the interpolated pole value from the above is performed, and the discrimination means ends the operation of the interpolated means when the interpolated pole value exceeds the predetermined value. Deviation amount detection device according to item (1). 4. Optical means for producing first and second optical images of a target object whose relative position changes depending on the focus adjustment state of the objective lens or the distance to the target object, respectively, Photoelectric conversion means including first and second photoelectric element arrays for photoelectrically converting the first optical image and the second optical image, respectively, an output pattern related to photoelectric output of the first photoelectric element array, and the second photoelectric element. The amount of shift relative to the output pattern related to the photoelectric output of the element array ..., L-1, L,
L + 1, ... (where L is an integer), the correlation calculation means for obtaining the correlation amount of both output patterns at the time of shifting, and at least three representative values are obtained from the plurality of correlation amounts,
Obtaining the slope of a straight line l1 connecting the maximum value and the minimum value of the representative values, and obtaining the straight line l2 having the same slope and the opposite sign, and passing through the remaining representative values. An interpolating means for interpolating the one shift amount from both straight lines to calculate an interpolated extreme value, and a misalignment detection propriety judging means for judging whether or not the misalignment detection is possible based on the magnitude of the interpolated extreme value. Controlling the correlation calculating means, the interpolating means, and the deviation detection possibility determining means, and when it is not determined that the deviation can be detected, the correlation calculating means further increases the shift amount to obtain the correlation amount, A deviation amount detecting device, comprising: a control unit that obtains an interpolated extreme value by the interpolation unit based on a plurality of correlation amounts and controls so as to repeat the determination operation of the displacement detection possibility determination unit. 5. The shift detection availability determination means determines a value associated with the contrast of the target object and determines the availability of the shift detection using the value associated with the contrast and the interpolated extreme value. Claims characterized by
The deviation amount detection device according to the item (4). 6. The interpolation means starts the predetermined shift amount L from a predetermined value, and successively increases or decreases gradually in a direction away from the predetermined value, or alternately shifts the increase and decrease gradually by the correlation calculation means. The operation of obtaining the interpolated extreme value from the obtained correlation amount is performed, and when the shift detection propriety determination means determines that the shift can be detected, the operation of the interpolated extreme value by shifting is sequentially ended. The deviation amount detecting device according to claim (4), which is characterized in that. 7. Optical means for producing first and second optical images of a target object whose relative position changes according to the focus adjustment state of the objective lens or according to the distance to the target object, respectively. Photoelectric conversion means including first and second photoelectric element arrays for photoelectrically converting the first optical image and the second optical image, respectively, an output pattern related to photoelectric output of the first photoelectric element array, and the second photoelectric element. The amount of shift relative to the output pattern related to the photoelectric output of the element array ..., L-1, L,
L + 1, ... (where L is an integer), the correlation calculation means for obtaining the correlation amount of both output patterns at the time of shifting, and the value E associated with the contrast of the target object are obtained, and the plurality of correlation amounts are obtained. The interpolated extreme value obtained by interpolating from
Interpolation means for calculating the standardized interpolated extreme value and the interpolated shift amount giving the standardized interpolated extreme value which does not depend on the contrast of the target object, and the standardized interpolated extreme value exceeds a predetermined value. Determination means for determining whether or not, and a control means for outputting the interpolation shift amount at that time as a signal representing the focus adjustment state or the distance when the standardized interpolation pole value exceeds the predetermined value. A deviation amount detecting device comprising: