JPH04230715A - Focus detector - Google Patents

Abstract

PURPOSE:To provide the focus detector (range finder) easy for optical adjustment, high in mass-productivity and precision. CONSTITUTION:In order to compensate influence of aberration of a focus detector optical system or adjustment deficiency, compensation data responding to positions of photoelectric element arrays 105 and 106 are stored, the part of an optical image on the photoelectric element array and the extent of contribution to relative deviation quantity which is the focus detection result are sought, and the relative deviation quantity which is the result of focal detection is compensated based on the before mentioned compensation data and the before mentioned contribution data.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a focus detection device for detecting the relative displacement of optical images of substantially the same object formed on a pair of photoelectric element arrays, and more particularly to a focus detection device for a camera or the like. It is used in distance measuring equipment.

0002

2. Description of the Related Art FIG. 1(a) shows an optical system of a focus detection device of a general camera, etc.
01, the image is formed near the field lens 102. This primary image of the subject 100 is formed as a secondary image by first and second re-imaging lenses 103 and 104 onto first and second photoelectric element arrays 105 and 106, respectively. The relative positional relationship of the secondary images on this array 105, 106 is detected from the image output of the array.

In order to perform such focus detection with high precision, each point of the optical image distributed on the photoelectric element array 105 and each corresponding point of the optical image distributed on the photoelectric element array 106 must be aligned. When in focus, the points must be relatively perfectly aligned with each other, and when out of focus, the corresponding points must be shifted by the same amount depending on the amount of out of focus. Of course, it is also possible to set each corresponding point to be shifted by a certain amount when in focus.

0004

However, due to aberrations or insufficient adjustment of the focus detection optical system consisting of the field lens 102, re-imaging lenses 103, 104, etc., the above conditions are not satisfied, and for example, when focusing, the image As shown in 1(b), the image point 107a, 1 at the center point 100a of the subject 100
08a is imaged at the same position (for example, the center) of each array 105, 106, but the image point 107b of the peripheral point 100b is
, 108b are imaged with relative shifts.

For this reason, a difference occurs between the focus detection results based on the point images 107a and 108a and the focus detection results based on the point images 107b and 108b, making correct focus detection impossible. In order to eliminate such differences in the relative position of each array and the optical image on it depending on the location of the array (hereinafter referred to as positional deviation), a focus detection optical system with extremely small aberrations is adopted. Moreover, optical adjustments must be made extremely carefully, which increases the cost of the focus detection optical system and significantly reduces mass productivity.

The above-mentioned problems are not limited to the pupil-splitting type focus detection device shown in FIG. 1, but also occur in devices that photoelectrically detect a pair of optical images of the same object and measure distance or focus from their relative positions. This is common when performing detection. An object of the present invention is to provide a highly accurate focus detection device (distance measuring device) that is easy to optically adjust and has high mass productivity.

In order to achieve the above object, the present invention corrects the effects of aberrations and insufficient adjustment of the focus detection optical system.
In addition to storing correction data according to the location of the photoelectric element array, determining which part of the optical image on the photoelectric element array contributed to the relative shift amount that is the focus detection result, and calculating the correction data and the contribution. The relative shift amount, which is the focus detection result, is corrected based on the degree data.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. In FIG. 2, a pair of photoelectric element arrays 105 and 106, each consisting of an image sensor such as a CCD, are juxtaposed in a photoelectric conversion unit 201. An image of the same subject is formed on each photoelectric element array by a focus detection optical system similar to that shown in FIG. The image outputs a1...aN from the array 105 are sequentially and time-series A/D converted by the A/D converter 202 and stored in the data memory means 204 in the microcomputer 203, and in exactly the same way, the image outputs b1 from the array 106 are ...bN is also stored in the data memory means 204 via the A/D converter 202.

The image shift calculation means 205 calculates the relative shift amount between the two image outputs, that is, the relative shift amount between the light images on the pair of arrays, based on the pair of image outputs a1...aN, b1...bN. Of course, the data used for the image shift calculation does not necessarily have to be the direct image output of the array, but may be image output obtained by appropriately filtering or sampling these outputs.

The correction data storage means 207 stores correction data for correcting the output of the image shift calculation means 205. This correction data is determined as follows. That is, after the focus detection optical system has been adjusted to some extent during the manufacture of the focus detection device, the array 105 in the in-focus state is
, 106 is measured, and the relative positional deviation amount for each image point on the array is stored in the storage means 207 as the correction amount.

To give a specific example, as shown in FIG. 1(b), when the position of the photoelectric elements of the arrays 105 and 106 in the arrangement direction is x and the center of each array 105 and 106 is xo, the array 105 , 106 is relatively easy to adjust so that the corresponding image points of the light images on 106 coincide at the time of focus at the center xo of the array. Therefore, when adjusted in this manner, the relative positional deviation amount of each point has a characteristic S(x) shown by a solid line at an in-focus point, and a characteristic Z(x) at an out-of-focus point, as shown in FIG. 3, for example.

Here, the in-focus characteristic S(x) shows that the amount of relative positional deviation is zero at the array center xo due to the above adjustment, and increases linearly from there toward the periphery of the array. The focus characteristic Z(x) is translated from the characteristic S(x) by an amount ZT corresponding to the degree of out-of-focus. This positional shift S(x) of the in-focus point is stored in the correction data storage means 207.

As this storage method, it is not preferable to store all the values of S(x) for each location x in the array because the amount of stored data increases. Therefore, when the amount of positional deviation S(x) can be approximated by a straight line passing through the origin xo as shown in FIG.
) can be specified, so data representing this slope can be stored in the storage means 207. In this case, the correction data storage means 207 can be extremely simplified, and as shown in FIG.
Adjust this potentiometer to represent the slope of 40
The output of 1 is sent to the microcomputer 2 via the A/D converter 402.
03 can be configured.

4 as a correction data storage means 207.
A ROM storing the above-mentioned inclination as shown in (a) may also be used. In this ROM, each of the output terminals P1 to P8 is also connected to the power line Vcc and the earth line Et, and each output terminal P1 to P8 is connected to the power line and the earth line depending on the data to be stored. It is processed so that it connects to only one side. Further, the correction data storage means 20
7 may also be a ROM in the microcomputer 203.

If the correction data S(x) is not a straight line but a curved line as shown in FIG. Bye. Alternatively, correction data regarding several predetermined positions xi may be stored, and if correction data for an intermediate position is required, for example, Lagrange interpolation may be used.

Returning to FIG. 2 again, the correction calculation means 206 corrects the output Z(x) of the image shift calculation means 205 using the correction data S(x) of the correction data storage means 207. )-S(x) to calculate the corrected image shift amount ZT. This corrected image shift amount ZT is converted into a defocus amount (the amount of shift between the subject image and a predetermined image formation of the photographic lens) by the correction calculation means 206, and is sent to the display drive means 208. This means 208 displays the focus adjustment state based on the defocus amount and drives the photographing lens to the in-focus position.

This effect will be described below. pair of arrays 1
Image output a1...aN, b1...b from 05, 106
N is stored in the data memory means 204 after A/D conversion. The image outputs a1...aN stored in the data memory means 204 are illustrated in FIG. 6(a). Image shift calculation means 2
05, this image output is divided into five regions X-2, X-1, X0, X1, for example, as shown in FIG. 6(b) or (c).
, X2, and in exactly the same way, image output b1...bN is also divided into five regions The partial image shift amount Z(xi) regarding xi is calculated individually.

The correction calculation means 206 calculates the partial correction amount S(xi) for the location xi from the contents of the correction data storage means 207. This partial correction amount S(xi) is obtained by substituting x=xi into the function S(x) as shown in FIG. Thereafter, the correction calculation means 206 subtracts the partial correction amount S(xi) from the partial image deviation amount Z(xi) to obtain the corrected partial image deviation amount ZT.
(xi) is obtained. That is, ZT(xi)=Z(xi)-S
(xi) is obtained.

In this way, the correction calculating means 206 obtains the corrected partial image shift amount ZT(xi) for each partial region Xi, and calculates, for example, the simple average of these values ΣZT(xi)/5 as the final image shift amount. Then, this is converted into a defocus amount and sent to the display driving means 208. Note that various methods for determining the final image shift amount ZT from the partial image shift amount ZT(xi) regarding each partial area Xi can be considered depending on the purpose, and some examples will be shown below. (A1) As mentioned above, the corrected partial image shift amount ZT(xi
) is defined as the final image shift amount ZT. (A2) The average value of the remaining corrected partial image deviation amounts after excluding the maximum and minimum corrected partial image deviation amounts ZT(xi) is set as the final image deviation amount ZT. (A3) When the corrected partial image shift amounts ZT(xi) are arranged in descending order, the center one is set as the final image shift amount ZT. (A4) Let the corrected partial image shift amount ZT regarding the partial region with the largest information amount E(x), which will be described later, be the final image shift amount ZT. (A5) The average value of the corrected partial image shift amounts ZT(xi) for the plurality of partial regions having relatively large amounts of information, or the average value obtained by adding weights according to the information amounts, is set as the final image shift amount ZT.

[0020] As an algorithm for calculating the partial image shift from the data of each partial area Xi, for example, a means of Fourier transforming the image output and comparing the phases (Japanese Unexamined Patent Application Publication No. 1983-1999) is used.
104859) or means for calculating the shift amount that provides the maximum correlation (Japanese Unexamined Patent Publication No. 57-45510). When the number of photoelectric conversion elements included in the partial region Xi is small, it is more accurate to use the Fourier transform method described above.

[0021] When an object with depth is imaged on the array, if the amount of image shift is calculated using the image output of the entire area of the array, it is completely unknown which part of the depth object will be automatically focused. becomes. Such problems regarding depth objects can be solved as follows by calculating the partial image shift amount Z(xi) for each partial area Xi as described above. (B1) If the one with the smallest image shift amount is selected from among the plurality of partial image shift amounts Z(xi) and the final image shift amount ZT is calculated based on this, the defocus amount for the closest part of the depth object can be calculated. On the other hand, by selecting the largest partial image shift amount, the defocus amount for the far part can be obtained, and by selecting the intermediate one, the defocus amount for the intermediate distance part can be obtained. (B2) If there is a plurality of partial image shift amounts Z(xi) that have approximately the same value, select that value and calculate the final image shift amount ZT based on it, and the object that occupies a relatively wide area can be captured. It is possible to obtain the defocus amount for . (B3) Partial area Xi with the largest amount of information, which will be described later
Select the partial image shift amount in , and based on this, the final image shift amount Z
If we calculate T, the subject that has the most information for focus detection,
Generally, it is possible to obtain the defocus amount for a subject with the best contrast.

Next, a specific example of calculating the amount of partial image shift as described above will be explained using a flowchart. Figure 7
In step (2), the partial image deviation amount Z(xi) and the information amount E(xi) in each partial area Xi are calculated by the image deviation calculation means 205. Here, the amount of information E(xi) represents the reliability of the corresponding partial image shift amount Z(xi), and the larger the value of this information amount, the higher the accuracy of the corresponding partial image shift amount. Specifically, as for the amount of information, if image shift calculation is performed by phase comparison after Fourier transform, then the value related to the amplitude after Fourier transform (Japanese Patent Laid-Open No. 5
r1, r1', r2, r2' of Publication No. 5-98710
is applicable. ) can be used, and when the image shift calculation is a correlation method, an autocorrelation value Wm, which will be described later, can be used.

[0023] In step (2), each of the partial areas Xi
The amount of information E(xi) is compared with a predetermined threshold Eth, and a partial area Xi having an amount of information E(xj) larger than this threshold is selected. In step (3), the correction data storage means 2
The partial correction amount S(xj) in the selected partial area xj is calculated from the content S(x) of 07, and the partial image deviation amount Z(xj) in the selected partial area Xj is calculated as the partial image deviation amount Z(xi ) to choose from.

In step (2), the corrected partial image shift amount ZT(xj) regarding the selected area Xj is calculated as ZT(xj)=
Determine from Z(xj)-S(xj). In step ■, the corrected partial image shift amount ZT (xj
) is smaller than a predetermined value ΔZ, specifically, it is determined whether the difference between the maximum value and the minimum value of the corrected partial image shift amount ZT (xj) is smaller than a predetermined value. , when the subject is small, it is determined that the subject has no depth and the process moves to step (2), and when it is not small, it is determined that the subject is deep and the process moves to step (2).

[0025] In step (2), for example, the above (A1) to (
The final deviation amount ZT is determined by one of the processes in A5). In step (2), the final image shift amount ZT is determined, for example, by any one of the processes (B1) to (B3) described above. In the above embodiment, a plurality of partial image shift amounts are calculated from a pair of image outputs, but next, a second embodiment will be described in which a single image shift amount is calculated.

FIG. 8(a) shows the data memory means 20 of FIG.
A pair of data strings A1...AN, B1...BN stored in
Shows one side. This data string may be the image output itself of the photoelectric element array as described above, or the image output obtained by filtering or sampling it. In FIG. 2, the image shift calculating means 205 calculates a pair of data strings A1 stored in the data memory means 204.
...AN, B1...BN, one data string A1...AN
is shifted by a predetermined amount with respect to the other data string B1...BN, and the correlation amount C(L) for each shift amount L is determined.

[0027] That is,

[0028]

[Math 1]

[0029] The shift amount L that minimizes this function C(L)
Find m as the amount of image shift. Since the image shift amount Lm obtained by the correlation calculation includes errors caused by insufficient adjustment of the optical system as described above, this image shift amount Lm must be corrected using the correction data S(x). No. However, this image shift amount is the data string A1...AN, B1...B
Since the calculation is made from the entire area of N, the problem is which area of the correction data S(x) is used as the correction amount.

This problem is solved as follows. Data strings A1...AN, B1...BN do not all contribute equally to the correlation amount C(L) as shown in FIG. 8(a).
), the part Y where the data string changes rapidly contributes greatly, and the part where the change is gradual makes a small contribution. That is, it depends on the contrast of each part. Therefore, by finding the degree of contribution (hereinafter referred to as contribution degree) for each location x of the data strings A1...AN, B1...BN, and finding the correction amount from the contribution correction data S(x) corresponding to this location, good. This degree of contribution Wm can be determined, for example, from the difference between adjacent data in a data string.

That is, Wm=|Am-Am+1| or |Bm-Bm+1
｜. This value Wm is shown in FIG. 8(b). Of course as Wm |
Am-Am+1|+|Bm-Bm+1| can also be used. The correction amount ST is

[0032]

[Math 2]

Here, Sm is s(x
). Therefore, the corrected image shift amount ZT can be found from the following equation. ZT=Lm-ST Note that the amount of positional deviation depends on the location between the relative position of the first photoelectric element array and the optical image thereon and the relative position of the second photoelectric element array and the optical image thereon. If is significantly large,
It may be difficult to correct the amount of image shift with high precision using the correction data S(x). Therefore, by changing the pitch of the photoelectric elements of the first and second photoelectric element arrays depending on the location, the amount of positional deviation is corrected to a certain extent in advance,
Even so, it is desirable to store the remaining positional deviation amount as correction data.

[0034]

As described above, according to the present invention, the relative relationship between the optical image on the first photoelectric element array and the optical image on the second photoelectric element array corresponding to each other in a substantially focused state is determined. In addition to storing correction data related to the amount of positional deviation of the target, it is also possible to evaluate which part of the optical image contributed to what extent to the relative deviation amount that is the focus detection result, and to Since the corresponding degree of contribution is calculated and the amount of relative shift is corrected using this correction data and the degree of contribution data, highly accurate focus detection (distance measurement) is possible.

[Brief explanation of the drawing]

FIG. 1(a) is an optical diagram showing the relationship between a general focus detection optical system and a photoelectric element array. (b) is a front view showing the optical image position on the photoelectric element.

FIG. 2 is a block diagram showing an embodiment of the present invention.

FIG. 3 is a graph showing an example of relative positional deviation amount.

FIGS. 4(a) and 4(b) are circuit diagrams showing specific configuration examples of correction data storage means.

FIG. 5 is a graph showing another example of relative positional deviation amount.

FIG. 6(a) is a graph showing an example of image output. (b) and (c) are diagrams showing how the image output is divided into a plurality of regions.

FIG. 7 is a flowchart showing specific operations of the first embodiment.

FIG. 8(a) is a graph showing another example of image output. (b) is a graph showing the intensity of change in image output.

[Explanation of symbols of main parts]

105, 106...Photoelectric element array 205...Image shift calculation means 206...Correction calculation means 207...Correction data storage means

Claims (3)

[Claims] (1) First and second photoelectric element arrays;
a focus detection optical system that forms a pair of optical images having parallax with respect to a subject on a second photoelectric element array; and a pair of optical images of each of the photoelectric element arrays based on the photoelectric outputs of the first and second photoelectric element arrays. and an image shift calculation means for detecting a relative shift amount of the image shift calculation means, and performs focus adjustment based on the output of the image shift calculation means, wherein the first photoelectric element array is substantially in focus. storage means for storing correction data related to relative positional deviation amounts for each corresponding image point between the optical image on the second photoelectric element array and the optical image on the second photoelectric element array; contribution calculation means for evaluating which part of the optical image contributes to what extent in detecting the amount of misalignment, and calculating a degree of contribution according to the part of the optical image; and the correction data in the storage means. and correction calculation means for correcting the detected relative shift amount based on the contribution data of the contribution calculation means, and the focus adjustment is performed based on the correction result by the correction calculation means. A focus detection device characterized by performing the following. (2) The focus detection device according to claim 1, wherein the contribution degree of the contribution degree calculation means is data related to the contrast of each portion of the optical image. (3) The focus detection device according to claim 2, wherein the contrast corresponds to a difference between the adjacent photoelectric outputs in a portion of the optical image.