JPH05264892A - Automatic focusing device - Google Patents

Abstract

PURPOSE:To obtain an automatic focusing device capable of continuously performing a stable range finding operation even in the case that the influence of flare light by a light source outside a range finding field is given to an object in a range finding field by correcting a light quantity difference between two images. CONSTITUTION:The object images divided into two by a range finding optical system 23 are separately image-formed on two photoelectric conversion element trains 30(L) and 30(R) again. Then, after the reliability is decided at first, the light quantity difference between two object images which are formed again is obtained. In the case that the object includes both of far and close objects and the contrast of the object is low, the light quantity difference between two images becomes small, and when the light quantity difference between two images is equal to or exceeding a prescribed value, it is decided that there is a high possibility that the influence of the flare light is given. Then, in the case that the light quantity difference between two images is equal to or exceeding the prescribed value, a distance between the images is obtained after correcting the light quantity difference, and then, the reliability with reference to the obtained distance between the images is decided. Thus, the accurate range finding performance is obtained without making it impossible for the camera to perform range finding even in the case the influence of the flare light is given.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic focusing device used in a camera or the like, and more particularly, it divides a subject image obtained through a photographing optical system into two images and re-images them, and photographs from the image interval. The present invention relates to a device for obtaining the defocus amount of an optical system.

[0002]

2. Description of the Related Art Generally, an automatic focusing device for dividing a subject image obtained by passing through a photographing optical system into two images, re-imaging, and obtaining a defocus amount of the photographing optical system from the image interval. There is. Such an automatic focus adjusting device determines the defocus amount of the photographic optical system by comparing the two re-formed images and determining the image interval having the highest coincidence.

Therefore, in order to improve the distance measurement accuracy,
It is required that the two re-imaged images have the same light intensity and the same waveform, and that the images have higher coincidence.
On the contrary, there has been proposed an automatic focus adjustment device that determines whether or not the detected image distance is reliable by verifying the coincidence of these two images. For example, Japanese Patent Publication No. 62-7
Japanese Patent No. 523 discloses a method of determining reliability by comparing a comparison value of two images with a contrast value of the images themselves. Also, JP-A-60-3751
Japanese Patent Publication No. 3 discloses a method of determining reliability based on a plurality of comparison values of two images. With these methods,
If it is determined that the detected image interval is not reliable, the camera determines that distance measurement is impossible and suspends the distance measurement operation.

[0004]

However, the above-described automatic focusing apparatus uses a photographing optical system composed of a plurality of lens groups of a camera, and there is a strong light source in or near the screen to be photographed. It is known that a flare phenomenon (hereinafter referred to as flare) occurs due to reflection or the like on the lens surface.

Optical designs and lens coatings have been conventionally tried to prevent the flare. However, as the size and weight of the camera are reduced and the cost is reduced, the flare does not affect the image quality. Has been accepted. However, when such flare occurs,
As shown in FIG. 1 and FIG. 12, accuracy is excessively affected in distance measurement. FIG. 11 is a diagram showing the principle of the phase difference method.

The image of the object O is formed on the film equivalent surface 4 after passing through the taking lenses 1, 2 and 3. The image formed on the film equivalent surface 4 is divided into two by the re-imaging lens 6 and is re-formed as two images L and R on the photoelectric conversion element array 7. The photoelectric conversion element array 7 converts the re-formed object images L and R into electric signals, and compares the electric signals of the two converted images to compare the re-formed object images L and R. Find the image distance. Next, the subject O outside the distance measuring field
FIG. 12 shows a situation in which there is a subject M that is significantly brighter than that of FIG.

The image of the subject M passes through the taking lenses 11, 12, and 13 and is imaged on the film equivalent surface 14, as shown by the solid line in FIG. However, the visual field mask 15 placed near the film equivalent surface 14 to limit the distance measuring visual field.
And is not re-imaged on the photoelectric conversion element array 17.

However, as shown by the dotted line in FIG. 12, when internal reflection occurs between the photographing lenses 11 and 12, in most cases, there is no regular image forming relationship, and the flare light is photoelectrically converted. It is projected onto the conversion element 17. The amount of light reflected on the inner surface is extremely small, and if the amounts of light of the objects O and M are about the same, there is no problem at all. However, when the light quantity of the subject M outside the distance measuring field is more than several hundred to several thousand times the light quantity of the subject O in the field of view, the re-formed object image L, There is some effect on R. FIG. 13 is a diagram showing an electric signal of the re-formed object image converted by the photoelectric conversion element array, and the influence on the object image will be described based on this figure.

In FIG. 13A, when there is no flare light, the object images L and R are detected as the same light intensity and the same waveform, but as shown in FIGS. 13B and 13C, the flare light is detected. When the image occurs, the subject images L and R have different light intensities,
Produces different waveforms. For such an image signal, 2
When the reliability of the detected image interval is further determined by obtaining the image interval, it is often determined that the reliability is not high,
The camera is in a state where it cannot measure the distance.

Further, even if it is determined that there is reliability,
Accuracy cannot be obtained with such image intervals detected with different waveforms. For example, if the waveform is as shown in FIG. 13B, the image interval of the subject is affected by the image interval of flare light. Therefore, it is an object of the present invention to provide an automatic focus adjustment device that can obtain accurate distance measurement performance without the camera being unable to perform distance measurement even under the influence of flare light.

[0011]

In order to achieve the above object, the present invention divides the subject light that has passed through the photographing optical system into two images, and defocuses the photographing optical system based on the image distance between the two images. In the automatic focus adjusting device for obtaining the amount, it is determined by the determination means whether the divided two images are affected by the flare phenomenon generated by the taking lens, and the determination means determines that the flare phenomenon is affected. In this case, the output of the photoelectric conversion means for photoelectrically converting the two images is corrected by the light amount difference between the two images, and the image interval is adjusted based on the photoelectric conversion signal corrected by the correction means. Provided is an automatic focus adjusting device including a calculating means for calculating.

[0012]

Embodiments of the present invention will now be described in detail with reference to the drawings. First, the basic concept of the automatic focus adjustment device of the present invention will be described.

According to the present invention, it is determined whether or not flare light influences the re-imaged subject light image, and if it is determined that the flare light has an influence, the light amount difference between the two re-formed images is determined. By correcting, the effect of flare light is removed. FIG. 2 shows and describes a method of correcting the light amount difference between the two images affected by the flare light.

As shown in FIG. 2 (a), the two areas L and R of the electric signal of the re-formed object image converted by the photoelectric conversion element array after passing through the optical system have the same area N. In a range,
The total sum W _L , W _R of the light quantity is obtained.

[0015] Then for large image of light intensity, (W _R -
If the light amount is corrected by W _L ) / N, the two images L and R will have the same light intensity and the same waveform as shown in FIG.
The image spacing detected in the corrected image is sufficiently reliable. An example in which such a difference in light amount between the two images causes inconvenience is shown in FIG.
Shown in. FIG. 3A is an example of a subject whose brightness continuously changes.

When the photographic optical system is defocused on such an object, the two re-formed images are re-formed at the image interval x _{1 as} shown in FIG. 3 (b). When the light amount difference correction shown in FIG. 2 is similarly performed on the image of FIG. 3B, the image shown in FIG. 3C is obtained, and the increment detected by the corrected image is x. It becomes ₂ , and the wrong distance measurement value is detected. Therefore, the correction of the light amount difference between the two images should be performed only when it is determined that the flare light affects the re-formed object light image. A method for determining whether flare light affects two re-formed images will be described below. FIG. 4 shows an example of a so-called multi-division distance measurement pattern.

In FIG. 4, the screen 50 is divided into six areas, and a photometric area 51 corresponding to the distance measuring field of view, a photometric area 52 corresponding to a main subject in the vicinity thereof, and an area 53 for photometrically measuring the peripheral portion of the screen. , 54, 55, 56.

In the photometric pattern shown in FIG. 4, when the amount of light detected in the peripheral photometry region 52 is several tens of times brighter than the amount of light detected in the photometry region 51 corresponding to the distance measuring field, It is considered that the subject light image re-formed on the distance measuring optical system is affected by flare light.

Further, the photometric areas 53, 54, 5 in the peripheral portion of the screen
If the amount of light detected in any one of 5, 5 is a few hundred times brighter than the amount of light detected in the photometric area 51, it is similarly determined that there is an influence of flare light.

That is, the determination standard of how much the difference in the light amount between the photometry area and each peripheral area influences the flare light is determined in advance by the configuration and characteristics of the photographing optical system, and the distance measurement area is determined. It is possible to determine whether or not there is an influence of flare light by comparing the light amounts of the area including the area and the other predetermined area. Next, a determination method other than the determination using the multi-division photometric pattern described above will be described.

When the image interval is obtained from the re-imaged subject image affected by the flare light shown in FIGS. 13B and 13C, the reliability of the image interval is determined as described above. In most cases, reliability is not obtained, but in the case of the image shown in FIG. 2B, sufficiently high reliability is obtained. Therefore, when the reliability of the image interval once detected cannot be obtained, it can be assumed that the flare light is affected.

However, a re-formed object image when a near and far object coexists on the screen, a re-formed object image of a large defocus amount that exceeds the re-imaging range of the distance measuring optical system, or a photoelectric image In the case of a re-formed object image of a low-contrast object which exceeds the capability of the conversion element, the reliability cannot be obtained even if the reliability is judged by the image interval detected by the multi-division photometric pattern.

Therefore, after first determining the reliability, the light amount difference between the two re-formed object images as described above is obtained. Here, in the case of a near-far co-existing subject or a low-contrast subject, the difference in light amount between the two images is small, and when the difference in light amount between the two images is equal to or greater than a predetermined value, it is determined that the flare light is likely to be affected. .. When the light amount difference between the two images is equal to or larger than a predetermined value, the light amount difference is corrected as described above, the image interval is calculated, and the reliability of the calculated image interval is determined.

Therefore, when the image is affected by flare light, the image of FIG. 2 (a) is corrected to the image of FIG. 2 (b),
Since the image interval detected from the corrected image is used, high reliability can be obtained. Also, in the case of a subject with a large defocus amount, the light amount difference between the two images is large, but since the original waveform is different between the two images, even if the image interval is detected with an image in which the light amount difference is corrected, There is no credibility. In this way, it is possible to determine whether or not the flare light is affected. Next in FIG.
A specific configuration of a camera equipped with the automatic focusing apparatus according to the first embodiment of the present invention will be shown and described.

In this structure, the object light O is a photographing lens 21 composed of five lens groups, that is, 17 lenses.
And then enters the main mirror 22a. The main mirror 22a is a half mirror, and half of the amount of incident light passes through the main mirror 22a and enters the distance measuring optical system 23 via the sub mirror 22b.

The distance-measuring optical system 23 is placed near the film equivalent surface, and the field-of-view mask 2 for limiting the distance-measuring field of view.
4. The infrared cut filter 25 that cuts infrared light unnecessary for distance measurement, the condenser lens 26 and the reflection mirror 27, and the light rays that have passed through two pupils on the photographing optical system among the light rays incident on the distance measurement optical system. A re-imaging diaphragm 28 and a re-imaging lens 29 for separating and re-imaging on a photoelectric conversion element, a photoelectric conversion element 30.
It is composed of

The remaining ½ amount of light incident on the main mirror 22a is reflected by the main mirror 22a and enters the finder optical system 33. The finder optical system 33 includes a focusing screen 34, a condenser lens 35, a prism 36, a mold roof mirror 37, an eyepiece lens 38, and a photometric lens 39 and a photometric element 40 that are arranged side by side next to the eyepiece lens 38.

The subject light that has passed through the taking lens 21 is imaged on the focusing screen 34. The photographer can observe the formed image through the condenser lens 35 and the eyepiece lens 38. Similarly, the photometric element 40 measures the amount of light of the subject image formed on the focusing screen 34 through the condenser lens 35 and the photometric lens 39. As described above, in the photometric element 40, the photometric area within the screen is divided into six photometric areas.

FIG. 5 is a diagram showing an electric signal processing system of the automatic focusing apparatus having the configuration of FIG. The subject optical image divided into two by the distance measuring optical system described above is re-imaged on the two photoelectric conversion element arrays 30 (L) and 30 (R), respectively. The photoelectric conversion element array 30 (L), 30
(R) is composed of 64 pixels (not shown) arranged in a matrix. Each of these pixels generates an electric signal corresponding to the amount of light of the object imaged on the pixel, and transmits it as a digital value to the CPU 61. The pixel data input to the CPU 61 is EE
The non-uniformity correction unit 64 performs non-uniformity correction based on the non-uniformity correction data stored in advance in the PROM 62. Here, the non-uniformity correction is to correct the non-uniformity of the light amount distribution due to the condenser lens 26, the re-imaging lens 29, etc. and the non-uniformity of the photoelectric conversion characteristic of each pixel. The data is measured at the time of manufacture and the EEPROM 62
The correction standard is stored in advance.

Next, the non-uniformity-corrected pixel data is filtered by the filter correction unit 65 to remove electric noise generated in the photoelectric conversion element and electric noise generated in other parts of the camera and input through the power supply line. Correction is applied. This filter correction is performed, for example, by providing a high-pass filter or setting a limit on pixel data.

Next, the filtered pixel data is stored in the pixel data storage unit 66. Correlation calculation is performed on the stored pixel data by a correlation calculation unit 67, which will be described later, and the image interval detection unit 68 reimages the pixel data.
After the image interval of the images is obtained, the reliability of the image interval is determined by the reliability determination unit 69. If the reliability of the image distance obtained here is sufficiently high, the lens drive amount calculation unit 70 measures the lens drive amount.

On the other hand, the object light amount signal measured by the photometric element 60 is the interface IC for each photometric area.
A / D conversion circuit 7 in the CPU 61, which is amplified by 63.
Input to 1. The A / D converted photometric value is 6
It is averaged for each one area and stored in the photometric value storage unit 72. Next, the flare determination unit 73 compares the previously stored photometric value reference value with the detected photometric value to determine whether or not there is an influence of flare light. In this determination, if it is determined that there is an influence of flare light, the light amount correction unit 7
4, the light amount of the pixel data stored in the pixel data storage unit 66 is corrected. A flare determination unit 7 that determines whether or not the detected photometric value is affected by flare light.
3 will be described.

The operation from photometry to determination of the influence of flare light will be described with reference to the flowchart of FIG. Here, in the photometric values detected in the six areas shown in FIG. 4, the photometric value “B” is detected from the photometric area 51 corresponding to the distance measuring field.
S ", similarly from the photometric area 52 to" BA ", the photometric area 53
To “BL”, photometric area 54 to “BR”, photometric area 5
It is assumed that "BU" is detected from 5 and "BD" is detected from the photometric area 56.

As described above, the photometric values detected for each of the six areas are detected 10 times as shown in steps S1 to S4, averaged, and stored in the photometric value storage section 72 (step). S5).

Next, it is determined whether or not the photometric value BA in the peripheral portion is brighter by 4 EV or more with respect to the photometric value BS in the central portion (step S6), and when it is bright (YES), the flare flag is set to 1 (step S6). S18), and it is determined that there is an influence of flare light. On the other hand, if it is dark in step S6 (N
O), the photometric value BR, BL of whichever is brighter on the left or right of the screen
One of them is set as the photometric value B (steps S7, S8, S1
0), it is determined whether or not the photometric value B is brighter than 6 EV by the photometric values BS and BA (steps S9 and S11). In this determination, if the photometric value B is brighter than the photometric values BS and BA by 6 EV or more (YES), the process proceeds to step S18, the flare flag is set to "1", and it is determined that the flare light has an influence.

On the other hand, when the photometric value B is darker than the photometric values BS and BA (NO), one of the brighter photometric values BU and BD at the top and bottom of the screen is set as the photometric value B (step S1).
2, S13, S14), this photometric value B is the photometric value BS, B
It is determined whether or not A is brighter than 8 EV (step S1).
5, S16). In this determination, the photometric value B is the photometric value BS, B
If 8 EV or more brighter than A (YES), step S
In step 18, the flare flag is set to "1" and it is determined that there is an influence of flare light. On the other hand, when the photometric value B is darker than the photometric values BS and BA (NO), it is determined that there is no influence of flare light (NO), and the flare flag is set to "0" (step S17). As described above, (1) BA is 4E from BS in each photometric area.
If it is brighter than V, (2) BL or BR is BA and BS
If it is brighter than 6 EV, (3) BU or BD is BA
And when it is 8 EV or more brighter than BS,

If any one of the conditions is satisfied,
It is determined that there is an effect of flare light. Here, the photometric values of the left and right areas (BL, BR) and the photometric values of the upper and lower areas (B
U, BD) is different in the judgment conditions from the distance measurement optical system in the same direction as the direction in which the subject light image is divided into two images. The left and right regions are affected by flare light and the image interval is detected. This is because it tends to be difficult. The flare light received from the upper and lower regions is likely to have an even effect on the two re-formed images, and the conditions may be severe.

Note that the flare light generation conditions differ greatly depending on the configuration and characteristics of the photographing optical system, so the judgment conditions are not limited to this. Further, in the case of the photographing optical system having the zoom function as in the present embodiment, the determination condition may be changed depending on the focal length because the positional relationship of the lenses greatly differs depending on the focal length.

In this embodiment, the relative value of the photometric value is used for the determination. However, if the flare light generation condition is considered to be limited to a high brightness light source such as sunlight, the absolute value is used for the determination. You may.

Next, with reference to the flow chart of FIG. 7, the operation until the influence of flare light is corrected and the lens driving force amount is obtained will be described. Here, the reference numerals of FIG. 5 are used as the reference numerals of the respective constituent members.

First, the two photoelectric conversion element arrays 30 described above are used.
The pixel data output from (L) and 30 (R) are input and stored (step S20). The non-uniformity correction unit 64 performs non-uniformity correction on the stored pixel data (step S2
1), filter correction is performed by the filter correction unit 65 (step S22).

Next, the flare flag is judged, and if it is "1", it is judged that there is an influence of flare light, and the above-mentioned light amount correction is performed on the stored pixel data (step S25). Next, the correlation calculation unit 67 performs the correlation calculation (step S24), and based on the result, the image distance detection unit 68 detects the image distance (step S2).
6). With respect to the image interval detected here, the reliability determination unit 69 calculates a reliability value (step S27),
The determination is made (step S28). In this determination, if it is determined to be reliable (YES), the lens drive amount calculation unit 70 calculates the lens drive amount (step S29), and if it is determined to be unreliable (NO), It is determined that distance measurement is impossible (step S30). Next, the light quantity correction unit 7 of the present embodiment
The light amount correction in No. 4 will be described.

First, re-imaging is performed and two photoelectric conversion element arrays 3 are formed.
The pixel data of the two images obtained at 0 (L) and 30 (R) is L
_(I) and R _(I) (I is a pixel number, which is 1 to 64 in order from the left end). The light quantity is obtained for each of these two images.

[0044]

[Equation 1]

In this case, the central portion 48 of each photoelectric conversion element array
It is calculated by adding pixel data for each pixel. The periphery is not used because the re-imaging diaphragm may block the optical path when the periphery is defocused. Next, the light amount difference is corrected.

[0046]

[Equation 2]

In this case, the brighter photoelectric conversion element output is subtracted and corrected. However, in the case of the method in which the photoelectric conversion element output is based on the brightest pixel output reference, the darker photoelectric conversion element output is used. On the other hand, addition correction may be performed.

[0048]

[Equation 3]

The pixel data corrected here is stored in the pixel data storage unit 66 shown in FIG. If the difference between the light amounts W _L and W _R of the two images is less than a predetermined value, it may be determined that there is no influence of flare light and no correction is performed. Next, the correlation calculation performed by the correlation calculation unit 67 of the present embodiment will be described with reference to the flowchart of FIG.

First, the variable SL has an initial value "5" and the variable SR.
The initial value "37" is set to (step S31). Next, the variable J is set to "8" (step S32). The variable SL is a leading pixel number of a small block that is compared with the other pixel data R _(I) by a correlation operation from the pixel data L _(I) of one of the two image data, Similarly, the variable SR is the small block start pixel number of R _(I) . The variable J is a variable for counting the number of times the small block has moved. Further, the correlation calculation is performed by the following equation to obtain the correlation output F _(s) (step S33).

[0051]

[Equation 4] However, the number of pixels of the small block is 27 pixels.

Next, the minimum value of the correlation output F _(s) is detected, that is, F _(s) and F _MIN are compared (step S3).
4) In this comparison, if F _(s) is smaller than F _MIN (Y
Substituting F _(s) into ES) and F _MIN , the variables SL,
Substitute SR into variables SLM and SRM (step S3)
5).

Next, "1" is subtracted from each of the variables SR and J (step S36). Then, it is determined whether or not the variable J has become "0" (step S37). If this determination is not "0" (NO), the process returns to step S32 and the correlation calculation is repeated. That is, the variable SL is fixed and the variable S
Correlation is performed while shifting R 8 times. If the variable J is "0" in the above judgment (YES), "4" is added to the variable SL and "3" is added to the variable SR (step S38).
The correlation calculation is repeated until the variable SL becomes "29" (step S39). By the above procedure, the correlation calculation can be efficiently performed between the two images.

As a result, the positions SLM and SRM of the small blocks which give the minimum value F _MIN of the obtained correlation output F _(s) indicate the image positions having the highest correlation, that is, the image distance between the two images. However, this image interval is on a pixel-by-pixel basis, and the accuracy is low in calculating the lens drive amount. Therefore, interpolation is performed and the two-image interval is obtained with an accuracy of pixel unit or less. First, the correlation outputs FM and FP when "± 1" pixels are shifted from the image interval having the highest correlation are obtained by the following equation (step S40).

[0055]

[Equation 5] Therefore, the obtained space S ₀ between the two images is as follows (step S41).

[0056]

[Equation 6]

Next, the reliability SK of the image interval obtained by the equations (10) and (11) is obtained (step S42). Here, the required reliability will be described with reference to FIGS.

First, the divided two images are detected by the detected image interval S.
_{If 0} is the correct image distance, the detected image distances match. Therefore, the correlation output F _(s) becomes "0" at the image interval S ₀ as shown in FIG. 9 (a).

However, as described above, in the case of an image affected by flare light, an image in which distant objects coexist, a large defocused image, or an image of a low-contrast object, two divided images are obtained. Therefore, the correlation output F _(s) does not become "0" at the detected image interval S ₀ as shown in FIG. 9B. From the characteristics shown in FIG.
The reliability value SK can be obtained by the following equation.

[0060]

[Equation 7]

That is, in the case of high reliability, FIG.
As shown in, the numerator of the equation (12) or the equation (13) has a size similar to that of the denominator, and the reliability value SK can be regarded as about “1”. However, when there is no reliability, the numerator becomes larger than the denominator and the value of SK becomes a value larger than “1” as shown in FIG. 9B.

Therefore, the reliability of the detected image interval S ₀ can be determined by comparing the SK obtained by the equation (12) or the equation (13) with the predetermined value SK0 (step S43). In this determination, if it is determined to be reliable (YES), from the detected image distance S ₀ ,
The lens drive amount is obtained (step S44). However, if it is determined that there is no reliability (NO), it is determined that distance measurement is impossible (step S45), and the correlation calculation is ended.

Actually, the image signals of the two divided images may not match due to the error of the distance measuring optical system, the conversion error of the optical conversion element, the influence of electric noise, etc., and the reliability judgment value SK0 may be a value larger than "1" (for example, 2-3).

Therefore, if the determination value SK0 is limited, it is often determined that the distance cannot be measured, and the commercial value of the camera is impaired. Therefore, the determination value is set to a large value depending on the conditions. For example, when the brightness is low, the matching value is deteriorated due to the influence of the dark current generated in the photoelectric conversion element, and therefore the determination value is set to be large.

Further, in a camera provided with auxiliary light for illuminating the subject at low brightness, the judgment value when the auxiliary light is not lit is set small and the judgment value when the auxiliary light is lit is set large,
Depending on the image distance detected when the auxiliary light is not turned on,
Avoid driving the lens.

Here, when the reliability value SK is smaller than the determination value SK0, the lens drive amount is immediately obtained at the detected image interval. If it is larger than the judgment value SK0 and smaller than the second judgment value SK1, the lens drive amount may be calculated after confirming that there is no change in the detected image interval after a plurality of distance measurements. ..

When the reliability value SK is larger than the judgment value SK0 and smaller than the third judgment value SK3, the positional relationship between the two images having the correlation in the correlation calculation may be shifted to re-correlate. This is because reliability deteriorates when a plurality of subject images with different distances are included in a correlated small block. The positional relationship between correlated small blocks is shifted or the number of pixels in the small block is changed. The object image can be narrowed down by reducing. Next, with reference to the flow chart of FIG. 10, an automatic focus adjusting device as a second embodiment according to the present invention will be described which does not use a photometric means.

2 detected in the same manner as in the first embodiment described above
Correlation calculation is performed on the image data of the images (step S51), and the image interval between the two images is detected (step S5).
2) A reliability value is calculated for the detected image interval (step S53). Next, the reliability is determined based on the calculated reliability value (step S54). If there is reliability in this determination (YES), the lens drive amount is immediately calculated (step S63). Here, the following image can be considered as the subject image that is determined to be unreliable. (1) Image affected by flare light, (2) Image in which subjects at different distances coexist, (3) Low contrast subject,
(4) An image with a large defocus amount.

If it is determined in step S54 that the image is unreliable (NO), it is assumed that the images are affected by flare light, and the light amounts of the two images are calculated based on equations (1) and (2). The light amount difference ΔW between the two images is calculated (step S5).
5).

[0070]

[Equation 8]

Next, in the case of a sensitizer or a low-contrast subject in which subjects at different distances coexist, the difference in light amount between the two images Δ
When W is small and the image is influenced by flare light or the defocus amount is large, the light amount difference ΔW between the two images shows a large value. Therefore, the light amount difference ΔW between the two images is determined by the determination value Δ
By comparing with W ₀ (step S56), the possibility of being affected by flare light can be increased. If the light amount difference between the two images is less than or equal to the predetermined value in this comparison (NO), it is determined that distance measurement is impossible (step S58).

Here, when the light amount difference is obtained by the equations (1) and (2), the determination value ΔW ₀ becomes about 1000. However, the generation condition of flare light and the degree of influence of the re-formed object image largely vary depending on the configuration and characteristics of the photographing optical system, and the determination value ΔW ₀ is determined according to the photographing optical system. If the camera has a zoom mechanism,
The determination value ΔW ₀ may be changed according to the focal length. Further, if the flare light generation state varies from device to device, the flare light is generated at the time of factory shipment under a predetermined condition, and the determination value ΔW ₀ is measured from the image data, and E
It can also be written in the EPROM.

In the comparison in step S56, if the light amount difference between the two images is equal to or larger than the predetermined value (YES), the CPU 61 is determined based on the equations (3), (4) or (5), (6). The light amount correction is performed on the pixel data stored in (step S57). Next, the above-described correlation calculation (step S59), image interval detection (step S60), and reliability calculation (step S61) are repeatedly performed on the pixel data whose light amount has been corrected.

As described above, when there is an influence of flare light, the image interval obtained from the pixel data whose light amount is corrected has high reliability. On the other hand, in the case of an image of a subject with a large defocus amount, the original two images do not match,
There is no credibility.

Therefore, by determining the reliability of the image interval obtained from the pixel data whose light amount has been corrected (step S6
2) It can be determined whether or not the image is affected by flare light. At the same time, the image interval obtained in step S60 becomes the correct image interval.

If it is determined in step S62 that there is reliability (YES), the lens drive amount is obtained from the detected image interval (step S63). If there is no reliability (NO), distance measurement is performed. It is determined to be defective (step S6)
3),

When there is no possibility of an image having a large defocus amount, for example, when the defocus detection range of the distance measuring optical system is sufficiently large with respect to the defocus amount of the photographing optical system, the light amount difference is large. After the determination (step S56),
The lens driving amount may be immediately obtained. Alternatively, the second correlation calculation (step S59) may be omitted, and the image position S of the small block obtained in the first correlation calculation (step S51) may be omitted.
The image interval may be obtained from pixel data in which the light amount is corrected for LM and SRM. Further, the present invention is not limited to the above-described embodiments, and it goes without saying that various modifications and applications can be made without departing from the scope of the invention.

[0078]

As described above in detail, according to the present invention, even if the flare light from the light source outside the distance measuring field affects the object image within the distance measuring field, the light amount difference between the two images is corrected. Thus, it is possible to provide an automatic focus adjustment device capable of continuing stable distance measuring operation.

[Brief description of drawings]

FIG. 1 is a diagram showing a specific configuration of a camera equipped with an automatic focusing device as a first embodiment according to the present invention.

FIG. 2 is a diagram showing an electric signal of a re-formed object image converted by a photoelectric conversion element array for explaining the principle of the present invention.

FIG. 3 is a diagram showing an electric signal of a subject image in which the brightness reconverted by the photoelectric conversion element array continuously changes, for explaining the principle of the present invention.

FIG. 4 shows an example of a multi-division distance measurement pattern.

5 is a diagram showing an electric signal processing system of the automatic focusing apparatus having the configuration shown in FIG. 1;

FIG. 6 is a flowchart showing an operation from photometry to determination of an influence of flare light.

FIG. 7 is a flowchart showing an operation of correcting the influence of flare light to obtain a lens driving force amount.

FIG. 8 is a flowchart showing a procedure of a correlation calculation performed by the correlation calculation unit of the first embodiment.

9A and 9B are diagrams showing a characteristic of a correlation output.

FIG. 10 is a flow chart showing an operation of an automatic focus adjusting device which does not use photometric means as a second embodiment according to the present invention.

FIG. 11 is a diagram showing a principle in distance measurement.

FIG. 12 is a diagram showing the principle in the case where there is a subject M which is significantly brighter than the subject outside the distance measuring field of view by the phase difference method.

13 (a), (b), and (c) are diagrams showing electric signals of a re-formed object image converted by a conventional photoelectric conversion element array.

[Explanation of symbols]

1, 2, 3, 11, 12, 13, 21, ... Shooting lens,
4, 14 ... Equivalent surface of film, 5, 15, 24 ... Field mask, 6, 16 ... Re-imaging lens, 7, 17, 30, 30
(L), 30 (R) ... Photoelectric conversion element array, 22a ... Main mirror, 22b ... Sub mirror, 23 ... Distance measuring optical system, 25
... infrared cut filter, 26 ... condenser lens, 27
... Reflecting mirror, 28 ... Re-imaging diaphragm, 29 ... Re-imaging lens, 33 ... Finder optical system, 34 ... Focusing screen, 35 ... Condenser lens, 36 ... Prism, 37 ... Mold roof mirror, 38 ... Eyepiece, 3
9 ... Photometric lens, 40, 60 ... Photometric element, 50 ... Screen,
51, 52, 53, 54, 55, 56 ... Photometric area, 61
... CPU, 62 ... EEPROM, 63 ... Interface IC, 64 ... Non-uniformity correction unit, 65 ... Filter correction unit, 6
6 ... Pixel data storage unit, 67 ... Correlation calculation unit, 68 ... Image interval detection unit, 69 ... Reliability determination unit, 70 ... Lens drive amount calculation unit, 71 ... A / D conversion circuit, 72 ... Photometric value storage unit, 7
3 ... Flare determination unit, 74 ... Light amount correction unit, L, R ... Subject image, O ... Subject light.

Claims (1)

[Claims] 1. An automatic focusing apparatus that divides subject light that has passed through a photographing optical system into two images and obtains a defocus amount of the photographing optical system based on an image distance between the two images. Judgment means for judging whether or not there is an influence of the flare phenomenon generated by the taking lens, and photoelectric conversion means for photoelectrically converting the two images when it is judged by the judgment means that there is an influence of the flare phenomenon. The automatic focus is characterized by comprising: a correction unit that corrects the output of the image by the light amount difference between the two images; and a calculation unit that calculates the image interval based on the photoelectric conversion signal corrected by the correction unit. Adjustment device.