JPH11344662A - Automatic focusing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an automatic focusing device which can perform fast focusing operation. SOLUTION: An AF stop 102 has different color filters (G filter 102a and M filter 102b) arranged at an aperture part made eccentric with the center of an AF lens 101 and for image pickup operation by AF, image data of AF areas in an image frame corresponding to respective pieces of luminous flux passing through the color filters 102a and 102b of the AF stop 102 are obtained by colors; and a cross correlation arithmetic part 106 calculates the cross correlation between the image data by the colors and a distance and direction calculation part 108 calculates the distance to and the direction of the focusing position of the AF lens 101 according to the cross correlation to drive the AF lens 101 to the focusing position.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an automatic focusing device, and more particularly, to an automatic focusing device applied to an image input device using an image pickup device such as a video camera and a digital camera.

[0002]

2. Description of the Related Art Conventionally, various methods such as a hill-climbing method, an autocorrelation method, and a cross-correlation method have been proposed as a focus position determining method in an automatic focusing apparatus. Hereinafter, these conventional focus position determination methods will be described.

The hill climbing method is disclosed in, for example,
-214868 (automatic focusing device) and JP-A-3-2
JP-A-16078 (automatic focusing device).

[0004] An "automatic focus control device" described in the above-mentioned Japanese Patent Application Laid-Open No. 3-214868 is an image pickup device that converts optical information of a subject image formed through a lens on an image pickup surface of the image pickup device into an electric signal, A / D conversion means for converting the converted electric signal into a digital signal, memory means for temporarily storing the digital signal for each frame, and performing two-dimensional orthogonal transform on an image formed by the digital signal in a frequency domain. Encoding means for detecting the magnitude of each frequency component included in the image, focus detection means for detecting the amplitude of a specific high-frequency component in the frequency components, and a control signal from the focus detection means. Lens driving means for adjusting the position of the lens by adjusting the position of the lens so that the amplitude of the specific high-frequency component is maximized. And requests.

[0005] Further, an "automatic focusing device" described in Japanese Patent Application Laid-Open No. Hei 3-216078 discloses an A / D converter for converting a supplied video signal into a digital signal, and an A / D converter.
An orthogonal transformer that converts a video signal transmitted by the converter into a frequency component domain, and a control unit that generates a control signal for performing a focusing operation based on information of the frequency component transmitted by the orthogonal transformer. It is provided.

As the autocorrelation method, for example, JP-A-7-154668 (automatic focus adjusting device), JP-A-7-159885 (automatic focus adjusting device), and JP-A-7-270674 (automatic focus adjusting device) An image input device having an adjusting unit) and an image input device described in JP-A-7-287162 (focus detection device).

[0007] The "automatic focus adjustment device" described in the above-mentioned Japanese Patent Application Laid-Open No. 7-154668 is based on information from a photoelectric conversion element on which a subject image is formed via a photographing optical system.
The two types of focusing means are driven to perform focusing of the photographing optical system. The first type of focusing means sequentially performs the focus evaluation of the subject image with the movement of the imaging lens, and calculates the evaluation value. The focal point where the maximum value is reached is set as the focal point, and the focusing means of the second system positions a stop having two apertures in the photographing optical system, divides one subject image into two optical paths, and performs photoelectric conversion. An image is formed on an element, and focusing or non-focusing is determined based on a difference in an imaging position. First, focusing is performed by focusing means of an autocorrelation method which is a second method, and then focusing is performed. Focusing is performed by the focusing means of the hill-climbing method, which is the first method.

[0008] The "automatic focus adjustment device" described in the above-mentioned Japanese Patent Application Laid-Open No. 7-159885 is a first focusing device which focuses a subject from information of a plurality of subject images formed through different optical paths. A focusing unit, and a second focusing unit that focuses the subject based on information of one subject image to be formed. The focusing operation is performed by using the focusing units in a first and a second order. In the apparatus, the setting of a distance measuring frame used for focusing operation by the first focusing means without causing a near-far conflict and the characteristics of a signal extraction filter suitable for each shooting scene are determined.

Further, the "image input apparatus having automatic focus adjusting means" described in Japanese Patent Application Laid-Open No. 7-270674 is an image forming means for forming a subject image on a subject image detecting means for detecting a subject image. An aperture member that adjusts the amount of light flux passing through the imaging unit; an exposure amount adjustment unit that has a shutter member that adjusts the exposure time; and different focus determination units based on signals from the subject image detection unit. When adjusting the focus of the imaging unit using signals from a plurality of focus determination units, the exposure amount adjustment unit adjusts the exposure state according to the type of the determination unit and the brightness of the subject image. is there.

The "focus detection device" described in Japanese Patent Application Laid-Open No. 7-287162 divides a pupil of a photographing system into a plurality of regions by a diaphragm having a plurality of apertures, and When forming a plurality of image information based on the luminous flux that has passed through the area on the surface of the color imaging unit and detecting the in-focus state of the imaging system at a plurality of in-focus positions using image signals from the color imaging unit, After detecting the plurality of focus positions, the focus detection method selection means performs focusing using another second focusing means different from the first focusing means, or performs the first focusing. The focus operation is performed by selecting whether or not to perform a focusing operation, and the in-focus state of the photographing system is detected.

As the above cross-correlation method, for example, those described in JP-A-5-103251 (autofocus device) and JP-A-6-250077 (autofocus device using a single lens) are known.

The "autofocus device" described in the above-mentioned Japanese Patent Application Laid-Open No. 5-103251 is provided between a light receiving portion for forming an image of an object to be focused and the light receiving portion and the object to be focused. A light amount adjusting means for changing an incident light amount of a part of the light from the object to be focused on the light receiving section, and switching a change position of the incident light amount on the light receiving section by the light amount adjusting means. Means for switching to at least two positions in the plane of the lens, obtaining a correlation of the video signal from the light receiving unit at these at least two positions in a correlator, and determining the direction of the focus shift and the amount of the shift by the determining means. The distance between the lens and the light receiving section is controlled based on the judgment output of the judging means.

The "autofocus device using a single lens" described in the above-mentioned Japanese Patent Application Laid-Open No. Hei 6-250077 discloses an autofocus device in which the optical system of the autofocus device includes a single lens and a front or rear side of the lens. With the mask arranged, and separated by the mask converged by the lens.
By irradiating a CD sensor and moving an objective lens while correlating two output signals output from the CCD sensor, a laser beam is automatically focused on a workpiece.

[0014]

However, in the focusing method based on the hill-climbing method, the A
Although the F evaluation value is obtained, it is difficult to detect whether the in-focus point exists before or behind this AF lens position. Therefore, in order to detect the in-focus position while avoiding the local maximum point, the AF evaluation value of the entire lens moving range is required.
In addition, there is a problem that an operation of returning to the in-focus position is added, and it takes time to obtain an accurate focus position.

In the focusing method based on the autocorrelation method, two or more different light beams are imaged in one frame.
Although the F information can be obtained, since the correlation is based on the same data, the result shows a very high peak value at the origin position (focal point) of the autocorrelation coefficient. For this reason,
In the vicinity of the focal point, there is a problem that another method must be used.

In the focusing method based on the cross-correlation method, two or more different light beams are imaged in separate frames. However, this requires the preparation of two or more frames. Does not produce a very high peak value at the origin position. From this, AF information can be obtained from the vicinity of the focal point to the distant place without being buried in this peak value. However, the amount of calculation of the cross-correlation calculation for the entire frame exceeds the practical range, and there is a problem in that it takes time.

The present invention has been made in view of the above, and has as its object to provide an automatic focusing device capable of performing a high-speed focusing operation.

[0018]

In order to solve the above-mentioned problems, an automatic focusing apparatus according to a first aspect of the present invention includes an AF lens for focusing a subject to be focused and an AF lens for focusing the subject to be imaged. Color imaging means for converting the subject image into an electric signal and outputting it as image data;
An aperture stop is provided at an eccentric position different from each other with respect to the center of the lens, and a different color filter is disposed at each of the openings, and an AF stop for limiting a light flux from the subject to be focused is provided. A part of the image frame picked up by the color image pickup means is set as an AF area, and in the image pickup for AF, the image of the AF area in each image frame corresponding to each light beam passing through each color filter of the AF diaphragm is set. Data is obtained, a cross-correlation between the image data in the AF area is calculated, and a distance and a direction to the in-focus position of the AF lens are calculated based on the cross-correlation. Is driven.

According to a second aspect of the present invention, in the automatic focusing apparatus according to the first aspect, one of the different color filters comprises a primary color filter and the other comprises a complementary color filter.

According to a third aspect of the present invention, in the automatic focusing apparatus according to the second aspect, the primary color filter comprises a G filter and the complementary color filter comprises an M filter.

According to a fourth aspect of the present invention, there is provided an automatic focusing apparatus, comprising: an AF lens for focusing a subject to be focused; and an image signal obtained by converting a subject image of the focused subject into an electric signal. A color image pickup means for outputting as a light beam, and apertures are respectively provided at eccentric positions different from each other with respect to the center of the AF lens, and different color filters are disposed in the respective apertures. AF stop for limiting, and a part of an image frame taken by the color
The area is set as an area, and in the imaging for AF, the AF
Acquiring image data of the AF area in each image frame corresponding to each light beam passing through each color filter of the aperture,
Calculating a cross-correlation between the image data, calculating a distance and a direction to a focus position of the AF lens based on the cross-correlation, and calculating an approximate focus position of the AF lens;
Further, in the vicinity of the approximate in-focus position, A
The final focus position is determined based on the high frequency components of the image data in the F area.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A preferred embodiment in which an automatic focusing apparatus according to the present invention is applied to a digital camera will be described below in detail with reference to the accompanying drawings.

(Embodiment 1) FIG. 1 shows a configuration diagram of an AF processing system of a digital camera according to Embodiment 1. Referring to FIG. 1, reference numeral 101 denotes an AF lens arranged on the optical axis, which is used to form a subject image of a subject to be focused on the primary color image sensor 103. 10
Reference numeral 2 denotes an AF stop, which is disposed on the rear side of the AF lens 101 and restricts a light beam (light amount) passing through the AF lens 101. AF diaphragm 1 in the present embodiment
02, two openings are provided at different eccentric positions with respect to the center of the AF lens 101, and a G filter 102 is provided in each opening.
a and an M filter 102b (FIG. 3
reference). A color CC 103 converts the formed subject image into an electric signal and outputs it as R, G, B image data.
1 shows a primary color image sensor made of D or the like. The primary color image sensor 103 is capable of random access of pixels, and is capable of transferring image data of an arbitrary area in an image frame. The AF lens 101, the AF stop 102, and the primary color image sensor 103 constitute an imaging optical system.

Reference numeral 112 denotes a frame R output from the primary color image sensor 103 in AF imaging.
Image data (R data) of the AF area in the frame B, and A in the frame B output from the primary color system image sensor 103.
5 shows a mixing operation unit that outputs image data of M colors (complementary colors) by performing a mixing operation on image data (B data) in the F area. 1
Reference numeral 04 denotes G image data of an AF area in an image frame (frame G) corresponding to a light beam that passes through the G filter 102a of the AF aperture 102 and is output from the primary color system image sensor 103 in imaging for AF. The image memory (for the G filter AF area) to be stored is shown.
Reference numeral 5 denotes an image memory (for an M filter AF area) that stores image data (M data) of the AF area output from the mixing operation unit 112. 106 is the image memory 10
4 shows a cross-correlation calculation unit for obtaining a correlation coefficient between image data of the AF areas stored in reference numerals 4 and 105, respectively. The cross-correlation calculation unit 106 performs a two-dimensional FFT for performing a two-dimensional FFT calculation on the first image data stored in the image memory 104.
An FT unit 200, a two-dimensional FFT unit 201 for performing a two-dimensional FFT operation on the second image data stored in the image memory 105, and a two-dimensional FFT unit 201 and a two-dimensional FFT unit 201, respectively. A complex multiplication unit 202 that performs a complex multiplication on the set of image data, and a two-dimensional IFFT that performs a two-dimensional IFFT on the result of the complex multiplication by the complex multiplication unit 202
And a unit 203.

Reference numeral 107 denotes a cross-correlation coefficient memory for storing the correlation coefficient calculated by the cross-correlation calculation unit 106. 10
Reference numeral 8 denotes a distance direction calculator 108 that calculates the distance and direction to the in-focus position of the AF lens 101 based on the correlation coefficient calculated by the cross-correlation calculator 106.

Reference numeral 109 denotes an AF lens driving mechanism for moving the AF lens 101 on the optical axis and adjusting the focus based on the calculation result of the cross-correlation calculator 106. Reference numeral 110 denotes an aperture switching mechanism that switches the operation of the AF aperture 102 for the AF operation and the photographing operation, and controls the operation of the AF aperture 102. Reference numeral 111 denotes a frame image memory in which image data of one captured frame is stored for each color in the case of normal shooting.

FIG. 2 shows an image frame of the primary color system image sensor 103 and its AF area. In the primary color system image sensor 103, R, G, and B sensors are arranged in a band shape or a staggered shape. In the figure, the frame of the image data by the R sensor is a frame R, the frame of the image data by the G sensor is a frame G, the frame of the image data by the B sensor is a frame B, and the frames R, G, and B include An AF area (focus position) is set. This AF area
The configuration may be arbitrarily set by the photographer at the time of photographing, or may be set in advance.

FIG. 3 is an explanatory diagram for explaining the capture of a light beam by the AF diaphragm 102.
2. Primary color image sensor 103, image memory 10
4, 105, and the mixed operation unit 112 are schematically shown. As shown in the figure, the AF diaphragm 102 is provided with two sets of openings at different eccentric positions with respect to the center (optical axis) of the AF lens 101, and a G filter 102a and an M filter 102b are arranged in each opening. Has been established. G filter 10
The light flux of the G component (light flux G) passes from 2a, and the light flux of the M component (light flux M) passes from the M filter 102b to form an image on the primary color image sensor 103.

The G sensor outputs the image data (G component) of the AF area in the frame G of the received light flux G to the image memory (for the G filter AF area) 104. The R sensor outputs the image data (R component) of the AF area in the frame R of the received light flux M to the mixing operation unit 112,
Further, the B sensor combines the image data (B component) of the received light flux M in the AF area in the frame B with the mixing operation unit 112.
Output to The mixing operation unit 112 performs a mixing operation on the input R component image data and B component image data, and outputs the M component image data to the image memory (for the AF area of the M filter) 105.

Next, the principle of the AF operation of the digital camera will be described with reference to FIG. FIG. 4 is an explanatory diagram for explaining the principle of AF of the digital camera, and schematically shows an AF lens 101, an AF stop 102, and a primary color image sensor 103. In FIG.
3B shows a light beam 1 (light beam G) passing through the G filter 102a of the AF stop 102, and FIG.
2 illustrates a light beam 2 (light beam M) passing from the M filter 102b of No. 02.

First, in the imaging for AF, as shown in FIGS. 3A and 3B, the light flux 1 from the G filter 102a and the M filter 102b of the AF diaphragm 102 are respectively detected.
(Light flux G) and light flux 2 (light flux M) pass simultaneously, and light flux 1
(Light flux G) and light flux 2 (light flux M) are imaged on the primary color image sensor 103, and the primary color image sensor 103
Is captured by In the primary color image sensor, the G sensor detects A in the frame G of the received light flux G.
The image data (G component) of the F area is output to the image memory (AF area for filter G) 104. The R sensor is
The image data (R component) of the received light flux M in the AF area in the frame R is transferred to the mixing operation unit 112, and the B sensor receives the image data (B component) of the AF area in the frame B of the received light flux M. ) Is transferred to the mixing operation unit 112. The mixing operation unit 112 performs a mixing operation on the input R-component image data and B-component image data, and transfers the M-component image data to the image memory (the AF area for the filter M) 105. That is, each light beam is separately captured in one shooting, and two sets of image data of the AF area are output.

Thereafter, the mutual correlation coefficient is obtained from the two sets of image data, and the extent of the two image data is determined from the direction and distance from the origin position to the peak point (black point) of the correlation value. Detect the direction of the shift.

FIG. 5 is a view showing an example of the relationship between the cross-correlation coefficients between the image data of the AF area of the G filter (light flux G) and the image data of the AF area of the M filter (light flux M). In the figure, (a) shows the case of the front pin,
(B) shows the case where the subject is in focus, and (c) shows the case where the subject is in the back focus. From these two sets of image data (31, 32), a cross-correlation coefficient (33) is obtained, and from the direction and distance from the origin position to the peak point (black point) of the correlation value, Is detected. Based on the detection result, the distance and direction to the in-focus position of the AF lens 101 are calculated. In addition, the larger the degree of deviation,
This indicates that the camera is at a position far from the in-focus position, and indicates that the camera is in the in-focus position when there is no deviation.

Next, an outline of the overall operation of the digital camera will be described with reference to FIG. First, when the shutter button (not shown) is half-pressed by the operation of the operator, the imaging optical system of the digital camera operates as an AF sensor. First, in imaging for AF, the light beam 1 (light beam G) and the light beam 2 (light beam M) pass simultaneously from the G filter 102a and the M filter 102b of the AF stop 102, respectively, and the light beam 1 (light beam G) and the light beam 2 ( The light flux M) forms an image on the primary color image sensor 103 and is captured by the primary color image sensor 103. In the primary color image sensor 103, the G sensor outputs image data (G component) of the AF area in the frame G of the received light flux G.
Is output to the image memory (AF area for filter G) 104. The R sensor mixes the image data (R component) of the received light flux M in the AF area in the frame R with the mixing operation unit 112.
The B sensor transfers the image data (B component) of the AF area in the frame B of the received light flux M to the mixing operation unit 112. The mixing operation unit 112 performs a mixing operation on the input R-component image data and B-component image data, and stores the M-component image data in an image memory (A for the filter M).
F area) 105. Different luminous fluxes are captured by one imaging.

Next, the cross-correlation calculating section 106 calculates a cross-correlation coefficient of two sets of image data (image data of the AF area of the G filter and the M filter) stored in the image memories 104 and 105. . When the two sets of image data are directly calculated in the spatial domain, the larger the AF area, the more enormous product-sum calculation is required, and the longer the calculation time is. The image data of the G filter 102a and the M filter 102b in the AF area are respectively represented by f (u, v) and g (u,
v) and the cross-correlation coefficient is h (u, v), the cross-correlation coefficient h (u, v) is represented by the following equation (1).

H (u, v) = f (u, v) ★ g (u, v) (1) where ★ indicates a symbol of a correlation operation.

In the present embodiment, the number of operations is greatly reduced by replacing the space domain with the multiplication in the frequency domain.

Specifically, the two-dimensional FFT units 200 and 20
1, the G filter 102a and the M filter 102 of the AF area stored in the image memories 104 and 105, respectively.
b, two-dimensional Fourier transform is performed on the image data f (u, v) and g (u, v), respectively, and the frequency domain data F
(U, V) and G (U, V).

Subsequently, the complex multiplication unit 202 performs a complex multiplication as shown in the following equation (2) on the frequency domain data F (U, V) and G (U, V), A cross-correlation coefficient H (U, V) is calculated.

H (U, V) = F (U, V) × G (U, V) ^* (2) where G (U, V) ^* is the complex conjugate of G (U, V) It is.

Then, the two-dimensional IFFT section 203 performs a two-dimensional Fourier inverse transform on the cross-correlation coefficient H (U, V) in the frequency domain to calculate a cross-correlation coefficient h (u, v). This makes it possible to calculate the cross-correlation coefficient h (u, v) with a small amount of calculation. The cross-correlation coefficient h (u, v) is stored in the cross-correlation coefficient memory 107.

The distance direction calculator 108 calculates the distance or direction between the point of the peak of the cross-correlation coefficient h (u, v) stored in the cross-correlation coefficient memory 107 and the origin (focal point) (FIG. 5). ), And the distance and direction from the current position of the AF lens 101 to the in-focus position are calculated. AF optical system drive circuit 109
Is the AF lens 10 calculated by the distance direction calculation unit 108
AF lens 1 based on the distance and direction to the in-focus position
01 moves the focus position. With the above operation, the AF operation ends. Note that in order to increase the focusing accuracy, the focus may be once moved to the focusing position, and then the above-described AF operation may be repeated to further improve the focusing accuracy.

When the AF operation is completed, the aperture switching mechanism 11
With the value of 0, the AF stop 102 is removed from the imaging optical system, and preparation for actual shooting of the subject is performed. Then, when the movement of the AF lens 101 is completed, a state where the shutter button is fully pressed is awaited.

Thereafter, when the shutter button is fully depressed, the shutter is opened for a fixed period of time in conjunction therewith, and the subject image from the subject is received by the primary color image sensor 103. Then, the subject image is photoelectrically converted by the primary color system image sensor 103 and transferred to the frame image memory 111 for each color and temporarily stored as image data. Then, the image data stored in the frame image memory 111 is transferred to a subsequent image processing system (not shown), and after undergoing image processing such as white balance, is stored or displayed on a monitor.

As described above, in the first embodiment, the AF diaphragm 102 has different color filters (the G filter 102a and the M filter 102) in the openings arranged at different eccentric positions with respect to the center of the AF lens 101. 102b) are arranged, and in imaging for AF, image data of an AF area in an image frame in an image frame corresponding to each light beam passing through each color filter 102a, 102b of the AF diaphragm 102 is acquired for each color, and a cross-correlation calculating unit 106 calculates the cross-correlation coefficient between the image data for each color, and the distance direction calculation unit 108 calculates the distance and direction to the in-focus position of the AF lens 101 based on the cross-correlation coefficient. Since the AF lens 101 is driven to the in-focus position, the distance and direction to the in-focus position can be obtained by one shooting for AF, and a high-speed in-focus operation can be performed. The ability. That is, two different light fluxes are respectively allocated to different frames, image data of the AF area of each frame is obtained, a correlation coefficient is obtained for the two sets of image data, and the AF lens 101 is determined based on the cross-correlation coefficient. Since the distance and the direction to the focus position are calculated, it is possible to quickly focus on an image area (AF area) to be focused without requiring a distance measuring detector (AF sensor module). Become.

Also, in the first embodiment, since the primary color filter and the complementary color filter are used as the color filters,
When a primary color system image sensor is used, the light beam that has passed through the primary color filter can be sensed by a certain color sensor, and the light beam that has passed through the complementary color filter can be sensed by the other two color sensors. It becomes easy to take in.

In the first embodiment, since the primary color filter is a G filter and the complementary color filter is an M filter, when a primary color image sensor is used, the luminous flux passing through the G filter is detected by the G color sensor. However, the light beam that has passed through the M filter can be sensed by the R and B color sensors, and it becomes easy to separate and take in the two light beams.

(Embodiment 2) FIG. 6 shows a configuration diagram of an AF processing system of a digital camera according to Embodiment 2. FIG.
The configuration of the AF processing system of the digital camera according to the second embodiment shown in FIG.
01 is provided with a high-frequency detector 113 for detecting high-frequency components of frequency data F (U, V) and G (U, V) in the image data subjected to the two-dimensional Fourier transform at 01. Since the configuration is the same as that in FIG. 1, the components having the same configuration are denoted by the same reference numerals as those in FIG.

In the digital camera of the second embodiment, the AF operation of the first embodiment, that is, the cross-correlation coefficient between the image data of the AF area corresponding to the light flux G and the image data of the AF area corresponding to the light flux M Is calculated, and the distance and direction to the in-focus position of the AF lens 101 are calculated based on the cross-correlation coefficient, and the approximate in-focus position is calculated.
Near the in-focus position, the final in-focus position is determined by the hill-climbing method. More specifically, AF detection is performed by the high-frequency detection unit 113.
The point at which the high-frequency component of the image data of the area has a peak is calculated, and the distance direction calculating unit 108 determines that the point at which the high-frequency component of the image data of the AF area has a peak is the focus position. Then, the AF optical system driving circuit 109 moves the AF lens 10 to the in-focus position calculated by the distance direction calculating unit 108.
Move 1

As described above, in the second embodiment, the AF diaphragm 102 has different color filters (the G filter 102a and the M filter 102) provided in the openings disposed at different eccentric positions with respect to the center of the AF lens 101. 102b) are arranged, and in imaging for AF, image data of an AF area in an image frame in an image frame corresponding to each light beam passing through each color filter 102a, 102b of the AF diaphragm 102 is acquired for each color, and a cross-correlation calculating unit 106 calculates the cross-correlation coefficient between the image data for each color, and the distance direction calculation unit 108 calculates the distance and direction to the in-focus position of the AF lens 101 based on the cross-correlation coefficient. , Find the approximate focus position,
Further, in the vicinity of the approximate focus position, the high-frequency detection unit 113
Is to determine the final focus position based on the high-frequency component of the image data of the AF area obtained by imaging, so that the focusing error (convergence to the maximum point which is one of the drawbacks of the hill-climbing method) And focusing can be performed with higher precision. Further, a frequency component can be obtained by a two-dimensional FFT unit used when calculating a cross-correlation coefficient,
A filter circuit for detecting a high-frequency component, which is normally required in the hill-climbing method, becomes unnecessary.

It should be noted that the present invention is not limited to the above-described embodiment, and can be appropriately changed without changing the gist of the invention.

The automatic focusing device of the present invention can be widely applied to image input devices using an image pickup device such as a video camera and a digital camera.

[0053]

As described above, according to the automatic focusing apparatus of the first aspect, the openings are respectively provided at different eccentric positions with respect to the center of the AF lens, and the respective openings have different colors. An AF aperture is provided for limiting a light flux from the subject to be focused, in which filters are arranged, respectively.
A part of an image frame to be imaged is set as an AF area, and in imaging for AF, image data of an AF area in each image frame corresponding to each light beam passing through each color filter of the AF diaphragm is obtained. , The cross-correlation between the image data of the AF areas is calculated, and the distance and direction to the in-focus position of the AF lens are calculated based on the cross-correlation.
Since the lens is driven to the in-focus position, the distance and direction to the in-focus position can be obtained by one shooting for AF, and a high-speed focusing operation can be performed.

According to the automatic focusing device of the second aspect, in the automatic focusing device of the first aspect, one of the different color filters is a primary color filter and the other is a complementary color filter. When a primary color system image sensor is used, the light beam that has passed through the primary color filter can be sensed by a certain color sensor, and the light beam that has passed through the complementary color filter can be sensed by the other two color sensors. It becomes easy to take in.

According to the automatic focusing device of the third aspect, in the automatic focusing device of the second aspect, the primary color filter is a G filter and the complementary color filter is an M filter. When used, the luminous flux passing through the G filter can be sensed by the G color sensor, and the luminous flux passing through the M filter can be sensed by the R and B color sensors, so that the two luminous fluxes can be easily separated and taken in. Becomes

According to the automatic focusing apparatus of the fourth aspect, the openings are respectively provided at eccentric positions different from each other with respect to the center of the AF lens, and different color filters are respectively disposed in the respective openings. An AF aperture is provided to limit the luminous flux from the subject to be focused, a part of an image frame to be captured is set as an AF area, and in imaging for AF, each luminous flux passing through each color filter of the AF aperture is provided. The image data of the AF area in each image frame corresponding to each is acquired, the cross-correlation between the image data of the AF areas is calculated, and the distance and direction to the focusing position of the AF lens are calculated based on the cross-correlation. Then, the approximate focus position of the AF lens is determined, and the final focus is determined based on the high-frequency component of the image data of the AF area obtained by imaging near the approximate focus position. Since it was decided to determine the position, thereby enabling high-speed and high-accuracy focusing operation.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a digital camera to which an automatic focusing device according to a first embodiment is applied.

FIG. 2 is a diagram showing an image frame of a primary color image sensor of FIG. 1 and an AF area thereof.

FIG. 3 is an explanatory diagram for describing light beam capture by an AF stop in FIG. 1;

FIG. 4 is an explanatory diagram illustrating the principle of AF of the digital camera in FIG. 1;

FIG. 5 is a diagram illustrating an example of a cross-correlation coefficient relationship between image data of a G filter and an M filter in an AF area.

FIG. 6 is a block diagram showing a configuration of a digital camera to which the automatic focusing device according to the second embodiment is applied.

[Explanation of symbols]

 Reference Signs List 101 AF lens 102 AF stop 102a G filter 102b M filter 103 Primary color system image sensor 104 Image memory (for AF area of filter G) 105 Image memory (for AF area of filter M) 106 Cross-correlation calculator 107 Cross-correlation coefficient memory 108 Distance direction calculation unit 109 AF optical system drive circuit 110 Aperture switching mechanism 111 Frame image memory 112 Mixing operation unit 113 High frequency detection unit 200 Two-dimensional FFT unit 201 Two-dimensional FFT unit 202 Complex multiplication unit 203 Two-dimensional IFFT unit

Claims (4)

[Claims] 1. An AF lens for focusing a subject to be focused, a color imaging means for converting a subject image of the focused subject to be formed into an electric signal and outputting it as image data, An aperture provided at an eccentric position different from the center with respect to the center, and a different color filter disposed at each of the apertures; and an AF stop for limiting a luminous flux from the subject to be focused; A part of the image frame picked up by the means is set as an AF area. In the image pickup for AF, the image data of the AF area in each image frame corresponding to each light beam passing through each color filter of the AF diaphragm is set. Respectively, calculating the cross-correlation between the image data of the AF areas, and based on the cross-correlation, up to the in-focus position of the AF lens. And calculates the distance and direction, an automatic focusing apparatus characterized by driving the AF lens in-focus position. 2. The automatic focusing device according to claim 1, wherein one of the different color filters is formed of a primary color filter, and the other is formed of a complementary color filter. 3. The automatic focusing apparatus according to claim 2, wherein said primary color filter comprises a G filter, and said complementary color filter comprises an M filter. 4. An AF lens for focusing a subject to be focused, a color imaging means for converting a subject image of the subject to be formed into an electric signal and outputting it as image data, and an AF lens. An aperture provided at an eccentric position different from the center with respect to the center, and a different color filter disposed at each of the apertures; and an AF stop for limiting a luminous flux from the subject to be focused; A part of the image frame picked up by the means is set as an AF area. In the image pickup for AF, the image data of the AF area in each image frame corresponding to each light beam passing through each color filter of the AF diaphragm is set. The cross-correlation between the image data is calculated, and the distance and direction to the in-focus position of the AF lens are calculated based on the cross-correlation. Out, and calculates the focus position of the outline of the AF lens, further, in-focus position near the outline, obtained by imaging A
An automatic focusing device for determining a final focusing position based on a high-frequency component of image data in an F area.