JPH05210040A - Camera provided with sight-line detection device - Google Patents

Abstract

PURPOSE:To decrease the operation of an button and to allow a photographer to concentrate on photographing by detecting a focus by fixing a gazing position decided to be gazing first by a gazing decision means. CONSTITUTION:By a CPU 1, the detection processing of spectacles, the detection processing of sight-line, the detection processing of the processing of photometry, the detection processing of rotational amount, the detection processing of operation, the driving control of a lens or the like are executed. Besides, the position of the line of the sight of the photographer observing a finder is detected based on a signal from a detection means for the light of sight 40. Then, it is decided whether a state set by an operation means 62 is a sight-line single focus detection mode (EG-S mode) or not. When it is the EG-S mode and a focus detection area set at this time is a small area, the focus detection position is taken as the focus detection position of the last time and the focus detection area is taken as the focus detection area of the last time, that means, the small area around the focus detection position. Therefore, when the variation of the position of the sight-line is temporarily extremely small in the EG-S mode, the size of the focus detection position and the focus detection area is fixed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a camera with a visual axis detecting device for controlling the photographing state of the camera based on the visual axis position of the photographer.

[0002]

2. Description of the Related Art Various applications of a line-of-sight detection device to a camera have been proposed (Japanese Patent Laid-Open Nos. 2-5 and 1-241).
511). First, the principle of line-of-sight detection will be briefly described with reference to FIGS. 39 and 40. Four bright spots are observed in the intraocular optical system. These bright spots are images of reflected light from the front surface of the cornea to the back surface of the crystalline lens, which are called Purkinje images, and are sequentially called 1, 2, 3, and 4 images from the front surface of the cornea.

The sight line detecting device detects the sight line position by using the Purkinje 1 image and the Purkinje 4 image. The Purkinje 1 and 4 images are observed while being changed by the rotation of the eyeball, and when the center is seen, as shown in FIG. 39 (A), the Purkinje 1 image and the Purkinje 4 image are observed to overlap, By looking at the right or left, FIG. 39 (B) or (C)
As shown in FIG.

The movement amounts of the Purkinje 1 image and the Purkinje 4 image are as shown in FIG.
If the rotation angle is about 0 °, it changes almost linearly with respect to the rotation angle. Therefore, the eyeball rotation angle can be known by measuring these movement amounts.

The line-of-sight detection device is incorporated in the viewfinder of the camera, detects these bright spots with a light receiving element such as a two-dimensional CCD, finds the barycentric position of each bright spot, and rotates the eyeball from the relationship of FIG. By knowing the corner, it is possible to detect which position in the finder the photographer is looking at.
In addition to those using Purkinje 1 and 4 images,
A method using the Purkinje 1 image and the center of the pupil and a scleral reflection method are known.

[0006]

A camera having a conventional line-of-sight detection device starts the line-of-sight position detection by operating a specific button (release button, button for exclusive use of line-of-sight detection, etc.), and focuses in the AF area of the line-of-sight position. The detection and the automatic exposure (AE) were performed. However, a specific button operation is required to fix the AF area determined by the line of sight, and when a subject moves in the AF area, a similar button operation is required each time.

Generally, the variation of the line of sight for a moving subject is large, and the variation of the line of sight for a stationary subject is small. In addition, a continuous mode (hereinafter referred to as AF-C mode) that always drives the shooting lens according to the focus detection result for a moving subject, and a focus detection result before focusing for a still subject are used. There are two focus detection modes, a single (one-shot) mode (hereinafter referred to as AF-S mode) that drives the taking lens according to the above, and prohibits the taking lens from being driven after focusing once. The selection of the detection mode was not linked to the gaze variation amount. An object of the present invention is to provide a camera with a line-of-sight detection device that solves the above-mentioned problems, has few button operations, and can concentrate on shooting.

[0008]

In order to solve the above problems, a first solution means of a camera having a line-of-sight detection device according to the present invention is a line-of-sight detection unit for detecting the line-of-sight position of a photographer, and the line-of-sight detection. Focus detection is performed by fixing the gaze determination unit that determines whether or not the photographer is gazing based on the detection results of the past multiple times of the unit, and the gaze position first determined to be the gaze by the gaze determination unit. And a focus detecting means having a single line-of-sight focus detection mode. A second solution means is that the focus detection means further comprises:
Has a line-of-sight continuous focus detection mode in which the focus detection area is moved along with the movement of the detected gaze position to constantly perform focus detection, and the line-of-sight single focus detection mode and the line-of-sight continuous focus detection mode are selected. It is characterized in that it has a mode selection means for performing.

A third solving means is a visual axis detecting means for detecting the visual axis position of the photographer, and a gaze for judging whether or not the photographer is gazing based on a plurality of past detection results of the visual axis detecting means. Determining means, focus detecting means for performing focus detection in a focus detecting area corresponding to the gaze position determined by the gaze determining means, and detecting in-focus or in-focus of the area, and focus detecting area of the focus detecting means When the focus position continues to be in the gaze position for a predetermined time after focusing, the focus detection area fixing means for fixing the focus detection area by the focus detection means thereafter is included.

A fourth solution means is to detect the line of sight of the photographer and to detect the focus in a focus detection area corresponding to the line of sight position, and once focus is achieved, the focus detection area at the time of focusing is performed. And a focus detection means for performing focus detection.

A fifth solving means is a visual axis detecting means for detecting the visual axis position of the photographer, and a gaze for judging whether or not the photographer is gazing based on the detection results of the past plural times of the visual axis detecting means. The configuration includes a determination unit and a line-of-sight control unit having a single line-of-sight mode for fixing the gaze position first determined to be gazed by the gaze determination unit. In a sixth solution means, the line-of-sight control means further has a line-of-sight continuous mode for moving the line-of-sight position based on the latest line-of-sight position, and selects the line-of-sight single mode or the line-of-sight continuous mode. It is characterized by having a line-of-sight mode selection means. In the seventh solution,
The gaze determination means is that the distance between the last and present gaze positions is less than or equal to a predetermined value, the difference between the previous and present eyeball rotation angles is less than or equal to a predetermined value, and the moving speed of the gaze is less than or equal to a predetermined value. It is characterized in that it is determined that the user is gazing when at least one of the conditions is satisfied or when the condition is satisfied and the time satisfying these conditions is a predetermined value or more. In the eighth solution means, there is a focus detection means for detecting whether the focus detection area on the photographing screen is in focus or out of focus, and the first and second AF modes for controlling the focus detection means.
An F control means and an AF mode selection means for selecting at least one AF mode of the AF control means are provided, and the gaze mode selection means includes the AF mode selection means for selecting the first AF mode. In this case, the line-of-sight single mode is selected, and when the second AF mode is selected, the line-of-sight continuous mode is selected. In a ninth solving means, the first AF mode is an AF single mode in which the driving of the photographing lens is prohibited once the focusing of the photographing lens is once detected, and the second AF mode is the focus mode. It is characterized by being in the AF continuous mode in which the focus of the taking lens is constantly adjusted based on the output of the detecting means. A tenth solution means includes a moving body determination means for determining whether or not the subject is a moving body, and the line-of-sight mode selection means selects the line-of-sight continuous mode based on the determination result of the moving body determination means. It is characterized by doing.

An eleventh solving means is a line-of-sight detecting means for detecting the line-of-sight position of the photographer, and a gaze for judging whether or not the photographer is gazing based on a plurality of past detection results of the line-of-sight detecting means. It is configured to include a determination unit and a line-of-sight control unit having a line-of-sight multi-mode that is fixed at the previous gaze position until the gaze determination unit determines the gaze position.
In a twelfth solving means, the line-of-sight control means further includes
It is characterized by having a line-of-sight continuous mode for moving the line-of-sight position based on the latest line-of-sight position, and having line-of-sight mode selection means for selecting the line-of-sight multi-mode and the line-of-sight continuous mode.

[0013]

According to the present invention, the detected movement of the line of sight is caused by a predetermined condition (hereinafter, gaze at a specific position).
This is called the EG-S mode lock condition. ) Single line-of-sight mode (EG-S mode) that fixes the line-of-sight position when
And at least two line-of-sight modes such as a line-of-sight continuous mode (EG-C mode) for continuously detecting the line of sight. If the line-of-sight mode can be switched in association with the switching of the AF mode, the AF area can be easily switched with a few button operations.

[0014]

【Example】

(First Embodiment) The present invention will be described in detail below with reference to the drawings and the like. FIG. 1 is a block diagram showing a first embodiment of a camera having a visual line detection device according to the present invention. The lens barrel 10 is replaceable with respect to the camera body 20 and can be detachably mounted. A taking lens 11 is built in the lens barrel 10. The photographing lens 11 is a lens whose focus can be adjusted by moving in the optical axis direction.

When the lens barrel 10 is attached to the camera body 20, the photographing light flux coming from the subject passes through the photographing lens 11 and is guided to the main mirror 21 provided in the camera body 20. A part of the photographing light flux is reflected to the finder side by the main mirror 21 and passes through the display means 23, the screen 24, the pentaprism 25, and the eyepiece lens 26, so that the screen image is observed by the photographer. The other part of the photographing light flux is the main mirror 21.
Is transmitted, is reflected by the sub-mirror 22, and is guided to the focus detection means 30 as a light beam for focus detection.

Further, inside the camera body 20, in the vicinity of the eyepiece 26, in addition to the line-of-sight detecting means 40, the eyeglasses detecting means 50 and the shutter means 27, which will be described later, well-known camera internal mechanisms such as unillustrated winding means are arranged. Has been done. The photometric unit 29 performs spot photometry or center-weighted photometry of the central portion within the photographic screen by the finder light flux branched by the half mirror 28.

FIG. 2 is a perspective view showing the structure of the focus detecting means incorporated in the camera of the first embodiment. The focus detection unit 30 includes a field mask 31 having a two-dimensional opening 31A, a field lens 32, a diaphragm mask 33 having a pair of openings 33A and 33B, and a pair of re-imaging lenses 34A and 34B. The photoelectric conversion means 35 includes light receiving portions 35A and 35B in which light receiving elements are two-dimensionally arranged.

The exit pupil 12 of the taking lens 11 has an optical axis 1
A pair of regions 12A and 12B that are symmetric with respect to 3 is included, and the light flux passing through these regions 12A and 12B forms a primary image near the field mask 31. This field mask 3
As shown in FIG. 3, 1 has an opening corresponding to the focus detectable range M. The primary image formed in the opening 31A of the field mask 31 is reflected by the field lens 32, the pair of openings 33A and 33B of the diaphragm mask 33, and the pair of re-imaging lenses 34A and 34B, and the pair of photoelectric conversion means 35. Are formed as a pair of secondary images on the light receiving portions 35A and 35B. The photoelectric conversion means 35 outputs a pair of secondary images to the light receiving portion 35.
The defocus amount of the taking lens 11 can be detected by detecting the relative positional relationship in the arrangement direction of A and 35B using the subject image signal generated by the photoelectric conversion means 35.

Light receiving portions 35A and 35B of the photoelectric conversion means 35
3 covers the area M on the screen N, as shown in FIG. 3, and this area M is the range in which focus detection is possible. The above-mentioned positional relationship can be detected in the designated position P and the designated size area Q on the photographing screen, and inside the focus-detectable area M, the focus of any position and size can be obtained. Focus detection can be performed by the detection area. For example, it becomes possible to arbitrarily change the focus detection area based on the result of line-of-sight position detection described later.

The CPU 1 shown in FIG. 1 carries out calculations for carrying out eyeglass detection processing, line-of-sight detection processing, focus detection processing, photometric processing, rotation amount detection processing, operation detection processing, lens drive control, shutter control, display control, etc., which will be described later. It is a processing means. The outputs of the light receiving portions 35A and 35B of the focus detection means 30 are CP
It is connected to U1.
Focus detection calculation is performed to obtain the defocus amount from the positional relationship of the images. The CPU 1 controls the motor 60 according to the obtained defocus amount to drive the taking lens 11 to the in-focus position. Also, the CPU 12 on the lens side
CPU 1 built in the lens barrel 10 on the body side
In response, various lens data (focal length, etc.) is transmitted.

Further, the focus detection area and the focus detection result are
It is displayed on the finder screen by the display means 23. The display of the display unit 23 is, for example, as shown in FIG.
As shown in FIG. 4, the focus detection area Q is shaded and displayed, and after focusing, as shown in FIG. 4, the inside of the focus detection area Q ′ can be transparent and only be framed.

When the release button 61 is pressed halfway from the released state, the operation of the CPU 1 is reset and the photographing standby state is set. When the release button 61 is fully pressed, the CPU 1 activates the exposure operation by the shutter means 27 in the normal mode. However, in the AF priority mode, the exposure operation by the shutter means 27 is not activated until the subject is in focus even when the shutter button is fully pressed.

The operating means 62 is a means for selecting a single line-of-sight focus detection mode and a continuous line-of-sight focus detection mode, which will be described later.
The operation mode is switched according to the selection of 2. here,
The gaze single focus detection mode is the release button 61.
In this mode, the focus detection area is fixed to the gaze position detected for the first time after the half-pushing operation is performed. In the line-of-sight continuous focus detection mode, the focus detection area is always set to the gaze position.

The rotation amount detecting means 63 is a means for detecting the rotation amount of the body 20, and the information on the rotation amount is the CPU 1
Sent to. 5 to 7 are views showing rotation amount detecting means used in the camera according to the embodiment, FIG. 5 is a front view, FIG. 6 is a plan view showing an encoder, and FIG. 7 is a detecting portion. It is a figure. The rotation amount detecting means 63 is arranged on the bottom surface of the body 20. At this time, the body 20 is attached to the tripod seat 71 of the tripod 70 with the tripod female screw 64B and the tripod male screw 64.
It is attached by A. On the upper surface of the tripod base 71, as shown in FIG. 6, an encoder including a high reflection part 73 and a low reflection part 72 in the circumferential direction is formed around the tripod male screw 64A.

As shown in FIG. 7, the rotation amount detecting means 63 is of a reflection detecting type including a light projecting portion 63A and a light receiving portion 63B. When the body 20 rotates around the tripod screw 64, The amount of rotation is detected in the form of the number of reflected light pulses as the movement of the tripod base 71 relative to the encoder. The rotation direction can be detected, for example, by disposing the two light receiving parts at different positions and determining the phase relationship between the two pulse signals. The rotation amount detection unit 63 is not limited to this, and may be any unit that can detect the rotation amount and the rotation direction of the body 20.

FIG. 8 is a view showing the line-of-sight detecting means used in the camera according to the embodiment. The line-of-sight detection means 40 includes an infrared light surface emitting element 41, a half mirror 42, a lens 43,
It is composed of a dichroic mirror 44 and the like. The infrared light emitted from the infrared light emitting element 41 is reflected by the half mirror 42 and reflects the infrared light installed in the lens 43 and the eyepiece lens 26.
4 and is projected onto the eye 45 of the finder observer. In this optical system, the shape and position of the optical member are set such that the light emitting surface of the infrared light emitting element 41 overlaps with the viewfinder screen in shape and position.

The infrared light projected on the eyes 45 of the observer is reflected by the retina 46, returns to the eyepiece lens 26 again, and follows the path opposite to that when it is emitted, and is reflected by the infrared light reflecting dichroic mirror 44. , Lens 43, half mirror 42
And is received by the surface light receiving element 47. The surface light receiving element 47 may be a two-dimensional position sensor or a two-dimensional image sensor. The operation of the infrared light surface emitting element 41 is controlled by the CPU 1, and the output of the surface light receiving element 47 is sent to the CPU 1 for processing.

In the above structure, as shown in FIG.
Since the reflection efficiency at the line-of-sight position is higher than in other directions, the amount of light received at the position on the surface light-receiving element 47 corresponding to the position viewed by the observer on the screen 24 of the finder becomes larger than in other regions. .. In order to remove the influence of infrared light from the outside incident from the viewfinder, the surface light receiving element 47 when the surface light emitting element 41 emits light and when it does not emit light.
It is possible to detect the line-of-sight position on the screen N by taking the difference (shown in FIG. 9) in the distribution of the amount of received light of the image and the position R indicating the maximum amount of received light in the distribution. Further, when the absolute value of the maximum amount of received light is small, it can be determined that the line of sight cannot be detected because the finder is not observed. In the above structure, beam scanning may be performed two-dimensionally instead of the surface emitting element 41. Furthermore, line-of-sight detection other than the above configuration may be performed.

FIG. 10 and FIG. 11 are views showing an example of the configuration of the glasses detecting means of the camera according to the embodiment, FIG. 10 is a plan view and FIG. 11 is a front view. Glasses detection means 50
Is composed of an infrared light emitting element 51, lenses 52 and 54, a surface light receiving element 55 and the like. The infrared light emitted from the infrared light emitting element 51 is projected onto the finder observer via the lens 52. The light projecting portion S (shown in FIG. 11) is set at a portion below the eyes 45 where the glasses 53 are present in order to avoid reflection by the eyeball. Therefore, the eyeglasses detection means 50 uses the eyepiece lens 26 in the body 20.
It is better to place it at the bottom of.

When the observer wears the glasses 53 at the normal finder observation position, as shown in FIG.
The light is reflected by the surface of the eyeglasses 53 and is received by the lens 54 and the surface light receiving element 55. The operation of the infrared light emitting element 51 is C
Controlled by PU1, the output of the surface light receiving element 55 is CP
It is sent to U1 and processed. The light projection angle with respect to the glasses 53 is set so that the light enters the surface of the glasses 53 at a shallow angle so that the reflectance is high. Further, the surface light receiving element 55 is a two-dimensional position sensor or a two-dimensional position sensor.
It may be a three-dimensional image sensor.

In the above structure, when the glasses 53 are present, the reflection in the specific direction is higher than in other directions, so that the light receiving amount in the small area on the surface light receiving element 55 is higher than that in other portions. It becomes larger by projecting. In order to remove the influence of infrared light incident from the outside, the difference in the light receiving amount distribution of the surface light receiving element 55 when the surface light emitting element 51 emits light and when it does not emit light is taken, and the maximum light receiving amount of the distribution is a predetermined value or more. Therefore, it can be determined that the glasses 53 are present when the user concentrates on one small area. Further, even when the normal glasses 53 are not worn, since there is some reflection due to the skin of the observer,
When the maximum amount of received light is extremely small, it can be determined that the viewfinder is not observed.

Further, in the above structure, one infrared light emitting element 51 is arranged. However, the reflection direction varies depending on the distance between the finder and the glasses 53 and the angle of the front surface of the glasses 53. Since there is a risk of false detection,
The direction of the projection beam may be two-dimensionally scanned.

As shown in FIG. 12, the glasses detecting means 50 includes infrared light emitting elements 51A, 51B and 51C, and a lens 5.
The light may be projected in a plurality of directions by 2A, 52B and 52C. Further, as shown in FIG. 13, one infrared light emitting element 51 and lenses 52A, 5 formed integrally with each other.
2B and 52C may be used to project light in a plurality of directions.

FIG. 14 is a flowchart showing the operation of the CPU of the camera according to the embodiment. In step 100, the CPU 1 starts the operation by half-pressing the release button 11. In step 101, it is detected based on a signal from the glasses detection means 50 that the photographer observing the viewfinder wears the glasses 53. In step 102, if the eyeglasses 53 are detected, the line-of-sight detection after step 103 may be erroneously detected.
The line of sight is not detected and the process proceeds to step 112. If no glasses are detected, the process proceeds to step 103.

In step 103, the line-of-sight position of the photographer observing the viewfinder is detected based on the signal from the line-of-sight detecting means 40. In this flowchart, the line-of-sight position is detected once in one focus detection sequence, but the line-of-sight position detection may be performed at a timing independent of the focus detection sequence (for example, a timer interrupt every predetermined time). Good. In step 104, it is judged whether or not the sight line position cannot be detected, and if it cannot be detected, the process proceeds to step 109. If it can be detected, the process proceeds to step 105.

In step 105, the variation amount of the line-of-sight position is detected. Here, a method of detecting the variation amount of the line-of-sight position will be described. FIG. 15 is a diagram showing how the line-of-sight position varies. The left side of the screen N shows the case where the variation of the line-of-sight position is large, and the right side shows the case where the variation is small. R (X0, Y0) = R0: latest line-of-sight position R (X1, Y1) = R1: line-of-sight position when detecting the line-of-sight position one time before ... R (Xn, Yn) = Rn: line-of-sight position n times before Gaze position at detection

(Variation amount detection method 1) In the variation amount detection method shown in FIG. 16, the integrated value of the distance over which the line-of-sight position moves within a predetermined time is used as the variation amount parameter P. That is, In the equation (1), the line-of-sight position Ri is taken for those within a predetermined time from the present (R0 to Rk).

(Variation amount detection method 2) In the variation amount detection method shown in FIG. 17, the area F of the circumscribing polygon that is convex outward and includes all the line-of-sight positions within a predetermined time is used as the variation amount parameter P. That is, P = F (2) In the formula (2), the line-of-sight position Ri is taken for those within a predetermined time from the present (R0 to Rk).

(Variation amount detection method 3) In the variation amount detection method shown in FIG. 18, the area or radius G of the circumscribing circle including all the line-of-sight positions within a predetermined time is used as the variation amount parameter P. That is, P = G (3) In the equation (3), the line-of-sight position Ri is taken for those within a predetermined time from the present (R0 to Rk).

(Fluctuation amount detection method 4) In the fluctuation amount detection method shown in FIG. 19, the sum or large value H of the differences Xs and Ys between the maximum and minimum values of the line-of-sight position in the x and y directions within a predetermined time.
Is the variation parameter P. That is, P = H = Xs + Ys or MAX (Xs, Ys) Xs = MAX (X0, X1, ..., Xk) -MIN (X0, X1, ..., Xk) Ys = MAX (Y0, Y1 ,. .., Yk) -MIN (Y0, Y1, ..., Yk) (4) In the formula (4), the line-of-sight position Ri is taken for those within a predetermined time from the present (R0 to Rk).

(Variation amount detection method 5) In this variation amount detection method, the sum or maximum value σ of the standard deviations σx and σy of the line-of-sight position in the X and Y directions within a predetermined time is used as the variation parameter P.
And That is, P = σ = σx + σy or MAX
(Σx, σy) In the equation (5), the line-of-sight position Ri is taken for those within a predetermined time (R0 to Rk) from the present.

Returning to FIG. 14, in step 106, the line-of-sight fluctuation amount P is compared with the predetermined value K1, and if the line-of-sight fluctuation amount P is large, the process proceeds to step 109, and if it is small, the process proceeds to step 107. In step 107, the line-of-sight variation P
If is small, the photographer determines that the photographer is in a gaze state, and detects the gaze position based on the line-of-sight position S (X, Y) (see FIG. 15).

The gaze position detecting method is as follows. (Gaze position detection method 1) The position of the center of gravity of the area F of the convex circumscribing polygon is defined as the gaze position S (X, Y) on the outside including all the gaze positions within a predetermined time. (Gaze position detection method 2) The center position of the circumscribing circle including all the gaze positions within a predetermined time is defined as the gaze position S (X, Y). (Gaze position detection method 3) X, Y of the gaze position within a predetermined time
The average of the maximum value and the minimum value in the direction is set as the gaze position S (X, Y). That is, X = (MAX (X0, X1, ..., Xk) + MIN (X0, X1, ..., Xk)) / 2 Y = (MAX (Y0, Y1, ..., Yk) + MIN (Y0 , Y1, ..., Yk)) / 2 (6) In the equation (6), the line-of-sight position Ri is taken for those within a predetermined time from the present (R0 to Rk). (Gaze position detection method 4) X, Y of the gaze position within a predetermined time
Gaze position S (X, Y)
).

Returning to FIG. 14 again, in step 108, it is judged whether or not the gaze position S (X, Y) is outside the focus detectable range M, and if it is outside the focus detectable range M, step Proceed to 109 (see FIG. 20). Gaze position S (X, Y)
Is within the focus detectable range M, step 11
Go to 0. In step 109, the elapsed time after entering this step is measured. If the elapsed time is within the predetermined time, the process proceeds to step 111, and if it is the predetermined time or more, the process proceeds to 112.

In step 110, the focus detection position P is set as the gaze position S (X, Y) and the process proceeds to step 113. Therefore, when there is little variation in the line-of-sight position of the photographer, focus detection is performed at the gaze position. In step 111, the focus detection position P is set as the previous focus detection position, and step 115
Proceed to. Therefore, if the line-of-sight position is undetectable, or the amount of line-of-sight variation is large, or if the gaze position is outside the focus detectable range and within a predetermined time after entering the above state, the focus immediately before entering the above state Focus detection is performed at the detection position (see FIG. 20). In step 112, the focus detection position P is set to the center position, and the process proceeds to step 115. Therefore, if the line-of-sight position cannot be detected, or the amount of change in line-of-sight is large, or if the gaze position is outside the focus detectable range, and if more than a predetermined time has elapsed since the above state, or if it is determined that glasses are present , The focus is forcibly detected at the center position of the screen (see FIG. 20).

Next, at step 113, the line-of-sight variation P
Is compared with a predetermined value K2 (<K1). If the line-of-sight variation amount P is large, the process proceeds to step 115.
Go to 4. In step 114, the focus detection area Q is set to a small area around the focus detection position P (see FIG. 21), and the AF single mode (AF-S) in which the focus is locked after the automatic focus adjustment mode is focused. Mode), the process proceeds to step 116. Therefore, if the change in the line-of-sight position is extremely small, spot-like focus detection is performed at the gaze position, and it is determined that the subject is stationary or the composition has not been changed, and the focus is adjusted after focusing. It will be locked mode.

In step 115, the focus detection area Q is set to a large area around the focus detection position P (see FIG. 22).
At the same time, the automatic focus adjustment mode is set as the AF continuous mode (AF-C mode) for continuing the servo without focus lock after focusing, and the routine proceeds to step 116. Therefore, if there is very little change in the line-of-sight position, or if the line-of-sight position cannot be detected or the amount of change in the line-of-sight position is large, or if the line-of-sight position is outside the focus-detectable range or if there are glasses, the wide range around the line-of-sight position Focus detection is performed in the area. In addition, it is determined that the subject is moving or the composition has been changed, and the lens drive mode is continued even after focusing.

In step 116, the set focus detection area is displayed on the screen by the display means 23, and step 1
Proceed to 17. In step 117, the set AF
It is determined whether or not the mode is AF-C mode. If the mode is AF-C mode, the process proceeds to step 118. If the mode is AF-S mode,
Proceed to step 119. In step 118, the focus focus lock flag is reset, and the process proceeds to step 120. In step 119, if the focus focus lock flag is set, focus detection and lens driving are not performed, and the process proceeds to step 126. If not, the process proceeds to step 120. Step 120
Then, a well-known focus detection calculation is performed using the output signal from the photoelectric conversion means 35 corresponding to the set focus detection area to obtain the focus detection result (defocus amount), and step 1
Proceed to 21. In step 121, the focus detection result is displayed on the screen by the display means 23, and the process proceeds to step 122.

In step 122, it is determined whether or not the focus detection result (defocus amount) is in focus. If it is in focus, the process proceeds to step 124, and if it is out of focus, the process proceeds to step 123. move on. In step 123, the motor 60 is controlled according to the focus detection result (defocus amount) to drive the taking lens 11 to the in-focus point, and then step 1
Proceed to 24. In step 124, the set AF
It is determined whether or not the mode is the AF-C mode. If the mode is the AF-C mode, the process proceeds to step 126, and if the mode is the AF-S mode, the process proceeds to step 125. At step 125, the focus lock flag is set and the routine proceeds to step 126.

In step 126, it is determined whether or not the setting state of the operating means 62 is the line-of-sight single focus detection mode (EG-S mode). If it is the line-of-sight single focus detection mode, the process proceeds to step 127, and the line-of-sight is selected. If it is the continuous focus detection mode, the process returns to step 101.
Therefore, in the line-of-sight continuous focus detection mode, the gaze position is always detected, and when the gaze position is detected, focus detection is performed at that position (see FIG. 23).

In step 127, if the focus detection area set this time is a small area, the operation proceeds to step 128,
If it is not a small area, the process returns to step 101. Therefore, even in the line-of-sight single focus detection mode, the gaze position is always detected unless the change in the line-of-sight position is extremely small.
When the gaze position is detected, focus detection is performed at that position.

At step 128, the focus detection position P is set as the previous focus detection position, and the routine proceeds to step 129. In step 129, the focus detection area is set to the previous focus detection area, that is, a small area around the focus detection position P, and step 1
Return to 16. Therefore, in the line-of-sight single focus detection mode, once the change of the line-of-sight position becomes extremely small, the size of the focus detection position and the size of the focus detection area are fixed (see FIG. 3).

(Second Embodiment) FIG. 24 is a flow chart showing the operation of the second embodiment of the camera having the visual axis detection device according to the present invention. In this embodiment, steps 126 to 129 of the flow chart of the embodiment shown in FIG.
The process proceeds from step 119, 123, 124, 125 to step 130.

That is, in step 130, based on the result of focus detection, if not in focus, the process returns to step 101, and if in focus, the process proceeds to step 131. Therefore, the gaze position is always detected during non-focus, and when the gaze position is detected, focus detection is performed at that position (see FIG. 23).

At step 131, the focus detection position P is set as the previous focus detection position, and the routine proceeds to step 132. In step 132, the focus detection area is set to the previous focus detection area, that is, a small area around the focus detection position P, and step 1
Proceed to 33. In step 133, the set focus detection area is displayed on the screen by the display means 23, and step 1
Proceed to 34. In step 134, a well-known focus detection calculation is performed using the output signal from the photoelectric conversion means 35 corresponding to the set focus detection area, the focus detection result (defocus amount) is obtained, and the process proceeds to step 135. Step 13
In step 5, the focus detection result is displayed on the screen by the display means 23, and the process proceeds to step 136. In step 136, the motor 60 is controlled according to the focus detection result (defocus amount), the photographing lens 11 is driven to the in-focus point, and the process returns to step 131.

Once the focus is obtained by the above operation, the focus detection position and the size of the focus detection area at the time when the focus is obtained are fixed and used thereafter. Also,
If step 136 is omitted, the lens drive can be prohibited once the focus is achieved.

(Third Embodiment) FIG. 25 is a flow chart showing the operation of the third embodiment of the camera having the visual axis detection device according to the present invention. In FIG. 25A, step 20
At 0, the CPU 1 starts the operation by half-pressing the release button 11. In step 201, a well-known focus detection calculation is performed using the output signal from the photoelectric conversion means 35 corresponding to the focus detection area having the position and size fixed and set in advance, and the focus detection result (defocus amount) Is obtained, the process proceeds to step 202. In step 202, the focus detection result is displayed on the screen by the display means 23, and the process proceeds to step 203. In step 203, the motor 60 is controlled according to the focus detection result (defocus amount), the photographing lens 11 is driven to the in-focus point, and the process proceeds to step 204. In step 204, based on the result of focus detection, if the object is not in focus, the procedure returns to step 201, and if the object is in focus, the procedure proceeds to step 205. Therefore, focus detection is always performed during non-focusing, and the lens is driven according to the result.

In step 205, the line-of-sight position of the photographer observing the viewfinder is detected based on the signal from the line-of-sight detecting means 40. In this embodiment, the line-of-sight position is detected once by one focus detection sequence, but the line-of-sight position detection may be performed at a timing independent of the focus detection sequence (for example, a timer interrupt every predetermined time). Good. In step 206, if the detected line-of-sight position does not match the focus detection position, step 20
5, the process proceeds to step 207 if they match. However, a margin is added to the position matching judgment. Further, the gaze position may be compared instead of the gaze position.

In step 207, the elapsed time after entering this step is measured. If the elapsed time is within the predetermined time, the process returns to step 205, and if it is more than the predetermined time, the process proceeds to step 208. In step 208, the exposure operation by the shutter means 17 is activated.

After focusing by the above operation, if the line of sight is held for a certain time or more at the focus detection position where the focus is obtained, the exposure operation can be automatically performed. Although the focus detection position is fixed in the above description, the focus detection position may be determined in the single line-of-sight focus detection mode.

FIG. 25B shows a modified example of FIG. 25A, in which step 208 is omitted and in step 207, the focus detection position at which the focus is obtained after the focus is obtained. When the line of sight is held for a certain period of time or more, the focus is locked in step 209.

Before focusing, the focus detection position is changed according to the line-of-sight position, and instead of step 208, the focus detection position and the size of the area are fixed and the process returns to step 201. By doing so, the focus detection region may be locked when the line of sight is held for a certain period of time or more at the focus detection position where the focus is obtained after focusing.

(Fourth Embodiment) FIG. 26 is a flow chart showing the operation of the fourth embodiment of the camera having the visual axis detection device according to the present invention. In this embodiment, step 204 and subsequent steps in the flow chart of the embodiment shown in FIG. 25A are changed, and the process proceeds from step 203 to step 210.

In step 210, if the result of focus detection indicates that the subject is not in focus, the process returns to step 201, and if the subject is in focus, the process proceeds to step 211. Therefore, during non-focus, the focus is always detected, and the lens is driven according to the result. In step 211, the elapsed time after entering this step is measured. If the elapsed time is within the predetermined time, the process returns to step 201, and if it is longer than the predetermined time, 212 is performed.
Proceed to. Therefore, the focus is always detected for a predetermined time after focusing, and the lens is driven according to the result. In step 212, the exposure operation by the shutter means 17 is activated, and the process proceeds to step 213. In step 213,
It waits for the end of the exposure operation, and returns to step 201.

When the in-focus state is maintained for a certain period of time after the in-focus state has been achieved by the above-described operation, the exposure operation can be automatically performed. However, it is possible to solve the drawbacks that there is no room to decide the composition and that the number of operation members increases and the operation becomes complicated when the exposure operation and the focus adjustment operation are performed independently.

(Fifth Embodiment) FIG. 27 is a flow chart showing the operation of the fifth embodiment of the camera having the visual axis detection device according to the present invention. In step 300, the CPU 1 starts the operation by half-pressing the release button 11. In step 301, a well-known focus detection calculation is performed using the output signal from the photoelectric conversion means 35 corresponding to the focus detection area having the position and size fixed in advance, and the focus detection result (defocus amount) , And go to step 302. In step 302, the focus detection result is displayed on the screen by the display means 23, and the process proceeds to step 303. In step 303, the motor 60 is controlled according to the focus detection result (defocus amount), and the photographing lens 11
To the focal point, and the process proceeds to step 304. In step 304, as a result of focus detection, if the object is not in focus, the procedure returns to step 301, and if the object is in focus, the procedure proceeds to step 305. Therefore, during non-focus, the focus is always detected, and the lens is driven according to the result. Step 305
Then, the line-of-sight position of the photographer who is observing the viewfinder is detected based on the signal from the line-of-sight detecting means 40. In this embodiment, the line-of-sight position is detected once in one focus detection sequence, but the line-of-sight position detection may be performed at a timing independent of the focus detection sequence (for example, a timer interrupt every predetermined time). Good.

In step 306, the line-of-sight position detected immediately after focusing is set as the initial line-of-sight position R (X0, Y0), and the flow advances to step 307. The gaze position may be used instead of the gaze position. In step 307, the amount of rotation of the camera is initialized by the transition to the focused state, and step 3
Go to 08. Therefore, the rotation angle of the camera body can be detected by accumulating the pulse signals from the rotation amount detecting means 63 after that according to the rotation direction.

In step 308, the line-of-sight position R (X, Y) of the photographer observing the viewfinder is detected based on the signal from the line-of-sight detecting means 40 (see FIG. 28), and the process proceeds to step 309. In step 309, the line-of-sight rotation amount Wa on the object side is detected according to the line-of-sight displacement amount (X−X0) from the initial line-of-sight position and the focal length f of the photographing lens obtained from the lens CPU 12. That is, Wa = TAN ^-1 ((X-X0) / f) (7) or may be expressed using the magnification B and the subject distance d. That is, Wa = TAN ⁻¹ ((X−X0) / (d · B)) (8)

At step 310, the rotation amount Wc of the camera body 20 with respect to the focused object 90 is detected by integrating the pulse signals from the rotation amount detection means 63 after focusing according to the rotation direction (see FIG. 29), and proceeds to step 311. In step 311, the line-of-sight rotation amount Wa
If does not match the camera rotation amount Wc, the process returns to 301, and if it does match, the process proceeds to 312. However, it is preferable that a margin is provided for the matching determination. If the displacement of the line-of-sight position in the Y direction (Y-Y0) is greater than or equal to the predetermined value, the process returns to step 301 regardless of the rotation amount. In step 312, the photometric value from the photometric means 29 is held at the value immediately after focusing, and the process returns to step 312.

By the above operation, while the amount of rotation of the camera after focusing and the amount of rotation of the line of sight match, that is, when the composition is changed but the same subject is seen, photometry is automatically performed. Since the value and focus adjustment state are locked in the same state as when focusing, focus lock after focusing as in the past,
AE lock operation, focus lock after focusing,
There is no need to switch between AE lock mode and non-AE lock mode. Further, in this embodiment, the rotation amount in the horizontal direction of the screen is detected, but the rotation amount in the vertical direction of the screen may be detected. Further, the configuration of the camera body rotation amount detecting means is not limited to the above-mentioned embodiment, and a gyro or the like may be used.

(Sixth Embodiment) FIG. 30 is a flow chart showing the operation of the sixth embodiment of the camera having the visual axis detection device according to the present invention. In step 400, the CPU 1 starts the operation by half-pressing the release button 11. In step 401, the line-of-sight position of the photographer observing the viewfinder is detected based on the signal from the line-of-sight detecting means 40. In this embodiment, the line-of-sight position is detected once in one focus detection sequence, but the line-of-sight position detection may be performed at a timing independent of the focus detection sequence (for example, a timer interrupt every predetermined time). Good. Further, the gaze position may be used instead of the gaze position. Step 40
In step 2, if the detected line-of-sight position does not match the position of the scale 232 (see FIG. 32) displayed by the display unit 23 on the finder screen, step 40
Return to 1 and if they match, proceed to step 403. However, it is preferable to give some margin to the position coincidence determination. Further, the gaze position may be compared instead of the gaze position. In step 403, the motor 60 is controlled according to the distance display value of the line-of-sight position (3 m in FIG. 32) to drive the taking lens 11 to a position corresponding to the distance set by the line-of-sight, and the process returns to step 401. In addition to the scale 232, PF (power focus) and PZ (power zoom) are displayed on the viewfinder screen.
There is a mode selection display 231 for switching between, and it is possible to select by the line of sight in the same manner as the operation described above. The operation shown in FIG. 30 is an operation when the PF is selected.

With the above operation, by selecting the desired distance on the scale on the viewfinder screen with the line of sight, it is possible to adjust the distance without manual operation, so that both hands can be used like when a telephoto lens is held by hand. It is convenient when it is blocked.

(Seventh Embodiment) FIG. 31 is a flow chart showing the operation of the seventh embodiment of the camera having the visual axis detection device according to the present invention. In step 500, the CPU 1 starts the operation by half-pressing the release button 11. In step 501, the line-of-sight position of the photographer observing the viewfinder is detected based on the signal from the line-of-sight detecting means 40. In this embodiment, the line-of-sight position is detected once in one focus detection sequence, but the line-of-sight position detection may be performed at a timing independent of the focus detection sequence (for example, a timer interrupt every predetermined time). Good. Further, the gaze position may be used instead of the gaze position.

In step 502, if the detected line-of-sight position does not match the position of the scale 232 displayed by the display means 23 on the viewfinder screen (see FIG. 33), the process returns to step 501, and if it does match. If so, go to step 503. However, it is preferable to give some margin to the position coincidence determination. Further, the gaze position may be compared instead of the gaze position.

In step 503, the power zoom motor (not shown) is controlled according to the focal length display value (3 m in FIG. 33) of the position of the line of sight, and the zoom lens (not shown) sets the focal length set by the line of sight. Drive to the position corresponding to, and return to step 501. The operation of FIG. 31 is an operation when PZ is selected by the line of sight on the mode selection display 231.

With the above operation, the desired focal length on the scale on the viewfinder screen can be selected with the line of sight to adjust the focal length without manual operation, so that the telephoto lens can be held by hand. It is convenient when both hands are blocked.

(Eighth Embodiment) FIG. 34 is a flow chart showing the operation of the eighth embodiment of the camera having the visual axis detection device according to the present invention. In step 600, the CPU 1 starts the operation by half-pressing the release button 11. In step 601, the line-of-sight position of the photographer observing the viewfinder is detected based on the signal from the line-of-sight detecting means 40. In this embodiment, the line-of-sight position is detected once in one focus detection sequence, but the line-of-sight position detection may be performed at a timing independent of the focus detection sequence (for example, a timer interrupt every predetermined time). Good. Further, the gaze position may be used instead of the gaze position.

In step 602, the focus detection area is divided according to the detected line-of-sight position. That is, when the line-of-sight position exists near the boundary lines v1 and h1 (see FIG. 36) when dividing the focus detectable area M into a plurality of predetermined focus detection areas, the main subject detects a plurality of focus detection areas. Since focus detection is performed by dividing the focus detection area into other areas, it is easy to be affected by other subjects.Therefore, when dividing the focus detection area into multiple focus detection areas so that the line-of-sight position is located approximately in the center of the divided focus detection areas. Change the line (see Figure 35). Step 6
In 03, focus detection is performed in each of the plurality of divided focus detection areas, and a predetermined algorithm is selected from the plurality of focus detection results (for example, a weighted average obtained by increasing the weight of the result of the focus detection areas in which the eye gaze position exists). ) Finally produces one result and returns to step 601.

With the above operation, when the focus detectable area is divided into a plurality of areas for focus detection, the focus detection area centering on the line-of-sight position can be set at all times. An accurate focus detection result can be obtained for a moving main subject.

(Ninth Embodiment) FIG. 37 is a block diagram showing a ninth embodiment of a camera having a visual axis detection device according to the present invention, and FIG. 38 includes an optical system of the camera according to the embodiment. It is a block diagram.

The camera with the line-of-sight detection device of this embodiment is
As shown in FIG. 37, a line-of-sight detection unit 1101 that detects the line-of-sight position of the photographer, and a single line-of-sight that locks the line-of-sight position when the detection results of the past multiple times of the line-of-sight detection unit 1101 satisfy a predetermined gaze condition. Mode (EG-S mode)
And a line-of-sight continuous mode for updating the line-of-sight position based on the latest line-of-sight position of the line-of-sight detecting means 1101 (EG-C
Mode) and changes the shooting state of the camera based on the line-of-sight position.

The line-of-sight detecting means 1101 uses the "Purkinje 1
This is a means for performing known line-of-sight detection by using a method such as “image + Purkinje 4 images”, “Purkinje 1 image + pupil center”, and “white eye / black eye boundary”.

When the camera is an AF camera, the line-of-sight mode selecting means 1103 for selecting at least one of the line-of-sight modes and the focus detecting means 1104 for measuring the distance to the object.
An AF control unit 1105 that controls the focus detection unit 1104 in a plurality of distance measurement modes, an AF mode selection unit 1106 that selects at least one of the distance measurement modes of the AF control unit 1105, and the like.

At this time, the line-of-sight mode selecting means 1103
Selects at least one of the line-of-sight modes based on the result of the AF mode selection means 1106.

When the line-of-sight control means 1102 and the AF control means 1105 are set to a predetermined mode, the photographing control means 1107 outputs a focus control signal to the lens driving means 1108 and the display means 1109 also detects the line-of-sight detection means. The AF area closest to the line-of-sight position detected in 1101 is selected and displayed.

Next, a more detailed description will be given with reference to FIG. The subject light that has passed through the taking lens 1205 is split into two directions by the main mirror 1204. The light transmitted through the main mirror 1204 is further bent in the optical path by the sub mirror 1214 and guided to the AF detection element 1202. Further, the light reflected by the main mirror 1204 is guided toward the finder 1210 and is reflected by the screen 1213.
The subject image is formed at.

The AF sensor 1202 has a plurality of focus detection areas, and C based on the distance measurement signal detected here.
The PU 1201 generates a lens matching signal, and drives the taking lens 1205 via the lens driving circuit 1203. The AE light-receiving element 1211 has a photometric area corresponding to the AF area, and the CPU 1201 generates a weighted and averaged photometric output based on the line-of-sight position, and controls the exposure time of the shutter device 1206.

In this embodiment, the line-of-sight detecting means 1101
Are incorporated in the finder 1210, and the optical system for detecting the line of sight shares the optical system of the finder. Infrared L
The light emitted from the ED 1207 is collimated by the optical system 1209 and projected onto the cornea of the eyeball 1212. The optical system 1209 is a half mirror for separating the projected light and the reflected light, a lens for forming an image of the reflected light on the light receiving element 1208,
It is composed of a dichroic mirror or the like that separates the projected light and the reflected light from the light from the finder 1210. A two-dimensional CCD or the like is used as the line-of-sight light receiving element 1208, and detects the bright spots of the above-mentioned Purkinje 1 image, 4 image and the like.

The half-push switch SW1 is a switch which is turned on by the first stroke of the release button. The full-press switch SW2 is a switch that is turned on by the second stroke of the release button. The AF mode selection switch SW3
This switch is for switching the AF mode. When the switch is on, the AF-C mode is selected, and when it is off, the AF-S mode is selected. Here, the AF-S mode is a mode in which driving of the lens based on the result of the focus detection unit 1101 is prohibited after the focus is once obtained. In the AF-C mode, the lens is detected based on the latest detection result of the focus detection unit 1101. It is a mode to continue driving. The line-of-sight mode selection switch SW4 is a switch for switching the line-of-sight mode, and selects the EG-C mode when it is on and the EG-S mode when it is off. The switches SW1 to SW4 are connected to the CPU 1201.

The camera of this embodiment is shown in FIGS.
It is controlled by the CPU 1201 according to the program shown in the flowchart shown in FIG.

Next, the operation of this embodiment will be described centering on the flow of the CPU 1201. The main flow shown in FIG. 41 is executed by turning on a power switch (not shown). First, in S1501, press the half-press switch SW1.
Is turned on, and if it is turned on, S1502
If it is not on, loop and wait until it turns on.

In S1502, the line-of-sight mode selection subroutine shown in FIG. 42 is executed. In S1601, it is tested whether or not the AF mode changeover switch SW3 is on, that is, the AF-C mode or the AF-S mode is tested, and if it is the AF-C mode, S1602
Go to and change the AF mode to AF-C mode and the line-of-sight mode to EG-C.
Set to mode. If it is the AF-S mode, the process advances to S1603 to set the AF mode to the AF-S mode and the line-of-sight mode to the EG-S mode.

In this embodiment, the AF mode is switched by a hardware switch, but it is not always necessary and it may be done by software according to a program stored in a memory such as RAM, ROM, EEPROM or the like. .. Also,
Even when the AF mode is automatically switched by the focus detection unit 1104, the line-of-sight mode may be automatically switched in association with the switching.

Next, returning to the main flow shown in FIG. 41 and proceeding to S1503, the line-of-sight detection subroutine shown in FIG. 48 is executed. In S1701, a known eye gaze detection operation is performed. In this embodiment, the line-of-sight position is detected using Purkinje images 1 and 4. First, the infrared LED which is the light emitting element 1207 of the line-of-sight detecting means 1101 is turned on.

The line-of-sight light receiving element 1208 is a two-dimensional CCD.
The CPU 1201 acquires data including the position information of each bright spot. In S1702, the line-of-sight position is calculated based on the data from the line-of-sight light receiving element 1208. The center of gravity positions I1 and I4 of each bright spot are calculated, and the eyeball rotation angle is obtained from the difference I4-I1. In this embodiment, the Purkinje 1 and 4 image systems have been described, but other line-of-sight detection systems may be used as long as they detect the line-of-sight position.

In S1703, is the current line-of-sight mode EG-S mode?
Test for EG-C mode, S1 for EG-S mode
Proceed to 704, and if in EG-C mode, proceed to S1706. S1704
EG-S mode tests whether or not the line-of-sight position is locked.If locked, returns from the line-of-sight detection subroutine, and if not locked, proceeds to S1705, where EG-S mode lock conditions are met. Whether or not there is, S1705
Go to. Details regarding the EG-S mode lock determination will be described later.

In S1706, the line-of-sight position is updated to the latest line-of-sight position. Here, the latest line-of-sight position is the line-of-sight position detected at the closest time. Since the detected line-of-sight position includes an error, the results of the multiple line-of-sight positions in the past may be statistically processed, and the result may be updated as the latest line-of-sight position.

Next, the program proceeds to S1504 shown in FIG. 41 and executes the AF subroutine. FIG. 44 shows the AF subroutine.
Shown in. S1801 stores the AF sensor 1202 and transfers data. In S1802, the AF area closest to the line-of-sight position determined by the line-of-sight detection subroutine is selected. Figure 45
In such a case, A1 is selected as the AF area. Here, the AF area is composed of three areas A1, A2, and A3, and the x mark indicates the updated line-of-sight position.

In S1803, the defocus amount for the selected AF area is calculated by a known correlation calculation. S1
In 804, it is tested whether or not the calculated defocus amount falls within the in-focus area. If the in-focus area is in focus, the flow returns as it is, and if it is out of focus, the flow advances to S1805. In S1805, the AF mode is tested whether it is the AF-S mode or the AF-C mode. If the AF-C mode, the process proceeds to S1807, and if the AF-S mode, the process proceeds to S1806.

In S1806, it is tested whether or not the AF is locked. If the AF is locked, the lens is not driven and the process returns. If the AF is not locked, the process proceeds to S1807, and the lens driving circuit 1203 is driven by a necessary amount. The photographing lens 1205 is driven via the. Here, the AF-S mode is
Once the focus is detected, the shooting lens is locked, and the AF-C mode is the mode in which the shooting lens is continuously driven by the latest defocus amount.

Next, S150 of the main flow shown in FIG.
Proceed to step 5 and calculate the photometric value with emphasis on the selected AF area position. For example, in the case of FIG. 45, A
It is also possible to calculate a weighted average photometric value for the AE area corresponding to the F area A1 or perform spot photometry using only the photometric value of the AE area corresponding to the AF area A1 to calculate the photometric value.

At S1506, it is tested whether or not the release button is fully pressed (fully pressed switch SW2 is on).
If it is fully pressed, the process proceeds to S1506, the mirror is raised, the shutter travels, the film is wound, and the like, and the process returns to S1501. If it is not fully pressed, nothing is done and the process returns to S1501.
After that, this cycle is repeated.

Here, the above-mentioned EG-S mode lock determination will be described with reference to FIGS. 46 and 47. The EG-S mode lock determination subroutine shown in FIG.
At S, a test is made as to whether or not the S-mode lock condition is satisfied. If the S-mode lock condition is satisfied, the process proceeds to S11003 to set the lock flag, and if not satisfied, the process proceeds to S11002 to clear the lock flag. Table 1 shows the EG-S mode lock conditions.

Table 1 (EG-S mode lock condition) First condition: The distance between the previous and present line-of-sight positions is less than or equal to a predetermined value. Second condition: The difference between the eyeball rotation angles of the previous time and this time is less than or equal to a predetermined value. Third condition: The moving speed of the line of sight is less than or equal to a predetermined value. Fourth condition: The first to third conditions are equal to or longer than a predetermined time.

First, the first condition will be described with reference to FIG. In FIG. 47, SPn is the latest line-of-sight position, SP (n-
1) indicates the previous line-of-sight position and SP (nm) indicates the line-of-sight position m times before. SP (n) represents the coordinate position on the screen 1213 of the camera. SP (n-3) -SP (n-4), SP (n-2) -SP (n-3)> L, SP (n-1) -SP (n-2), SP (n) -SP If (n-1) ≤ L, the line-of-sight position is not locked until the n-2th time, but the line-of-sight position is locked at the n-1th time.

If the predetermined value L is too small,
Since the person cannot lock due to involuntary eye movement (even if he thinks that a person is gazing at one point, the eyeball vibrates slightly around the gazing point), etc. It should be slightly larger than the value corresponding to the amount of vibration.

In the case of the second condition, the judgment of the first condition is changed from the distance between two points to the eyeball rotation angle. However, since this predetermined value L depends on the sampling time for line-of-sight detection, it is the third condition that it is standardized at time intervals.

That is, the line-of-sight position is locked when (SP (n) -SP (n-1)) / T (n) <LS is satisfied. Here, T (n) is the time interval from the previous sight line detection to the current sight line detection.
This time interval is measured by a timer built in the CPU 1201. The predetermined value LS is about 3 to 10 ° / sec, usually 5 °
/ Sec is good. Generally, the eyeball rotation angle required to see the field of view (24 × 36 mm) of a single-lens reflex finder is ± 15.
It is about °.

The fourth condition is to lock for the first time when the conditions 1, 2 and 3 are satisfied for a predetermined time or more. The above is summarized as shown in Table 1 above.

(Tenth Embodiment) FIG. 48 is a flowchart showing a line-of-sight mode selection subroutine according to the tenth embodiment. The tenth embodiment is applied to the one in which the AF mode is switched based on the result of the moving body determination unit that determines whether the subject is a moving body or is still based on the focus detection results of the past multiple times. It is a thing. When the subject is a moving body, the subject is likely to move within the field of view, and thus the AF area where the subject is present is likely to change. Therefore, if the line-of-sight mode is the EG-S mode, there is a possibility that a photo opportunity will be missed. Therefore, as shown in FIG. 48, the line-of-sight mode selection means 110
Make up 3.

In S11201, the moving body determination known in Japanese Patent Laid-Open No. 2-50140 is performed. Next, in S11202, if the subject is a moving body, the process proceeds to S11203, and if the subject is stationary, the process proceeds to S11204. S11203
Now, set the line-of-sight mode to EG-C mode. In S11204, the line-of-sight mode is set to the EG-S mode. S11203, S1
Set the line-of-sight mode with 1204 and return.

The moving body determination cannot be made unless the focus detection is performed after the distance measuring area has been determined. Therefore, as a default, the moving body determination is performed in the A2 area in the case of FIG.
This setting may be set in advance by an operation member or the like.
The initial line-of-sight mode may be set to the EG-S mode, or may be initially set by an external operation member.

(Eleventh Embodiment) An eleventh embodiment will be described with reference to FIG. This embodiment has a line-of-sight multi-mode (EG-M mode) that locks the line-of-sight position until a new gazing point is obtained as the line-of-sight mode. EG
In -S mode, the gaze position is fixed once the gaze position is obtained, but in this EG-M mode, when a new gaze position is obtained, the gaze position is moved to that position.
-Different from S mode. Also, the difference from the EG-C mode is whether the gaze position moves continuously or jumping when the gazing point moves in a jumping manner.

FIG. 49 is a flow chart for explaining the multimode operation of the eleventh embodiment. In step S11301, it is determined whether or not the user is gazing. This gaze position determination uses the same method as the lock determination in EG-S mode. The description is omitted here. Subsequently, in step S11302, it is determined whether or not the gaze position has moved. If the gaze position has moved, the process proceeds to step S11303, and the eye gaze position is updated in this step.

In step S11302, if there is no movement of the gazing point or if a new gazing point cannot be obtained,
Progressing to S11304, the line-of-sight position is fixed. With a few exceptions, the gaze point is always updated this way.

FIG. 50 is a plan view showing the movement pattern of the line-of-sight position corresponding to each mode of the embodiment. In this case, the first gaze position is point A and the gaze position after a while is point B. In EG-S mode, the line-of-sight position remains fixed at C1, in EG-C mode it moves like C1 → C2 → C3 → C4 → C5, and in EG-M mode it jumps like C1 → C5. Move. Further, in the first embodiment, the AF mode is AF-C.
In the mode, the line-of-sight mode was EG-C mode.
-It may be transformed to become M mode. By doing so, it is not necessary to capture blinking and careless movement of the gaze position, which is useful for improving the processing capability.

(Twelfth Embodiment) FIG. 51 is a block diagram showing a twelfth embodiment of the present invention. Line-of-sight detection means 20
03 is a means for detecting the latest gaze position. The gaze point determining means 2001 is means for determining a gaze position based on the gaze position detected by the gaze detecting means 2003 a plurality of times in the past. The line-of-sight control unit 2002 is a unit that calculates the control line-of-sight position that controls the imaging control unit 2004 based on the determination result of the gazing point determination unit 2001. The photographing control means 2004 is
This is a means for performing imaging control based on the control line-of-sight position calculated by the line-of-sight control means 2002. For example, it is used for selecting a necessary AF area from a plurality of AF areas, determining a weighting coefficient of multi-AE, and the like.

FIG. 52 is a block diagram showing the twelfth embodiment. The subject light that has passed through the taking lens 1205 is split into two directions by the main mirror 1204. Main mirror 1204
The light that has passed through is further bent in the optical path by the sub-mirror 1214 and guided to the AF sensor 1202. Also, the main mirror 12
The light reflected at 04 is guided toward the viewfinder 1210,
A subject image is formed on the screen 1213. AF sensor 12
02 has a plurality of focus detection areas, the CPU 1201 generates a lens matching signal based on the distance measurement signal detected here, and the photographing lens 1205 is passed through the lens drive circuit 1203.
To drive. The AE light-receiving element 1211 has a photometric area corresponding to the AF area, and the CPU 1201 generates a weighted and averaged photometric output based on the line-of-sight position to control the exposure time of the shutter device 1206. The operation member 1215 has a plurality of switches (not shown). The switch SW1 is a switch that is turned on by the first stroke of a release button (not shown) (hereinafter referred to as a half-press switch). Switch SW2
Is a switch that is turned on by the second stroke of the release button (hereinafter referred to as a full-press switch). The AF mode selection switch SW3 is a switch for switching the AF mode, and selects the AF-C mode when it is on and the AF-S mode when it is off.

The camera of this embodiment has a CPU according to the program shown in the flowchart of FIG.
Controlled by 1201. Next, the operation of this embodiment will be described focusing on the flow of the CPU 1201. The main flow shown in FIG. 53 is executed by turning on a power switch (not shown). First, press the S2301 halfway switch
Test if SW1 is on and if yes, S230
Go to 2 and loop and wait until it is on.

In S2302, a visual axis detection subroutine is executed. Next, this visual axis detection subroutine will be described with reference to FIG. In S2401 of FIG. 54, a known line-of-sight detection operation is performed. In this embodiment, the line-of-sight position is detected using Purkinje images 1 and 4. First, the infrared LED which is the light emitting element 1207 of the line-of-sight detecting means 2003 is turned on. Line-of-sight light receiving element
The 1208 is composed of a position detecting element such as a two-dimensional CCD and has a C
The PU 1201 acquires data including position information of each bright spot.
S2402 stores the previously detected line-of-sight position in SP1. Here, SP1 represents the two-dimensional position coordinate of the screen 1213 with the center of the screen as the origin at the previous line-of-sight position. That is, SP1 =
It is represented by (x1, y1). S2403 calculates the line-of-sight position based on the data acquired by the line-of-sight detection circuit 1201, and SP {= (x,
y)}. That is, the barycentric positions I1 and I4 of each bright spot are calculated, and the eyeball rotation angle is obtained from the difference I4-I1. The screen 12 seen by the photographer from the eyeball rotation angle obtained here.
13 Find the position coordinates on. In this embodiment, the Purkinje 1 and 4 image methods have been described, but other visual axis detection methods may be used as long as the visual axis position can be detected. S2404 is
After the line-of-sight detection is started, the EG-S mode subroutine that locks the line-of-sight position to the position where the user first gazes is executed. Upon completion of this subroutine, the line-of-sight position (hereinafter referred to as the control line-of-sight position) used when changing the control state of the camera by the line-of-sight is output to the variable EP. Variable here
EP represents a two-dimensional position coordinate (EP = (xe, ye)) on the screen 1213, similar to SP and SP1. Details of the EG-S mode subroutine will be described later. Here, the visual axis detection subroutine is ended, and the flow proceeds to the AF subroutine of S2303.

The next step S2303 is an AF subroutine, in which the AF area is determined based on the variable EP representing the control line-of-sight position output by the line-of-sight detection subroutine of S2302. Based on this, the lens is driven. This AF subroutine is shown in FIG. S180
1 stores the AF sensor, transfers data, and performs AD conversion. S1802 selects the AF area closest to the control line-of-sight position determined in the line-of-sight detection subroutine. Figure 4
In the case shown in 5, A1 is selected as the AF area. Here, the AF area is composed of three areas A1, A2, and A3, and the x mark indicates the control line-of-sight position represented by the variable EP. S1803 calculates the defocus amount for the selected AF area by a known correlation calculation.
In step S1804, it is determined whether or not the calculated defocus amount falls within the in-focus area. If the in-focus area is in focus, the process directly returns. If not, the process proceeds to step S1805. The S1805 determines whether the AF mode is the AF-S mode or the AF-C mode. If the AF-C mode is selected, the process proceeds to S1807. If the AF-S mode is selected, the S1806 is selected.
Go to. In S1806, whether the AF is locked or not is tested. If the AF is locked, the lens is not driven and the process returns. If the AF is not locked, the process proceeds to S1807 to drive the lens by a necessary amount.

Next, the procedure advances to S2304 in FIG. 53, and a photometric value that focuses on the selected AF area position is calculated. For example, in the case of FIG. 45, the AE area corresponding to the AF area A1 is calculated with a weighted average photometric value,
The SPOT metering using only the metering value of the AE area corresponding to the AF area A1 may be performed to calculate the metering value. S2
305 determines whether or not the release button is fully pressed, and if it is fully pressed, the process proceeds to S2306, release operations such as mirror up, shutter running, and film winding are performed, and the process returns to S2301. If it is not fully pressed, nothing is done and the process returns to S2301. After that, this cycle is repeated. The above is a rough flow of this embodiment.

Now, the flow of the EG-S mode, which is one of the line-of-sight control modes, will be described with reference to FIG. The EG-S mode is a mode in which the control line-of-sight position is fixed at the first gaze position after half-pressing. S2501 determines whether or not the control line-of-sight position is currently locked.
The EG-S mode subroutine is returned, and the control line-of-sight position is not updated. If the control line-of-sight position is not locked, the process proceeds to S2502, where a gaze point determination subroutine for determining whether or not the line of sight of the photographer is gazing is executed.
If gazing, proceed to S2503 and set the line-of-sight lock flag. If not gazing, proceed to S2504.
S2504 is a variable that represents the latest gaze position for control.
Store in EP. Then, the EG-S mode subroutine is ended.

The gazing point determination subroutine shown in FIG. 56 will be described. S2601 determines whether or not the absolute value of the difference between the variable SP1 representing the previous line-of-sight position and the variable SP representing the current line-of-sight position is smaller than a predetermined value L, and if smaller, it is determined that the photographer is gazing. Proceed to S2602, and if it is large, proceed to S2603 assuming that the photographer is not gazing. That is, whether or not the control line-of-sight position is updated is determined by determining whether the absolute value of the difference between the previous and present line-of-sight positions (hereinafter referred to as the line-of-sight movement amount) is larger or smaller than a predetermined value. Here, || represents an absolute value, and | SP-SP1 | represents the following equation. ｜ SP-SP1 ｜ ＝ ((x-x1) 2＋ (y-y1) 2) 1/2

In the EG-S mode, the transition of the control line-of-sight position when the line-of-sight moves as shown in FIG. 47 will be described. In FIG. 47, SPn is the latest line-of-sight position,
SP (n-1) represents the last line-of-sight position, and SP (nm) represents the line-of-sight position m times before. SP (n) represents the coordinate position on the viewfinder screen of the camera. SP (n-3) -SP (n-4), SP (n-2) -SP (n
-3)> L, SP (n-1) -SP (n-2), SP (n) -SP (n-1) <= L, the control line-of-sight position is changed up to the n-2th time. Without locking
Lock the control line-of-sight position at the (n-1) th time. Here, if the predetermined value L is too small, the subject is locked due to involuntary eye movements (the eye slightly vibrates around the gazing point even if one thinks that a human is gazing at one point). Since it becomes impossible to do so, it is preferable that the value is slightly larger than the value corresponding to the amount of vibration of the eyeball such as the involuntary eye movement. Further, SP1 and SP are represented by two-dimensional coordinates on the screen, but they are not always necessary and may be represented by an eyeball rotation angle.

(Thirteenth Embodiment) In the thirteenth embodiment, whether or not the photographer is gazing is determined more accurately. The thirteenth embodiment is different only in the gazing point determination subroutine, and the description of the other subroutines will be omitted. FIG. 57 is a flowchart showing a gaze point determination subroutine. Here, in order to determine the gaze position, the moving speed of the line of sight is calculated, and the moving speed is a predetermined value.
If it is smaller than S, it is determined that the photographer is gazing. In the twelfth embodiment, the amount of movement proportional to the line-of-sight movement speed can be calculated when the detection interval of the line-of-sight detection is substantially constant, but the line-of-sight moves at the same moving speed when the detection intervals are significantly different. However, when the detection interval is short, the amount of movement is small, so there is a possibility that it may be erroneously determined to be a gaze. Therefore, in the thirteenth embodiment, a more accurate gaze position determination can be performed. The S2701 is
The absolute value of the difference between the current line-of-sight position SP and the previous line-of-sight position SP1 is further divided by the time interval between these to calculate the line-of-sight movement speed, and whether the calculated line-of-sight movement speed is smaller than the predetermined value S is determined. It is judged that if it is small, it is watching
When it is large, it is determined that the user is not watching, S2702
Continue to 2703.

(Fourteenth Embodiment) FIG. 58 is a flow chart showing another example using a gaze determination subroutine. In this embodiment, the gaze position is determined to be gazing when the number of times the line-of-sight movement amount is smaller than a predetermined value and continuously exceeds the predetermined number. S2801 determines whether or not the amount of movement of the line of sight is smaller than a predetermined value L.
Proceed to step 2, and if larger, proceed to S2803. In S2802, the value of the counter Count that counts the number of times the eye movement amount is smaller than the predetermined value is incremented by 1, and the process proceeds to S2804. In S2803, the line of sight is moving, so the counter Count is cleared to 0.
Go to 4. S2804 determines whether or not the number of times the line-of-sight movement amount is smaller than a predetermined value is larger than a predetermined value C1, and if it is larger than the predetermined value C1, it is determined that the user is gazing, and the process proceeds to S2805. Assuming that he is not watching, proceed to S2806. In other words, even if the user is gazing in the same way, he gazes for an important thing for a longer time, so that it is possible to prevent locking at an unimportant control line-of-sight position.

(Fifteenth Embodiment) In the fourteenth embodiment,
Although the number of times was counted, in the fifteenth embodiment, when the detection intervals of the line-of-sight detection are different, the determination is made not by the number of times but by the sum of the detection intervals. FIG. 59 is a flowchart showing a gaze determination subroutine according to the 15th embodiment. S2901 determines whether or not the amount of movement of the line of sight is smaller than a predetermined value L, and if smaller, proceeds to S2902,
The current detection interval TL is added to the gaze time TT, the result is stored in the gaze time TT, and the process proceeds to S2904. If it is larger, the process proceeds to S2903, the watching time TT is cleared to 0, and the process proceeds to S2904. In S2904, it is determined whether or not the gaze time TT is longer than the predetermined time TC, and if it is longer, it is determined that the user is watching, and if it is smaller, the process proceeds to S2905, and it is determined that he is not watching, and the process proceeds to S2906.

(Sixteenth Embodiment) In the above-described embodiments, when it is determined that the user is gazing once, the visual line mode in which the control visual line position is fixed thereafter is used. When moving frequently, it was necessary to half-press again to move the control line-of-sight position. Therefore, in the sixteenth embodiment, once the gaze point is found, the control line-of-sight position is fixed at that position, but when a new another gaze position is found, the gaze line for moving the control line-of-sight position to that position is obtained. It is designed to perform line-of-sight mode control called multi-mode (EG-M mode).

This embodiment is realized by replacing the EG-S mode subroutine of S2404 of FIG. 54 with the EG-M mode subroutine of FIG. Therefore, FIG.
Will be described. FIG. 60 is a flow chart showing the subroutine of the line-of-sight multi-mode. S3001 proceeds to S3002 when the subject is gazing in the gaze determination subroutine, and proceeds to S3003 when not gazing. As the gaze determination subroutine, any of the subroutines shown in FIGS. 56 to 59 can be used. S3002 sets a gaze point present flag indicating that the first gaze point has been found. S300
In step 3, it is determined whether or not the user has gazed last time, and if it is gazing, the process returns, and if not, the process proceeds to S3004.
In S3004, since it is determined to be a gaze in the gaze determination, the gaze flag is set. S3005 stores the current line-of-sight position SP in the control line-of-sight position EP. In S3006, since the gaze determination does not determine the gaze, the gaze flag is cleared. S300
Step 7 determines whether or not the first gazing point is found. If the first gazing point is not found, the process proceeds to S3005,
Update with the current line-of-sight position SP, and if already found, return. In this embodiment, the control eye gaze position EP is updated with the latest eye gaze position SP until the first gaze position is found. Here, as a modified example, initialization may be performed so that the first gaze position is always in the center. That is, EP ← 0 (=
0, 0), set the point of interest flag.

EG-S mode, EG-M mode, and operation of updating the control line-of-sight position in the line-of-sight continuous mode (EG-C mode) that updates the control line-of-sight position with the latest detected line-of-sight position. Will be described with reference to FIG. In this example, it is assumed that the photographer first looks at point A and then looks at point B. In EG-C mode, change the control line-of-sight position from C1 → C2
→ Update C3 → C4 → C5. In EG-S mode, position C1
The control line-of-sight position is fixed by and is not changed thereafter. However, in the EG-M mode, after once fixed at the position C1, when the photographer gazes at the point B, the control line-of-sight position is changed to the position C5 and fixed there.

(17th Embodiment) In the 17th embodiment,
Whether or not gaze is determined using the amount of movement of the line of sight, which is the difference between the previous and current line-of-sight positions, but once the control line-of-sight position is once fixed in EG-M mode, as shown in FIG. The difference between the fixed control line-of-sight position EP and the current line-of-sight position SP is a predetermined value
By determining whether or not it is L or more, it is determined whether or not to be a gaze. FIG. 61 is a flowchart showing the seventeenth embodiment. In S3101, it is determined whether or not the user is gazing. If the gazing is in progress, the process proceeds to S3102, and if the gazing is not in progress, the process proceeds to S3103. S3102 determines whether or not the absolute value of the difference between the control gaze position EP representing the current gaze position and the current gaze position is smaller than a predetermined value L. In that case, it is determined that he is not gazing. The reason for doing this is that it is not considered to be gazing if the position deviates greatly from the gazing position this time. In S3103, when the absolute value of the difference between the current and previous line-of-sight positions is smaller than the predetermined value, it is determined that the user is gazing, and when the absolute value is large, it is determined that he is not gazing. Therefore, even if the line of sight moves slowly, the position of the control line of sight can be accurately fixed. Further, in these EG-M mode examples, the control line-of-sight position was fixed to the first gaze position determined to be gaze, while the gaze point for the same position, the control line-of-sight position You may make it calculate | require by the average value of a gaze position. In other words, if you are watching the last time and you are watching this time, EP → (EP +
The control line-of-sight position may be updated as in (SP) / 2.

(Eighteenth Embodiment) Next, an eighteenth embodiment will be described with reference to FIG. FIG. 62 is a block diagram of the eighteenth embodiment. In the eighteenth embodiment, the line-of-sight detecting means 2003 for obtaining the line-of-sight position of the photographer, the gaze point determining means 2001 for determining whether or not the photographer is gazing, and the control line-of-sight position for controlling the camera are set. Gazing point determination means 2001
The line-of-sight control means 2002 for controlling the line-of-sight mode that outputs according to the result and the line-of-sight control mode selection means 2005 for selecting at least one of the plurality of line-of-sight modes are configured. Here, the line-of-sight mode is the EG-C mode, EG-S mode, or EG-M mode described in the twelfth to seventeenth embodiments. In the eighteenth embodiment, the line-of-sight control mode selection means 2005 is selected by the line-of-sight control mode selection switch SW4 included in the operation member 1217 of FIG. 63 and 6
4 is a flow chart showing the operation of the line-of-sight control mode selection means. The main flowchart is the same as that in FIG. 53, and therefore its explanation is omitted. The line-of-sight mode is selected each time the line-of-sight control mode selection switch SW4 is pressed.
4 Interrupt changes cyclically from EG-S mode to EG-C mode to EG-M mode.

FIG. 63 is a flow chart for explaining the SW4 interrupt. S3201 cyclically selects the line-of-sight mode. For example, the line-of-sight control mode selection switch SW4
If the line-of-sight mode is the EG-S mode before is pressed, the line-of-sight mode is set to the EG-C mode. S3202 clears the line-of-sight lock flag because the line-of-sight mode has been changed. The reason for this is that a change in the line-of-sight mode has a high possibility of changing the gaze position of the photographer.
S3203 clears the watching flag. S3204 clears the gaze position present flag. Further, S3205 changes the control line-of-sight position EP to the center position of the screen which is the initial position. Here, storing 0 in the control line-of-sight position represents moving the control line-of-sight position to the center position. Execute the above flow and return SW4 interrupt.

FIG. 64 is a flow chart showing the visual axis detection subroutine in the eighteenth embodiment. S3301
Drives the line-of-sight detection circuit to detect the line-of-sight. S3302
Updates the previous line-of-sight position SP1. In S3303, a known line-of-sight detection calculation is performed to calculate the current line-of-sight position SP. S3
304 is S3 in EG-S mode depending on the selected gaze mode.
Go to 305, go to EG-C mode to S3306, go to EG-M mode
The process proceeds to S3307, the control line-of-sight position is calculated based on each line-of-sight mode, and the line-of-sight detection subroutine is returned.

(Nineteenth Embodiment) Next, a nineteenth embodiment will be described. In the thirteenth embodiment, the line-of-sight movement speed was calculated by the difference between the last and present line-of-sight positions in the calculation of the line-of-sight movement speed, but an accurate line-of-sight movement speed calculation due to an error when the line-of-sight detection interval becomes short. I can't. Even when the line of sight is actually gazing, the human eye vibrates finely in the vicinity of the gazing position (this is referred to as involuntary eye movement).

FIG. 66 is a diagram for explaining involuntary eye movements.
In this figure, the horizontal axis represents time and the vertical axis represents the line-of-sight position.
Gaze movement speed based on current and previous gaze position and gaze movement speed based on current and previous gaze position It turns out that is small. Even if the user is gazing, the eye movement speed in a short time interval is not necessarily small due to the involuntary eye movement, but the average speed in a predetermined time interval is small. Therefore, when the calculation time interval of the line of sight is short, it is more accurate to determine the gaze position at the average speed. Therefore, the time interval required to calculate the line-of-sight movement speed is set to be a predetermined time interval or more, instead of the difference between the line-of-sight positions of the previous time and this time. In other words, if the time interval between this time and the previous time is less than or equal to the predetermined time, it is determined whether the time interval between the current time and the time before the previous time is the predetermined time or more. If it is less than or equal to the predetermined time, the time interval with the previous line-of-sight position data is compared with a predetermined value, and the line-of-sight position data and the current line-of-sight position data that makes the time interval larger than the predetermined time are calculated. Calculate the line-of-sight movement speed.

The nineteenth embodiment will be described with reference to the flowchart of FIG. S3501 updates variables SP1, SP2, and SP3 that indicate the line-of-sight position for the past three times. S3502 updates variables TL1 and TL2 that indicate the time interval of the line-of-sight detection. S3503
Drives the line-of-sight detection circuit to detect the line-of-sight. S3504 is
Performs a known line-of-sight detection calculation to calculate the current line-of-sight position SP,
Calculate the last and current gaze detection time interval TL. S3505
Determines whether the time interval TL between the previous time and this time is greater than or equal to the predetermined value TS, and if it is greater than or equal to S3506, if less than S3507.
Go to. S3506 calculates the line-of-sight movement speed S from the previous and present line-of-sight positions and their time intervals. S3507 determines whether or not the time interval (TL + TL1) between this time and the time before last is equal to or larger than a predetermined value TS. If it is greater than or equal to S3508, the process proceeds to S3508, and if less than this, the process proceeds to S3509.

At S3508, the line-of-sight moving speed S is calculated by | SP2-SP | / (TL + TL1). S3509 is ｜ SP3-SP ｜ / (TL + T
L1 + TL2) calculates the eye movement speed S. S3510 is
It is determined whether or not the line-of-sight movement speed S is less than or equal to a predetermined value SS.
S3511 sets a gaze point present flag indicating that there is a gaze position. The gaze point present flag is a flag indicating whether or not the first gaze position has been found after half-pushing, and is cleared in an initialization flow (not shown). S3512
Is a gaze flag to determine whether or not you were watching last time,
If the user is watching, the process returns without updating the control line-of-sight position EP. If not watching the previous time, proceed to S3513. S3513 is
A gaze flag indicating that a gaze is being performed this time is set. S3514 updates the control line-of-sight position EP with the current line-of-sight position SP. Since S3515 is not gazing at this time, the gazing flag is cleared. In S3516, it is determined whether or not there is a gaze position. If the gaze position exists, the flow is returned without updating the control line-of-sight position, and if there is no gaze position, the process proceeds to S3514. This is because the control line-of-sight position is updated with the latest line-of-sight position until the first gaze position is found.

In this embodiment, the line-of-sight position data is searched and used so that the time interval for calculating the line-of-sight movement speed becomes equal to or longer than the predetermined time interval, but if the line-of-sight detection interval is known in advance, it is as follows. It is also possible to calculate the line-of-sight movement speed based on the line-of-sight position before a predetermined number of times and the line-of-sight position of this time, without performing a search.

[0141]

As described above, according to the present invention,
As the line-of-sight mode, the line-of-sight single mode (EG-S) that locks when the line of sight gazes, the line-of-sight continuous mode (EG-C) that continuously updates the line-of-sight position for control, and It has a line-of-sight multi-mode (EG-M) that locks the line-of-sight position to the control gaze position, and since the line-of-sight mode can be switched in conjunction with AF mode, it is possible to concentrate on shooting with less button operation. There is.

[Brief description of drawings]

FIG. 1 is a block diagram showing a first embodiment of a camera having a visual axis detection device according to the present invention.

FIG. 2 is a perspective view showing a configuration of a focus detection unit incorporated in the camera according to the first embodiment.

FIG. 3 is a diagram showing a display example of a focus detection area of the camera of the first embodiment.

FIG. 4 is a diagram showing a display example of a focus detection area of the camera of the first embodiment.

FIG. 5 is a front view showing rotation amount detection means used in the camera of the first embodiment.

FIG. 6 is a plan view showing an encoder of a rotation amount detecting means used in the camera of the first embodiment.

FIG. 7 is a diagram showing a detection unit of a rotation amount detection means used in the camera according to the first embodiment.

FIG. 8 is a diagram showing a line-of-sight detection means used in the camera of the first embodiment.

FIG. 9 is a diagram for explaining the reflection efficiency at the line-of-sight position of the camera of the first embodiment.

FIG. 10 is a plan view showing a configuration example of a glasses detection means of the camera of the first embodiment.

FIG. 11 is a front view showing a configuration example of a glasses detection means of the camera of the first embodiment.

FIG. 12 is a plan view showing a modified example of the glasses detection means of the camera of the first embodiment.

FIG. 13 is a plan view showing a modified example of the glasses detection means of the camera of the first embodiment.

FIG. 14 is a flowchart showing the operation of the CPU of the camera of the first embodiment.

FIG. 15 is a diagram showing a mode of variation of the line-of-sight position.

FIG. 16 is a diagram for explaining a method of detecting a variation amount of a line-of-sight position.

FIG. 17 is a diagram for explaining a method of detecting a variation amount of a line-of-sight position.

FIG. 18 is a diagram for explaining a method of detecting a variation amount of a line-of-sight position.

FIG. 19 is a diagram for explaining a method of detecting a variation amount of the line-of-sight position.

FIG. 20 is a diagram for explaining a method of setting a focus detection area.

FIG. 21 is a diagram for explaining a method of setting a focus detection area.

FIG. 22 is a diagram for explaining a method of setting a focus detection area.

FIG. 23 is a diagram for explaining a method of setting a focus detection area.

FIG. 24 is a flowchart showing the operation of the second embodiment of the camera having the visual axis detection device according to the present invention.

FIG. 25 is a flowchart showing the operation of the third embodiment of the camera having the visual line detection device according to the present invention.

FIG. 26 is a flowchart showing the operation of the fourth embodiment of the camera having the visual axis detection device according to the present invention.

FIG. 27 is a flowchart showing the operation of the fifth embodiment of the camera having the visual axis detection device according to the present invention.

FIG. 28 is a diagram illustrating a method of detecting a line-of-sight position of a camera according to a fifth example.

FIG. 29 is a diagram illustrating a rotation amount detection unit of a camera according to a fifth example.

FIG. 30 is a flowchart showing the operation of the sixth embodiment of the camera having the visual line detection device according to the present invention.

FIG. 31 is a flowchart showing the operation of the seventh embodiment of the camera having the visual line detection device according to the present invention.

FIG. 32 is a diagram illustrating a scale displayed by the display unit of the camera of the seventh embodiment.

FIG. 33 is a diagram illustrating a scale displayed by the display unit of the camera of the seventh embodiment.

FIG. 34 is a flowchart showing the operation of the eighth embodiment of the camera having the visual line detection device according to the present invention.

FIG. 35 is a diagram illustrating a method of dividing a focus detection area of a camera according to an eighth example.

FIG. 36 is a diagram for explaining a problem of a camera having a conventional visual line detection device.

FIG. 37 is a block diagram showing a ninth embodiment of the camera having the visual line detection device according to the present invention.

FIG. 38 is a configuration diagram including an optical system of a camera according to a ninth example.

FIG. 39 is a diagram for explaining the principle of line-of-sight detection.

FIG. 40 is a diagram for explaining the principle of line-of-sight detection.

FIG. 41 is a flowchart showing the main flow of the CPU according to the ninth example.

FIG. 42 is a flowchart showing a gaze mode selection subroutine according to the ninth embodiment.

FIG. 43 is a flowchart showing a gaze detection subroutine according to a ninth embodiment.

FIG. 44 is a flowchart showing an AF subroutine according to the ninth embodiment.

FIG. 45 is an explanatory diagram of AF area selection according to the ninth embodiment.

FIG. 46 is a flowchart showing a subroutine for EG-S mode lock determination according to the ninth embodiment.

FIG. 47 is an explanatory diagram of EG-S mode lock conditions according to the ninth embodiment.

FIG. 48 is a flow chart showing a gaze mode selection subroutine according to the tenth embodiment.

FIG. 49 is a flowchart for explaining the operation of the EG-M mode of the eleventh embodiment.

FIG. 50 is a plan view showing the movement pattern of the line-of-sight position corresponding to each mode of the eleventh embodiment.

FIG. 51 is a block diagram showing twelfth to nineteenth embodiments.

FIG. 52 is a configuration diagram showing twelfth to nineteenth embodiments.

FIG. 53 is a diagram showing a main flow of the twelfth embodiment.

FIG. 54 is a flowchart showing a visual axis detection subroutine controlled in the EG-S mode.

FIG. 55 is a flowchart showing a subroutine of EG-S mode.

FIG. 56 is a flowchart showing a gaze determination subroutine of the twelfth embodiment.

FIG. 57 is a flowchart showing a gaze determination subroutine of the thirteenth embodiment.

FIG. 58 is a flowchart showing a gaze determination subroutine of the fourteenth embodiment.

FIG. 59 is a flowchart showing a gaze determination subroutine of the fifteenth embodiment.

FIG. 60 is a flowchart showing a subroutine of EG-M mode.

FIG. 61 is a flowchart showing a gaze determination subroutine of the sixteenth embodiment.

FIG. 62 is a block diagram showing a seventeenth embodiment.

FIG. 63 is a flowchart showing a subroutine of SW4 interrupt.

FIG. 64 is a flow chart showing a gaze detection subroutine of the eighteenth embodiment.

FIG. 65 is a flow chart showing a visual axis detection subroutine of the nineteenth embodiment.

[Fig. 66] Fig. 66 is a diagram for describing involuntary eye movements.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 CPU 11 Photographing lens 20 Body 30 Focus detection means 40 Eye-gaze detection means 50 Glasses detection means 63 Rotation amount detection means 1101 Eye-gaze detection means 1102 Eye-gaze control means 1103 Eye-gaze mode selection means 1104 Focus detection means 1105 AF control means 1106 AF mode selection means 1107 Photographing control means 1108 Lens drive means 1109 Display means 1201 CPU 1202 AF sensor 1203 Lens drive circuit 1204 Main mirror 1205 Photographing lens 1206 Shutter device 1207 Light emitting element 1208 Eye-gaze light-receiving element 1209 Eye-gaze detecting optical system 1210 Finder 1211 AE light-receiving element 12 Eye 1213 Screen 1214 Sub-mirror 1216 Photometric element SW1 Half-press switch SW2 Full-press switch SW3 AF mode Selection switch SW4 Line-of-sight mode selection switch

Claims (12)

[Claims] 1. A gaze detecting means for detecting a gaze position of a photographer, and a gaze determining means for judging whether or not the photographer is gazing based on a plurality of past detection results of the gaze detecting means, A camera having a line-of-sight detection device including a focus detection unit having a single-line-of-sight single focus detection mode for performing focus detection by fixing a gaze position first determined to be a gaze by the gaze determination unit. 2. The focus detecting means further has a line-of-sight continuous focus detection mode in which a focus detection area is moved along with a movement of the detected gaze position to constantly perform focus detection. The camera having the line-of-sight detection device according to claim 1, further comprising mode selection means for selecting a detection mode and the line-of-sight continuous focus detection mode. 3. A gaze detecting means for detecting a gaze position of the photographer, and a gaze determining means for judging whether or not the photographer is gazing based on a plurality of past detection results of the gaze detecting means, Focus detection is performed in a focus detection area corresponding to the gaze position determined by the gaze determination means, and focus detection means for detecting in-focus or out-of-focus of the area, and the focus detection area of the focus detection means is predetermined after in-focus. A camera having a line-of-sight detection device including focus detection area fixing means for fixing the focus detection area by the focus detection means after that when the eye continues to be in the gaze position during time. 4. A line-of-sight detecting means for detecting a line-of-sight position of a photographer, focus detection in a focus detection region corresponding to the line-of-sight position, and focus detection in a focus detection region at the time of focusing after once focused. A camera having a line-of-sight detection device including focus detection means. 5. A line-of-sight detection unit that detects the line-of-sight position of the photographer; a gaze determination unit that determines whether or not the photographer is gazing based on the detection results of the past multiple times of the line-of-sight detection unit; A camera having a line-of-sight detection device including a line-of-sight control unit having a single line-of-sight mode for fixing a gaze position first determined to be gaze by the gaze determination unit. 6. The line-of-sight control means further has a line-of-sight continuous mode for moving the line-of-sight position based on the latest line-of-sight position, and a line-of-sight mode for selecting the line-of-sight single mode or the line-of-sight continuous mode. A camera having the line-of-sight detection device according to claim 5, further comprising a selection unit. 7. The gaze determination means is configured such that the distance between the last and present gaze positions is less than or equal to a predetermined value, the difference between the previous and present eyeball rotation angles is less than or equal to a predetermined value, and the gaze movement speed is It is determined that the user is gazing when at least one condition out of being less than or equal to a predetermined value is satisfied, or when the condition is satisfied and the time for satisfying these conditions is at least a predetermined value. A camera having the line-of-sight detection device according to item 5. 8. A focus detection means for detecting focus or out-of-focus of a focus detection area on a photographing screen, an AF control means having first and second AF modes for controlling the focus detection means, AF mode selection means for selecting at least one AF mode of the AF control means, wherein the line-of-sight mode selection means is the single line-of-sight mode when the AF mode selection means selects the first AF mode. 7. The camera having the line-of-sight detection device according to claim 6, wherein the line-of-sight continuous mode is selected when the first AF mode is selected and the second AF mode is selected. 9. The first AF mode is an AF single mode in which the driving of the photographing lens is prohibited once the focus of the photographing lens is detected, and the second AF mode is the focus detection means. 9. The camera having the line-of-sight detection apparatus according to claim 8, wherein the camera is in an AF continuous mode in which the focus of the photographing lens is constantly adjusted based on the output of the camera. 10. A moving body determination means for determining whether or not the subject is a moving body, wherein the line-of-sight mode selection means selects the line-of-sight continuous mode based on the determination result of the moving body determination means. A camera having the line-of-sight detection device according to claim 6. 11. A line-of-sight detection unit that detects the line-of-sight position of the photographer, and a gaze determination unit that determines whether or not the photographer is gazing based on the detection results of the past multiple times of the line-of-sight detection unit, A camera having a line-of-sight detection device including a line-of-sight control unit having a line-of-sight multi-mode that is fixed at the previous gaze position until the line-of-sight determination unit determines the gaze position. 12. The line-of-sight control means further has a line-of-sight continuous mode for moving the line-of-sight position based on the latest line-of-sight position, and a line-of-sight mode for selecting the line-of-sight multi-mode and the line-of-sight continuous mode. A camera having a line-of-sight detection device according to claim 11, further comprising a selection unit.