JP3437343B2 - Eye gaze detection device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an eye-gaze detecting device for obtaining an eye-gaze position of an observer by using individual difference data of the observer.

[0002]

2. Description of the Related Art Conventionally, various devices (for example, eye cameras) that detect what position on a viewing surface a photographer is observing, that is, a so-called line of sight (visual axis) have been provided. For example, in JP-A-1-274736,
Projects a parallel light flux from the light source to the anterior segment of the photographer's eye,
The gazing point is obtained by using the corneal reflection image by the reflected light from the cornea and the imaging position of the pupil. Further, this publication discloses an example in which a gazing point detection device is provided in a single-lens reflex camera and automatic focus adjustment of a photographing lens is performed using gazing point information of a photographer.

FIG. 7 is a diagram for explaining the principle of line-of-sight detection.

In the figure, 15 is the photographer's eyeball, and 16 is
Is a cornea and 17 is an iris. The method of detecting the line of sight will be described below with reference to this figure.

The infrared light emitted from the light source 13b illuminates the cornea 16 of the observer's eyeball 15. At this time, the corneal reflection image d (virtual image) formed by a part of the infrared light reflected on the surface of the cornea 16 is condensed by the light receiving lens 12 and imaged at the position d ′ on the image sensor 14. Similarly, the light source 13
The infrared light emitted by a illuminates the cornea 16 of the eyeball 15. At this time, the corneal reflection image e formed by a part of the infrared light reflected on the surface of the cornea 16 is condensed by the light receiving lens 12 and is imaged at the position e ′ on the image sensor 14.

The luminous flux from the ends a and b of the iris 17 is
Position a on the image sensor 14 via the light receiving lens 12
Images of the end portions a and b are formed on ′ and b ′. Light receiving lens 1
When the rotation angle θ of the optical axis of the eyeball 15 with respect to the optical axis of 2 is small, and the x coordinates of the ends a and b of the iris 17 are xa and xb, the coordinate xc of the center position c of the pupil 19 is xc≈ ( It is expressed as xa + xb) / 2.

Further, the x coordinate of the midpoint of the corneal reflection images d and e and the x coordinate xo of the center of curvature o of the cornea 16 substantially coincide with each other.
Therefore, the x-coordinates of the corneal reflection image generation positions d and e are xd,
xe, where OC is the standard distance between the center of curvature o of the cornea 16 and the center c of the pupil 19, the rotation angle θx of the optical axis 15a of the eyeball 15 is OC * SINθx≈ (xd + xe) / 2−xc. The relational expression (1) is substantially satisfied. Therefore, the rotation angle θ of the optical axis 15a of the eyeball 15 can be obtained by detecting the position of each characteristic point of the eyeball 15 (corneal reflection image and the center of the pupil) projected on the image sensor 14.

From the equation (1), the rotation angle of the optical axis 15a of the eyeball 15 is β * OC * SINθx {(xpo-δx) -xic} * pitch ... (2) β * OC * SINθy {(ypo-δy ) -Yic} * pitch ... (3) Here, θx is the rotation angle of the eyeball optical axis in the zx plane, and θy is the rotation angle of the eyeball optical axis in the yz plane. (Xpo, ypo) is the coordinates of the midpoint of the two corneal reflection images on the image sensor 14, and (xic, yic) is the coordinates of the pupil center on the image sensor 14. pitch is the pixel pitch of the image sensor 14. Further, β is an imaging magnification determined by the position of the eyeball 15 with respect to the light receiving lens 12, and is substantially obtained as a function of the interval between two corneal reflection images.

Δx and δy are correction terms for correcting the coordinates of the midpoint of the corneal reflection image, and a correction term for correcting an error caused by illuminating the eyeball of the photographer with divergent light instead of parallel light, Further, regarding δy, a correction term for correcting an offset component caused by illuminating the eyeball of the photographer with divergent light from the lower eyelid is also included.

Next, the gazing point of the photographer is obtained from the relationship between the eyeball rotation angle and the coordinates of the gazing point. The relationship between the rotation angle of the eyeball and the coordinates of the gazing point is defined by the rotation angle of the eyeball optical axis of the photographer (θx, θy)
Then, the gazing point (x, y) on the observer's observation plane is x = m * (θx + Δ) ………… (4) y = m * θy ………… when the camera is in the horizontal position. (5) is required. Here, the x direction indicates the horizontal direction with respect to the photographer when the camera is in the horizontal position, and the y direction indicates the vertical direction with respect to the photographer when the camera is in the horizontal position. m is a conversion coefficient for converting the rotation angle of the eyeball into coordinates on the focus plate, and Δ is an angle formed by the eyeball optical axis 15a and the visual axis (gazing point). Since the conversion coefficients m and Δ at this time have individual differences, it is necessary to obtain in advance the individual difference correction data including the conversion coefficients.

The individual difference correction data can be calculated by having the photographer look at a predetermined coordinate and obtaining the rotation angle of the optical axis of the eyeball at that time. The detection operation for obtaining individual difference correction data is called calibration,
Individual difference correction data obtained by calibration is called calibration data.

When the rotation angle of the eyeball when gazing at (x1,0) on the observation plane is (θx1,0) and the rotation angle of the eyeball when gazing at (x2,0) is (θx2,0), Individual difference data can be calculated by the following formula.

M = (x1−x2) / (θx1−θx2) (6) Δ = (x2 · θx1−x1 · θx2) / (x1−x2) (7) Input of individual difference data Is prohibited in a posture in which the line of sight is not detected. For example, in a camera or the like, the photographer holds the camera upside down and does not take a picture, and therefore individual difference correction data cannot be input in the upside-down state. Further, due to the hardware configuration, it is impossible to input the individual difference data in the vertical position such as when the visual axis cannot be detected in the vertical position. As a calibration proposal, we have calibration data in the vertical and horizontal directions (Japanese Patent Application No. 6-7).
6567) and so on. Further, as a proposal of the vertical position and the horizontal position, the visual line detection is prohibited at the vertical position, the display is turned off (Japanese Patent Application No. 4-2602588), the posture is detected, and the illumination is switched (Japanese Patent Application No. 4-204). 2602290) and the like.

[0014]

However, in regard to the individual difference data, although the rotation angle data of the eyeball must be obtained in the same posture, no consideration is given to the posture change during the calibration. In the conventional example, it is merely determined whether or not the posture is prohibited. Conventionally, in the line-of-sight detection device, which is prohibited from being used in the vertical position, if the orientation of the camera is changed from the horizontal position to the vertical position during the calibration, the calibration will fail. Instead of being
Calibration is simply a failure due to the prohibited vertical position. In the case of the line-of-sight detection device that can be used in the vertical position, the calibration can be performed even if the posture changes from the horizontal position to the vertical position during the calibration.

Further, in order to increase the accuracy of the individual difference data, the eyeball rotation angle data may be detected a plurality of times and statistically processed. In such a case, in particular, if the attitude of the camera changes during the acquisition of the individual difference data, incorrect individual difference correction data will be input, and the correct line-of-sight detection operation cannot be performed.

(Object of the Invention) An object of the present invention is to input incorrect correction data for individual differences due to a change in the posture of the apparatus during the acquisition of data for individual differences, thereby causing an erroneous line-of-sight detection operation. An object of the present invention is to provide a line-of-sight detection device that can prevent such a situation.

[0017]

In order to achieve the above object, the present invention provides a posture detecting means for detecting the posture of a visual line detecting device, an individual difference data acquiring means for obtaining individual difference data, and the individual detecting means. When the individual difference data is obtained by the difference data acquisition unit, a storage unit that stores the posture of the apparatus at the start of acquisition of the individual difference data and a posture that is stored in the storage unit and a posture during acquisition of the individual difference data are compared. However, when the postures are different, the individual difference data determination means for outputting that the individual difference data acquisition has failed is provided, and the posture of the apparatus is stored prior to the acquisition of the individual difference data of the observer. When the change in the posture is detected during the acquisition of the individual difference data of the observer, the fact that the acquisition of the individual difference data has failed is output.

[0018]

BEST MODE FOR CARRYING OUT THE INVENTION The present invention will be described in detail below based on the embodiments shown in the drawings.

FIG. 1 is a block diagram of a main part of an electric circuit built in a camera body according to each embodiment of the present invention.

In FIG. 1, a central processing unit (hereinafter referred to as CPU) 100 of a microcomputer, which is a camera control means incorporated in a camera body, includes a visual axis detection circuit 101, a posture detection circuit 102, and an automatic focus detection. Circuit 103, signal input circuit 104, LCD drive circuit 105,
LED drive circuit 106 and IRED drive circuit 107
Are connected. Further, although not shown, signals are transmitted to the focus adjusting circuit and the diaphragm driving circuit arranged in the taking lens through mount contacts.

As a storage means attached to the CPU 100
The EEPROM 100a stores a film counter and other shooting information, and also stores individual difference correction data obtained by performing calibration.

The visual axis detection circuit 101 outputs the eyeball image output from the image sensor 14 (CCD-EYE) to the CPU 100.
Send to. The CPU 100 A / D-converts the eyeball image signal from the image sensor 14 by the A / D conversion means inside the CPU 100, extracts this image information from each feature point of the eyeball image necessary for the sight line detection according to a predetermined algorithm, and further, The rotation angle of the photographer's eyeball is calculated from the position of each feature point.

The attitude detection circuit 102 is composed of a plurality of pairs of photosensors, LEDs (HV1, HV2) and a shield movable by gravity, and the shield moves to change the outputs of the photosensors HV1, HV2. The posture is detected by.

FIG. 2 shows a circuit diagram of the posture detection circuit 102.

The photosensors HV1 and HV2 are 90 degrees apart from each other.
They are arranged at an angle of about °, and each is also arranged at an angle from the horizontal axis. In this example, the shields (BALL1, BALL2) block the LED outputs of both sensors at the horizontal position, but the posture sensor does not necessarily have such a position configuration.

Next, the posture detection output will be described with reference to FIG.

3A-1 and 3A-2 show the case where the camera is in the horizontal position, and the photosensors HV1 and HV are shown.
Both 2 are turned off, and the output is HVSW1 = "H",
HVSW2 = “H”. 3 (b-1), (b-
In 2), the grip is in the downward vertical position and the photo sensor HV
The shield BALL1 on the first side moves, and the photo sensor HV1
Turns on. Therefore, the output is HVSW1 = "L", H
VSW2 = “H”. Similarly, (c-1) of FIG.
(C-2) is a shield BALL on the photo sensor HV2 side
2 moves, and the photo sensor HV2 is turned on. Therefore, the output is HVSW1 = "H", HVSW2 = "L".
Becomes Finally, although not shown, the output when the camera is upside down is HVSW1 = “L”, HVSW = 2.
It becomes "L".

In this way, the posture of the camera can be detected by the outputs of HVSW1 and HVSW2.

Returning to FIG. 1, the automatic focus detection circuit 10 is described.
3 is a photoelectric conversion voltage obtained from a plurality of line sensors
It is sent to U100, and in the CPU100, the line sensor signal is sequentially A / D converted by the built-in A / D conversion means,
Focus detection is performed by performing a predetermined calculation.

The signal input circuit 104 inputs signals from various switches, and when the state of the switches changes, C
It causes the PU 100 to generate an interrupt signal. SW-1 is a switch that is turned on by the first stroke of the release button to start photometry, AF, and line-of-sight detection operations. SW-2 is a release switch that is turned on by the second stroke of the release button.
-MDIAL is a mode dial that selects the various shooting modes of the camera, as well as the calibration mode and the lock position of the camera. SW-CALDEL is an individual difference correction data erasing switch. SW-DIAL1 and SW-DIAL2 are electronic dial switches that are input to the up / down counter of the signal input circuit and count the rotary click amount of the dial.

The LCD drive circuit 105 displays various information on the in-finder LCD 24 and the monitor LCD 42 and also drives a buzzer (not shown) to generate a buzzer sound.

The LED drive circuit 106 superimposes and illuminates a distance measuring point (focus detection point) in the finder 21 (LED-L2, LED-L1, LED-).
This is a circuit for lighting C, LED-R1, LED-R2). LED-L2, LED-L1, LED-C,
LED-R1 and LED-R2 are focus detection marks FP-L2 (left) and FP-L1 (not shown) in the viewfinder, respectively.
(Middle left), FP-C (middle), FP-R1 (middle right), F
It corresponds to P-R2 (right), and the focus detection mark in the finder is illuminated by turning on the LED 21.

The IRED drive circuit 107 is an infrared LED (IRED-) for illuminating the eyeball of the photographer with infrared light.
0, IRED-1, IRED-2, IRED-3, IR
ED-4, IRED-5, IRED-6, IRED-
This is a circuit for lighting 7). This circuit incorporates a circuit section for preventing overcurrent and a safety circuit section for preventing accidents due to energization for a long time. Two IREDs are used as a pair and illuminate the eyeball from below.

FIG. 8 is a diagram showing the position of a specific IRED.
Shown in.

When the camera is in the horizontal position, IRED-0 and IRED-1 are used for naked eyes, and IRED for glasses.
ED-4 and IRED-5 are used as a pair respectively, and in the vertical position on the grip, IRED- for naked eyes is used.
3, IRED-0, IRED-7, I for eyeglasses
RED-4 is used as each pair. Also,
In the vertical position under the grip, IRED-1, for the naked eye,
IRED-2, or IRED-5, IR for glasses
ED-6 is used as a pair, respectively. That is,
In any posture, the naked eye uses an IRED pair with a narrow interval, and the glasses or the eyeballs with a long distance use an IRED pair with a wide interval.

The line-of-sight detection device and individual difference data acquisition means of the present invention include a CPU 100, a line-of-sight detection circuit 101, and an IRE.
It is composed of a D drive circuit 107, an image sensor 14, an LCD drive circuit 105, an LED drive circuit 106, and an eyeball rotation angle detection means. The eyeball rotation angle detection means is CP
It is composed of a program inside the U100. Further, the CPU 100 sends a signal to each functional member of the line-of-sight detection device to perform control, and receives signals from various detection circuits to perform signal processing thereof.

The attitude detecting means of the present invention includes an attitude detecting circuit 102, a pair of photosensors, and an LED (HV).
1, HV2). The storage means is realized by the RAM in the CPU 100, and the individual difference data determination means is realized by the program written in the CPU 100.

Next, the calibration operation and display will be described with reference to the flowchart of FIG.

When the mode dial SW-MDIAL of the camera is set to the calibration mode, the program control moves to step # 1. You can change the calibration number by rotating the electronic dial (SW-DIAL1) in the calibration mode. When there is a change in the electronic dial in the calibration mode, the process of changing the calibration number is performed by interruption.

Next, the parameters are initialized (step # 2). When the initial setting is completed, the camera posture is detected and the posture of the camera is detected in four postures (horizontal position, vertical position:
The grip is classified into the upper position and the vertical position: the grip is located in the lower position (upside down) (step # 3). Next, determine whether there is calibration data for that posture (step #
4) If there is data, calibration number (CA
The L number is turned on (step # 5). If there is no data, the CAL number is blinked (step # 6).

Next, it waits until the switch SW-1 is turned off (step # 7). If the switch SW-1 remains ON, steps # 3 to # 7 are repeated.

When the switch SW-1 is turned off,
Superimpose the calibration index in the viewfinder. The right focus detection point is used as the first index when held in the horizontal position, and the focus detection point that is always above when used in the vertical position is used as the first index.

Therefore, it is determined whether the posture is the vertical position below the grip (step # 8), and if so, LED-L2
The left distance measuring mark FP-L2 is displayed by blinking (step # 10). If the camera is in any other posture, the LED-R2 is blinked and the right distance measuring mark FP-R2 is displayed. In this state, it waits until the switch SW-1 is turned on (step # 11). Switch SW-1 is off
During this period, steps # 3 to # 11 are repeated.

If the electronic dial is rotated during this time, the display of the calibration number changes.

When the switch SW-1 is turned on (step # 11), the LED blinking in step # 9 or step # 10 is turned on (step # 12), and then the calibration data is acquired. Call the subroutine "CAL line-of-sight calculation" (step # 1
3). Details will be described with reference to the flowchart of FIG. Then, when this "CAL line-of-sight calculation" ends, LE
D is turned off (step # 14), and it is determined whether the line-of-sight calculation has failed (step # 15). If the calibration is unsuccessful, the flag "CAL_ERR" is set to 1, so the determination is made with that. If it is successful, the buzzer "beep" informs the success (step # 16).

Next, it waits until the switch SW-1 is turned off (step # 17), and when it is turned off, the second index is blinked. In the case of the vertical position under the grip, the blinking LED is different, so judge that (Step # 1
8) If the vertical position is under the grip, the LED-R2 is made to blink, and the right focus detection mark FP-R2 is displayed (step # 19). Otherwise, the LED-L2 is made to blink and the left focus detection mark FP-L2 is displayed (step # 2).
0). With this, the lower focus detection point in the vertical position and the left focus detection point in the horizontal position blink.

Here, waiting for the switch SW-1 to be turned on again (step # 21), when it is turned on, the LED is changed from blinking to lighting (step # 22), and the subroutine "CAL line of sight detection" is performed (step # 21). 23). When the line-of-sight calculation is completed, turn off the LED (step # 24),
It is determined by the flag "CAL_ERR" whether the line of sight has failed (step # 25). If it is successful, a buzzer sound "pip" is given (step # 26). Then, the completion of calibration is displayed (step # 27), and the switch SW-1 is turned off (step # 2).
8). When it is turned off, the process returns (step # 31).

Finally, the case where the line-of-sight calculation fails in the subroutine "CAL line-of-sight calculation" will be described.

In this case, step # 15 or step # 15
It branches from 25 to step # 29. A failure display is displayed here. Both the "CAL" display and the number will blink, and the buzzer will be displayed continuously.
・ Make the mistake recognized by sounding "." Then, it waits until the switch SW-1 is turned off (steps # 30, #
28) and returns (step # 31).

After the return, if the mode dial of the camera is still in the calibration mode,
Run the program from step # 1.

Here, the deletion of the individual difference correction data will be briefly described.

Calibration data deletion switch SW-CAL when the camera mode is the calibration mode
By turning on DEL, the individual difference correction data for each posture can be erased. Turn on the switch SW-CALDEL
Then, the signal input circuit 104 causes the CPU 100 to generate an interrupt signal, and the CPU 100 executes the interrupt subroutine “data erase”. After executing the interrupt subroutine "data erase", the program is executed from step # 1.

Next, the subroutine "CAL line-of-sight calculation" (steps # 13 and # 23 in the flowchart of FIG. 4) which is the main part of the present invention will be described with reference to the flowchart of FIG.

When the subroutine "CAL line-of-sight calculation" is called (step # C01), it is first determined whether the camera posture is the upside down position. If it is upside down, branch to NAC_CAL_Fail and watch the calibration operation. The flag CAL_ERR indicating that the calibration has failed is set to 1 (step # C21).

On the other hand, if they are not upside down, the process branches to step # C03 to set the parameters required for calibration and to set the threshold value of the eyeball distance (step # C0).
4). The eye distance threshold varies depending on the camera posture (vertical and horizontal). Next, the line of sight is stabilized for a predetermined time (10
Wait for 0 to 200 msec) (step # C05). This is because if the calibration operation is performed immediately, the line of sight is not stable (the eyes are fluttering), and accurate calibration data cannot be obtained.

Next, it is determined whether or not the glasses determination has been completed (step # C06). The glasses determination is performed only at the beginning of the calibration operation.

If the eyeglasses determination is not completed, step #
Branch to C07. Then, the attitude of the camera is stored prior to the eyeglass determination. Here, the camera posture before the start of the acquisition of the calibration data can be recorded. The attitude of the camera is the output HVSW of the photo sensors HV1 and HV2.
1, HVSW2 as it is as 2-bit information CPU
It is stored in a predetermined RAM in 100. The posture information may be stored as a flag for each posture (such as a horizontal position flag, a vertical position flag 1, a vertical position flag 2, and an upside down position flag), and converted into that flag. If only comparing the postures, the information of the posture detection switch may be stored as it is, but to use it for other control etc., convert it to a flag corresponding to each position and store it. Is easy to handle.

Next, eyeglasses determination is performed (step # C0
8). At the beginning of calibration, it is determined whether the photographer is wearing glasses or naked eyes, and the illumination to be used and the storage control method are determined. When the glasses are judged, RAM
"Attitude memory" and "glasses determination" will not be performed from the next time. It is determined whether or not the eyeglasses determination has been correctly performed (step # C09), and if the eyeglasses determination fails, NAC_NG
Since the flag is set to 1, (Step # C1
0) to calibration failure processing (NAC_CAL_Fail)
Jump to.

If the eyeglasses determination is successful, the process branches to step # C11. Here, the attitude of the camera is detected. If the camera posture is different from the stored posture (step # C12), NA
The C_NG flag is set to 1 (step # C13), and the process jumps to the calibration failure process (NAC_CAL_Fail) (step # C14).

If the posture obtained in step # C11 is the same as the stored posture, the line-of-sight detection is performed for acquisition of the calibration data in the subroutine "CalSample" (step # C15). In this subroutine "Cal Sample", the line of sight is detected a predetermined number of times and statistical processing is performed to increase the accuracy of the data. At this time, the control of the sensor and the illumination are performed by the subroutine "glasses determination". Calibration is failed (CAL_ERR = 1) when a sufficient number of line-of-sight calculations cannot be obtained.
become.

When the data acquisition of the calibration is completed, it is judged whether or not it is the second point of the calibration (step # C16), and if the calibration is not failed at the second point (CAL_ERR = 0), the process branches to step # C18. If there is calibration data before acquiring this calibration data, read R from the EEPROM.
Read in AM. Next, new calibration data is calculated from the data acquired this time and the data up to now (step # C19). Finally, save the new calibration data in the EEPROM (step # C20) and return.

Finally, the calibration failure process (NA
C_CAL_Fail) will be described.

The process branches to either the case where the camera position is first inverted, the case where the glasses cannot be determined, or the case where the posture changes during the calibration. In this case, calibration failure (CAL_ERR =
1) is set (step # C21), and the process returns (step # C22).

The subroutine "CAL line-of-sight calculation" is shown in FIG.
Is called twice from the calibration (steps # 13 and # 23 in FIG. 4), the difference between the first and second operations will be supplementarily described.

First, when the call is made from step # 13, the glasses determination is not completed in step # C06 in FIG. 5, so the posture is stored in step # C07. In step # 16, since it is not the second point of CAL, calibration data is not calculated or saved.

When called from step # 23 in FIG. 4 (the second calibration point), step # 23 in FIG.
In C06, the process proceeds to step # C11 without branching. Therefore, in the next step # C12, the calibration 2
Posture and CAL called in step # 13 of FIG.
The posture stored during the line-of-sight calculation, that is, the first posture of calibration is compared. Furthermore, after data acquisition,
In step # C16, the process branches to steps # C17 and # C18 to calculate and save new calibration data.

In this way, the posture of the camera is stored at the beginning of calibration, and the posture of the first point of calibration is compared with the posture of the second point to determine whether or not there is a posture change. It prevents the calibration from failing and inputting incorrect calibration data.

(Second Embodiment) In the first embodiment, the posture of the calibration first point and the posture of the second point are compared, but the subroutine "CalSample" (FIG. 5) is used.
The posture may be compared each time the line-of-sight calculation is performed in step # C15).

FIG. 6 is a flow chart for explaining the second embodiment of the present invention.

This drawing is a detailed explanation flowchart of the subroutine "CalSample" of step # C15 in FIG.

When this subroutine is called, first,
Initial setting is performed when performing the CAL line-of-sight calculation (step # 60). Set the number of attempts to detect the line of sight in calibration, set the address to store the acquired data,
A counter that counts the number of times it can be detected, and "CAL_
Clear the "ERR flag". Next, the posture of the camera is detected (step # 61), and this posture is detected in step # 6 of FIG.
It is compared whether the posture is different from the posture stored in C07 (step # 62). If the posture is different, the process branches to step # 75, and the flag "CAL_ERR" indicating the failure of calibration.
Is set to 1 (step # 75), and the subroutine is immediately returned (step # 76).

On the other hand, if the posture has not changed, calibration data is input (step # 63), and the result of the line-of-sight detection is determined. NA if gaze detection is successful
Since the C_OK flag is set to 1, it is determined (step # 65). If successful, the line-of-sight data detected at step # 66 is stored in a predetermined address, and the number of successful data is counted at step # 67.

Next, the counter for counting the number of trials is subtracted (step # 68). If not 0, steps # 61 to # 68 are repeated a predetermined number of times. When it is repeated a predetermined number of times, the loop is exited and the individual difference data is calculated (step # 69). Here, the data for one index can be calculated. Next, the flag “CAL_STEP2” is used to determine whether the first calibration point is the second point. If the first point is the first point, the process branches to step # 71. If the second point is the step # 72, the address for storing the calculation data is set. To do. Store the calculated data when the address is set (step #
73). Next, it is determined whether the number of successful data (counted in step # 67) is more than a predetermined number,
If it does not reach the specified number, set the "CAL_ERR flag" to 1. Finally, the subroutine returns (step # 7).
4).

As described above, the attitude of the camera is stored at the beginning of the calibration, and whether or not the attitude is changed during the calibration is compared every time. If the attitude is changed, the calibration is failed and an error is made. It prevents the calibration data from being input.

Incidentally, the correspondence relationship between each configuration of each of the above-described embodiments and each configuration of the present invention is as described above, but the present invention is not limited to the configurations of these embodiments, and claims It goes without saying that any structure may be used as long as it can achieve the function shown in the section or the function of the embodiment.

(Modification) The present invention is described as applied to a camera such as a single-lens reflex camera, a lens shutter camera, a video camera, etc. However, other optical devices and other devices,
Furthermore, it can be applied as a constituent unit.

[0077]

As described above, according to the present invention,
The posture of the device is stored prior to the acquisition of the individual difference data of the observer, and if the change in the posture is detected during the subsequent acquisition of the individual difference data of the observer, it means that the acquisition of the individual difference data has failed. Is output.

Therefore, it is possible to prevent erroneous line-of-sight detection operation from being performed by inputting incorrect individual difference correction data due to a change in the posture of the apparatus during acquisition of individual difference data. It will be possible.

[Brief description of drawings]

FIG. 1 is a block diagram of a main part of an electric circuit built in a camera body according to each embodiment of the present invention.

FIG. 2 is a circuit diagram showing a configuration of a posture detection circuit shown in FIG.

FIG. 3 is a diagram for explaining a posture of a camera and a state of a posture detection sensor at that time.

FIG. 4 is a flowchart showing a calibration operation of the camera of the first embodiment of the present invention.

5 is a flowchart showing an operation of CAL line-of-sight calculation executed in step # 13 of FIG.

FIG. 6 is a Cal of a camera according to a second embodiment of the present invention.
It is a flow chart which shows Sample operation.

FIG. 7 is a diagram for explaining the gaze detection principle.

FIG. 8 is a diagram for explaining an arrangement relation of an eyeball illumination IRED included in the camera according to each embodiment.

[Explanation of symbols]

14 Image sensor 100 Central processing unit of microcomputer (CP
U) 101 Line-of-sight detection circuit 102 Posture detection circuit 103 Automatic focus detection circuit 104 Signal input circuit 105 LCD drive circuit 106 LED drive circuit 107 IRED drive circuit HV1 Photo sensor HV2 which is one of the constituent elements of the posture detection sensor Photo sensor, one of the components

─────────────────────────────────────────────────── ─── Continuation of the front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) A61B 3/00-3/16 G02B 7/28

Claims (4)

(57) [Claims] 1. An eye gaze detection device for obtaining the eye gaze position of an observer using individual difference data of the observer, the posture of the device is stored prior to acquisition of the individual difference data of the observer, and subsequent observation is performed. A line-of-sight detection device, which outputs that the acquisition of the individual difference data has failed when the change in the posture is detected during the acquisition of the individual difference data of the individual. 2. An eye-gaze detecting device for obtaining an eye-gaze position of an observer using individual difference data of the observer, and an attitude detecting means for detecting an attitude of the device, and an individual difference data acquiring means for obtaining the individual difference data. And a storage unit for storing the attitude of the apparatus at the time of starting the acquisition of the individual difference data in obtaining the individual difference data by the individual difference data acquisition unit, and the attitude and the individual difference data being acquired in the storage unit. And a personal difference data determination unit that outputs that the personal difference data acquisition has failed when the postures are different from each other. 3. An eyeball rotation angle detection means for obtaining an eyeball rotation angle of an observer, and an individual difference correction means for performing individual difference correction using the individual difference data for the eyeball rotation angle, wherein the eyeball rotation angle detection means is provided. The eye gaze detection device according to claim 2, wherein the eye gaze position of the observer is detected by performing the individual difference correction on the obtained eyeball rotation angle by the individual difference correction means. 4. An eyeball rotation angle detection means for obtaining an eyeball rotation angle of an observer, wherein the individual difference data acquisition means is an individual based on the eyeball rotation angle information of the observer obtained by the eyeball rotation angle detection means. The line-of-sight detection device according to claim 2, wherein the difference data is acquired.