JPH07146432A - Sight line detecting device - Google Patents

Abstract

PURPOSE:To detect a line of sight speedily and precisely by detecting the center gravity of the white of the eye of a photographer by processing a difference between a different light emission pattern and the detecting signal. CONSTITUTION:Light emitting LED17a and 17b are arranged in an almost symmetric position by sandwiching a plane containing the finder optical axis of a finder optical system 16 between them, and a PSD 19 receives reflected light of projected light of at least one of the light emitting LED17 and 17b, reflected from a prescribed area of eyes of a photographer. A CPU12 controls the light emission timing in a combined light emission pattern of the light emitting LED17a and 17b and the light receiving timing of the PSD19 in synchronism with the light emission timing, and extracts information on the sight line direction by processing a difference obtained by subtracting light receiving output by the light emission pattern of the light emitting LED17a and 17b from light receiving output of the PSD19 by the combined light emission pattern.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a line-of-sight detection device for detecting the direction of the line-of-sight of a photographer and setting camera information with the line-of-sight signal.

[0002]

2. Description of the Related Art Conventionally, various kinds of information are input to a camera by using, for example, a dial or a button, and the operating environment becomes complicated as the input information increases. For example, a technique of detecting the line-of-sight direction of a photographer looking into a viewfinder and inputting information to the camera from the information of the line-of-sight is disclosed in Japanese Patent Laid-Open No. 63-194237, Japanese Patent Laid-Open No. 3-87818, and the like.

According to Japanese Unexamined Patent Publication No. 1-160537, Japanese Unexamined Patent Publication No. 2-206425, etc., a technique of detecting using a cornea, an iris, a pupil, a sclera (white eye), etc. Is disclosed.

Further, a technique relating to determination of distance measurement information using the sight line detection position is also disclosed in JP-A-4-307506 and JP-A-5-88075. In addition to the above, in the signal processing of the sensor, it is common to perform the differential processing of the signal at the time of lighting / non-lighting in synchronization with the light emitting LED.

Then, for example, "Design Method of Pupil Imaging Optical System for Detecting Line of Sight" (Akira Banno, IEICE Transactions D-II vol.J74-D-II No.6 pp736-747 1991 6
(Corresponding to JP-A-2-138673)), the signals detected by light projection from different positions, that is, light projection at a red eye occurrence position and light projection at a red eye non-occurrence position are subjected to difference processing. A technique for extracting the red-eye signal from the face and detecting the pupil is also shown.

[0006]

However, in the line-of-sight detection, it is necessary to know information such as "where (position) to which direction (rotation)". As described above, the conventional detection method using, for example, corneal reflected light or iris edge as the information of a plurality of eyes requires an area sensor and an advanced image processing algorithm.
In the detection method using the ratio of black eyes, it is necessary to fix the position viewed by the photographer.

When used for a camera, it becomes difficult to fix the position where the photographer looks into, that is, the finder position or the eye point.
Furthermore, it also puts an excessive burden on the photographer.
Further, if the photographer can freely look into the position to some extent, the information can be detected using a plurality of pieces of information. However, the current area sensor detects a plurality of signals to detect the line-of-sight direction. Therefore, there is a problem that a sophisticated image processing algorithm including calibration that absorbs individual differences is required, and that line-of-sight detection of a compact camera or the like is expensive and difficult to use. In addition, when an attempt is made to detect the white-eye / black-eye ratio with a light-receiving element such as a photodiode, the reflectance of the iris portion forming the black eye and the reflectance of the white eye do not significantly differ with respect to infrared light. Became difficult and adjustment became difficult.

With respect to the above calibration problem,
There is also a method in which the initial position is displayed in the finder within the sequence (from the start of the detection sequence to the gaze information acquisition), and the gaze position is detected based on the change in the eye image caused by the display. It is desired to realize it with a cheap sensor such as PSD. However, in the case of a sensor having a low image resolution such as PSD, that is, a sensor having no pixel division, it becomes impossible to detect the change of the feature point according to the change of the line of sight due to the influence of the diffused light from the white eye / eyelid of the photographer.
In addition, the characteristic signal corresponding to the change of the line of sight of the photographer is
Since this occurs when D is lit, a detected component and a non-detected component occur at the same time. Therefore, when light projection such as red eye generation / non-red eye generation is used, if the eye rotation is detected instead of eye discrimination, new noise is generated by the red eye and fine signal detection cannot be performed. .

The present invention has been made in view of the above problems. An object of the present invention is to detect the center of gravity of the photographer's eyes by difference processing of different pattern projections and their detection signals.
It is an object of the present invention to provide an inexpensive and highly accurate visual axis detection device that performs quick and accurate visual axis detection.

[0010]

In order to achieve the above object, the line-of-sight detection apparatus according to the first aspect of the present invention is a first and a second object which are substantially symmetrical to each other with a plane including a finder optical axis of a finder optical system interposed therebetween. At least one of the first and second light projecting means arranged at two positions, and at least one of the first or second light projecting means is a predetermined region of the photographer's eye. The light receiving means for receiving the reflected light reflected from the light emitting means, the light projecting timing in the combined light projecting pattern of the first and second light projecting means, and the light receiving timing of the light receiving means synchronized with the light projecting timing are controlled. It is characterized by comprising a control means and a visual line information extraction means for extracting information on the visual line direction based on the output of the light receiving means.

The line-of-sight detection apparatus according to the second aspect is
The line-of-sight information extracting means subtracts the received light output of the first or second light projecting means from the received light output of the combined light projecting pattern of the first and second light projecting means. It is characterized by performing processing.

[0012]

That is, in the line-of-sight detection device according to the first aspect of the present invention, the first and second light projecting means are substantially symmetrical to each other with the plane including the finder optical axis of the finder optical system interposed therebetween.
Of at least one of the first or second light projecting means.
The two projected lights receive the reflected lights reflected from the predetermined area of the photographer's eye. Then, the control means controls the light projecting timing in the combined light projecting pattern of the first and second light projecting means and the light receiving timing of the light receiving means synchronized with the light projecting timing, and the line-of-sight information extracting means carries out the above. Information about the line-of-sight direction is extracted based on the output of the light receiving unit.

Further, in the visual axis detecting device according to the second aspect, the visual axis information extracting means is used to detect the first or second of the received light output by the combined light projecting pattern of the first and second light projecting means. Difference processing is performed by subtracting the received light output by the light projection pattern of the light projecting means.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Before describing the embodiments of the present invention, the principle of line-of-sight detection adopted in the present invention will be described. still,
There are various methods for detecting the line-of-sight direction, but here, a method of detecting the rotation of the eyeball by using the moving component of the center of gravity of the white eye (sclera) is adopted.

First, FIG. 13 is a diagram showing how the reflected light from the eyeball 90 of the light beam projected from the periphery of the optical axis is received and output. When light is projected onto the eyeball 90 and its reflected image is captured, a first Purkinje image (corneal reflected image) 95a having a high received light output is detected. It is difficult to detect the Purkinje images other than the first Purkinje image 95a because the amount of reflected light is weak and the position where the reflected image is formed is different. Then, the reflected light 95b from the fundus when the light is projected onto the eyeball 90 has a low output when the red eye (occurs when the light projected from the fundus returns to the detection system) does not occur, and the red eye When occurs, it is lifted though not shown in the figure. In FIG. 13, reference numeral 91 is an iris, 92 is a sclera (white eye), and 93 is a pupil.

Next, FIG. 14 is a diagram showing how the detected image changes as the eyeball 90 rotates. That is, FIGS. 14A to 14C show the state of the eyes and the light receiving element output when the eyes are moved to the left, the center, and the right (the photographer), respectively. It is clear from FIGS. 14A to 14C that the rate of occurrence of the white eye changes, and the rotation of the eyeball can be detected by using only one side of the white eye that has occurred. Also, in one shutter sequence of general photography (from near the 1st release to the 2nd release), the eye shift does not change greatly, and relative movement, that is, the change in the center of gravity of the white eye and the reference position are detected. You can know the direction the photographer is going to see. The reference position detection may be performed within the sequence.

An embodiment of the present invention based on such a principle will be described below. First, FIG. 1 is a diagram showing the configuration of a visual axis detection device according to a first embodiment of the present invention. As shown in FIG. 1, a light projecting unit 1 having a plurality of light projecting elements arranged with the optical axis sandwiched in the detection axis direction is connected to a light projecting control unit 2, and the projected light beam from the eye is projected. The light receiving unit 3 that receives the reflected light is connected to the feature extracting unit 4, and the feature extracting unit 4 receives the light receiving unit 3.
And the light projecting control unit 2 are connected to each other.

In such a configuration, the light projecting section 1 projects light to the photographer's eye by a plurality of light projecting elements based on the signal from the light projecting control section 2, and one light projecting control section 2 is provided. The above light projection pattern is created, and the light projection timing is further created. Then, the light receiving unit 3 receives the reflected light from the eye of the projected light beam, converts it into an electric signal, and outputs it to the feature extraction unit 4. Further, the feature extraction unit 4 performs signal processing (difference processing) on the basis of the signal to the light receiving unit 3 which is captured in synchronization with the signal and the light emitting signal in the light emitting timing and light emitting pattern of the light emitting control unit 2, and the line-of-sight Extract feature signals.

Next, the processing sequence of the first embodiment will be described with reference to the flowchart of FIG. Here, in the first embodiment, it is assumed that light is projected by two light projecting elements, and the light projected by the following patterns a, b, and c is determined based on the relationship shown in the following table.

[0020]

[Table 1]

When the present sequence is started, first, the pattern a is projected onto the eyeball (step S1), and the data Da is detected (step S2). Then, the pattern b is projected onto the eyeball (step S3), and the data Db is detected (step S4). Further, the pattern c is projected onto the eyeball (step S5), and the data Dc is detected (step S6). Then, the detected data Da,
Signal processing with Db and Dc signals (D0 = Db-Da, D
1 = Db-Dc) is performed (step S7), the center of gravity is detected from the data D0 and D1 (step S7), and all the sequences are finished (step S9).

As described above, in the line-of-sight detection apparatus according to the first embodiment, from the light-receiving output of the light-receiving unit by the combined light-projecting pattern of the two light-projecting elements, the light-projecting pattern of each of the light-projecting elements is obtained. A difference process is performed by subtracting the received light output to extract information about the line-of-sight direction.

Next, FIG. 3 is a diagram showing the structure of a visual axis detecting device according to the second embodiment of the present invention. In this embodiment, the focus area of the horizontally arranged three-point AF is selected using the line-of-sight signal. Further, this embodiment also has a configuration using two light emitting LEDs as in the first embodiment.

Multi-point distance measuring circuit (multi-AF circuit) 11,
The I / F circuit 20 is connected to a central processing unit (CPU) 12, which has a display circuit 13 and a light projecting LE.
D17a, 17b and the drive circuit 15 are connected. The drive circuit 15 is also connected to the lens 14. Then, the light emitting LEDs 17a and 17b and the finder optical system 1
Reference numeral 6 is connected to the line-of-sight detection optical system 18. The display circuit 13 is also connected to the finder optical system 16. The visual axis detection optical system 18 is connected to a visual axis sensor 19, and the visual axis sensor (PSD) 19 is connected to the CPU 12 via an interface circuit (I / F circuit) 20.
It is connected to the. The PSD 19 is a device that detects the position of the center of gravity of the light amount distribution with a current ratio that depends on brightness.

In such a structure, the multi-AF circuit 11 measures the distance to a plurality of points and sends the distance measurement information to the CPU 12, and when the line-of-sight detection system receives a command from the CPU 12, a predetermined display is displayed from the display circuit 13. Lighted LED
Light is projected from the eyes 17a and 17b to the eyeball 90 via the line-of-sight detection optical system 18 and the finder optical system 16.

The reflected light from the eyeball 90 is again received by the line-of-sight sensor 19 via the finder optical system 16 and the line-of-sight detecting optical system 18. The output signal from the PSD 19 is taken into the CPU 12 via the I / F circuit 20, converted into a signal corresponding to the position of the center of gravity and a signal corresponding to the brightness, and then input to the CPU 12.

This CPU 12 is based on the information about the line of sight obtained by the detection of the center of gravity after signal processing based on the signal taken in by the I / F circuit 20 and the distance measurement information of a plurality of points obtained by the multi AF circuit 11. The defocus amount of the focus area corresponding to the line-of-sight position is determined. Further, the lens 14 is driven via the drive circuit 15 with the set defocus amount.

Next, FIG. 4 is a diagram showing a detailed structure of the visual axis detecting optical system 18. As shown in FIG.
A half mirror 22 is disposed on the optical path of the light projected from the (IRED: infrared band projection LED) 17a, 17b via a projection lens 21, and the projected light is reflected by the half mirror 22. To be done. A prism 23 is arranged on the optical path of the reflected light, and the reflected light is reflected by the reflecting surface 23a of the prism 23 and guided to the eyeball 90 of the photographer through the finder 24. The reflected light from the eyeball 90 of the photographer enters the prism 23 through the finder 24 through the same optical path as the incident light, and is reflected by the reflecting surface 23a. Then, the reflected light passes through the half mirror 22 and is received by the line-of-sight sensor 19 via the light receiving lens 25.

On the other hand, the subject light is incident on the prism through the display liquid crystal 26 which is superposed on the subject light and indicates a display, passes through the reflecting surface 23a thereof, and further passes through the viewfinder 24 to the eyeball 90 of the photographer. Be guided. In this configuration, although the finder optical system is also used as a light emitting and receiving part,
It is also possible to share only one of the light receiving systems (such as installing the light projecting system outside the finder optical system) or not sharing both of them with the finder optical system.

Next, FIG. 5 is a diagram showing an equivalent relationship between the light emitting LEDs 17a and 17b, the light receiving sensor 19 and the eyeball 90. The optical axes of the projected light fluxes of the light projecting LEDs 17a and 17b and the PS
The optical axes of the received light beams of D19 do not match, and the projected light beams are
Is configured to be illuminated from the side, and the direct reflection light from the fundus light is difficult to enter the PSD 19.

Next, FIG. 6 is a diagram showing a state of a detection signal of a reflection signal from the eyeball by the projected light flux. Two floodlight LE
The reflected image from the eyeball by D17a, 17b is the projected light L
The reflected light image from the white eye and the reflected image from the cornea can be formed by the light emission of the EDs 17a and 17b. When both of the projecting LEDs 17a and 17b emit light, the reflected image from the cornea has two peaks (see FIG. 6 (a)), and when only one projecting LED emits light, the reflected image from the cornea. Has only one peak (see FIGS. 6B and 6C). In the figure, signals are projected and added in the detection direction.

When the difference processing between the two light emitting LEDs 17a and 17b emitting one light and the one emitting one is performed, FIG.
As shown in (d) and (e), the center of gravity of the movement of the white eye on one side can be detected. However, since the barycentric position is determined by the balance of the white-eye component, the corneal reflection component, and the CD component (generated due to the difference in the total light amount), which have different total light amounts between the two light projections and the one light projection, The amount of change in the detected center of gravity does not change linearly. This is as shown in FIG.

Hereinafter, with reference to the flowchart of FIG.
The main sequence of the first embodiment will be described (the initialization display uses those shown in FIGS. 12A and 12C). When the main sequence is started, first, ON / OFF determination of the 1st release is performed (step S11). If the first release is "OFF", step S
25, all the operations are completed, and the 1st release “O”
In the case of N ", initialization (setting the focus area to the center) is subsequently performed (step S12).

Then, a subroutine "line of sight detection" for detecting the line of sight is executed (step S13). Then, the line-of-sight coordinates X detected by this subroutine "line-of-sight detection"
a, Xb (center of gravity of the line of sight) are put into XA1, XA2 (step S14). Further, multi AF (AF data: DL, DC, DR: AF method from the left may be a known phase difference method or an active method) and AE (photometer known to the camera although not described in this embodiment) are performed. (Step S1
5). Then, the display for detecting the sight line reference position (display flashing or sound is preferably used together.
It may also be used as an end signal) (step S1)
6).

Then, the subroutine "line-of-sight detection" is executed again (step S17), and the line-of-sight coordinates Xa, Xb (centroids of line-of-sight) detected by this subroutine "line-of-sight detection" are put into XB1, XB2 (step S18). further,
The camera information (information not displayed in step S16) is displayed in the viewfinder (step S19), and the subroutine "focus area setting" is executed (step S20).

Then, the ON / OFF judgment of the 1st release is performed (step S21), and the 1st release "OF".
If it is F ", the process proceeds to step S25, and if it is the first release" ON ", the second release is determined (step S22). If it is not the second release" ON ", the process returns to step S21 and the second release is performed. "O
In the case of N ″, the lens is driven according to the defocus amount of the set focus area (determined by the distance measurement data L) (step S23).

In this way, the camera sequence such as the exposure sequence and the winding sequence is performed (step S2).
4) Then, this sequence is finished (step S25). Next, with reference to the flowchart of FIG. 9, a sequence of a subroutine "visual axis detection" for detecting a visual axis will be described.

When the subroutine "line-of-sight detection" sequence is started, first, a variable is initialized (variable i) (step S31), and the IRED is projected to the photographer's eyes in "pattern 1" (step S32). Then, the PSD output signal is set to IL1, in synchronization with the projection of this IRED.
It is detected as IR1, I (1) = IL1 + IR1 (step S33). Then, the IRED is projected in the eyes of the photographer in the "pattern 2" (step S34), and the IRE
The PSD output signal is set to IL2, IR2, in synchronization with the light projection of D.
It is detected as I (2) = IL2 + IR2 (step S
35). Further, the IRED is projected in the eyes of the photographer in the pattern 3 (step S36), and the PSD output signal is IL3, IR3, I (3) = IL3 in synchronization with the projection of the IRED.
It is detected as + IR3 (step S37).

Next, signals I (1) to I (I) relating to brightness
The determination of (3) (blinking, abnormality of glasses, etc.) is performed (steps S38, S40, S41). That is, E2a <I
(2) <E2b is not satisfied (step S38), the count value i is incremented (step S39), and the process returns to step S32. Further, when E1a <I (1) <E1b (step S40), the determination of I (3) is similarly made (step S46).

If E3a <I (3) <E3b (step S46), the detected current is subjected to the difference processing by the following equation to detect the center of gravity (S47, S48), and the process proceeds to step S51. In the following equation, h represents a predetermined coefficient and g represents a non-detectable predetermined value.

Current processing: IAL = IL2-IL1, IA
R = IR2-IR1 IBL = IL2-IL3, IBR = IR2-IR3 Centroid processing: Xa = h * (IAR-IAL) / (IAR +
IAL) Xb = h * (IBR-IBL) / (IBR + IBL) Further, when E3a <I (3) <E3b is not satisfied (step S46), the detected current of one of the light projections is subjected to the difference processing by the following equation to determine the center of gravity. It is detected and the process proceeds to step S51 (steps S49 and S50).

Current processing: IAL = IL2-IL1, IA
R = IR2-IR1 Centroid processing: Xa = h * (IAL-IAL) / (IAL +
IAL) Xb = g If E3a <I (3) <E3b (step S4)
1), the detection current of one of the light projections is subjected to difference processing by the following equation to detect the center of gravity, and the process proceeds to step S51 (step S43,
S44).

Current processing: IBL = IL2-IL3, IB
R = IR2-IR3 Centroid processing: Xa = g Xb = h * (IBR-IBL) / (IBR + IBL) Thus, the count value i is determined (step S4).
2), if i> k (k is a predetermined value of 1 or more), the process returns to step S39, and if i> k, the center of gravity value Xa =
g, Xb = g (g is a predetermined value) is set (step S4).
5) Then, this subroutine is exited (step S51).

Here, FIG. 10 is a diagram showing an example of the timing of light projection / detection. As shown in FIGS. 10A and 10B, P is synchronized with the simultaneous light emission of the light projecting LEDs 17a and 17b.
The SD19 detects a signal (1), the PSD 19 detects a signal (2) in synchronization with the light emission of only the light projecting LED 17a, and the PSD 19 detects a signal (3) in synchronization with the light emission of only the light projecting LED 17b. In this embodiment, (1) and (2),
The difference processing of the signals of (2) and (3) is performed, and these patterns are set as one sequence, and this pattern is repeated a predetermined number of times k. Although the signal detection of the PSD 19 is shown as the current detection type, it may be configured as the integral detection type. In addition, the projection LED 17
Of course, the order of a and 17b may be changed.

The sequence of the "focus area setting" subroutine for setting the focus area will be described below with reference to the flowchart of FIG. When entering the sequence of this subroutine, first, the barycentric coordinates XA1, XA
2, the effectiveness of XB1 and XB2 is determined (step S5).
1 to S57).

That is, when XA1 = g and XA2 = g (step S51), the line-of-sight information is not used,
After setting the distance measurement data L by the multi-AF algorithm (step S58), the line-of-sight mode mark blinks to display a warning, and the process proceeds to step S72 (step S5).
9). Then, when XA1 = g and XA2 = g are not satisfied (step S52), but XB1 = g and XB2 = g are satisfied, the process similarly proceeds to step S58 and subsequent steps. Further, XA1 = g (step S53), and XB2 = g
Also (step S57), the process proceeds to step S58 and thereafter.

If XB2 = g is not satisfied (step S57), the amount of change X2 (X2 = XA2-XB2) from the reference position due to the projection of one of the light projecting LEDs is calculated (step S66), and this amount of change is calculated. The position of X2 is determined (steps S67 and S68). Then, dM2 ≦ X2 ≦ dP2
In the case of (step S67), the focus area is set to the center and the distance measurement data L is set to DC (step S70). Further, although not dM2 ≦ X2 ≦ dP2,
If X2> dP2 (step S68), the focus area is set to the left and the distance measurement data L is set to DL (step S69). If X2> dP2 is not satisfied (step S68), the focus area is set to the right and the distance measurement data L is set to DR (step S7).
1).

On the other hand, although XB1 = g is not satisfied (step S5)
4), when XA2 = g (step S55), the change amount X1 (X
1 = XA1-XB1) is calculated (step S60), and the position of the change amount X1 is determined (steps S61 and S6).
2). If dM1 ≦ X1 ≦ dP1 (step S61), the focus area is set to the center and the distance measurement data L is set to DC (step S70). Further, when X1> dM1 (step S62), the focus area is set to the right and the distance measurement data L is set to DR (step S71). If X1> dM2 is not satisfied (step 62), the focus area is set to the left,
The distance measurement data L is set to DL (step S69).

Although not XA2 = g, XB2 = g
If it is (step S56), the process shifts to the process of step S60 and after. If XB2 = g is not satisfied (step S56), the amount of change X from the reference position is X.
1 (= XA1-XB1) and X2 (= XA2-XB2) are obtained (step S63), and the positions of the change amounts X1 and X2 are determined (steps S64 and S65). And X
If 1 <dM (step S64), the focus area is set to the right and the distance measurement data L is set to DR (step S71). Further, when X2 <dP (step S65), the focus area is set to the left and the distance measurement data L is set to DL (step S69). Also, X
If not 2> dP (step S65), the focus area is set to the center and the distance measurement data L is set to DC (step S70).

The defocus amount of the photographing lens is calculated from the distance measurement data L thus set (step S72),
Exit this subroutine (step S73). here,
FIG. 12 is a diagram showing a finder display.

12 (a) and 12 (c) show the case of displaying in the center of the screen by superimposing.
(B) and (d) are cases where the frame is displayed. In addition, the line-of-sight mode display lights up in the line-of-sight detection mode,
If the line of sight cannot be detected, a warning is given by blinking. The display method is preferably dynamic display by flashing, size change, or the like. Further, there are other displays (focus display, strobe light emission display, etc.) other than the above around the viewfinder. Incidentally, FIG. 12D shows a case where the initialization display of the focus display is also used.

Although the embodiments of the present invention have been described above, it is needless to say that the present invention is not limited to these and various improvements and modifications can be made. For example, a sensor (line sensor or the like) other than PSD may be used as long as the center of gravity can be detected. Further, the number of the light emitting LEDs is not limited to two as long as it is one or more, and the light emitting pattern may be two or more. Further, if there are a plurality of distance measuring points, there is no need to be caught by three points, and it is sufficient that there are a plurality of distance measuring points. Further, when the line-of-sight detection is not required as the camera sequence (when the distance measuring point is within the depth, etc.), the line-of-sight detection sequence may be omitted.

Further, the reliability (blinking) is determined by the sum of the detection signals of the light emission, but it is better to perform it for each light emission. Furthermore, the contact between the camera and the face other than the blink (in the camera finder). It is more preferable to combine the judgment of "pressing eyes" and the judgment of eyelashes. Furthermore, the line-of-sight detection, the line-of-sight display, and the line-of-sight detection may be performed in parallel with a sequence of cameras such as AF and AE. Also, the PSD detection integration time may be different between when both the light emitting LED 17a and the light emitting LED 17b emit light, and when one emits light, the light emission intensity may be changed. Further, the integration time may be changed, and the light emission intensity may be changed in the case of current detection. The light projecting LEDs 17a and 17b may be asymmetric with respect to the optical axis, and it is not necessary to project the light projecting timing symmetrically.

As described above in detail, in the line-of-sight detection device of the present invention, the light-projecting LEDs 17a and 17b are composed of a plurality of light sources arranged with the optical axis sandwiched between the detection axes, and the light-projection control section. Signal is emitted to the eye of the photographer, the light receiving unit receives the reflected light from the light emitting unit from the eyes in synchronization with the light emitted from the plurality of light emitting units, and the feature extracting unit emits different light. The output signal of the light receiving unit received in the pattern can be signal-processed to extract the characteristic signal regarding the line of sight. Further, the light emission control unit controls one or more light emission patterns and light emission timings of the plurality of light emission units, and the feature extraction unit synchronizes between outputs of the light receiving units detected in synchronization with different light emission patterns. By performing the difference processing, it is possible to determine the effectiveness in the light projection pattern according to the magnitude of the difference value.

[0055]

According to the present invention, the center of gravity of the photographer's eyes is detected by the different pattern projection and the difference processing of the detection signals thereof, and the line of sight is swiftly and accurately detected. It is possible to provide an inexpensive and highly accurate visual line detection device.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a visual line detection device according to a first exemplary embodiment of the present invention.

FIG. 2 is a flowchart showing a processing sequence of the first embodiment.

FIG. 3 is a diagram showing a configuration of a visual line detection device according to a second exemplary embodiment of the present invention.

FIG. 4 is a diagram showing a detailed configuration of a line-of-sight detection optical system 18.

FIG. 5 is a diagram showing an equivalent relationship among light emitting LEDs 17a and 17b, a light receiving sensor 19, and an eyeball 90.

FIG. 6 is a diagram showing a state of a detection signal of a reflected signal from an eyeball by a projected light beam.

7 is a diagram showing the amount of change in the detected center of gravity with respect to the change in movement of the eyeball 90. FIG.

FIG. 8 is a flowchart showing a main sequence of the first embodiment.

FIG. 9 is a flowchart for explaining a sequence of a subroutine “visual axis detection” for detecting a visual axis.

FIG. 10 is a diagram showing an example of detection timing of light projection / light reception.

FIG. 11 is a flowchart showing a sequence of a subroutine “focus area setting” for setting a focus area.

FIG. 12 is a diagram showing a finder display example.

FIG. 13 is a diagram showing how received light of reflected light from an eyeball 90 of a light beam projected from around the optical axis is received.

FIG. 14 is a diagram showing how a detected image changes due to rotation of an eyeball 90.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Projection part, 2 ... Projection control part, 3 ... Light receiving part, 4 ... Feature extraction part, 11 ... Multi AF circuit, 12 ... CPU, 13 ... Display circuit, 14 ... Photographing lens, 15 ... Drive circuit, 16 ... finder optical system, 17 ... projecting LED, 18 ... line-of-sight optical system, 19 ... position detecting element, 20 ... interface circuit.

Claims (2)

[Claims] 1. A first and second light projecting means arranged at least one at substantially symmetrical first and second positions with a plane including the finder optical axis of the finder optical system interposed therebetween, and A combination of a light receiving means for receiving at least one of the first or second light projecting means, which is reflected light from a predetermined region of the photographer's eye, and the first and second light projecting means. Control means for controlling the light projecting timing in the light projecting pattern and the light receiving timing of the light receiving means synchronized with the light projecting timing; and a line-of-sight information extracting means for extracting information about the line-of-sight direction based on the output of the light receiving means, A line-of-sight detection device comprising: 2. The line-of-sight information extracting means receives from the light-receiving output by the combined light-projecting pattern of the first and second light-projecting means and receives the light by the light-projecting pattern of the first or second light-projecting means. The line-of-sight detection apparatus according to claim 1, wherein the difference processing is performed by subtracting the output.