JPH1196379A - Eye position detector - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve accuracy in the position detection of an eye by dividing the detection area of the eye into plural small areas and detecting the position of the eye when curve data extracted by the lateral continuity of points extracted from a density change state to be the candidate data of the eye are confined in one area. SOLUTION: A retrieval area setting means CL1 divides inputted image data into plural parts and sets the plural retrieval areas of the eye. A curve data extracting means CL5 extracts the curve data extended in the lateral direction of a face while continuing extraction points in the direction of a pixel stream. A curve data confirming means CL6 confirms the presence/absence of curve data in the retrieval areas of the eye. When the curve data are not confirmed by the curve data confirming means CL6, a retrieval area change commanding means CL1. outputs a command to a retrieval area changing means CL1 for changing that area to the other retrieval area of the eye among the plural divided areas. When the curve data are confirmed, a position detecting means CL3 for the eye detects the position of the eye.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an eye position which can be used for automatically controlling a mirror and a seat and for specifying an eye position necessary for detecting a drowsy driving state from a driver's eye open / closed state. It relates to a detection device.

[0002]

2. Description of the Related Art As a conventional device for detecting the eye position of a driver or the like, there is, for example, one described in Japanese Patent Application Laid-Open No. Hei 7-181012.

[0003] This eye position detection device performs a binarization process on grayscale image data photographed by a camera installed in front of a driver, and performs a density projection, a labeling process, and the like on the binarized image to determine an eye position. It is configured to detect.

Further, as an eye position detecting apparatus for a gray-scale image, there is an apparatus described in JP-A-8-101904.

This eye position detecting device detects the density of pixels along a vertical pixel row of the face, determines one element for each local increase in density in the pixel row, and sets the element as an extraction point. In addition, the extraction points adjacent to each other in the pixel column direction of adjacent pixel columns are connected to detect the position of the eye from the curve data extending in the horizontal direction of the face.

[0006]

However, in the above-mentioned eye position detecting device, as shown in FIG. 23, a pattern (for example, a pattern which can be seen from the rear window, etc.) ), The data of the background, in addition to the data of the eyes, eyebrows, nostrils, etc., appear, which makes not only the selection formula for detecting the eyes very complicated, but also the accuracy of the eye position detection. There was a problem that it decreased.

[0007] Also, in the eye position detecting apparatus for a grayscale image, as shown in Fig. 24, a large number of curve data based on the background pattern still appear, and the same problem still remains.

SUMMARY OF THE INVENTION It is an object of the present invention to provide an apparatus capable of maintaining the eye position detection accuracy even if there are a lot of noise patterns in the background portion of the eye position detection target person. .

[0009]

According to a first aspect of the present invention, there is provided image input means for inputting image data including a face, and a plurality of eye search areas by dividing the input image data into a plurality of pieces. Eye search area setting means for setting, an eye search area changing means for sequentially switching the search areas of the eyes, and a pixel search along the vertical pixel row of the face in each of the eye search areas to be sequentially switched. Detecting the density, specifying the pixel of the one-way peak of the density value as an extraction point on the condition that the differential value of the density change before the one-way peak of the density value on the pixel row exceeds a predetermined value. Point extraction means, curve data extraction means capable of extracting curve data in which the extraction points extend continuously in the pixel column direction in the lateral direction of the face, and curve data confirmation for confirming the presence or absence of curve data in the eye search area Means and confirmation of said curve data When the curve data is not confirmed by the step, the search area change command means output to the eye search area change means to change to the plurality of divided eye search areas, and the curve data confirmation means An eye position detecting means for detecting an eye position from the relative positional relationship of the curve data in the eye search area when the curve data is confirmed.

According to a second aspect of the present invention, in the eye position detecting apparatus according to the first aspect, the curve data at the uppermost end of the eye search area is detected when the curve data is confirmed by the curve data confirming means. An uppermost curve data detecting means, a first stability condition determining means for the uppermost curve data which is determined to be stable on condition that the uppermost curve data appears at the same position a plurality of times, and a first stability condition. A reduced first area setting means for setting a reduced first area based on the uppermost curve data when the appearance of the uppermost curve data is stabilized by the determination means, and an appearance in the reduced first area Eye position detecting means for detecting the position of the eye from the curve data.

According to a third aspect of the present invention, in the eye position detecting apparatus according to the first or second aspect, when the appearance of the uppermost curve data is not stable as determined by the first stability condition determining means, the uppermost edge is detected. First eye search area cutting means for cutting the upper part of the eye search area on the basis of the curve data of the eye, wherein the uppermost curve data detecting means re-opens the uppermost curve data within the cut eye search area. Is detected.

According to a fourth aspect of the present invention, in the eye position detecting apparatus according to the second or third aspect, when the appearance of the uppermost curve data is stabilized by the determination of the first stability condition determining means, the uppermost edge is detected. A reduced second region setting means for reducing a region in the vertical direction and setting a reduced second region expanded in the horizontal direction based on the curve data of Curve data extraction means for extracting two, curve data feature quantity determination means for determining at least one of the positional relationship and length of the curve data extracted by the curve data extraction means, and extraction by the curve data extraction means A second stability condition determining means for determining that the obtained curve data is stable on the condition that it appears a plurality of times at the same position and the same length;
The area setting means sets the reduced first area based on the curve data when the appearance of the curve data is stable as determined by the second stability condition determining means.

According to a fifth aspect of the present invention, in the eye position detecting apparatus according to the fourth aspect, when the appearance of the curve data in the reduced second region is not stable by the second stability condition determination means, the curve data is detected. Second eye search area cutting means for cutting the upper part of the eye search area on the basis of, and the uppermost curve data detecting means detects the uppermost curve data again in the eye search area after the cut. It is characterized by the following.

According to a sixth aspect of the present invention, there is provided the eye position detecting apparatus according to the fourth or fifth aspect, wherein the eye position of the reduced first area is determined according to a state of curve data satisfying a stability condition in the reduced second area. It is characterized by having a reduced first area setting correction means for correcting the setting.

According to a seventh aspect of the present invention, in the eye position detecting apparatus according to the sixth aspect, the reduced first area setting correcting means comprises:
The setting of the first reduced area is corrected when the length of the curve data satisfying the stability condition in the second reduced area exceeds a predetermined value.

According to an eighth aspect of the present invention, in the eye position detecting apparatus according to any one of the fourth to seventh aspects, the reduced second area setting means sets a reference point based on a center coordinate of the uppermost curve data. A reduced second area is set.

According to a ninth aspect of the present invention, in the eye position detecting apparatus according to any one of the fourth to eighth aspects, an interval between two left and right curve data appearing in the reduced second region is smaller than a predetermined value. When, the curve data determining means for determining that the curve data has not been confirmed, the search area change command means, when the curve data determining means determines that the curve data was not confirmed, to the plurality of The information is output to the eye search area changing means in order to change the search area to another divided eye search area.

[0018]

According to the first aspect of the present invention, the eye detection region is divided into a plurality of small regions in advance, and one of the regions is located at the side of a point extracted from the density change state serving as eye candidate data. By determining whether or not the curve data extracted by the continuity in the direction is confirmed, and by detecting the position of the eye when the curve data is confirmed in the area, the selection formula for performing eye detection is determined. It can be simplified and the eye position detection accuracy can be improved.

That is, the whole image is divided, and the stable appearance of the curve data appearing at the top in the area is determined.
By sequentially narrowing the eye search area, it is possible to cope with differences in the position of the face due to individual differences between drivers and the like. High eye position detection is possible.

According to the invention of claim 2, in addition to the effect of the invention of claim 1, a reduced first area is set on condition that curve data appearing at the uppermost end of the eye detection area is stable, By detecting the position of the eye with respect to the curve data appearing in the area, the selection formula for detecting the eye can be simplified, and the accuracy of detecting the position of the eye can be further improved.

According to the invention of claim 3, claim 1 or 2
In addition to the effects of the invention described above, when the uppermost curve data does not appear, the upper part of the search area is cut, so that stable uppermost curve data can be detected.

According to the invention of claim 4, further, claim 2 is provided.
Or, in addition to the effect of 3, the reduced second area is set on condition that the curve data appearing at the uppermost end of the eye search area is stable, and the condition of the curve data appearing in the area is stable. Since the reduced first area is set in the range, the accuracy of eye position detection can be further improved.

According to the invention of claim 5, in addition to the effect of the invention of claim 4, when the appearance of the curve data is not stable, the upper part of the search area is cut again to reliably detect the stable curve data. it can.

According to the invention of claim 6, according to claim 4 or 5,
In addition to the effects of the present invention, the set position of the reduced first region is corrected according to the state of the curve data, and the eye position detection accuracy can be further improved.

According to the invention of claim 7, in addition to the effect of the invention of claim 6, when the length of the curve data exceeds a predetermined value, the set position of the reduced first area is corrected, so that, for example, Even in the case where the image is worn, the accurate reduced first area can be set, and the eye position detection accuracy can be further improved.

According to the eighth aspect of the present invention, in addition to the effects of any one of the fourth to seventh aspects, by setting the reduced second area based on the center coordinates of the topmost curve data, the eye The position detection accuracy can be further improved.

According to the ninth aspect of the present invention, in addition to the effects of any one of the fourth to eighth aspects, when the interval between the two left and right curve data appearing in the reduced second area is smaller than a predetermined value, Since it is determined that the curve data has not been confirmed, it is possible to further improve the eye position detection accuracy.

[0028]

FIG. 1 is a functional block diagram of an eye position detecting device according to an embodiment of the present invention. The eye position detecting device includes an eye search area setting and changing means CL1 and an image input device. Means CL2, density detecting means CL3, point extracting means CL4, curve data extracting means CL5, checking means CL6, uppermost curve data detecting means CL7, first stability condition judging means CL8, reduced second area Setting means C
L9, curve data determination means CL10, reduced first area setting means CL11, reduced first area setting correction means CL1
2 and eye position detecting means CL13.

The eye search area setting and changing means CL
1 sets or changes an eye search area as a search area setting means, a search area change means, a search area change command means, and first and second search area cut means. The image input means CL2 inputs a face image. The pixel row density detecting means CL3 detects the density of the vertical pixel row of the face image input from the image input means CL2. The point extracting means CL4 extracts points based on the increase in the density in the pixel row and its change state. The curve data extraction means CL5 continuously extracts points in the horizontal direction of the face from points adjacent to each other in the pixel column direction. The curve data checking means CL6 checks the number of curve data within the set eye search area as a curve data checking means and a curve data determining means. The uppermost curve data detecting means CL7 detects the uppermost curve data in the eye search area. The uppermost curve data stability condition determination means CL8 determines the stable appearance of the uppermost curve data. The reduced second area setting means CL9 sets a reduced second area based on the curve data when the appearance of the uppermost curve data is stable. The curve data determination unit CL10 determines the appearance state of the curve data in the reduced second area as a curve data extraction unit, a very small amount determination unit, and a second stability condition determination unit. The reduced first area setting means CL11 sets the reduced first area based on the curve group specified in the reduced second area. Setting and correcting means CL12 for the reduced first area
Corrects the setting of the reduced first area according to the state of the curve data specified in the reduced second area. The eye position detection means CL13 detects the position of the eye within the reduced first area.

FIG. 2 is a configuration block diagram according to an embodiment of the present invention.

As shown in FIG. 2, a camera 21 is provided as image input means CL2 installed on the instrument and for photographing the driver's face from the front. The input image from the camera 21 is as shown in FIG. 7 in this embodiment. In the side (X)
It consists of 512 pixels in the direction and 480 pixels in the vertical (Y) direction.

The input image taken by the camera 21 is
The data is stored in the image memory 23 as digital input image data via the AD converter 22.

An image data calculation circuit 24 is connected to the image memory 23.

The image data calculation circuit 24 includes a means for setting and changing the search area of the eye CL1, a density detecting means CL3 for the pixel row, a point extracting means CL4, the curve data extracting means CL5, Means for confirming the number of data CL6, means for detecting the uppermost curve data CL7,
The first stability condition determining means CL8, the reduced second area setting means CL9, and the curve data determining means CL10
And the reduced first area setting means CL11, and the reduced first area setting correction means CL12.

The image data calculation circuit 24 is connected to an eye position detection circuit 25 as eye position detection means CL13 for detecting the position of the eye within the reduced eye search area.

Next, the flow of operation in the above configuration will be described with reference to the flowcharts of FIGS. 3 to 6 and the explanatory diagrams of FIGS. 7 to 22.

First, in step S301, an eye search area is set. The set value of the eye search area in the initial state is one obtained by dividing the entire screen. An example of division in this embodiment will be described with reference to FIG. In FIG. 10, the entire screen is divided into four by I to IV, and is an area I in an initial state.

In step S302, each variable such as a counter, a flag, and a memory used for the subsequent determination is initialized. At this time, the initial value of the variable gksyflg that controls the switching of the eye search area is 1.

In step S303, it is determined whether or not the eye search area change flag is set. Step S30
After the initialization of each variable in step 2, since the eye search area change flag has not been set, the flow shifts to step S304.

In step S304, variables that need to be initialized each time one image data is determined are initialized. In step S305, pixel data in the eye search area is read out, and in step S306, eye candidate data serving as curve data (in the transition step, the curve data is referred to as continuous data).

The contents of the processing in steps S305 and S306 will be described with reference to the flowchart of FIG. 6 and the explanatory diagrams of FIGS.

First, in step S601, the face of the driver is photographed by the camera, and the input image for one frame is converted into a digital signal and stored in the image memory. In step S602, point extraction processing is performed on the eye search area I as shown in FIG. 7, and after the completion of one line, processing proceeds to the next adjacent line, and point extraction is performed on all lines in a predetermined direction. It is determined whether or not has been completed.

If it is determined in step S602 that point extraction has not been performed on all lines, the process proceeds to step S603. In step S603, arithmetic mean averaging of the density values of one line in a predetermined direction is performed. This processing is intended to eliminate a small variation in the change in the density value at the time of capturing the image data, and to capture a global change in the density value. FIG. 8A shows a processing result of the arithmetic averaging operation of the line data of Xa in FIG.

In step S604 of FIG.
A differential operation is performed on the arithmetic mean value that is the operation result of step 603. The result of this processing is shown in FIG.

In step S605 of FIG.
Point extraction is performed based on the differential value that is the result of operation 604. The point extracting method extracts points (p1 to p10) at which the differential value changes from negative to positive, and in FIG. 8A, points where the graph is convex downward. This point becomes a unidirectional peak of the density value on the pixel row. Next, the change in the density value until the point is reached (q
1 to q10) is equal to or less than a predetermined value, and FIG.
It is determined whether or not the differential value is equal to or less than a determination reference value (exceeding a predetermined value), and a Y coordinate value (a3, a4) is extracted for a point having a density change less than the predetermined value. After this processing is completed for one line, in step S606,
Switch to the processing of the next line.

If it is determined in step S602 that point extraction for all lines has been completed, point extraction as shown in FIG. 9A is performed. That is, Xa = X in FIG.
In the case of one line, three points A1, A2, and A3 are extracted. Then, step S607
To the extraction points (A1, A
2, A3...) Are compared, and if the Y coordinate value is within a predetermined value, the continuous data group number, continuous start line number, and continuous data number are stored as continuous data.

In this case, since the detection target is the eye, the feature amount thereof can be data that continues relatively long in the horizontal direction. Therefore, continuous data is extracted on the condition that the data continues for a predetermined value or more in the horizontal direction. it can. Therefore, step S
As a processing result in 607, G as shown in FIG.
1, continuous data of G2 can be recognized as curve data.

After extracting the continuous data in step S306,
It moves to step S307. In step S307, the variable gk controlling switching of the eye search area is set.
The state of syflg is determined. Since the initial value of gksyflg is 1 as described in step S302, the process proceeds to step S308. Step S30
At 8, it is determined whether or not continuous data exists in the eye search area. If there is one or more continuous data, the process proceeds to step S312. If there is no data, the process proceeds to step S309, and the undetected counter 1 of the continuous data is incremented. In step S310, it is determined whether or not the undetected counter 1 of the continuous data has exceeded 20. That is, after the undetected counter 1 is incremented, the process returns to step S304 again. If continuous data is not extracted in the same eye search area, the undetected counter is incremented. The process moves to step S311.

In step S311, a search area change flag 1 is set. After that, the process shifts to step S303 to determine whether or not the search area change flag is set. At this time, since the search area change flag is set, step S3
01, and the search area of the eye is changed. This processing will be described with reference to the explanatory diagram of FIG.

If the driver's face is almost at the center of the screen as shown in FIG. 11I, in step S306, continuous data of eyebrows and eyes are extracted as shown in FIG. 11I. However, depending on the driving posture and physique of the driver's face,
At the position indicated by II in FIG. 11, continuous data does not exist in the search area I of the eye. This situation can be reliably grasped by repeating the processing a plurality of times. This state is determined in step S310, and a search area change flag 1 is set as a trigger for switching to area II. In addition, even when the face is shifted leftward or the face is lower leftward, as shown in the examples of III and IV in FIG. 11, it can be sequentially switched by using the trigger that there is no continuous data in the eye search area. it can.

The initial position of the divided eye search area,
The switching order aims at detecting the position of the driver's left eye earlier and more reliably. The reason is that, in a normal driving posture, the area of the I where the left eye of the driver is most stochastically included is the area I, and the area of the II where the position of the face is lower due to the physique difference tends to be second. Because. In addition, the reason why the driver's left eye is focused on is that the image of the driver's right eye may be significantly degraded by the direct sunlight from the right side in a right-hand drive car, considering the light environment in the cabin. is there. Therefore, the setting and switching order of this area is left-right symmetric when the vehicle is a left-hand drive vehicle.

As shown in FIG. 9B, the number of pieces of continuous data in the divided eye search area is limited as described above. In contrast, the eye selection formula can be simplified, and the eye position detection accuracy can be improved.

Next, step S in the flowchart of FIG.
From 312 onward, a method for further improving the eye position detection accuracy will be described.

In step S312, the judgment value of each continuous data is read, and in step S313,
In this step, the slope of the continuous data is calculated from the upper and lower positions of the continuous start point and the upper and lower positions of the continuous end point, which are one of the determination values, and the continuous data having a large slope is deleted. This is because, assuming that the data is continuous data of the eyes, data having a face inclination or more can be deleted. This is to improve the eye detection accuracy by reducing at least one candidate data from the eye selection targets.

In step S314, step S307 is executed.
The state of the variable gksyflg that controls the switching of the eye search area is determined in the same manner as in (1). Here, too, since the initial value of gksyflg is 1, step S
Move to 315. In step S315, step S
After deleting the continuous data having a large inclination in 313, the same determination as in step S308 is performed.

Also in step S315, it is determined whether or not continuous data exists after deletion in the eye search area.
If there is one or more continuous data, step S4 in FIG.
Move to 01. If there is no one, step S3
The process proceeds to S16, and the undetected counter 2 of the continuous data is incremented.

In step S317, it is determined whether or not the undetected counter 2 of the continuous data has exceeded 20. That is, after the undetected counter 2 is incremented, the process returns to step S304 again, and if continuous data is not extracted in the same eye search area, the undetected counter is incremented. ,
It moves to step S318.

In step S318, a search area change flag 1 is set. After that, the process shifts to step S303 to determine whether or not the search area change flag is set. At this time, since the search area change flag is set, the flow shifts to step S301 to change the eye search area. The method of changing the eye search area has been described above with reference to FIG.

In step S401 of FIG. 4, as shown in FIG. 12, the uppermost continuous data in the eye search area is detected. After that, the process shifts to step S403 to determine whether or not the appearance of the uppermost continuous data is stable. The condition for stabilizing the appearance of the uppermost continuous data is that it is determined that the uppermost continuous data continuously appears a predetermined number of times at an equivalent position that allows an error in the extraction of the continuous data in the search area of the eye. If this condition is not satisfied, the process moves to step S419, and it is determined whether or not the appearance state of the uppermost continuous data satisfies the unstable condition.
The instability condition is a state in which the position of the topmost continuous data varies for each processing of each frame, and is counted up each time the continuous data position is different. It satisfies the conditions. Therefore, processing until the unstable condition is satisfied is performed in step S
Returning to 304, the process proceeds to extraction of continuous data of the next frame.

If the unstable condition is satisfied in step S419 before the appearance of the uppermost continuous data is stabilized in step S403, the process proceeds to step S404.
In step S404, the search area change flag 3 is set, and the process returns to step S303. In step S301, the eye search area is changed.

As shown in FIG. 13, the method of changing the search area of the eye is to cut and reduce the upper part of the search area based on the top position of the continuous data that appears in the search area of the same eye. . By this processing, a background pattern which tends to frequently appear in an unstable state or an area based on continuous data of splashed hair is deleted. When the continuous state at the uppermost end continues to be unstable as shown in FIG. 13, the search area of the eye is reduced stepwise, and the continuous data does not appear in the reduced area as shown in FIG. In this case, the eye search area I is switched from the eye search area I to the eye search area II in step S308 or step S315 in FIG. 3, and the eye search area can be switched to the eye position in the image data shown in FIG.

If the eyes or eyebrows can be captured as the uppermost continuous data in the eye search area in this way, the appearance of the uppermost continuous data is very stable, and the stability condition in step S403 can be easily set. Can be satisfied. At this time, gksyflg is set to 0, and the flow shifts to step S405. In step S405, it is determined whether or not gksyflg is 0. If gksyflg is 0, the process proceeds to step S406, and the eye detection accuracy is further determined based on the topmost continuous data position that has stably appeared. To improve,
A reduced second region that is reduced in the up-down direction and enlarged in the left-right direction is set as the determination region illustrated in FIG. In this area, it is determined whether the detected continuous data is an eye or an eyebrow from the appearance state of one or two continuous data appearing from the upper part in the same area.

After setting the reduced second area in step S406, gksyflg is set to 88, and gksyflg is set in step S407.
If syflg is 88, the flow shifts to step S408.

In step S408, one or two continuous data appearing in the reduced second area shown in FIG. 15 are detected, and depending on the length and positional relationship of the continuous data, the corresponding continuous data is determined by eyebrows or eyes or eyes. A determination value for determining whether the frame is a frame of glasses is calculated, and the process proceeds to step S409.

In step S409, if the eye or the eyebrow is detected, it is detected that continuous data stably appearing in a symmetrical positional relationship appears. FIG.
When the uppermost continuous data stably appearing in the eye search area is one side of the eye or the eyebrow as in the case of 長 い, the reduced second horizontal area in the left-right direction with respect to the continuous data is used. When two consecutive data from the upper part are detected, the other eye or eyebrow in the target relationship can be detected. Also in these two continuous data, as in the case of the uppermost continuous data, when they are eyes or eyebrows, they appear at the same position and at the same interval. Therefore, the stability condition can be easily satisfied as in the case of step S403.

In step S409, if the appearance of the corresponding continuous data is not stable, or even if it appears stably, if the interval between two continuous data is not equal to or more than the predetermined value L1 as shown in FIG. The continuous data to be processed is not regarded as eyes or eyebrows, and the process proceeds to step S420. FIG.
Shows an example in which a background pattern is detected as stable topmost continuous data. Even in such a case, according to the determinations in steps S408 and S409,
The corresponding noise data can be removed. The reason is that if the data is continuous data such as a background pattern or bouncing hair, it is difficult to stabilize even when it appears alone, and there is another continuous data at a position at a predetermined distance at the same height. Is extremely small.

In step S420, it is determined whether or not the appearance state of the symmetrical continuous data satisfies the unstable condition. As described above, the unstable condition is a state in which the position and length of continuous data that are symmetrical in the left and right directions vary according to the processing for each frame, and are counted up each time the continuous data and length are different. It is stated that the unstable condition is satisfied when the number of times exceeds a predetermined number. Therefore, the process until the unstable condition is satisfied returns to step S304 and shifts to extraction of continuous data of the next frame. When this process is repeated, since gksyflg is 88, step S30
7, S312, S313, S314, S407, S40
8, S409, and S420 are performed in this order, and the continuous data at the uppermost end is not detected.

If the unstable condition is satisfied in step S420 before the appearance of left-right symmetric continuous data is stabilized in step S409, the process proceeds to step S410.
In step S410, the search area change flag 3 is set, and
Returning to step S303, the eye search area is changed in step S301. The method of changing the eye search area at this time is to reduce the area based on the set position of the reduced second area set as the area for determining the appearance of symmetrical continuous data. By this processing, a background pattern which tends to frequently appear in an unstable state or an area based on continuous data of bouncing hair is deleted.

Also, gksylfg is set to 1 and step S3
By performing initialization at 02, the stability condition determination of the uppermost continuous data in the reduced eye search area is returned again. If the stable uppermost continuous data is detected again in the reduced eye search area, the stability condition of the symmetrical continuous data is determined again based on the continuous data, and the stability of the symmetrical continuous data is determined. The same determination is repeated until the condition is satisfied. Therefore, also in this case, as shown in FIG. 17, the eye search area is reduced stepwise, and when continuous data does not appear in the eye search area, the eye search is performed in step S308 or step S315. By switching from the region I to II, the eye search region can be switched to a position where the eye is present in the image data shown in FIG.

Next, as shown in FIG. 19, if the driver wears glasses and only one continuous data appears in steps S408 and S409, or if two continuous data appear, one of them is very emergency. The determination method when the length is long will be described.

In the case of a driver wearing glasses with black frames, the left and right continuous data in the reduced second area may be integrated. In such a case, instead of determining the interval between two continuous data, the determination can be made based on whether or not the length of the target continuous data is equal to or longer than a predetermined value.

When stable left-right symmetric continuous data is detected in step S409, the variable gksyflg controlling switching of the eye search area is set to 99. In step S411, after confirming that the state of gksyflg is 99, the process proceeds to step S412.

The reduced area in step S412 is the first reduced area.
As shown in FIG. 20, when the vehicle is an eye inside the vehicle, that is, a right-hand drive vehicle, the region is set to be small around the left eye of the driver. The center of the continuous data on the right side of the symmetric continuous data is used as the reference for setting the area.

Next, the correction when setting the reduced first area will be described with reference to FIG. When the driver wears glasses, the uppermost continuous data detected in the range of the eye search area is as shown in FIG. 21A, and the center coordinates in the horizontal direction of the uppermost continuous data are as follows. May not match eye position. If the reduced first area is set using these coordinate values as they are, the left eye of the driver may not be able to catch the center of the area, as shown in FIG. In such a case, it is determined that the continuous data at the uppermost end is longer than a predetermined value, it is recognized that the user is wearing glasses, and as shown in FIG. , The entire length L2 of the corresponding continuous data is calculated, and the reduced second region after correction is set by dividing L2 by a predetermined ratio a: b. With this correction, section D
Is corrected, and the eye can be detected at the center of the reduced first area. By performing the eye detection in the reduced first region of FIG. 22 which is set small around the left eye of the driver in this manner, the eye selection formula can be greatly simplified, and the detection accuracy can be improved. is there. In addition, in the detection before using this selection formula, since it is assumed that the frame is an eyebrow, an eye, or a frame of eyeglasses, one or two continuous data in a limited area is detected from above. This is a repetition of determination in a very simple form.

Finally, a method for specifying the position of the eye in the processing after step S413 will be described.

In step S413, the glasses frame
A determination formula is created based on the relative positional relationship of limited facial components such as eyebrows and eyes, and continuous data is selected according to which one of the expressions. In this selection, if the pattern does not match any of the patterns, steps S414 to S4
Then, the process proceeds to S16, where the undetected counter 3 of the continuous data is incremented, and the process proceeds to Step S417.

In step S417, the undetected counter 3
Is repeated until the value exceeds 20. In the processing flow at this time, since gksyflg is 99, determination of continuous data that is symmetrical at the uppermost end and left and right is not performed. If the undetected counter 3 exceeds 20 in step S417, the search area change flag 4 is set in step S418,
The process returns from step S303 to step S304, and the eye position detection is performed again from the initial state. It is rare in practice to enter this flow, and the probability that the detected object is not an eye is very high.

If there is continuous data that matches the eye selection formula in step S414, the process moves to step S415. In step S415, the stability condition value is calculated on the condition that the appearance position of the continuous data specified as the eye is detected a plurality of times at the same position.

Thereafter, the stability condition is determined in step S501 of FIG. If stability conditions are not met,
In step S502, a variation determination of coordinate data is performed based on a change in position coordinates of each eye of each frame. Step S502
If it is determined that there is no variation, step S3
Returning to 04, the processing when gksyflg is 99 is repeated. If it is determined in step S502 that the coordinates vary, the process proceeds to step S503, and the undetected counter 4 of the continuous data is incremented.

In step S504, it is determined whether or not the variation counter for counting the variation state has exceeded 10, and the same processing is repeated until the variation counter has exceeded 10. If so, the search area change flag 5 is set in step S505. Then, the process returns from step S303 to step S304, and the eye position detection is performed again from the initial state. When entering this flow, it can be said that the driver's face position is largely moving at the time of entering the final eye detection stage, and the eye position detection in this state includes many errors, so This is because it is preferable to return to the state and wait for the timing at which the face position calms down.

If the coordinate position of the continuous eye data is stabilized in step S501, the coordinates of the eye are output in step S506.

The overall processing flow has been described with reference to the flowcharts of FIGS. 3 to 5. In the present embodiment, the description has been made in the order of setting the reduced second area and then setting the reduced first area. As the content, even if the setting of the reduced second region is omitted, the accuracy of eye detection can be improved as compared with the conventional example.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of one embodiment of the present invention.

FIG. 3 is a flowchart illustrating the operation of the embodiment.

FIG. 4 is a flowchart showing an operation of the embodiment.

FIG. 5 is a flowchart showing the operation of the embodiment.

FIG. 6 is a flowchart showing the operation of the embodiment.

FIG. 7 is an explanatory diagram related to a process in a flowchart.

FIG. 8 is an explanatory diagram related to a process in a flowchart.

FIG. 9 is an explanatory diagram related to a process in a flowchart.

FIG. 10 is an explanatory diagram related to a process in a flowchart.

FIG. 11 is an explanatory diagram related to a process in a flowchart.

FIG. 12 is an explanatory diagram related to a process in a flowchart.

FIG. 13 is an explanatory diagram related to a process in a flowchart.

FIG. 14 is an explanatory diagram related to a process in a flowchart.

FIG. 15 is an explanatory diagram related to a process in a flowchart.

FIG. 16 is an explanatory diagram related to a process in a flowchart.

FIG. 17 is an explanatory diagram related to a process in a flowchart.

FIG. 18 is an explanatory diagram related to a process in a flowchart.

FIG. 19 is an explanatory diagram related to a process in a flowchart.

FIG. 20 is an explanatory diagram related to a process in a flowchart.

FIG. 21 is an explanatory diagram related to a process in a flowchart.

FIG. 22 is an explanatory diagram related to a process in a flowchart.

FIG. 23 is an explanatory diagram according to a conventional example.

FIG. 24 is an explanatory diagram according to a conventional example.

[Explanation of symbols]

CL1 Eye search area setting and changing means (search area setting means, search area changing means, search area change command means, first and second search area cutting means) CL2 image input means CL3 Density detecting means CL4 Point extracting means in each pixel column CL5 Curve data extracting means CL6 Curve data number checking means (curve data checking means, curve data judging means) CL7 Uppermost curve data detecting means in a predetermined area CL8 Uppermost curve Data first stability condition determination means CL9 Reduced second area setting means CL10 Curve data determination means in the reduced second area (curve data extraction means, feature quantity determination means, second stability condition determination means) CL11 Reduced first area Setting means CL12 Reduction first area setting correction means CL13 Eye position detection means 21 TV camera 22 A / D converter 3 an image memory 24 the image data calculation circuit 25 eye position detection circuit

Claims (9)

[Claims] 1. An image input means for inputting image data including a face, an eye search area setting means for dividing the input image data into a plurality and setting a plurality of eye search areas, An eye search area changing means for sequentially switching the search area of each eye, and a density value on the pixel row by detecting a density of a pixel along a vertical pixel row of a face in each of the eye search areas to be sequentially switched. Point extraction means for identifying a pixel of a one-way peak of the density value as an extraction point on condition that the differential value of the density change before the one-way peak exceeds a predetermined value; and Curve data extraction means capable of extracting curve data extending in the horizontal direction of the face continuously in the direction,
A curve data confirmation unit for confirming the presence / absence of curve data in the eye search region; and if the curve data confirmation unit does not confirm the curve data, the curve data confirmation unit changes to the plurality of divided eye search regions. A search area change instructing means for outputting to the eye search area changing means; and detecting the position of the eye from the relative positional relationship of the curve data in the eye search area when the curve data is confirmed by the curve data checking means. An eye position detecting device, comprising: an eye position detecting means. 2. The eye position detecting device according to claim 1, wherein when the curve data is confirmed by the curve data confirming means, the curve data at the uppermost end of the eye search area is detected. Detection means, first stability condition determination means for determining the uppermost curve data that is determined to be stable on the condition that the uppermost curve data appears at the same position a plurality of times, and determination by the first stability condition determination means. A reduced first area setting means for setting a reduced first area based on the uppermost curve data when the appearance of the uppermost curve data is stable; An eye position detecting device comprising an eye position detecting means for detecting the position of the eye. 3. The eye position detecting device according to claim 1, wherein the appearance of the uppermost curve data is not stable as determined by the first stability condition determination means, and the uppermost curve data is used as a reference. First eye search area cutting means for cutting the upper part of the eye search area is provided, and the uppermost curve data detecting means detects the uppermost curve data again in the cut eye search area. Eye position detecting device characterized by the following. 4. The eye position detecting device according to claim 2, wherein the uppermost curve data is used as a reference when the appearance of the uppermost curve data is stable as determined by the first stability condition determining means. Reducing second area setting means for reducing the area in the vertical direction and setting a reduced second area in the horizontal direction, and extracting one or two of the curve data appearing in the reduced second area from above. Curve data extraction means;
A characteristic amount determination unit for curve data that determines at least one of a positional relationship and a length of the curve data extracted by the curve data extraction unit; The second stability condition determination means for determining the stability of the curve data which is determined to be stable on the condition that it appears in the length, and the reduced first area setting means,
An eye position detecting device, wherein when the appearance of curve data is stabilized by the determination of the stability condition determining means, a reduced first area is set based on the curve data. 5. The eye position detecting device according to claim 4, wherein the curve data is used as a reference when the appearance of the curve data in the reduced second area is not stable as determined by the second stability condition determining means. A second search area cutting means for cutting an upper part of the eye search area, wherein the uppermost curve data detecting means detects the uppermost curve data again in the cut eye search area. Eye position detecting device. 6. The eye position detecting device according to claim 4, wherein the setting of the first reduced area is corrected in accordance with a state of curve data satisfying a stability condition in the second reduced area. An eye position detection device comprising a first area setting correction unit. 7. The eye position detection device according to claim 6, wherein the reduced first area setting correction unit has a length of curve data satisfying a stability condition in the reduced second area exceeds a predetermined value. An eye position detecting device for correcting the setting of the reduced first area. 8. The eye position detecting device according to claim 4, wherein the reduced second area setting means sets the reduced second area based on the center coordinates of the uppermost curve data. An eye position detection device characterized by setting. 9. The eye position detecting device according to claim 4, wherein when an interval between two left and right curve data appearing in the reduced second area is smaller than a predetermined value, the curve data is obtained. The search area change instructing means, when the curve data determining means determines that the curve data has not been confirmed, the other eyes divided into a plurality of pieces. An eye position detecting device for outputting to the eye search area changing means so as to change to the search area.