JP2002331835A - Direct sunshine anti-glare device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an anti-glare device capable of detecting the position of eyes of an occupant without a burden on the occupant, and dynamically performing anti-glare according to the position of the detected eyes. SOLUTION: This direct sunshine anti-glare device comprises one or a plurality of cameras installed to pick up the image of a face of the occupant, a detecting means for detecting the position of the eyes of the occupant from the image picked up by the cameras, a detecting means for detecting the brightness of the light incident on the face of the occupant, a shading means for shading a windshield by changing the shading range, and a shading control means for controlling the shading means by setting the shading range according to the position of the eyes of the occupant. Since the position of the eyes is detected from the picked-up image, anti-glare can be dynamically performed without any burden on the occupant.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a direct-light anti-glare device for automatically dimming an upper part of a windshield, and more specifically, to a direct-light anti-glare device for performing anti-glare according to the position of a passenger's eyes. Related to the device.

[0002]

2. Description of the Related Art A typical automobile is provided with a sun visor that blocks direct sunlight for passengers. This sun visor is provided in the upper front part of the driver's seat, and the light is shielded by a passenger manually moving the sun visor. A conventional technique for automating anti-glare in a vehicle is disclosed in Japanese Patent Application Laid-Open No. 5-203.
No. 906, Japanese Utility Model Publication No. 5-34013, Japanese Utility Model Application Publication No.
No. 43929.

In the anti-glare device described in Japanese Utility Model Laid-Open No. 5-34013, a sensor provided above the driver's seat detects the amount of sunlight. A light-shielding member provided in a strip shape on the upper part of the windshield is controlled in accordance with the amount of light detected by the sensor to perform anti-glare.

In the anti-glare device described in Japanese Utility Model Laid-Open No. 43929/1993, it is determined whether or not the driver's eyes are exposed to sunlight based on the position of the sun detected by a sensor on the dashboard. When it is determined that the driver's eyes are exposed to sunlight, the light-shielding member provided on the windshield is controlled according to the illuminance detected by the sensor to perform anti-glare.

The anti-glare devices disclosed in Japanese Utility Model Laid-Open No. 5-34013 and Japanese Utility Model Laid-Open No. 5-43929 have no means for detecting the position of the eyes of the occupant. Whether direct sunlight hits the passenger's eye or not depends on the passenger's body type,
It changes according to the posture change. Therefore, these anti-glare devices cannot execute dynamic anti-glare according to the position of the occupant's eyes.

In order to reliably perform anti-glare according to the position of the occupant's eyes, the anti-glare device described in Japanese Patent Application Laid-Open No. 5-203906 mounts an optical element at the occupant's eyes. Light reflected by the optical element is detected by a sensor. Since the reflected light corresponds to the light incident on the driver's eyes, the light shielding member of the windshield is controlled in accordance with the intensity of the reflected light to perform anti-glare.

The anti-glare device described in Japanese Patent Application Laid-Open No. 5-203906 has the advantage that dynamic anti-glare can be performed according to the position of the eyes, but the occupant needs to mount a special optical element. Had to be put on the passengers.

[0008]

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to detect the position of the occupant's eyes without imposing any burden on the occupant, and to dynamically reduce glare according to the detected eye position. It is an object of the present invention to provide an anti-glare device capable of performing the following.

[0009]

In order to solve the above problems, a direct light anti-glare device according to the present invention comprises one or more cameras installed for photographing the face of an occupant; Detecting means for detecting the position of the occupant's eyes from the photographed image, detecting means for detecting the brightness of light incident on the occupant's face, and shielding the windshield by changing a light shielding range. It is configured to include a light-shielding means that can be set, and light-shielding control means that sets the light-shielding range according to the position of the eyes of the occupant and controls the light-shielding means.

According to the present invention, since the position of the eyes is detected from the photographed image, the glare can be dynamically reduced without any burden on the occupant.

According to one embodiment of the present invention, the brightness detecting means of the direct glare prevention device is configured to detect the brightness from the face of the occupant photographed in the image.

According to this embodiment, the direct light anti-glare device can detect the brightness without using a sensor for directly detecting the illuminance of sunlight.

According to one embodiment of the present invention, the light blocking means of the direct light anti-glare device comprises a light blocking means using liquid crystal.

According to this embodiment, the transmittance of the light shielding means can be arbitrarily changed by electrical control.

According to one embodiment of the present invention, the plurality of cameras of the direct light anti-glare device constitute a stereo camera.

According to this embodiment, the position of the occupant's face and eyes is three-dimensionally detected by the stereo camera, so that an accurate light-shielding range can be set.

According to one embodiment of the present invention, the direct light anti-glare device comprises: a traveling direction judging unit for judging a traveling direction of a vehicle; a date and time judging unit for judging a current date and time; Anti-glare determining means for determining whether to anti-glare based on the current date and time.

According to this embodiment, it is possible to determine from the current date and time and the traveling direction of the vehicle whether direct sunshine will hit the occupant's face, and antiglare can be performed according to the determination.

[0019]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 shows one embodiment of a motor vehicle equipped with an anti-glare device according to the present invention. The vehicle in FIG. 1 includes an image input unit 11, a light shielding unit 13, and a control ECU 15. In this embodiment, a face image of the occupant in the driver's seat is captured by the image input unit 11, and the positions of the occupant's face and eyes are detected from the captured image. Further, an image of a part of the face is extracted from the captured image, and the brightness of the extracted image is compared with a predetermined reference value. When the brightness of the extracted image is higher than the reference value, the light-blocking range of the light-blocking unit 13 is set according to the detected position of the occupant's eyes, and light incident from the windshield is blocked.

The light-shielding portion 13 of this embodiment is provided above the windshield, and can shield external light. For example, the light shielding unit 13 is formed of a film-like liquid crystal filter whose transmittance can be changed according to an applied voltage. This liquid crystal filter is opaque because the direction of liquid crystal molecules is irregular and scatters light when no voltage is applied, but when a voltage is applied, the liquid crystal molecules are aligned according to the direction of voltage application. Therefore, light can be transmitted.

FIG. 2 showing the front inside the vehicle shows an example of the arrangement of the liquid crystal filter of the light shielding unit 13 on the windshield. As shown in the figure, the light shielding unit 13 is formed of a liquid crystal filter having a plurality of band-shaped patterns arranged in a band with respect to the horizontal direction, and is arranged above the windshield. The plurality of belt-shaped patterns can be controlled to have different transmittances by being driven separately. The number of the band-shaped patterns may be any number as needed, but is preferably about 1 to 5. By configuring the liquid crystal filter as a band-shaped pattern, it becomes possible to shield only a necessary portion of the upper part of the windshield from light.

The image input section 11 of this embodiment is composed of two video cameras installed to photograph a passenger.
FIG. 3 shows an embodiment of the image input unit 11. In this embodiment, a stereo method of the prior art is used to stereoscopically view an object to be photographed. The two video cameras are fixed at predetermined positions so that the passenger's face can be photographed stereoscopically.

Each video camera (41, 42) is controlled via a camera control unit (45, 46). Each camera control unit is connected through an external synchronization signal line, and the left and right video cameras are synchronized by the synchronization signal.

As shown in FIG. 3, an image photographed by the image input unit 11 is processed variously by an image processing board 48 connected to the control ECU 15. For example, the image processing board 48 not only functions as an input port for an NTSC video signal, but also includes a memory for storing images and a hardware circuit for performing complicated image processing, and can perform general image processing algorithms at high speed. Can be performed. The image processing algorithm processed by the hardware circuit includes processes such as an oblique projection mechanism, a Hough transform, a binary image matching filter, and an affine transform (rotation, enlargement, and reduction of an image).

In order to reduce shading on the face in the photographed image and improve the accuracy of image recognition, an infrared irradiation unit 19 may be used as illumination as shown in FIG.
The infrared irradiator 19 is installed on the front of the occupant so as to irradiate near-infrared light to the occupant's face. Infrared irradiator 19
Is used, the wavelength bands of the left and right video cameras are respectively limited by a wavelength limiting filter 43 that limits an appropriate wavelength.

FIG. 4 shows functional blocks executed by the control ECU 15. The image input unit 11 includes a passenger 17
Is photographed from the front, and the photographed image is sent to the face image analyzing unit 21. The face image analysis unit 21 searches the captured image for an image area in which the occupant's face is captured, and continuously detects and tracks an image area in which facial features such as eyes and mouth are captured. I do. The anti-glare determining unit 31 is configured to determine the date and time
7 and whether or not direct sunlight enters the eyes of the occupant, based on the data sent from the progress judging section 29. When it is determined that direct sunlight is incident on the eyes of the occupant, based on the image captured by the image input unit 11,
The brightness detector 33 determines the brightness of light illuminating the occupant. The light-blocking control unit 35 sets a range in which the light-blocking unit 13 blocks light according to the position of the occupant's eyes, and controls a band-shaped pattern of the light-blocking unit 13 according to the set light-blocking range.

The date / time judging section 27 provides the current time data together with the current date to the anti-glare judging section 31 to judge the position of the sun. The traveling direction determination unit 29 provides data on the current traveling direction of the vehicle in order to determine which direction the occupant is facing with respect to the position of the sun. The anti-glare determining unit 31 determines whether direct sunlight enters the eyes of the occupant based on these data.

In one embodiment, a typical car navigation system using a GPS (Global Positioning System) is used for the date and time determination unit 27 and the heading determination unit 29. In a car navigation system using GPS, not only the longitude and latitude information can be used to accurately determine the current location of a vehicle, but also
The exact current time including the date is also known from the time data sent from the PS satellite. If the current time including the longitude and latitude of the vehicle and the date is known, the current position of the sun with respect to the vehicle can be obtained. In addition, typical car navigation
The system also has a function of detecting the traveling direction of the vehicle. By judging whether or not the sun is present in the traveling direction of the vehicle, it can be determined whether or not direct sunlight enters the eyes of the occupant. Therefore, by using such a car navigation system in the anti-glare device of the present invention, it is possible to process a reliable anti-glare adjustment.

In another embodiment, the date and time determination unit 27 may be a typical clock. In this case, since the current position of the vehicle is not accurately known, the approximate position of the sun is determined based only on the date and time. The traveling direction determination unit 29 may use a general compass or the like.

FIG. 5 shows a flowchart of the anti-glare adjustment process, and the anti-glare adjustment process will be described in detail with reference to FIG.

In step 101, the anti-glare determining section 31 determines whether or not to start the anti-glare adjustment processing based on the data provided from the date and time determining section 27.

For example, if the current time is noon in Japan, the position of the sun is sufficiently high with respect to the horizon, and direct sunlight does not enter the occupant's face.
There is no need for anti-glare adjustment. In such a case, the anti-glare adjustment processing is not executed. Actually, even at the same time, the position of the sun changes according to the date.
Calculates the current position of the sun based on the time including the date from the date and time determination unit 27, and determines whether direct sunlight is incident on the face of the occupant when the occupant faces the sun. .

The traveling direction data of the vehicle from the traveling direction determination section 29 is used to determine whether or not the occupant is currently facing the sun. Even at the time when the direct sunlight enters the occupant's face, if the sun is located behind or beside the occupant, the direct sunlight does not enter the occupant's eyes.
Therefore, the anti-glare determining unit 31 determines whether direct sunlight enters the occupant's eyes based on the current position of the sun and the traveling direction of the vehicle.

The traveling direction data may be used to predict the direction in which direct sunlight is incident. This is used when a light-shielding portion divided into right and left as shown in FIG. 12 is used, and it is determined whether the direction of the sun is the left direction or the right direction based on the traveling direction data, and antiglare is required. The right and left light shielding portions may be controlled according to the directions.

In step 103, the face image analysis unit 21
The position and direction of the occupant's face in the vehicle are detected from the image frame captured by the image input unit 11. Here, by detecting the passenger's face position and face orientation,
It is possible to execute the control of the blocking unit 13 according to the posture and the body type of the occupant. Hereinafter, the operation of the face image analysis unit 21 will be described in detail with reference to FIGS.

As shown in FIG. 4, the face image analysis section 21 includes a face search section 23 and a face tracking section 25, and detects the position and orientation of the occupant's face from the image captured by the image input section 11. I do. FIG. 6 shows a flowchart of the processing of the face image analysis unit 21. The face search unit 23 and the face tracking unit 25 operate in association with each other, and can detect the face direction of the occupant in real time from continuously captured images.

Steps 201 to 203 shown in FIG.
5 shows a flowchart of a process of the face search unit 23. The processing of the face tracking unit 25 is from step 205 to step 2
The detection of the occupant's face orientation, indicated by 13, is realized by these steps. The face search unit 23 is used for initial initialization immediately after the start of the face image analysis unit 21 and for error recovery in face direction detection.

The face searching section 23 roughly searches an image area where a human face is shot from the image shot by the image input section 11. The processing here is performed by the face tracking unit 2
It can also be said to be pre-processing for No. 5. Before the processing of the face tracking unit 25, the face searching unit 23 roughly searches the input image for an image area where the face is photographed, so that the face tracking unit 25 performs detailed analysis of the face in the input image. Can be executed at high speed.

First, in step 201, a region where a human face is photographed is roughly searched from the whole photographed image. This is executed by template matching using the search template 47 stored in advance.

FIG. 7 shows an example of the search template 47. The image used for the search template 47 is an image obtained by partially cutting out a human face facing the front. This image includes one characteristic region of the human face such as eyes, nose, and mouth. Included in template. The search template 47 is stored in advance as a low-resolution image in order to increase the processing speed in template matching. Further, the search template 47 is a differential image in order to reduce the influence of illumination fluctuation. This template is created from a plurality of samples and stored in advance.

The search for the entire face in step 201 is performed on the image of either the right video camera or the left video camera. Hereinafter, an example of template matching will be described assuming that an image of the right video camera is used.

When template matching is performed using the image of the right video camera, an image area corresponding to the search template 47 is detected from the right image. Next, template matching is performed on the left image using the region in the right image matched here, and the three-dimensional position of the occupant's face is obtained from the result of the stereo matching.

The face tracking unit 25 determines in step 205
The search range of the feature point of the face is set, and the feature point of the face is searched from the image captured by the image input unit 11 in step 207. The template used for this search uses an image from the three-dimensional facial feature point model stored in the database in advance. FIG. 8 shows an example of the three-dimensional face feature point model 67.

3D face feature point model 6 in this embodiment
7 is generated from partial images (51 to 65) in which characteristic portions of a human face facing the front are locally cut from the image. For example, as shown in FIG.
Face images prepared in advance, such as left eyebrow head 51, right eyebrow head 59, left eye corner 53, left eye corner 55, right eye corner 61, right eye corner 63, mouth left end 57, mouth right end 65, etc. Generated locally. Each of these partial images is associated with three-dimensional coordinates representing the three-dimensional position of the object (in this example, the left and right eyebrows, the left and right outer and inner corners of the eye, and both ends of the mouth) photographed in the image. Stored in the database. In this specification, a partial image of a facial feature region having these three-dimensional coordinates is called a facial feature point, and a face model generated from the plurality of face feature points is referred to as a three-dimensional face feature point model 67. Call. These three-dimensional face feature point models 67 are generated from a plurality of samples and stored in a database in advance.

The face tracking section 25 uses the partial image of each feature point forming the three-dimensional face feature point model 67 as a template, and searches the input image for an image area corresponding to each feature point.

In step 205, the face tracking unit 25
Sets the search range of each feature point in the input image.
For example, consider a case in which an image area corresponding to the template of the left eyebrow 51 is searched from the input image. Step 20
When Step 205 is processed after Step 3, since the image area where the entire face is photographed is searched in advance in Step 201, the image area where the left eyebrow 51 is photographed is roughly known. Therefore, the face tracking unit 25 can set a search range for each feature point template based on previously detected image information.

In step 207, the face tracking unit 25
Performs template matching using the template of each feature point. The template matching of each feature point is executed based on the search range of each feature point set in step 205.

The template matching in step 207 uses either the left or right image in the same manner as the template matching in step 201. In this embodiment, it is assumed that the image of the right video camera is used in the template matching in step 207. In this case, as a result of the search for each feature point in step 207, a total of eight partial images of the left and right eyebrows, the left and right inner and outer corners of the eye, and both ends of the mouth captured in the right image are obtained.

In step 209, the image of each feature point obtained from the search in step 207 is used as a template.
Stereo matching is performed on the left image. Thus, three-dimensional coordinates of each feature point of the input image corresponding to each feature point of the three-dimensional face feature point model 67 are obtained.

In this embodiment, stereo matching is performed using the left and right eyebrows, the left and right outer and inner corners of the eye, and the images at both ends of the mouth in the searched input image as templates. As a result of the stereo matching, three-dimensional coordinates of the left and right eyebrows, the left and right outer and inner corners of the eye of the passenger, and both ends of the mouth are obtained. Therefore, the three-dimensional position of the occupant's face in the vehicle can be obtained from the three-dimensional coordinates of each feature point of these input images.

In step 211, three-dimensional model fitting is performed using the three-dimensional face feature point model 67, and the direction of the face is detected. Hereinafter, the three-dimensional model fitting will be described.

As described above, the previously stored 3
The three-dimensional face feature point model 67 is generated from feature points of a face facing forward. On the other hand, the face captured in the input image does not always face the front. If the face captured in the input image is not facing the front,
The three-dimensional coordinates (observation points) of each feature point of the input image obtained in step 207 are shifted from the three-dimensional coordinates of each feature point of the three-dimensional face feature point model 67 by an arbitrary angle and displacement. I have. Therefore, when the three-dimensional face feature point model 67 facing the front is arbitrarily rotated and displaced, the angle and displacement corresponding to each feature point of the input image correspond to the direction and position of the face in the input image.

The three-dimensional face feature point model 67 is arbitrarily rotated,
When a displacement is made to fit each feature point of the input image, the fitting error E is expressed by the following equation.

[0054]

(Equation 1) Here, N is the number of feature points, x _i is the three-dimensional coordinates of each feature point in the model, and y _i is the three-dimensional coordinates of each feature point from the input image. ω _i is a weighting coefficient for each feature point, and uses a correlation value in stereo matching when a three-dimensional position of the feature point is obtained from the input image. By using this correlation value, the reliability of each feature point can be considered. The rotation matrix is
R (φ, θ, ψ), and the translation vector is t (x,
y, z), and these are the variables in this equation.

Therefore, if the rotation matrix R and the translation vector t that minimize the fitting error E in the above equation are obtained, the face direction and the face position of the input image can be obtained. This calculation is performed by using a fitting method using a least squares method or a virtual spring model, or the like.

In step 213, it is determined whether or not the face direction has been correctly detected in step 211. If it is determined that the face direction was not detected correctly,
Returning to step 201, a series of processing is repeated using a new image frame. If it is determined that the face direction has been correctly detected, the detected position and direction of the occupant's face are provided to the anti-glare determination unit 31 together with the corresponding image frame.

In step 105 of FIG.
Determines whether the position and orientation of the face being photographed are within a predetermined range. For example, when the occupant is completely turned sideways and does not need anti-glare adjustment (when not looking in direct sunlight), the process does not proceed to the next step. Further, even when the face direction which causes a problem in the image processing after step 105 is photographed in the image, the process does not proceed to the next step.

If it is determined in step 105 that the position and orientation of the occupant's face are within the predetermined range, the process proceeds to the next step, in which the brightness of the light illuminating the face is detected by the brightness detection unit 33. Is detected by

The brightness detecting section 33 of this embodiment detects the brightness of the light illuminating the face based on the image of the occupant's face taken by the image input section 11. For example, when the direct sunlight illuminates the passenger's face, the captured passenger's face is captured as a bright image. On the other hand, when the direct sunlight does not illuminate the passenger's face, the passenger's face is captured as a relatively dark image. Brightness detector 3
3 compares the brightness of the image area where the face of the occupant is photographed with a predetermined brightness reference value to determine the brightness of the light illuminating the face.

More specifically, the brightness detector 3
3 extracts a specific area of a face having a relatively constant contrast from a captured image (for example, an image area of a face that does not include eyes, mouths, eyebrows, etc. is extracted). In this embodiment, after extracting the specific area of the face, the brightness detection unit 33 calculates the sum of all the gray scale values of the specific area, and averages the sum with the total number of pixels of the specific area. This averaged gray scale value is used as a value representing the brightness of a specific area of the face. The average gray scale value of the specific area is compared with a reference value of brightness, and the light shielding unit 13 is controlled by the light shielding controller 35 based on the comparison result.

For example, when an image is photographed with 256 gradations, the gray scale value of 0 is the brightest and the gray scale value of 255 is the darkest value. The average value is a value between 0 and 255. The reference value of brightness is
It is determined as a gray scale value (a value from 0 to 255) set according to the brightness at which the passenger feels dazzling.

In this embodiment, the specific area of the face is extracted as two images divided based on the position of the passenger's eyes. The first image is an image of a face area around the eyes, and the second image is an image of an area under the face below the eyes. Since the light-shielding part 13 shields the area around the eyes from light and does not shield the lower part of the face, the brightness of the light illuminating the occupant is usually determined by the area of the lower part of the face that is not shielded. The brightness of the image around the eyes is used to check whether the light shielding has been performed correctly.

In step 107 of FIG. 4, the area under the face is extracted from the image. This example is shown in FIG.

As shown in FIG. 9, in order to extract the image of the area under the face, an extraction range 73 based on the position of the mouth is set in advance, and the image area of the mouth is excluded from the extraction range 73. Only the face region 77 is extracted. The area 77 under the face extracted in this way is used to determine the brightness of the face. This image extraction process is executed using each feature point extracted by the face tracking unit 25.

In step 109, the brightness detection section 33
An average gray scale value of the extracted area 77 under the face is determined and compared with a predetermined reference value. The average grayscale value of the area 77 under the face is obtained by calculating the sum of the grayscale values of all the pixels of the area 77 and dividing the sum by the total number of pixels. The average grayscale value thus obtained corresponds to the brightness of the light currently illuminating the face.

In step 109, the average gray scale value of the lower part of the face is compared with the reference value. If the average gray scale value of the lower part of the face is brighter than the reference value, light is shielded in step 113. If the average gray scale value of the lower part of the face is not brighter than the reference value, there is no need to block light, so that the entire band-shaped pattern of the light shielding unit 13 is cleared so that light passes through the upper part of the windshield.

As described above, the reference value of the brightness in this embodiment is a value set according to the brightness at which the occupant feels dazzling. In one embodiment, the anti-glare device may include means for changing the reference value according to an individual. For example, in general, as the age increases, more light is required for visual observation. In the case of such a person, the anti-glare adjustment can be changed according to the individual by setting the gray scale value serving as the reference value to a small value (ie, setting a relatively bright threshold value).

If it is determined in step 109 that the average gray scale value of the lower part of the face is brighter than the reference value, the light-shielding control unit 35 sets the light-shielding range of the light-shielding unit 13 in step 113, and according to the light-shielding range. Shield. Figure 10 below
The setting of the light shielding range will be described with reference to FIG.

FIG. 10 is a schematic view of the occupant in the vehicle as viewed from the side, showing the difference in the light-shielding range of the light-shielding portion 13 depending on the body shape. The passenger 79 indicates a relatively tall passenger, and the passenger 81 indicates a relatively short passenger. The light-blocking area in which the light-blocking section 13 has to block light for a tall passenger 79 in the situation where the sun is ahead is indicated by reference numeral 83. On the other hand, short passenger 8
The shaded area for 1 is designated by reference numeral 85. Such a difference in the light-shielding range is not only a difference in the passenger's body shape,
It also occurs in response to changes in the passenger's posture. That is,
The light-blocking range in which the light-blocking section 13 must block light varies depending on the position of the sun and the position of the eyes of the occupant. In the anti-glare device of this embodiment, the three-dimensional position of the occupant's eyes is detected in real time by a stereo camera, and an optimal light-shielding range is set according to the position of the occupant's eyes.

FIG. 11 is a diagram showing the setting of the light shielding range of the light shielding unit 13. The light shielding unit 13 includes five band-shaped patterns indicated by reference numerals 85 to 93. These strip patterns can have their transmittance changed separately by separate voltage drive. When direct sunlight is shielded, the light is shielded in order from the uppermost band-shaped pattern 85 to the lower band-shaped pattern, and antiglare is performed. Reference numeral 17 in FIG. 11 indicates a passenger looking ahead of the windshield, and reference numeral 39 indicates the sun.

The light-shielding controller 35 determines which one of the uppermost band-shaped patterns 85 is to be shaded based on the position of the occupant's eyes and the sun. Since the current eye position of the occupant has been detected in advance by the face image analysis unit 21, this information is used for setting the light blocking range.

For example, as shown in FIG.
Sets the light-shielding range of the light-shielding portion 13 so as to shield the area around the eyes of the passenger. That is, the light-shielding range is set from the uppermost band-shaped pattern 85 to a band-shaped pattern 89 existing on a straight line connecting the position of the sun and the eyes of the occupant. In this light-blocking range, the occupant does not look directly at the sun (ie, is glare-proof). At this time, the strip patterns 91 and 9 below the strip pattern 89
3 is maintained so as to transmit light, so that the brightness around the eyes of the occupant becomes darker, while the lower part of the occupant's face maintains the original brightness. The brightness at the lower part of the face is always detected in step 109 in FIG. 5 in the case where the need for light blocking has been eliminated for some reason (for example, when the sun is blocked by a building).

After the light is shielded in step 113 according to the current position of the eyes of the occupant, it is verified whether or not the anti-glare is correctly performed by the light shielding. For this verification, a new light-shielded image is acquired from the face image analysis unit 21 in step 115.

In step 117, the brightness detection unit 33
An image around the eyes of the occupant is extracted from the image after shading.
This process is similar to the image extraction of the lower part of the face in step 107. As shown in FIG. 9, an extraction range 71 based on the eye position is set in advance to extract an image around the eye, and only the face region 75 excluding the eye portion from the extraction range 71 is set. Is extracted. The brightness of the region 75 around the eyes extracted in this way is used to verify whether or not the anti-glare has been correctly performed after the light is shielded.

In step 119, the brightness detection unit 33
An average gray scale value of the extracted area 75 around the eye is obtained and compared with a predetermined reference value. The average grayscale value of the area 77 around the eyes is obtained in the same manner as the method for obtaining the average grayscale value of the area under the face.

If it is determined that the brightness of the area 75 around the eyes is not brighter than the reference value, it is determined that the light shielding in step 113 has been successful. If the brightness of the area 75 around the eyes is brighter than the reference value, the light shielding has not been successful for some reason or the light shielding is insufficient. In this case, it is necessary to finely adjust the light shielding range.
Proceed to 21.

In step 121, fine adjustment of the light shielding range is performed by lowering the band-shaped pattern by one step. After the light-shielding range has been lowered by one step in step 121, the process proceeds to step 115 again to verify whether anti-glare has been correctly performed by the light-shielding, and a series of operations is repeated using a new image.

In this embodiment, the anti-glare adjustment is performed only for the driver sitting in the driver's seat, but is not limited to the driver. For example, if necessary, all persons in the vehicle, such as a person sitting in a passenger seat and a person sitting in a rear seat, may be targeted. As used herein, "passenger"
The term includes all persons in such a vehicle.

For example, as shown in FIG. 12, the light-shielding portion provided on the windshield 37 may be divided into left and right portions so that the driver and the person sitting in the front passenger's seat are separately deglazed. In this example, the light-shielding portion is provided separately in the right band-shaped pattern 95 and the left band-shaped pattern 97, and the light-shielding ranges can be individually controlled. In this case, the face of the driver and the face of the person sitting in the passenger seat are each photographed by the image input unit, and the light-shielding range is set according to the brightness of each face.

Although the present invention has been described with reference to a specific embodiment, the present invention is not limited to such an embodiment, and various modifications that can be easily made by those skilled in the art are also included in the scope of the present invention. included.

[0081]

According to the present invention, the position of the eyes of the occupant can be detected without imposing any burden on the occupant, and the glare can be dynamically reduced in accordance with the detected position of the eyes. Can be provided.

[Brief description of the drawings]

FIG. 1 shows an embodiment of a motor vehicle equipped with an anti-glare device according to the present invention.

FIG. 2 is an example of an arrangement of a liquid crystal filter of a light shielding portion on a windshield.

FIG. 3 is an example of an image input unit.

FIG. 4 is a diagram showing functional blocks executed by a control ECU 15;

FIG. 5 is a flowchart of an anti-glare adjustment process.

FIG. 6 is a flowchart of processing of a face analysis unit.

FIG. 7 is an example of a search template.

FIG. 8 is an example of a three-dimensional facial feature point model 67;

FIG. 9 is a diagram showing extraction of a region under the face and a region around the eyes.

FIG. 10 is a schematic view of a passenger in the vehicle as viewed from the side.

FIG. 11 is a diagram illustrating setting of a light-shielding range of a light-shielding unit.

FIG. 12 is a diagram showing a light-shielding portion having a divided strip pattern.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Image input part 13 Light shielding part 17 Passenger 21 Face analysis part 27 Date and time determination part 29 Travel direction determination part 31 Anti-glare determination part 33 Brightness determination part 35 Light shielding control part 37 Windshield

Claims (5)

[Claims] A plurality of cameras installed for photographing a face of the occupant; detection means for detecting a position of the occupant's eyes from an image photographed by the camera; Detecting means for detecting the brightness of the incident light; light-shielding means capable of changing the light-shielding range to shield the windshield; and setting the light-shielding range according to the position of the passenger's eyes,
A direct light anti-glare device comprising: a light-shielding control unit that controls the light-shielding unit. 2. The apparatus according to claim 1, wherein said brightness detecting means detects brightness from a passenger's face photographed in said image. 3. The apparatus according to claim 1, wherein said light shielding means is a light shielding means using a liquid crystal. 4. The apparatus of claim 1, wherein said plurality of cameras comprises a stereo camera. 5. A traveling direction determining unit that determines a traveling direction of a vehicle, a date and time determining unit that determines a current date and time, and
The apparatus according to claim 1, further comprising: anti-glare determining means for determining whether to perform anti-glare.