JP2006251926A - Direct light detecting device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately determine whether the eyes of a driver receive direct light. <P>SOLUTION: The facial image of the driver included in a video signal outputted from a video camera 10 is acquired by a facial image acquiring part 20. An eye coordinate output part 21 detects eye coordinates indicating the positions of the eyes, based on the facial image, and tracks the eye coordinates. In the meantime, an eye opening/closing determining part 22 learns the opening/closing state of the eyes of the driver, based on the facial image and the eye coordinates, determines it, and also acquires the opening degree of the eyes of the driver. An image pick-up state acquiring part 23 acquires the image pick-up condition of the video camera 10. An eye vicinity area light measuring part 24 measures the amount of light in the area in the vicinity of the eyes of the driver, based on the facial image, the eye coordinates, and the image pick-up condition. A direct light detecting part 25 determines whether the eyes of the driver receive the direct light or not, based on the opening degree of the eyes and the light amount. A direct light signal output part 26 outputs a direct light signal which indicates the determination result of the direct light detecting means 25. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an apparatus that detects a driving state of a driver who drives a vehicle, and more specifically, to a direct light detection apparatus that determines whether or not direct light is shining on the eyes of a driver driving a vehicle. .

  In recent years, various systems have been proposed that utilize various image analysis techniques to detect the driving state of a driver driving a vehicle and assist the driver in driving the vehicle comfortably. There are various methods for detecting the driving state of the driver. As a method using the image analysis method, there is a method of determining the driving state by detecting an image of the driver's eye position and the vicinity of the eye. .

  As a conventional technique of such a system, Patent Document 1 discloses a system that determines the open / closed state of the driver's eyes by detecting the position and shape of the driver's eyes, and thereby estimates the driver's arousal state. It is disclosed.

  In the system disclosed in Patent Document 1, the open / closed state of the driver's eyes is determined based on the vertical width of the eyes. In addition, as a method for detecting the opening of the eye, the point where the density change from the white part of the driver's skin to the black eye part of the eye becomes the maximum, and from the black eye part of the eye to the white part of the skin. Examples include a method for obtaining an interval from a point at which the density change of the eye becomes maximum, a method for obtaining a vertical width of the black-eye portion of the eye by setting a binarization threshold, and the like.

  Further, Patent Document 2 discloses an example of a system that automatically reduces glare felt by the driver by detecting the driver's eye position.

  In the system disclosed in Patent Document 2, the position of the driver's eyes is determined using a monitoring system in the vehicle, and direct light from the headlight of the oncoming vehicle or the oncoming vehicle directly hits the driver's eyes. In order to prevent this, the transmittance of a part of the windshield of the vehicle is controlled. Such a system utilizes electrochromic glass, liquid crystal processed glass, detection means for detecting the incident direction of light, and the like.

  Further, Patent Document 3 is given as a conventional technique. The system disclosed in Patent Document 3 includes one or a plurality of cameras installed to photograph the face of the passenger (driver), and the eye position of the passenger from the image photographed by the camera. When the brightness of light incident on the passenger's face is detected, the light shielding range for shielding the windshield is made variable. At this time, the light shielding range is set according to the eye position of the passenger, and the light shielding means is controlled. According to this system, since the eye position is detected from the photographed image, it is possible to perform glare prevention dynamically without giving any burden to the passenger.

An example of a system related to the conventional system described above is also disclosed in Patent Literature 4 and Patent Literature 5.
JP 2000-142164 A JP-A-9-039728 JP 2002-331835 A JP 9-044685 A JP-A-7-329566

  By the way, when estimating the degree of arousal from the eye open / closed state of the driver, the accuracy of the eye open / close determination is important. On the other hand, the driver not only feels dazzling but also has a narrowed expression even when he / she laughs, for example, so that the driver opens and closes according to the vertical width of the eye as disclosed in Patent Document 1. In the case of using the method for determining the above, there arises a problem of erroneously detecting that the driver feels dazzling when making a smile.

  In order to cope with such a problem, Patent Document 4 proposes a method of performing eye open / close determination using a relative positional relationship between face parts such as the distance between the eyes and the eyebrows. It is said that it can recognize facial expressions and facial expressions when they feel dazzling.

  However, in order to perform image analysis by the method disclosed in Patent Document 4, it is necessary to install an arithmetic device having a relatively high image analysis capability in the system, resulting in a problem that the cost of the product increases.

  In addition, as a system that automatically reduces the glare felt by the driver by detecting the position of the driver's eyes, a method that uses a face image to determine whether the driver's eyes are exposed to light ( Patent Documents 2 and 3), a method of installing a sensor such as an illuminometer in the vicinity of the eye (Patent Document 5), the time that direct light hits the driver based on the altitude and direction of the sun and the traveling direction of the vehicle A determination process using a method for estimating the vehicle speed (Patent Document 3) or the like is performed, and it has not been possible to reliably determine whether or not the driver really feels dazzling.

  Therefore, the present invention is proposed in view of the above-described conventional situation, and direct light detection capable of reliably determining whether or not direct light is hitting the eyes of the driver who drives the vehicle. An object is to provide an apparatus.

  The direct light detection apparatus according to the present invention acquires a driver's face image included in the video signal output from the imaging unit by the face image acquisition unit, and based on the face image, the eye coordinate acquisition unit The eye coordinates indicating the driver's eye position are detected and the eye coordinates are tracked, and the eye opening degree of the driver is calculated based on the face image and the eye coordinates by the opening / closing eye determination means. The open / closed state of the driver's eyes is determined from the eye opening. The direct light detection device acquires the imaging conditions of the imaging means by the imaging condition acquisition means to determine whether the driver's eyes are exposed to direct light, and the eye vicinity area photometry means. Based on the eye coordinates and the imaging conditions, when the amount of light in the region near the driver's eyes in the face image is measured, direct light hits the driver's eyes based on the amount of light and the eye opening. It can be determined whether or not.

  According to the direct light detection device according to the present invention, whether the direct light hits the eye using the eye opening detected by measuring the amount of light in the vicinity of the eye in consideration of the imaging condition of the imaging means and determining the open / closed eye. It is possible to determine whether or not the driver feels dazzle by direct light.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[First Embodiment]
The present invention is applied to a direct light detection device of a direct light detection system according to the first embodiment configured as shown in FIG. 1, for example. In the present embodiment, it is assumed that the direct light detection system is mounted on the vehicle for the purpose of assisting the driving operation of the driver who drives the vehicle.

[Configuration of direct light detection system]
As shown in FIG. 1, the direct light detection system includes a video camera 10 as an imaging means for capturing a driver's face image, an illumination device 11 that illuminates the periphery of the driver's face, and sunlight on the driver's face. And a direct light detection device 12 that determines whether or not direct light such as the above is applied.

  The video camera 10 has a light receiving element such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor) sensor, a driver circuit that controls the light receiving element, a mirror head on which an optical filter, a lens, and the like are mounted. The captured image (driver's face image) is output to the direct light detection device 12 as a video signal.

  The illuminating device 11 includes a light source such as an LED (Light Emitting Diode) or a halogen lamp, a driver circuit that controls the light source, an optical element such as an optical filter, and a lens, and applies near infrared light to the driver's face. Irradiate and illuminate.

  Here, FIG. 2 shows a state in which external light including direct light hits the face of the driver when the driver drives the vehicle. In a state where the driver's face is not exposed to direct light, that is, in a normal running state, as shown in FIG. 2A, the driver does not feel dazzling and has wide eyes. In a state where the driver's face is exposed to direct light, as shown in FIG. 2 (b), the driver feels dazzling and the opening of the eyes becomes smaller than in the normal traveling state. Under this state, when the driver uses a shading device such as a sun visor mounted on the vehicle, as shown in FIG. 2 (c), the driver's eyes are shielded and the eyes are open in a normal driving state. It is equivalent to.

  Next, the direct light detection device 12 will be described with reference to FIG. As shown in FIG. 3, the direct light detection device 12 according to the first embodiment includes a face image acquisition unit 20, an eye coordinate output unit 21, an open / closed eye determination unit 22, an imaging state acquisition unit 23, and the vicinity of the eyes. An area photometry unit 24, a direct light detection unit 25, and a direct light signal output unit 26 are provided.

  The face image acquisition unit 20 receives the video signal output from the video camera 10 and uses one or a plurality of image frames in a series of image frames included in the video signal, thereby using the face image of the driver. Extract and get. At this time, the face image acquisition unit 20 performs digital processing using the video signal input from the video camera 10, captures the face image as a digital image, and stores the digital image in the video memory. Here, the position in the image at which the driver's face is imaged is determined from the mounting position and angle of view of the video camera 10, and the face image acquisition unit 20 captures a digital image of the position in the image. Become.

  The eye coordinate output unit 21 detects eye coordinates indicating the driver's eye position based on the face image acquired by the face image acquisition unit 20, and also includes a plurality of face images obtained from a plurality of face images that move back and forth in time. Eye coordinates are tracked by comparing eye positions, and output to the open / close eye determination unit 22 and the near-eye region photometry unit 24.

  The open / close eye determination unit 22 learns the open / closed state of the driver's eyes based on the face image acquired by the face image acquisition unit 20 and the eye coordinates detected and tracked by the eye coordinate output unit 21. Then, the open / close state is determined based on the eye opening of the driver, and the eye opening is output to the direct light detection unit 25.

  The imaging state acquisition unit 23 acquires information related to imaging conditions of the video camera 10. As the information acquired by the imaging state acquisition unit 23, for example, information relating to the shutter speed and the aperture when the driver's face image is captured, and values stored in a memory (not shown) can be used. is there.

  The near-eye area photometry unit 24 is based on the face image acquired by the face image acquisition unit 20, the eye coordinates acquired by the eye coordinate output unit 21, and the imaging condition acquired by the imaging state acquisition unit 23. The amount of light in the near eye area is measured and output to the direct light detection unit 25.

  The direct light detection unit 25 determines whether the driver's eyes are exposed to direct light based on the eye opening degree acquired by the open / close eye determination unit 22 and the light amount acquired by the near-eye region photometry unit 23. Is output to the direct light signal output unit 26.

  The direct light signal output unit 26 outputs the determination result by the direct light detection unit 25 as a direct light signal to an externally connected device.

  As shown in FIG. 4, the eye coordinate output unit 21 includes an eye position detection unit 21 a that detects eye coordinates indicating the driver's eye position based on the face image acquired by the face image acquisition unit 20, and An eye position tracking unit 21b that tracks eye coordinates by comparing a series of face images moving back and forth in time (comparison between a plurality of image frames) may be provided. That is, each function performed by the face image acquisition unit 20 may be processed by the eye position detection unit 21a and the eye position tracking unit 21b sharing the functions.

  Further, as shown in FIG. 4, the open / close eye determination unit 22, based on the face image acquired by the face image acquisition unit 20 and the eye coordinates detected by the eye coordinate output unit 21, An open / closed eye learning unit 22a for learning the open / closed state, and an open / closed eye determination unit 22b for performing eye opening / closing determination based on a learning result by the open / closed eye learning unit 22a and acquiring an eye opening degree. May be. That is, each function performed by the open / close eye determination unit 22 may be processed by the open / close eye learning unit 22a and the open / close eye determination unit 22b sharing the functions.

"Eye position detection process"
Next, the eye position detection process in the eye position detection unit 21a will be described.

  As shown in FIG. 5, when the eye position detection process is started, the eye position detection unit 21a determines whether the one-side candidate detection process or the both-sides determination process is performed when position detection starts. The detected detection mode is confirmed (step S10).

  Then, the eye position detection unit 21a performs a one-eye candidate detection process (step S11) for detecting a candidate for one eye from the entire screen of the face image according to the confirmed detection mode. Then, the process proceeds to one of the binocular confirmation processing (step S12) in which both eyes are detected and both eyes are confirmed. Further, when the process in step S11 or step S12 is completed, the eye position detection unit 21a ends the eye position detection process.

  When the one-eye candidate detection process is started in step S11, the eye position detection unit 21a sets a process area for detecting one-eye candidates (step S20), as shown in the flowchart of FIG. Whether to perform continuous data detection of density change points existing in the processing area (step S21), narrowing down one eye candidate (step S22), and detecting one eye candidate in the next image frame Then, the process of determining whether to perform the binocular determination process (step S23) is sequentially performed.

  Here, the processing area set in step S20 is set as a right half region or a left half region of the face image as shown in FIG. Note that FIG. 7 shows an example in which the face image is set to the right half region. Then, in the vicinity of the right half or left half of the face image, for example, as shown in FIG. 8, continuous data (group A) that is a group of density change points adjacent in the horizontal direction indicating the outline of the eye, Continuous data (group B) that is a density change point group that is not an eye such as an eyebrow is detected.

  In step S21, the eye position detection unit 21a detects a change in the light amount density for each vertical line (FIG. 9A) in the processing area set in step S20 (FIG. 9B). A process of detecting the density change point group as continuous data by detecting a density change point group in which the detected density change points that change from white (light) to black (dark) continuously in the horizontal direction are detected. Perform (FIG. 9C). In FIG. 9A, only one of the plurality of vertical lines set in the processing area is shown, and the change in the light amount density for the one vertical line is shown in FIG. 9B. Shown in FIG. 9C shows a state in which both end points of a plurality of vertical lines are arranged while paying attention to both end points detected as “eye” in the change in the light intensity density shown in FIG. 9B. In FIG. 9C, both end points detected as “eye” are indicated by “◯” and “x”, respectively.

  In the process of narrowing down one eye candidate in step S22, if the continuous data detected in step ST21 by the eye position detection unit 21a is an upper eyelid line, it is assumed that it has an upward convex shape. Based on the shape recognition, only the density change points that match the convex shape are left as an image showing the eye.

  In step S23, whether the eye position detection unit 21a performs the one-eye candidate detection process shown in FIG. 6 again in the next image frame or the binocular determination process in step S12 in the next image frame. to decide.

  If it is determined that the binocular confirmation process is to be performed in step S12, the binocular confirmation process is performed with the density change point detected as continuous data having a continuously convex shape in the vicinity of the same coordinate and being detected as continuous data. Transition. In step S12, as shown in FIG. 10, processing for setting a processing area for detecting both eyes at the coordinates of one eye candidate (step S30) and continuous data of density change points in the processing area are detected. Processing (step S31), processing for determining both eyes by pattern recognition by comparing continuous data detected as a binocular candidate and a template prepared in advance (step S32), and one eye candidate in the next image frame The process of determining whether to return to the detection process, to continue the binocular determination process, or to perform the eye position tracking process (step S33) is sequentially performed.

  Here, as the processing area set in step S30, as shown in FIG. 11, an area including both eyes of the driver is set in the face image. In FIG. 10, a region for detecting the nose is also set, and the density change point detected from the nose is also shown in the region.

  In addition, the pattern recognition processing employed in step S32 prepares in advance image data that generates a template eye image from a plurality of eye images, and the template and the continuous data detected in step ST31. It is possible to use a method using a cross-correlation method for obtaining the correlation. Alternatively, in this step ST32, a plurality of eye images and non-eye face part images (glass frame, eyebrows, nose, etc.) are learned by a neural network in advance, and an image for determining whether or not an eye is present is detected by the neural network. A method for determining more can be used.

"Eye position tracking process"
Next, the eye position tracking process in the eye position tracking unit 21b will be described with reference to the flowchart shown in FIG.

  When the eye position tracking process is started, the eye position tracking unit 21b starts from the region near the eye coordinates detected by the previous eye position detection process or the binocular confirmation process as shown in FIG. Eyes are detected (step S40). Here, the detection position of the reference eye in the above-described eye position detection process refers to “eye (FIG. 9C)” detected by the eye position detection unit 21a, as shown in FIG. The upper end, the lower end, the left end, and the right end are set, and the center position is detected as the reference eye position.

  Next, the eye position tracking unit 21b determines whether or not the reference eye has been detected (step S41), and if it has been detected, as shown in FIG. 13B, the detected current reference eye is detected. The interocular distance from the coordinates to the coordinates of the eyes detected by the previous eye position detection process or the binocular confirmation process is obtained, a detection area for the opposite eye is set, and the opposite eye is detected from the detection area (step) S42).

  In step S41, if it is determined that the reference eye cannot be detected, the eye position tracking unit 21b changes the opposite eye to the reference eye (step S43) and detects the reference eye again (step S44). Next, the eye position tracking unit 21b determines whether or not the opposite eye can be detected as the reference eye (step S45). If it is determined that the eye cannot be detected, it is considered that tracking in the image frame has failed. (Step S46), the eye position tracking process is terminated. In Step S45, when it is determined that the eye position tracking unit 21b has detected the opposite eye as the reference eye, the process proceeds to Step S42.

  After detecting the opposite eye in step S42, the eye position tracking unit 21b determines whether or not the opposite eye has been detected (step S47), and if detected, the coordinates of the detected reference eye and the opposite eye are detected. Is determined (step S48), and the eye position tracking process is terminated. Further, when it is determined in step S47 that the opposite eye cannot be detected, the eye position tracking unit 21b obtains the interocular distance from the previous eye coordinates, and moves the interocular distance corresponding to the reference eye coordinates. The position is determined as the coordinates of the opposite eye (step S49), and the eye position detection process is terminated.

"Open / close eye judgment process"
Next, the opening / closing eye learning process and the opening / closing eye determination (determination) process by the opening / closing eye determination unit 22 will be described with reference to the flowchart shown in FIG.

  When the open / close eye learning process and the open / close eye determination process are started, the open / close eye determination unit 22 calculates 2 based on the gray value of the image from the image area near the eye coordinates obtained by the eye position tracking process by the eye position tracking unit 21b. By determining the threshold value and cutting out the eye region from the face image, the “left end”, “right end”, “upper end”, and “lower end” of the eye are obtained, and the open / close eye determination parameters are set (step S50). Here, the coordinates of the “left end”, “right end”, “upper end”, and “lower end” of the eye that are set as the open / closed eye determination parameters should be set so that the relative positional relationship differs between when the eye is opened and when the eye is closed. become.

  Next, the open / closed eye determination unit 22 obtains the distance (vertical width of the eye) between the “upper end” and the “lower end” of the eye as the eye opening (step S51). Here, from the open / closed eye determination parameters set in step ST50, the relative positional relationship between the “upper end” and “lower end” of the eye becomes closer as the open / closed state of the eye approaches the closed eye.

  Next, based on the shape of the upper eyelid by the “right end”, “left end”, and “upper end” of the eye, the open / close eye determination unit 22 coordinates “upper end” with respect to the line segment connecting the “right end” and “left end”. Is determined to be above or below. Then, the open / close eye determination unit 22 determines whether the eye shape is upward, downward (concave) or horizontal based on the determination result (step S52).

  Next, the open / close eye determination unit 22 determines whether or not the eye open / close learning has been completed (step S53), and if not, the eye open / close learning is performed (step S54). On the other hand, when the eye opening / closing learning is completed, the opening / closing eye determination unit 22b sets the eye height and the detected eye height width obtained from the eye learning result and the detected current eye height. Based on this, it is determined that the detected current eye height is equal to or smaller than the eye height threshold, and the eye is equal to or greater than the height threshold (step S55). The open / close eye determination unit 22 ends the series of processes after the process of step S54 or step S55.

  The eye opening / closing learning process in step S54 is performed by the opening / closing eye learning unit 22a as follows. That is, the opening degree value when the eye shape is upwardly convex from the vertical width of the eye and the eye shape obtained by the series of processing shown in the flowchart of FIG. The counter memory array prepared for each value is counted, and the total value of the left and right counter memories for each opening value is equal to or greater than a preset setting value (in this embodiment, “20”). The opening value is learned as an eye opening value. Further, a predetermined ratio (in this embodiment, “30%”) set in advance from the opening degree value of the eye opening is set as a threshold value for distinguishing between opening and closing.

“Near eye area metering”
Next, processing in the near eye region photometry unit 24 will be described.

The near-eye region photometry unit 24 calculates the value of the light amount d by the following equation (1) based on the eye coordinates (eye detection position) obtained by the processing in the eye coordinate output unit 21.

  In the above formula (1), as shown in FIG. 16A, the width of the photometric range set around the right eye and the left eye is w, and the height of the photometric range is h. As shown in FIG. 16B, this photometric range is set such that the entire pixel value is clearly different between the normal time and the time when direct light hits the eyes. Then, by performing the calculation according to the above formula (1), the near-eye region photometry unit 24 has a pixel value p (x corresponding to the light amount of each pixel (x, y) constituting the pixel group existing in the photometry range. , Y) to obtain an average pixel value and multiply the average pixel value by the reciprocal of the shutter speed s at the time of imaging obtained by the imaging state acquisition unit 23, thereby near the eye in the current image frame The amount of light d in the area is calculated.

  The near-eye region photometry unit 24 calculates the light amount d obtained by the above equation (1) in time series for a plurality of image frames, obtains the average of the plurality of light amounts d, and obtains the final light amount.

  Next, processing in the direct light detection unit 25 will be described with reference to FIG.

  The direct light detection unit 25 has the current eye vertical width obtained by the open / close eye determination process by the open / close eye determination unit 22 and the right eye / left eye obtained by the near eye region photometry process by the near eye region photometry unit 24. The amount of light around each eye is acquired, and each change is monitored (monitored) in time series. Further, the direct light detection unit 25 increases the amount of light of both eyes above a preset value, and when the vertical width of the eye (opening opening value) falls below a threshold value set in advance, It is determined that the eyes are exposed to direct light.

  At this time, the direct light detection unit 25 shows an eye vertical width threshold value (FIG. 17A) for determining whether the eyes are open or closed in a normal state where the direct light does not hit the eyes, as shown in FIG. In addition, it is corrected downward by the eye opening degree value when it is determined that direct light is in the eye. That is, the opening degree value of the eye opening when the direct light hits the eye (FIG. 17B) is lower than the opening degree value of the eye opening when the direct light does not hit the eye (FIG. 17A). The eye vertical width threshold is set lower by the opening degree value that is lower. As a result, it is possible to prevent erroneous determination that the eye is narrowed due to factors other than glare caused by direct light, such as a state where the driver laughs or sleeps, for example, as being closed.

  Next, processing in the direct light signal output unit 26 will be described.

  The direct light signal output unit 26 is connected to the direct light detection device 12 when it is determined that the direct light is applied to the driver's eyes based on the result of the direct light detection process by the direct light detection unit 25. Outputs a direct light signal to the device. As this external device, in order to reduce the driver's glare and fatigue, the notification means for prompting the driver to use a sun visor, sunglasses, etc., the automatic that blocks the light entering the periphery of the driver's eyes A light shielding device can be used.

  As this notification means, a visual notification method by displaying characters and image icons in a navigation screen or meter panel of a vehicle navigation system mounted on a vehicle, a voice message or notification sound from a speaker, etc. And the like that employ an auditory notification method by voice output. In addition, as an example of an automatic light shielding device, a sun visor mechanism that is stored on the ceiling surface of the vehicle interior when not in use and slides on the front window surface to shield light when in use, or a liquid crystal shutter installed on the front window surface And a light shielding mechanism.

[Effect of the first embodiment]
As described above in detail, according to the direct light detection system according to the first embodiment, the face image acquisition unit 20 that acquires the face image of the driver included in the video signal output from the video camera 10, and the face Based on the image, the eye coordinate indicating the driver's eye position is detected, the eye coordinate output unit 21 that tracks the eye coordinate, and the open / closed state of the driver's eye based on the face image and the eye coordinate. Based on learning and determination, an open / close eye determination unit 22 that acquires the eye opening of the driver, an imaging state acquisition unit 23 that acquires an imaging condition of the video camera 10, a face image, eye coordinates, and an imaging condition. A near-eye region metering unit 24 that measures the amount of light in the region near the driver's eyes, and a direct light detection unit that determines whether or not the driver's eyes are exposed to direct light based on the eye opening and the light amount 25 and direct light indicating the determination result by the direct light detection device 12 A direct light detection device including a direct light signal output unit 26 that outputs a signal can be realized, and the position of the driver's eye can be automatically detected, and the light amount and the eye in the vicinity of the eye can be detected. Based on the opening degree, it can be reliably determined whether or not the driver feels dazzle by direct light.

  In other words, in the existing technology, in the method of estimating the arousal level from the determination result of the eye open / closed state of the driver, although accuracy of the eye open / close determination is important, the eye is narrowed and the accuracy of the open / closed eye determination is reduced. Many drivers do not use a sun visor if they are dazzling enough to narrow their eyes. On the other hand, considering the driving psychology in an environment exposed to direct light for the driver, the driver's adaptation is considered when the driver's fatigue accumulates or sudden changes in ambient illuminance occur such as building shadows or entering tunnels. Then, it is desirable to use a sun visor, and moreover, since it is easy to erroneously detect that the eyes are closed in order to determine the open / closed state of the eyes, it is desirable that the driver use the sun visor. On the other hand, according to the direct light detection system according to the first embodiment described above, the eye opening degree value based on the determination result of the open / closed state of the eye is obtained by learning, and the amount of light hitting the eye is estimated to obtain direct light. It is possible to detect whether or not the driver is in contact with the eyes, and whether or not the driver feels dazzled by direct light.

-Notify (signal output) the detection result of direct light.

・ Or, further, perform automatic shading control (auto sun visor operation).

  Further, the eye coordinate output unit 21 detects the eye position from the entire face image by the eye position detection unit 21a, and when the eye position can be detected, the eye position tracking unit 21b performs the eye position detection in the previous image frame. From the position, the position of the eye in the current image frame can be tracked. Therefore, according to the direct light detection device 12, it is not necessary to process the entire face image for each of the image frames, and it is possible to increase the speed of the detection process and reduce the calculation capability required for image analysis.

  Further, the open / close eye determination unit 22 acquires and learns the eye opening based on the face image and the eye coordinates, sets the eye vertical width threshold based on the eye coordinates and the eye opening value, and sets the vertical width. By determining whether the driver's eyes are open or closed based on the threshold value, it is possible to automatically learn the shape of the driver's eyes and make an open / close determination, and to perform calibration such as setting individual differences among drivers in advance. Work can be made unnecessary. At this time, the open / close eye determination unit 22 may correct the vertical width threshold according to the eye opening degree when the direct light detection device 12 determines that the driver's eyes are receiving direct light. Good. Thereby, in the period when the driver's eyes are exposed to direct light and the driver feels dazzling, it is possible to optimally control the threshold of the eye vertical width for opening / closing determination.

  Furthermore, the near-eye region photometry unit 24 captures the video acquired by the imaging state acquisition unit 23 with respect to the average pixel value in the region set based on the eye coordinates for each of the right and left eyes of the driver. The light quantity may be calculated by a value obtained by multiplying the reciprocal of the shutter speed of the camera 10 by an average value for a predetermined period set in advance. As a result, not only the pixel value of the face image but also the camera parameters of the video camera 10 are taken into consideration in the light amount calculation, so that the light amount can be accurately calculated even when a large difference in light amount occurs.

[Second Embodiment]
Next, a second embodiment of the direct light detection device 12 will be described.

  As illustrated in FIG. 18, the direct light detection device 12 according to the second embodiment is different from the direct light detection device 12 according to the first embodiment in that a vehicle environment detection unit 50 is added. Therefore, in the following description regarding the second embodiment, the same or equivalent parts as those of the above-described first embodiment are denoted by the same reference numerals in the drawings, and detailed description thereof is omitted.

  The vehicle environment detection unit 50 acquires a traveling state and a date and time (current time) including the current position and traveling direction of the vehicle driven by the driver. In addition, although the apparatus which the vehicle environment detection part 50 makes acquisition target of each information is arbitrary, what is necessary is just to acquire required information from the apparatus in connection with the vehicle navigation system mounted in the vehicle, for example.

  In the second embodiment, in addition to the processing described in the first embodiment, the direct light detection unit 25 determines whether the vehicle is based on the traveling state of the vehicle detected by the vehicle environment detection unit 50 and the current time. A vehicle direction detection process is performed to determine whether or not the vehicle is facing the sun. The vehicle direction detection process by the direct light detection unit 25 can be realized by performing a calculation process described below.

That is, when the direct light detection unit 25 calculates the solar azimuth ψ and altitude α at the current time at a point of latitude φ and longitude λ where the vehicle is currently located, first, the first day (January) [Theta] o determined based on the number of continuous days dn from 1st).

Next, the solar declination δ and the equation of time Eq at the current time are obtained by the following equations (3) and (4).

Next, when the vehicle is located in Japan, the solar hour angle h is obtained from the Japan standard time JST by the following equation (5).

Next, based on the solar declination δ, latitude φ, and hour angle h calculated above, the solar azimuth ψ and altitude α are obtained by the following equations (6) and (7).

  The direct light detection unit 25 determines whether or not the traveling direction of the vehicle is facing the sun based on the solar azimuth ψ at the current time and the altitude α of the sun obtained by the above calculation process, and faces the sun. Only when it is determined that there is a direct light detection process. Thus, it is possible to perform direct light detection processing considering whether the direct light from the sun actually hits the driver's face from the positional relationship between the vehicle and the sun, for example, the driver laughed Reduces the possibility of misjudging a state in which the eyes are narrowed due to factors other than glare caused by direct light, such as a state or a drowsy state, as closed eyes due to direct light hitting the eyes .

[Effects of Second Embodiment]
As described above, the direct light detection system according to the second embodiment further includes the vehicle environment detection unit 50 that acquires the travel state and date / time including the current position and traveling direction of the vehicle driven by the driver, Based on the information acquired by the vehicle environment detection unit 50, the light detection device 12 is directly irradiated to the driver's eyes only when it is determined that the vehicle is traveling in the direction in which the sun is currently positioned. It can be determined whether or not. Thereby, the amount of calculation required for image analysis can be reduced, and erroneous detection due to direct light detection processing can be reduced.

[Third Embodiment]
Next, a third embodiment of the direct light detection device 12 will be described.

  As shown in FIG. 19, the direct light detection device 12 according to the third embodiment has a nose coordinate output unit 60, a nose vicinity region photometry unit 61, and a direct light detection device 12 according to the first embodiment. The difference is that a nose direct light detector 62 is added. Therefore, in the following description of the third embodiment, the same or equivalent parts as those in the first embodiment described above are denoted by the same reference numerals in the drawings and detailed description thereof is omitted.

  The nose coordinate output unit 60 obtains nose coordinates indicating the driver's nose position based on the face image acquired by the face image acquisition unit 20 and the eye coordinates acquired by the eye coordinate output unit 21. The nose vicinity region photometry unit 61 sets a nose vicinity region based on the face image acquired by the face image acquisition unit 20 and the nose coordinates acquired by the nose coordinate output unit 60, and the light amount in the nose vicinity region is determined. calculate. The nasal direct light detection unit 62 determines whether or not the driver has received direct light on the periphery of the nose based on the amount of light acquired by the nose vicinity region photometry unit 61.

  Further, the direct light signal output unit 26 in the third embodiment includes the eye coordinates obtained by the eye coordinate output unit 21, the direct light detection result by the direct light detection unit 25, and the nose coordinates obtained by the nose coordinate output unit 60. Then, based on the nose direct light detection result by the nose direct light detection unit 62, it is determined whether or not the direct light is applied to the peripheral portion of the driver's eyes, and when the direct light is applied, the direct light signal is output. .

  Next, the nose coordinate output processing by the nose coordinate output unit 60 will be described with reference to the flowchart shown in FIG.

  When the nose coordinate output process is started, the nose coordinate output unit 60 sets a nose search area for the face image acquired by the face image acquisition unit 20 based on the eye coordinates obtained by the eye coordinate output unit 21. (Step S60).

  Next, the nose coordinate output unit 60 sets a plurality of vertical lines in the nose search area, and performs binarization processing for each vertical line (step S61). At this time, as a threshold value in the binarization process, a pixel value that internally divides the maximum value and the minimum value of the pixel value into 2: 1 for each vertical line is used.

  Next, the nose coordinate output unit 60 performs a labeling process for setting the feature points of the nose as island-like labels (regions) for the image obtained as a result of the binarization process (step S62). As shown in FIG. 20, the nose feature points include a part where the left and right nostrils are located, a part where the left and right nostril shells start to bulge, and the like. ) To perform the labeling process.

  After the labeling process in step S62, the nose coordinate output unit 60 uses the label width / height ratio close to 1: 1 and the pixel value in the label as a reference to be within the set value. A possible label is selected, and a combination of two labels existing side by side is selected (step S63).

  Next, the nose coordinate output unit 60 determines whether or not there is a candidate label that can be determined as a nostril in the selection process in step S63 (step S64), and when only one candidate label combination exists, the nostril is detected. Therefore, the coordinates of the intermediate position of the pair of nostrils are determined as the nose coordinates (step S65). In addition, when the nose coordinate output unit 60 determines that there is no combination of candidate labels or that two or more candidate labels exist, it is assumed that the nostril has not been detected (step S66). And the nose coordinate output part 60 will complete | finish a nose coordinate output process, if the process in step S65 or step S66 is completed.

  Next, the nose vicinity area photometry processing by the nose vicinity area photometry unit 61 will be described with reference to FIG.

The nose vicinity region photometry unit 61 calculates the value of the light amount d by the following equation (8) based on the face image obtained by the face image acquisition unit 20 and the nose coordinates obtained by the nose coordinate output unit 60. To do.

  In the above equation (8), as shown in FIG. 21 (a), the width of the photometric range set in the region including the driver's nose is w and the height is h. As shown in FIG. 21 (b), this photometric range has a clear overall pixel value in normal times and when direct light hits the nose, as with the right and left eyes in FIG. 16 (b) described above. A different range is set. And by calculating according to said Formula (8), the nose vicinity area | region photometry part 61 is pixel value p (x corresponding to the light quantity of each pixel (x, y) which comprises the pixel group which exists in a photometry range. , Y) is used to obtain an average pixel value, and the average pixel value is multiplied by the reciprocal of the shutter speed s at the time of imaging obtained by the imaging state acquisition unit 23 to obtain the vicinity of the nose in the current image frame The amount of light d in the area is calculated.

  The nose vicinity region photometry unit 61 calculates the light amount d obtained by the above equation (8) in a time series for a plurality of image frames, obtains the average of the plurality of light amounts d, and obtains the final light amount.

  The nasal direct light detection unit 62 acquires the amount of light around the nose obtained by the nose vicinity region photometry unit 61 and monitors (monitors) the change in the amount of light in time series. Moreover, the nose direct light detection part 62 determines with direct light having hit the driver | operator's nose, when the light quantity around a nose rises more than the preset value.

  Here, FIG. 22 shows time-series changes in the light amount and the eye opening in the vicinity of the right eye, the left eye, and the nose obtained by the direct light detection device 12 according to the third embodiment. Here, FIG. 23A shows an example of a face image in a normal state where no direct light hits the driver's eyes, and FIG. 23A shows an example of a face image in a state where only one eye of the driver hits the driver's eyes. 23B shows an example of a face image in a state where the driver's eyes are exposed to direct light. FIG. 23C shows an example of a face image in a state where the driver uses a sun visor. Each is shown in (d).

  As can be seen from FIG. 22, from the normal state where the driver's eyes are not exposed to direct light (FIG. 23 (a)), only the driver's eye is exposed to direct light (FIG. 23 (b)). As the driver changes to a state where the driver's eyes are exposed to direct light (FIG. 23 (c)), the right eye, left eye, and nose increase in light, while the eye opening decreases. ing. In addition, when a sun visor, which is an existing light-shielding device mounted on a vehicle, is used (FIG. 23 (d)), it can be seen that the right and left eye light amounts decrease and the eye opening increases.

  In addition, when the amount of light increases only for one eye of the driver, the eye opening degree tends to decrease, but the eyes are not narrowed as compared with the normal state. It can be seen that the eye opening is decreasing. This is considered to indicate that in a state where only one eye is exposed to direct light, the driver does not feel much dazzling and has little influence on driving.

  Next, an example of mounting the direct light detection device 12 according to the third embodiment on a vehicle will be described with reference to FIG. The usage example shown in FIG. 24 is obtained by adding a light shielding device 70 and a light shielding control device 71 for controlling the operation of the light shielding device 70 to the direct light detection system shown in FIG. Therefore, among the units shown in FIG. 24, the same or equivalent components as those shown in FIG.

  As shown in FIG. 24, the light shielding control device 71 controls the operation of the light shielding device 70 based on the direct light signal output from the direct light detection device 12, and direct light that strikes around the eyes of the driver is transmitted by the light shielding device 70. It has a function of blocking light.

  Here, FIG. 25 shows an example of temporal changes in the light amount near the driver's eyes, the light amount near the nose, and the signal level of the direct light signal output from the direct light detection device 12. As shown in FIG. 25, when the light intensity near the eyes and near the nose rises above the threshold and it is determined that the direct light hits the driver's eyes, the direct light detection device 12 outputs a direct light signal, In response to this, the light shielding control device 71 outputs a control signal to the light shielding device 70. As a result, the light shielding device 70 operates to shield the surroundings of the driver's eyes.

  As the light shielding device 70, there is a mechanism that is stored in a vehicle interior ceiling (roof) portion in a normal state and the light shielding portion descends when the light is shielded, or a mechanism that shields an arbitrary portion by a liquid crystal shutter pasted in the front window. It is conceivable to provide light shielding by providing.

  When a mechanism of a type in which the light shielding part descends is used as the light shielding device 70, as shown in the lower part of FIG. 25, when it is determined that the direct light is hitting the driver's eyes, the light shielding part starts to descend. It is desirable to perform control by the light shielding control device 71 so as to lower the light shielding portion to a position where the light amount near the nose is lowered. Thereby, a driver | operator's eyes peripheral part can be light-shielded reliably.

  Further, the light shielding control device 71 may operate the light shielding device 70 so as to hold the light shielding portion at a slightly raised position when the light amount near the nose falls below the threshold value. As a result, a portion (nose peripheral portion) that does not require light shielding is not shielded by the light shielding device 70, and only the vicinity of the eyes can be reliably shielded from light.

  Thus, FIG. 26 shows a state transition diagram of the light shielding device 70 when the light shielding device 70 is controlled in the present embodiment. As the state of the light shielding device 70, “no light shielding” that does not require light shielding, “light shielding increase” in which the light shielding part is increased by the lowering control of the light shielding part, and light shielding part is decreased by the upward control of the light shielding part. There are five states: “reduce light shielding”, “optimal light shielding” indicating an optimal light shielding state, and “non-light shielding” indicating a state where light shielding is impossible. The state transition of the light shielding device 70 includes whether the amount of light in the eye part is appropriate (positive) or bright (bright), whether the amount of light in the nose part is appropriate (positive) or bright (bright), and The three parameters of whether the opening value is normal (positive) or decreased (decreasing) are judged by time series data as shown in FIG. This is done by controlling the device 70.

  In this embodiment, the direct detection of the direct light is detected only when the sun is in the vehicle front direction from the direction of the sun and the altitude and the traveling direction of the vehicle by further combining the vehicle environment detecting unit 50 described in the second embodiment. Processing may be performed.

[Effect of the third embodiment]
As described above, according to the direct light detection system according to the third embodiment, the nose coordinate output unit 60 (the nose coordinate acquisition unit) detects the nose coordinate indicating the driver's nose position based on the face image and the eye coordinates. ), A nose vicinity area photometry unit 61 (nose vicinity area photometry means) that measures the amount of light in the vicinity of the driver's nose based on the face image and the nose coordinates, and the driver based on the amount of light in the vicinity of the nose The nose direct light detection unit 62 for determining whether or not the direct nose is hitting the nose of the nose, and the direct light detection device 12 is whether the direct light hits the eye coordinates, the nose coordinates, and the driver's eyes. Based on the determination result of whether or not and the determination result by the nose direct light detection unit 62, it can be determined whether or not the driver's eyes are exposed to direct light. As a result, when the light shielding control is performed, it can be determined whether or not the periphery of the eye can be reliably shielded, and the timing at which the light shielding control is no longer necessary can be obtained from the amount of light around the nose.

  The above-described embodiment is an example of the present invention. For this reason, the present invention is not limited to the above-described embodiment, and various modifications can be made depending on the design and the like as long as the technical idea according to the present invention is not deviated from this embodiment. Of course, it is possible to change.

1 is a schematic block diagram of a direct light detection system including a direct light detection apparatus according to a first embodiment to which the present invention is applied. In the direct light detection system of 1st Embodiment of this invention, it is a schematic diagram which shows a mode that a driver | operator's face is exposed to direct light, (a) is a figure which shows the normal state where direct light is not hit, ( (b) is a figure which shows the state which has received the direct light, (c) is a figure which shows the state which used the sun visor. It is a functional block diagram which shows the structural example of the direct light detection apparatus which concerns on 1st Embodiment to which this invention is applied. It is a functional block diagram which shows another structural example of the direct light detection apparatus which concerns on 1st Embodiment to which this invention is applied. It is a flowchart which shows an example of the eye position detection process by an eye position detection part. It is a flowchart which shows an example of a single eye candidate detection process. It is a schematic diagram which shows an example of the process area set by one eye candidate detection process. It is a figure for demonstrating acquisition of the continuous data used as the candidate of an eye coordinate. It is a schematic diagram for demonstrating a single eye candidate detection process, (a) is a figure which shows the vertical line set in a process area, (b) is the state by which the density change of the light quantity was detected in the vertical line (C) is a figure which shows the state from which the "eye" contained in the face image was detected. It is a flowchart which shows an example of a binocular confirmation process. It is a schematic diagram which shows an example of the process area set by a binocular confirmation process. It is a flowchart which shows an example of an eye position tracking process. It is a schematic diagram for explaining eye position tracking processing, (a) is a schematic diagram for explaining detection of a reference eye, (b) is a schematic diagram for explaining detection of an opposite eye, (C) is a schematic diagram for demonstrating the detection of a reference | standard eye. It is a flowchart which shows an example of the process by an opening-and-closing eye determination part. It is a schematic diagram for demonstrating the learning process by an opening-and-closing eye determination part. It is a schematic diagram for demonstrating the process by an eye coordinate output part. It is a schematic diagram for demonstrating the process by a direct light detection part, (a) is a schematic diagram for demonstrating determining whether the direct light is hit by the opening degree value of eyes, (b) FIG. 5 is a schematic diagram for explaining correction of a threshold value of the vertical width of an eye according to an eye opening degree value. It is a functional block diagram which shows the example of 1 structure of the direct light detection apparatus which concerns on 2nd Embodiment to which this invention is applied. It is a functional block diagram which shows the example of 1 structure of the direct light detection apparatus which concerns on 3rd Embodiment to which this invention is applied. It is a flowchart which shows an example of the nose coordinate output process in the direct light detection apparatus which concerns on 3rd Embodiment to which this invention is applied. It is a schematic diagram for demonstrating the nose vicinity area | region photometry process in the direct light detection apparatus which concerns on 3rd Embodiment to which this invention is applied, (a) is a schematic diagram for demonstrating detecting a nose coordinate. (B) is a schematic diagram which shows the specific example of a photometry range. It is a graph which shows an example of the time-sequential change of the light quantity and eye opening degree in the direct light detection apparatus which concerns on 3rd Embodiment to which this invention is applied. In the direct light detection device according to the third embodiment to which the present invention is applied, it is a schematic diagram showing a state where the driver is exposed to direct light, and (a) is a normal state where the driver's eyes are not exposed to direct light. It is a schematic diagram which shows an example, (b) is a schematic diagram which shows an example of the state in which direct light is shining on only one eye of the driver, and (c) is that the direct light is shining on both eyes of the driver. It is a schematic diagram which shows an example of the state which exists, (d) is a schematic diagram which shows an example of the state which the driver is using the sun visor. It is a schematic diagram which shows an example which mounted in the vehicle the direct light detection apparatus which concerns on 3rd Embodiment to which this invention is applied. It is a graph which shows an example of the time change of the signal level of the direct light signal output in the direct light detection apparatus which concerns on 3rd Embodiment to which this invention is applied. It is a state transition diagram which shows the state transition of the light-shielding apparatus connected to the direct light detection apparatus which concerns on 3rd Embodiment to which this invention is applied.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Video camera 11 Illuminating device 12 Direct light detection apparatus 20 Face image acquisition part 21 Eye coordinate output part 21a Eye position detection part 21b Eye position tracking part 22 Opening / closing eye determination part 22a Opening / closing eye learning part 22b Opening / closing eye determination part 23 Imaging state acquisition Unit 24 near-eye region photometry unit 25 direct light detection unit 26 direct light signal output unit 50 vehicle environment detection unit 60 nose coordinate output unit 61 nose vicinity region photometry unit 62 nose direct light detection unit 70 light shielding device 71 light shielding control device

Claims (8)

Face image acquisition means for acquiring a driver's face image included in the video signal output from the imaging means;
Eye coordinate acquisition means for detecting eye coordinates indicating the driver's eye position based on the face image and tracking the eye coordinates;
Opening / closing eye determination means for calculating an eye opening of the driver based on the face image and the eye coordinates, and determining an opening / closing state of the driver's eye from the eye opening;
Imaging condition acquisition means for acquiring imaging conditions of the imaging means;
Based on the eye coordinates and the imaging conditions, a near-eye region photometry unit that measures the amount of light in the driver's eye near region of the face image;
A direct light detection device comprising: direct light detection means for determining whether or not direct light is shining on the driver's eyes based on the eye opening and the light amount. The eye coordinate acquisition means includes
Eye coordinate detection means for detecting the eye coordinates based on the face image;
The eye coordinate tracking means for tracking the eye coordinates by comparing the eye coordinates detected in each of the plurality of face images moving back and forth in time with each other. The direct light detection apparatus as described. The opening / closing eye determination means includes
Eye opening learning means for obtaining an eye opening value that is a learning result of the eye opening when the eye is in an open state based on a plurality of eye openings and an eye shape composed of a plurality of eye coordinates;
Opening / closing eye determination means for setting an eye vertical width threshold value based on the eye coordinates and the eye opening degree value, and performing opening / closing determination of the driver's eye based on the vertical width threshold value. The direct light detection device according to claim 1, characterized in that:   The near-eye area photometric means is the imaging means acquired by the imaging condition acquisition means at an average pixel value in an area set based on the eye coordinates for each of the driver's right eye and left eye. The direct light detection device according to claim 1, wherein a value obtained by multiplying a reciprocal of the shutter speed is calculated, and an average value of the calculated value for a predetermined period set in advance is calculated as the light amount.   The open / close eye determination means corrects the vertical width threshold according to an eye opening degree value when the direct light detection means determines that the driver's eyes are exposed to direct light. The direct light detection device according to claim 3. Vehicle environment acquisition means for acquiring the driving state and date and time including the current position and traveling direction of the vehicle driven by the driver;
The direct light detection means directly hits the driver's eyes when it is determined based on the information acquired by the vehicle environment acquisition means that the vehicle is currently traveling in the direction in which the sun is positioned. The direct light detection device according to claim 1, wherein it is determined whether or not the light is hit. Nose coordinate acquisition means for detecting nose coordinates indicating the driver's nose position based on the face image and the eye coordinates;
Based on the face image and the nose coordinates, a nose vicinity region photometry means for measuring the amount of light in the driver's nose vicinity region of the face image;
Further comprising nose direct light detection means for determining whether the driver's nose is exposed to direct light based on the amount of light in the vicinity of the nose,
The direct light detection means is based on the eye coordinates, the nose coordinates, the determination result of whether or not the driver's eyes are exposed to direct light, and the determination result by the nose direct light detection means, The direct light detection device according to claim 1, wherein it is determined whether or not the driver's eyes are exposed to direct light. Acquiring the driver's face image included in the video signal output from the imaging means, and detecting eye coordinates indicating the driver's eye position based on the face image, and tracking the eye coordinates The eye opening of the driver is calculated based on the eye coordinates and the face image, and the opening / closing state of the driver's eyes is determined from the eye opening.
Based on the eye coordinates and the imaging conditions of the imaging means, photometrically measure the amount of light in the vicinity of the driver's eyes in the face image, and based on the amount of light and the eye opening, the driver's eyes A direct light detection device, wherein it is determined whether or not direct light hits the light.