JP2007295946A - Detection system for alcohol influence and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a detection system for alcohol influence and a detection method for alcohol influence capable of taking measures to avoid the risk of an accident according to the degree of alcohol influence on a driver by estimating the degree of alcohol influence on the driver in a contactless state in a vehicle while the driver is driving as well. <P>SOLUTION: The detection system for alcohol influence comprises an image photographing part 2 for photographing the image of the driver of the vehicle; a data accumulation part 10 for storing the normal-time image of the driver of the vehicle in normal time when the driver is not under the influence of alcohol, or the biological data of the driver in the normal time; and data processing/analyzing part 8 for detecting the degree of alcohol influence on the driver of the vehicle by comparing/collating the image photographed by the image photographing part 2 and the biological data obtained by analyzing the image with the normal-time image or biological data in the normal time stored in the data accumulating part 10. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a drunk detection system and a drunk detection method, and more particularly, to a drunk detection system and a drunk detection method for detecting the degree of drunkness of a vehicle driver.

  In recent years, there have been an increase in accidents and the like resulting from drunk driving by drivers of vehicles such as automobiles. Therefore, for example, before the vehicle is driven, the administrative side also checks the driver's alcohol concentration and interviews the driver with a third party to confirm that the driver is not drunk (not drinking). It provides guidance and countermeasures to prevent accidents caused by drunk driving, such as requesting the transportation side to perform such activities. However, it is not possible to detect when the driver is drunk after starting the driving operation only by checking whether the driver is not drunk before driving the vehicle. Accidents caused by accidents cannot be sufficiently prevented.

Therefore, for example, a driver inspection system that measures the alcohol concentration in the breath of the driver, determines that driving is impossible when the alcohol concentration exceeds a reference value, and reports to the administrator server has been proposed (for example, a patent) Reference 1).
In addition, by taking a person's iris image and performing personal authentication, it is possible to prevent other people from impersonating the subject even if they are not facing the examiner, and to measure alcohol concentration and blood pressure with the sensor, A physical condition management device and a data management method thereof that are managed in association with corresponding personal information have been proposed (for example, see Patent Document 2).
In addition, the alcohol concentration is measured in the vehicle while the vehicle is stopped, and the measurement result is reported by e-mail together with the image of the measurer. When measuring the alcohol concentration, the oxygen saturation and blood pressure are measured as health management data, There has been proposed an alcohol detection system that broadcasts this measurement result to a mail reporting the measurement result of the alcohol concentration (see, for example, Patent Document 3).
JP 2005-296252 A JP-A-2005-245958 JP-A-2005-118177

However, according to the technique disclosed in Patent Document 1, the driver opens the window, and so on, so that the measurement result does not exceed the reference value even if the driver is actually drunk. There is a possibility of avoiding the judgment. There is also a problem that it cannot be distinguished from the exhalation of a passenger.
In addition, according to the method disclosed in Patent Document 2, since measurement is not performed during driving or getting on, it is not possible to detect a drunk state during driving.
Further, according to the technique disclosed in Patent Document 3, there is a problem that measurement can be performed only while the vehicle is stopped, and since the driver must operate the measuring device himself, the alcohol concentration, oxygen saturation, blood pressure, etc. Measurement is troublesome, and the driver can refuse measurement. In addition, since these sensors must be mounted on the body surface, there has been a problem of invasiveness to the driver.

  Accordingly, the present invention has been made to solve the above-described problems, and estimates the driver's alcohol level in a non-contact manner in a vehicle including during driving, and the risk is determined according to the level of alcohol level. It is an object of the present invention to provide a dampness detection system and a drunkenness detection method capable of taking measures to avoid.

In order to solve the above-mentioned problem, the drunk detection system according to claim 1, an image capturing unit that captures an image of a vehicle driver,
A normal data holding means for holding a normal image or normal biological data of a vehicle driver in a normal state that is not drunk;
Comparing and contrasting the image captured by the image capturing unit and the biometric data obtained by analyzing the image with the normal image or the normal biometric data stored in the normal data storage unit A data processing unit for detecting the degree of drunkness of the vehicle driver,
It is characterized by having.

  In the invention according to claim 1 having such a configuration, comparing and contrasting the image captured by the image capturing unit and the biological data obtained by analyzing the image with the normal image or the normal biological data. Thus, the data processing unit detects the degree of drunkness of the vehicle driver.

  According to a second aspect of the present invention, in the drunkenness detection system according to the first aspect, the image capturing unit captures images of the head and neck of a vehicle driver. .

  Therefore, in the invention described in claim 2, by acquiring images of the head and neck of the vehicle driver and analyzing the acquired images, biometric data necessary for detecting the degree of drunkenness is obtained.

Further, the invention according to claim 3 holds an evaluation function or an evaluation table that relates the biometric data and the degree of drunkness of the vehicle driver in the drunk detection system according to claim 1 or 2. A correlation data holding means,
The data processing unit is characterized by detecting a degree of drunkenness of a vehicle driver while referring to the evaluation function or the evaluation table held in the association data holding means.

  Therefore, in the invention according to claim 3, the data processing unit detects the drunkenness of the vehicle driver while referring to the evaluation function or the evaluation table.

  According to a fourth aspect of the present invention, in the drip detection system according to any one of the first to third aspects, an instruction signal relating to vehicle operation control is output based on the dampness level of the vehicle driver. It is characterized by having a transmission part.

  Therefore, in the invention according to the fourth aspect, an instruction signal related to the vehicle operation control is output from the transmission unit based on the degree of drunkenness of the vehicle driver.

  Furthermore, the invention described in claim 5 is a drunkenness detection system according to any one of claims 1 to 4, further comprising imaging calibration means for calibrating imaging characteristics of the image capturing unit. It is characterized by having.

  Therefore, according to the fifth aspect of the present invention, calibration is performed on the photographing characteristics of the image photographing unit by the photographing calibration means.

  Furthermore, the invention described in claim 6 includes an analysis calibration means for calibrating the evaluation function or the evaluation table in the drunkenness detection system according to any one of claims 3 to 5. It is characterized by having.

  Therefore, in the invention described in claim 6, the evaluation function or the evaluation table is calibrated by the analysis calibration means.

According to the seventh aspect of the present invention, there is provided a method for detecting an alcoholic spirit, an image photographing process for photographing an image of a vehicle driver,
By comparing and contrasting the captured image and the biometric data obtained by analyzing the image with the normal image or normal biometric data of the vehicle driver in a normal state that is not drunk, the driver's drunkness Data processing step to detect the degree,
It is characterized by having.

  In the invention according to claim 7 having such a configuration, the vehicle operation is performed by comparing and contrasting the captured image and the biological data obtained by analyzing the image with the normal image or the normal biological data. Data processing is performed to detect a person's alcohol level.

  According to an eighth aspect of the present invention, in the method for detecting drunkenness according to the seventh aspect, the image capturing step captures images of the head and neck of the vehicle driver.

  Therefore, in the invention described in claim 8, by acquiring images of the head and neck of the vehicle driver and analyzing the images, biometric data necessary for detecting the degree of drunkenness is obtained.

  According to a ninth aspect of the present invention, in the method for detecting an alcoholic beverage according to the seventh or eighth aspect, the data processing step includes an evaluation function or an evaluation for relating the biological data and the alcoholic beverage driver's alcoholic beverage degree. It is characterized by detecting the degree of drunkness of the vehicle driver while referring to the table.

  Therefore, in the invention according to the ninth aspect, data processing for detecting the drunkenness of the vehicle driver is performed while referring to the evaluation function or the evaluation table.

  According to a tenth aspect of the present invention, in the method of detecting dampness according to any one of the seventh to ninth aspects, an instruction signal relating to vehicle operation control is output based on the dampness level of the vehicle driver. It is characterized by having a transmission process.

  Therefore, in the invention described in claim 10, the instruction signal relating to the operation control of the vehicle is output based on the drunkenness degree of the vehicle driver.

  The invention described in claim 11 includes the photographing calibration step of calibrating the photographing characteristics in image photographing in the method for detecting drunkenness according to any one of claims 7 to 10. It is a feature.

  Therefore, in the invention described in claim 11, calibration is performed for the photographing characteristic of the image photographing unit.

  A twelfth aspect of the invention includes an analysis calibration step of calibrating the evaluation function or the evaluation table in the drunkenness detection method according to any one of the ninth to eleventh aspects. It is characterized by that.

  Therefore, in the invention described in claim 12, calibration is performed for the evaluation function or the evaluation table.

  According to the invention described in claim 1 and claim 7, it is possible to measure the drunkenness of the vehicle driver at the time of boarding the vehicle in a non-contact manner without bothering the driver himself / herself. Play.

  In addition, since the vehicle driver's normal data is held, it is possible to appropriately detect the influence on the vehicle driver due to alcohol.

According to the invention described in claim 2 and claim 8, there is an effect that it is possible to detect the drunkenness degree of the vehicle driver without requiring the labor of the vehicle driver.
In addition, there is an effect that it is possible to evaluate the alcohol level at a site where the influence of drinking is likely to occur.

  According to the invention described in claim 3 and claim 9, it is possible to relate the biometric data obtained by analyzing the image and the alcohol level, and to detect the alcohol level based on the image. There is an effect that it becomes possible.

  According to the invention described in claims 4 and 10, there is an effect that it is possible to avoid an accident caused by driving in a drunk state.

  According to the invention described in claims 5 and 11, there is an effect that the accuracy of image analysis can be improved.

  According to the invention described in Claim 6 and Claim 12, there exists an effect that the precision of detection of a drunkness degree can be improved.

  Hereinafter, with reference to FIG. 1 to FIG. 10, an embodiment of the alcoholic beverage detection system and the alcoholic beverage detection method according to the present invention will be described. However, the scope of the invention is not limited to the illustrated examples.

  In the present embodiment, the alcohol detection system 1 is installed, for example, in front of a driver's seat in a vehicle, on a dashboard, or the like.

FIG. 1 is a functional block diagram showing a schematic configuration of a drunk detection system 1 in the present embodiment.
As shown in FIG. 1, the drunk detection system 1 includes an image capturing unit 2, an illumination unit 3, a light stimulus illumination unit 4, a user interface unit 5, a display unit 6, a memory unit 7, and a data processing / analysis unit 8. , Parameter setting / management unit 9, data storage unit 10, I / O unit 11, external communication unit 12, vehicle control unit interface unit 13, and control unit 15 that comprehensively controls these components. Has been. The alcohol detection system 1 is configured to be able to transmit and receive information to and from various external devices 16 via the external communication unit 12. In addition, a vehicle operation control unit 17 that controls various operations of the vehicle is connected to the drunk detection system 1 via a vehicle control unit interface unit 13.

FIG. 2 shows an example of the arrangement of the image capturing unit 2, the illumination unit 3, and the light stimulus illumination 4 when the alcohol detection system 1 is installed in a vehicle such as an ordinary automobile.
As shown in FIG. 2, in the liquor detection system 1 according to the present embodiment, the image capturing unit 2 is composed of cameras 21 and 22 having different capturing targets (imaging site or analysis content). Are installed at different positions in consideration of the object to be photographed. Specifically, a camera 21, which is an image capturing unit 2 for capturing the face and eyes (head) of the vehicle driver, is installed in front of the driver's seat 101 of the vehicle and around the upper end of the windshield 103. Has been. In addition, a camera 22 that is an image capturing unit 2 that captures the driver's neck and chin is provided around the lower end of the windshield 103 and around the center of the dashboard 102. In addition, an illumination unit 3 that illuminates each imaging region (detection region) of the vehicle driver who is the subject at the time of imaging is disposed at one end of the lower end of the windshield 103. Further, near the center of the upper end of the windshield 103, a light stimulation illumination unit 4 that provides light stimulation to the eyes during pupil measurement (pupil imaging) is arranged.

  Each of the cameras 21 and 22 preferably has a position adjustment mechanism that can adjust the positions of the cameras 21 and 22 according to the position of the driver's head, the position and height of the driver's seat 101, and the like. . The positions of the cameras 21 and 22 may be fixed at the time of image shooting, or shooting may be performed while the positions of the cameras 21 and 22 are changed, and movement information with the changed positions may be acquired simultaneously with the image data and used for image analysis. Good. Each of the cameras 21, 22, the illumination unit 3, and the light stimulation illumination unit 4 is housed when not photographing, and has an arrangement in which an arm or the like is automatically extended and arranged at a predetermined position during photographing. Also good. In addition, the positions and the number of the cameras 21 and 22 as the image capturing unit 2, the illumination unit 3, and the light stimulation illumination unit 4 are not limited to those illustrated here. The image capturing unit 2 may be configured by a single camera, and the driver's face to neck may be captured by the single camera.

  Hereinafter, each component constituting the alcohol detection system 1 will be described in detail.

  As described above, the image capturing unit 2 captures the camera 21 for capturing the face and eyes of the driver of the vehicle and the neck and chin of the driver and acquires the pulse of the neck (cervical pulse data). And a camera 22 that obtains an image of the above. The cameras 21 and 22 include an image sensor (not shown) such as a CCD (Charge Coupled Device) and a CMOS (Complementary Metal-Oxide Semiconductor), for example, and obtain a moving image or a still image by photographing each part of the driver as a subject. For example, a color or monochrome video camera, a CCD camera, a CMOS camera, a digital still camera, or a camera module attached to a mobile phone or the like can be used.

In particular, the camera 21 for photographing the driver's face and eyes can capture a color image and a thermal image of the driver's face. The face image may be a still image or a moving image. In addition, the camera 21 can automatically adjust the shooting position and the line-of-sight direction of the camera 21 to take a moving image of the driver's eyes.
In addition, as a camera 21 for photographing a driver's face and eyes, for example, a color camera or a driver's face as a camera for photographing an image for acquiring facial color data (facial color data). A thermosensor (thermal image camera) having sensitivity in the near-infrared to infrared region of the subject may be provided as a camera that captures an image for acquiring skin surface temperature data (face temperature data). In addition, a camera that captures an image for acquiring color data of a face and a camera that captures an image for acquiring surface temperature data of the face and the like are configured by a combination of a monochrome camera and a bandpass filter. Also good. In the case where a thermosensor (thermal image camera) is provided, the thermosensor (thermal image camera) sets parameters such as sensor sensitivity, range, gain, or measurement temperature range to a constant value during shooting. . When these values fluctuate, it is necessary to correct the values inside the thermosensor so that a captured image similar to the case of the same parameter can be output. This is indispensable for maintaining the accuracy of data conversion from pixel values to temperature data.

The camera 22 that captures images of the driver's neck and jaws to acquire carotid pulse data captures a moving image of the detected part of the subject for at least 2 seconds. In this way, by setting the shooting time to 2 seconds or more, it is possible to obtain a moving image for approximately two cycles of the pulse. The longer the shooting time, the more accurate the pulse can be detected. However, it is necessary to improve the reliability of image analysis for the movement of the subject.
The image to be taken may be either a color image or a monochrome image. For example, in the case of a color image, the R channel image is used for image analysis to be described later. In the case of a monochrome image, photographing is performed using an optical filter having a high transmittance in the red to near infrared region. When the pulsation of the carotid artery cannot be detected from the image, imaging is performed so that it can be detected by turning on the illumination unit 3 and shading the carotid artery as described later.

  The cameras 21 and 22 are preferably constituted by cameras having high sensitivity in the near infrared region and the infrared region. However, when the cameras 21 and 22 are provided with a cut filter (for example, an IR cut filter) in the near-infrared region, the image is taken by removing this. Further, the position, angle, aperture, shutter speed, or the like of each camera 21, 22 may be configured to be controllable by the control unit 15.

  The illumination unit 3 is a lamp that emits white to incandescent visible light, and illuminates the subject when the surroundings of the driver who is the subject are dark at the time of photographing. The illumination part 3 is comprised by light sources, such as LED (Light Emitting Diode), for example. In addition, the light source which comprises the illumination part 3 is not limited to what was illustrated to this. Further, the light source of the illumination unit 3 may be a point light source or a surface light source.

  An infrared light source for pupil extraction may be separately provided. Moreover, you may provide the separate illumination which illuminates a neck and a chin as the illumination part 3 separately. In the case of providing such an infrared light source or dedicated illumination, the infrared light source or dedicated illumination is configured so as to be housed in the alcoholic detection system 1, and an arm or the like is automatically extended at a predetermined position during photographing. It can be set as the structure arrange | positioned. Further, the position, angle, illumination intensity, and the like of this infrared light source and dedicated illumination may be configured to be controllable by the control unit 15.

Further, in the case of providing dedicated illumination for exclusively illuminating the neck and chin, it may be configured such that illumination light is applied to the detection part of the subject to provide a shadow during imaging. The camera 22 is used to acquire data for capturing the pulse by photographing the neck and jaw, and such a pulse (pulsation) is most prominent in the vicinity of the carotid artery. Appear. Therefore, by applying illumination light to the subject's jaw and neck periphery to create a shadow, it is possible to observe the change in the shadow state of the part of the moving image where the pulse of the neck muscle is applied.
That is, in order to detect subtle movements of the skin surface near the carotid artery, for example, a dedicated illumination for illuminating the neck and chin is applied to the subject from a direction in which shadows can be easily taken (for example, an oblique direction with respect to the front of the subject). It arrange | positions so that light can be irradiated. The angle in the direction in which the illumination light is applied can be, for example, about 30 degrees from the front direction of the subject. However, since the optimum angle changes depending on the physique of the subject and the distance relationship between the illumination unit 3 and the camera 22, the angle in the direction in which the illumination light is applied is not limited to 30 degrees.

  For example, in the case where the subject detection site is the right neck, it is possible to photograph in the most preferable state of the change in shadow when illumination light is applied obliquely from the left front. In addition, when the detection part of the subject is a dent on the side of the throat Buddha, if the illumination unit 3 is placed on the same side as the image capturing unit 2, direct illumination is applied and shadows cannot be produced. When the light is applied or the illumination light is applied obliquely from the rear right, it is possible to photograph the most preferable state of the change of the shadow. In other words, light may be applied from an oblique direction opposite to the detection part from the front direction of the subject from any direction (left side when the detection part is located on the right side of the subject center). However, for example, if it is tilted too far to the left, the entire right side will be completely shaded, so it is better to tilt it slightly to the left with respect to the front of the subject. In addition, it is desirable that the height of the light source of the illumination unit 3 is approximately the same as that of the throat Buddha.

  Alternatively, a grid or pattern image may be formed by the light source of the illuminating unit 3 and projected onto the imaging region (detection region) of the subject. Thereby, it is possible to detect a movement of a living body such as a pulse by distortion of a lattice or a pattern in a captured image.

  Further, the illumination unit 3 may be configured to move the light source position in conjunction with the motion vector of the subject extracted from the captured image. At this time, the relative positional relationship of the light source is kept constant with respect to the detection site for creating the shadow. Further, when the illumination unit 3 is a dedicated illumination that illuminates the neck and chin as described above, the arm is moved as much as the guaranteed motion vector. Moreover, when it is set as the structure which arranged LED as a light source of the illumination part 3 in the one-dimensional form or the two-dimensional form, it can respond by switching so that the position of the light source which light-emits may be shifted | deviated by it.

  Or, normal illumination for the subject and illumination for shading may be alternately illuminated. Illumination can be performed without giving a sense of incongruity to the subject if illumination is performed at such a speed that the alternate illumination (flashing) is not understood by humans (about 20 cycles / second or more). An LED is preferable for imaging while intermittently flashing, but other light sources may be used as long as the same purpose can be achieved.

  In addition, when using the same drunk detection system 1 by several persons, since the optimal illumination angle differs for every user, for example, the optimal illumination angle for every user is memorize | stored in the parameter setting and management part 9, and user interface It is desirable to switch the position of the light source of the illumination unit 3 using manual input in the unit 5 or personal authentication for authenticating the user. In this case, the optimum illumination angle for each user can be obtained by, for example, placing the user in an appropriate position in advance, pressing the part that feels the most pulsation with his / her hand, and detecting it.

The light stimulating illumination unit 4 is a lamp that applies light stimulation to the driver's eyes with visible light during pupil measurement (pupil photographing), and irradiates flash light (instant light) with stable intensity and irradiation time. It is configured with a possible flash light source (not shown).
When photographing the pupil, in order to observe the response of the pupil due to stimulation, the driver is irradiated with flash light, which is light stimulation, from the light source of the light stimulation illumination unit 4 at regular intervals. However, imaging of the pupil using flash light is performed while the vehicle is stopped. It is preferable that the light emission timing and the light emission time of the light source of the light stimulation illumination unit 4 can be set by the parameter setting / management unit 9 of the alcoholic beverage detection system 1 described later.

  Next, the user interface unit 5 is, for example, a touch panel configured integrally with the display unit 6. Note that the user interface unit 5 may have a configuration other than the touch panel, such as various operation buttons. In addition, it is desirable to provide a configuration that can communicate with a user's voice, gesture, and gesture (including sign language and other advanced communication means) by providing sound equipment such as a speaker and a microphone.

  In the present embodiment, the user operates the user interface unit 5 to determine how much alcohol is consumed and how much alcohol is in the alcohol state, and how much is allowed to be in the alcoholic state. It is possible to set a threshold value (upper limit value) as to whether or not to determine that the state is drunk. The set threshold value (upper limit value) is stored in, for example, the parameter setting / management unit 9.

The upper limit value can be determined arbitrarily by the user of the drunkenness detection system 1, but the user can freely set, for example, not accepting a value corresponding to 0.15 mg or more in 1 liter of exhalation as the upper limit value. The range of values may be limited so that a predetermined value or more cannot be set as the upper limit value. Further, when the driver and the manager who manages the driver or the system operator of the drunk detection system 1 are different from each other, only the administrator or the system operator can set the upper limit value, and the set upper limit is set. A security function can be provided so that the driver cannot arbitrarily change the value.
In addition, as an example of the upper limit value, for example, a value corresponding to 0.15 mg in 1 liter of exhaled breath, which is the lower limit of drunken driving violation in the Road Traffic Act (revised June 1, 2002), It may be a value corresponding to a blood alcohol concentration of 0.05 mg (about one large beer bottle). Further, an upper limit value may be set so that it is determined that the person is drunk when other than 0 (when alcohol is consumed even a little).

The display unit 6 can be configured by an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube) display, an organic EL (Electro Luminescence) display, a plasma display, a projection display, or the like. , Image data being processed by the data processing / analysis unit 8 or image data held by the data storage unit 10, parameters managed by the parameter setting / management unit 9, which will be described later, and drunk detection Information related to the operation status of the system 1 and each component constituting the system, image processing results, information given from the external device 16, and the like are displayed.
As described above, the display unit 6 may be configured by a touch panel, and the display unit 6 and the user interface unit 5 may be integrated with each other.

  The memory unit 7 includes, for example, a RAM (Random Access Memory), a ROM (Read Only Memory), a DIMM (Dual Inline Memory Module), a DRAM (Dynamic Random Access Memory), an SDRAM (Synchronous DRAM), and the like. The data storage / analysis unit 8 or the like temporarily transfers the necessary data from the data storage unit 10 or the like and temporarily stores it, thereby allowing the drunkenness detection system 1 to operate at high speed and stably. . Note that the memory unit 7 of the present embodiment needs to have a memory space having a capacity necessary for executing image processing and control in the alcoholic detection system 1 in real time.

  The data processing / analysis unit (data processing unit) 8 is image data of the driver's face, eyes, etc. acquired by photographing with the camera 21, and image data of the driver's neck, etc. acquired by photographing with the camera 22. Extraction and generation of biometric data such as face color data, face temperature data, pupil data, and carotid pulse data by processing and analysis, etc., and the driver's spirit level is determined based on the generated biometric data It is detected and it is determined whether or not the driver is drunk.

  First, biometric data extraction / generation processing performed by the data processing / analysis unit 8 will be described.

The data processing / analysis unit 8 extracts and generates face color data at a predetermined position (detection site) in the face from the color image of the driver's face acquired by photographing with the camera 21.
Specifically, the data processing / analysis unit 8 performs the following processing.
As will be described later, the parameter setting / management unit 9 stores patterns and position information of the respective detection parts in advance. When a color image of the driver's face is captured by the camera 21, the data processing / analysis unit 8 reads out the pattern and position information of each detection region from the parameter setting / management unit 9, and the image captured by the camera 21 Matching with the pattern and position information of each detection part is performed. When a detection site is specified for the image captured by the camera 21 in this manner, the image data of the detection site is analyzed to extract and generate face color data.
Note that a plurality of detection sites from which data is extracted are preferable. Moreover, as a detection site | part, the site | part which is easy to show the influence of blood circulation, such as a lip, a cheek, an ear | edge, a forehead, etc. is good, for example. In the present embodiment, four locations of the lips, cheeks, ears, and forehead of the face are used as detection regions, and face color data is extracted and generated by analyzing the image data of the portions.

  For analysis of an image for extracting and generating face color data, for example, a method of analyzing using a relationship table between pixel values and colors, a method of analyzing using an RGB-XYZ conversion formula based on the sRGB standard, etc. Various known methods can be used. The data processing / analysis unit 8 finally uses color space coordinate data that separates lightness and other elements (hue, saturation) such as L *, u *, v * or L *, a *, b *. Face color data is extracted and generated in a format.

  Further, the data processing / analysis unit 8 extracts thermal image data at a predetermined position (detection site) in the face from which data is extracted from the facial image from the thermal image data of the driver's face acquired by photographing with the camera 21. Face temperature data is extracted and generated by analyzing and automatically converting to temperature data. Note that the face temperature data referred to here is temperature data of the face surface, not body temperature. In the present embodiment, as in the case of face color data, four locations of the lips, cheeks, ears, and forehead are detected in the face, and the image data of the location is analyzed to extract face temperature data. It is designed to generate.

  Here, the pixel value of the thermal image is, for example, a value of 0 to 255 if it is 8 bits, and does not directly represent the temperature. However, if each parameter in the camera 21 is set to a predetermined value, the relationship between the pixel value of the thermal image and the temperature can be uniquely determined. Therefore, for example, a table in which the pixel values of the thermal image and the temperatures are associated with each other is automatically created in advance and stored in the parameter setting / management unit 9 described later, and a thermal image of the face is captured by the camera 21. Then, the data processing / analysis unit 8 reads out this table from the parameter setting / management unit 9 and refers to this table to extract and generate face temperature data by converting the pixel value of the thermal image of the face into temperature. It has become. Note that the extraction / generation method of the face temperature data is not limited to the one exemplified here, and various known methods can be used.

  Further, the data processing / analysis unit 8 analyzes time-series changes such as the diameter of the pupil from the moving image of the driver's eyes acquired by photographing with the camera 21, and sets various parameters regarding the time-series changes of the pupil. Extract and generate. Specifically, first, the data processing / analysis unit 8 extracts the pupil region from the image data of each frame of the moving image of the driver's eye acquired by the camera 21, and determines the pupil size and the pupil center position. get.

  Here, the method of extracting the eye region image and the method of extracting the image of the pupil region are, for example, stored in advance in the parameter setting / management unit 9 to be described later, and the pattern and position information of the eye region and the pupil region are photographed. By performing template matching between the eye moving image and the pattern and position information of the eye region / pupil region, the eye region image is obtained from the image data of each frame of the driver's eye moving image acquired by the camera 21, A technique such as extracting a pupil region can be used. The camera 21 captures not only a moving image of the eye but also a moving image of the face portion including the eyes, and the data processing / analysis unit 8 extracts an eye region image from the captured moving image of the face portion. A pupil region may be extracted.

  Further, in the present embodiment, for example, a template is generated by acquiring data in advance for the ratio of white eyes to black eyes, etc., and stored in the parameter setting / management unit 9. When the pupil region is extracted, the data processing / analysis unit 8 analyzes the extracted pupil region data using the template stored in the parameter setting / management unit 9, and calculates the pupil size and the pupil center position. , And various parameters (biological data) based on the time series change of the pupil size and the pupil center position are extracted. Note that the method of extracting the pupil region and the method of acquiring the pupil size and the pupil center position are not limited to those exemplified here, and various known methods can be used.

  Here, the time series change of the pupil appears, for example, as shown in FIG. FIG. 3A is a graph showing a time-series change in the diameter of the pupil when light stimulation is applied three times. When a light stimulus is applied to the eyes by causing the illumination unit 4 for light stimulation to emit light, as shown in FIG. 3A, when a light stimulus is applied (when applying a stimulus (a), when applying a stimulus (b), At the time of stimulus application (c)), the diameter of the pupil changes greatly. FIG. 3B shows in more detail the time series change of the pupil at the time of stimulus application (b) shown in FIG.

  Examples of parameters extracted and generated by the data processing / analysis unit 8 from such a time series change of the pupil include a stimulus response start diameter Pa, a stimulus response end diameter Pb, a stimulus response peak diameter Pc, and a stimulus response start to peak. Time Ta, stimulus response peak to end time Tb, stimulus response start to end time Tc, stimulus response start to peak-to-peak diameter change Sa, stimulus response end to peak-to-peak diameter change Sb, no-stimulus time average value & Dispersion value (see “without stimulation” in FIG. 3A), pupil center position (lateral direction) dispersion value (see thick line graph in FIG. 4), pupil center position (vertical direction) dispersion value (thin line in FIG. 4) For example). Note that the parameters extracted and generated from the time series change of the pupil are not limited to those exemplified here. Further, the data processing / analysis unit 8 may extract and generate all of them, or may extract and generate only a part of them.

  Further, the data processing / analysis unit 8 analyzes the change in the state of the shadow in the moving image of the driver's neck and chin (detected part) obtained by photographing with the camera 22, thereby generating carotid pulse data of the carotid artery Is extracted and generated.

That is, the data processing / analysis unit 8 of the present embodiment calculates the average pixel value of the shadow portion at the detection site for each frame of the moving image, and accumulates the average pixel value for each photographing time (elapsed time). Then, for example, by counting the number of peaks (or valleys) in one minute when the time-series change of the average pixel value is represented in the graph, the carotid pulse data of the carotid artery portion of the driver as the subject ( The pulse rate is detected.
In addition, the method of acquiring the jugular pulse data is not limited to the method exemplified here, and various known methods can be used.

  Next, the detection / determination process of the driver's alcohol level performed by the data processing / analysis unit 8 will be described with reference to FIG.

  In the present embodiment, as will be described later, the data storage unit 10 stores a normal image captured when the driver is not drunk, face color data obtained by analyzing the normal image, and face temperature. Normal biological data such as data, pupil data, and jugular pulse data are stored as normal data.

  As shown in FIG. 5, the data processing / analysis unit 8 newly captures each detection part of the driver, and when various biological data is acquired from the captured image, this is used as analysis target data. The difference data is generated by detecting the difference between the data to be analyzed and the normal data.

As a method for detecting the difference, the difference between the two (including the absolute value) is calculated ((analysis target data)-(normal data)), or the ratio between the two is calculated ((analysis target data) / (normal data). )) Is used. In the difference between the two, if positive and negative are significant, difference data is generated with values including positive and negative. For example, in the difference in face color data, the sign of the difference between L *, a *, and b * means the direction of the difference in color and the like, and therefore, difference data is generated with a value including the sign.
In the following description of the present embodiment, a case where difference data is generated by calculating a difference between analysis target data and normal data will be described.

  In addition, as will be described later, the parameter setting / management unit 9 stores an evaluation function or an evaluation table for evaluating the degree of alcoholism created in advance. When the data processing / analysis unit 8 generates the difference data between the analysis target data and the normal data, the data processing / analysis unit 8 reads the evaluation function or the evaluation table from the parameter setting / management unit 9 and uses the evaluation function or the evaluation table to read the difference data. To evaluate the driver's level of drunkenness. Further, the parameter setting / management unit 9 stores a threshold value (upper limit value) that is determined as a drunken state set by the user by operating the user interface unit 5 as described above. The analysis unit 8 determines whether or not the detected alcohol level exceeds this threshold value, and determines whether or not the driver is in an alcoholic state. Then, the data processing / analysis unit 8 outputs a determination result as to whether or not it is in an alcoholic state to the control unit 15.

  The parameter setting / management unit 9 sets parameters related to the control of each component of the alcoholic beverage detection system 1, such as imaging in the image capturing unit 2, extraction of various data in the data processing / analysis unit 8, analysis processing, etc. The set parameters are managed.

  Further, in the present embodiment, as described above, an evaluation function or an evaluation table for associating various biological data and the degree of alcohol driving of the vehicle driver and evaluating the degree of alcohol driving is created in advance, and the parameter setting / management unit 9 The parameter setting / management unit 9 functions as an association data holding unit.

The evaluation function is created as follows, for example.
That is, in the liquor detection system 1 according to the present embodiment, an image obtained when the driver is not drunk (normally) is obtained in advance by the same method as that for acquiring the analysis target data described above. By analyzing the normal image, face color data, face temperature data, pupil data, and carotid pulse data (normal biological data) are acquired. Similarly, each of the biometric data (biological data at the time of drinking) is acquired by analyzing an image at the time of drinking obtained by taking an image of the driver at the time of drinking.
In addition, biological data at the time of drinking measures the amount of alcohol consumed or the alcohol concentration in expiration, and the state of low alcohol consumption or low alcohol concentration in expiration to the state of high alcohol consumption or high alcohol concentration in expiration A plurality of data up to are acquired in association with the amount of alcohol consumed during measurement or the alcohol concentration in expiration. Then, for example, each data is analyzed in the data processing / analysis unit 8, and a difference between normal data and drinking data is detected, for example, to detect a difference between normal and drinking data of each biological data. The relationship between this difference and the amount of alcohol consumed or the alcohol concentration in the breath is statistically analyzed.
Since the degree of alcohol and the amount of alcohol consumed or the concentration of alcohol in the breath have a correlation, by analyzing the difference between the normal time of each data and the time of drinking and the relationship between the amount of alcohol consumed or the alcohol concentration in the breath It is possible to acquire an evaluation function of each biological data such as face color data, face temperature data, pupil data, and carotid pulse data, and a degree of alcohol. An evaluation table may be created from this evaluation function as necessary and stored in the parameter setting / management unit 9.

Here, it is generally said that drinking and various biological data have the following relationship, for example.
Blood circulation → get better.
Skin surface temperature → Increases Skin color → Turns red.
Pulse rate → Increase Breathing → Faster Pupil movement → Decrease, Visual field narrows, Response slows down

Among such biological data, the relationship between the difference between the normal surface temperature of the face (face temperature data) and the time of drinking (hereinafter referred to as “surface temperature difference”) and the amount of alcohol consumed or the alcohol concentration in the breath. An example is shown in FIG. In FIG. 6A, the surface temperature difference increases as the amount of alcohol consumption increases or the alcohol concentration in the expiration increases.
FIG. 6B shows an example of the relationship between the difference between the normal pulse rate of the cervical pulse (cervical pulse data) and alcohol consumption (hereinafter referred to as “pulse difference”) and the amount of alcohol consumed or the alcohol concentration in expiration. Shown in In FIG.6 (b), a pulse difference becomes large, so that the amount of alcohol consumption increases or the alcohol concentration in expiration | expired_air becomes high.
FIG. 6C shows an example of an evaluation function for evaluating the degree of alcohol. An evaluation function as shown in FIG. 6C is obtained by analyzing the relationship between the surface temperature difference and the amount of alcohol consumed or the alcohol concentration in expiration and the relationship between the pulse difference and the amount of alcohol consumed or the alcohol concentration in expiration. . By using this evaluation function, if you analyze the difference between normal and normal drinking of biological data such as surface temperature difference, pulse difference, etc., you can determine the amount of alcohol or the alcohol concentration in the breath corresponding to each analysis result, Furthermore, it is possible to detect a degree of alcohol consumption correlated with the amount of alcohol consumed or the alcohol concentration in the expiration.

The data storage unit 10 includes image data input from the outside, image data captured in the past, data that has already undergone image processing, analysis, and the like, or temporary data that is being processed and analyzed To manage and maintain. Further, in the present embodiment, the data storage unit 10 analyzes the normal image captured when the driver is determined not to be drunk as a result of the analysis by the drunkenness detection system 1 and the normal image. It functions as normal data holding means for managing and holding normal biological data (hereinafter referred to as “normal data”) such as face color data, face temperature data, pupil data, and carotid pulse data obtained by the above. The normal data can be regarded as data related to the driver's normal (normal) physical condition.
As described above, when the data processing / analysis unit 8 analyzes the biological data (analysis target data) by comparing and comparing with the normal data, if the driver is determined not to be drunk, the analysis is performed. All data including the target data when it is determined that the driver is not drunken is averaged and statistically analyzed to obtain normal data. That is, when it is determined that the analysis target data is not in a drunken state, the normal data is updated to the data including the analysis target data, and the data storage unit 10 holds the updated normal data.

  The I / O unit 11 includes a vital sensor (such as a thermometer, a weight scale, a body fat percentage meter, a sphygmomanometer, an electrocardiograph, a skin age meter, a bone densitometer, a spirometer) as a biological data acquisition means, a CF card It is configured to be able to connect devices that handle portable devices such as SD cards and USB memory cards, and various data necessary for the operation setting of the drunkenness detection system 1 may be input or output from these devices. . Note that the I / O unit 11 is not an essential component of the alcoholic detection system 1 and may be configured without the I / O unit 11.

  The external communication unit 12 is configured to perform information communication with the external device 16 by a wired or wireless communication unit. In addition, since the drunk detection system 1 of this embodiment can also handle image data, it is desirable that the communication mode be capable of high-speed transmission as much as possible.

  The external communication unit 12 is connected to an external device 16 so as to be able to communicate with each other via a network. When it is determined in the drunk detection system 1 that the driver of the vehicle is drunk, a message to that effect is sent to the control unit 15, and the control unit 15 is in a drunk state via the external communication unit 12. A signal and a message for notifying that are transmitted to the external device 16.

  The network in the present embodiment is not particularly limited as long as it means a communication network capable of data communication. For example, the Internet, a LAN (Local Area Network), a WAN (Wide Area Network), a telephone line network, ISDN ( An integrated services digital network (CIT) line, a CATV (Cable Television) line, an optical communication line, and the like can be included. Moreover, it is good also as a structure which can communicate not only by wire but by radio | wireless.

  The external device 16 is configured by, for example, a personal computer. For example, if the alcohol detection system 1 is installed in a taxi vehicle, the external device 16 is a computer installed in a taxi company to which the taxi belongs. is there. Moreover, you may comprise the external apparatus 16 by the portable terminal registered previously, such as a mobile telephone of the vehicle driver's family. In addition to or in addition to the external device 16, the data processing device (not shown) that functions as a database of these data, in addition to analyzing data such as the image data obtained by the alcoholic detection system 1 However, it is good also as a structure connected to the drunk detection system 1 via the external communication part 12. FIG.

  In this embodiment, the vehicle control connection interface unit 13 transmits an instruction signal related to vehicle operation control to the vehicle operation control unit 17 that controls the function and operation of the vehicle according to the determination result of the drunkenness detection system 1. It functions as a transmission part.

When the determination that the driver is in a drunk state is sent to the control unit 15, the control unit 15 transmits the vehicle when it is determined to be in the drunk state via the vehicle control connection interface unit 13. An instruction signal corresponding to the situation is transmitted to the vehicle operation control unit 17. For example, if it is determined that the vehicle is drunk, and the current vehicle status is stopped with the engine turned off, an instruction signal indicating that the engine cannot be turned on is connected to the vehicle control. It is transmitted from the interface unit 13 to the vehicle operation control unit 17. Further, when the current vehicle state is stopping with the engine running, an instruction signal “stop engine” is transmitted from the vehicle control connection interface unit 13 to the vehicle operation control unit 17. Further, for example, during operation, an instruction signal such as “switch to automatic steering, retreat to a safe place and stop the engine” is transmitted from the vehicle control connection interface unit 13 to the vehicle operation control unit 17.
The signal transmitted from the control unit 15 via the vehicle control connection interface unit 13 is only the determination result that the driver is in a drunk state, and the vehicle operation control unit 17 receives the determination result. It is also possible to determine what kind of control is to be performed and send an instruction corresponding to this to each part of the vehicle.

The control unit 15 includes a CPU (Central Processing Unit) that performs various arithmetic processes, a ROM (Read Only Memory) that stores various control programs such as a system program, and a RAM (Random Access Memory) that temporarily stores various data. The control unit 15 reads out various programs stored in the ROM and the like, expands them in the RAM, executes various processes in accordance with the expanded programs, and the later-described alcohol. Each component of the band detection system 1 is driven and controlled.
In particular, in the present embodiment, the ROM of the control unit 15 stores a shooting calibration control program for performing shooting calibration described later, an analysis calibration control program for performing analysis calibration, and the like in addition to the system program. The control unit 15 functions as an imaging calibration unit and an analysis calibration unit that perform imaging calibration and analysis calibration in cooperation with the imaging calibration control program and the analysis calibration control program. Yes.
In addition, since the liquor detection system 1 of the present embodiment also handles moving images, it is desirable that the control unit 15 be configured by a chip that can operate and control as fast as possible.

  Here, imaging calibration and analysis calibration performed by the control unit 15 in the present embodiment will be described.

  For example, under a usage environment in which a change in the surrounding environment is large, a change in shooting characteristics of the image shooting unit 2 may occur over a long time. The shooting calibration is for acquiring a change in shooting characteristics of the image shooting unit 2 on a daily basis, correcting an image according to the characteristics, and acquiring a stable image.

  As shown in FIG. 7, in this embodiment, for example, a color chart 104 for photographing calibration is provided on the ceiling portion 105 of the vehicle. One end of the color chart 104 is fixed to the ceiling portion 105 of the vehicle, and is rotatable from a position substantially parallel to the ceiling portion 105 to a position substantially perpendicular to the ceiling portion 105 with the fixed position as an axis. It is arranged at a position substantially parallel to the ceiling portion 105 at the time of shooting calibration, and is automatically rotated to a position substantially perpendicular to the ceiling portion 105 at the time of shooting calibration. Further, the color chart 104 is in front of the seat 106 of the driver's seat 101 when it is almost perpendicular to the ceiling portion 105 and in the vicinity of the position where the driver's face is located when the driver sits on the seat 106. It is configured to be arranged.

  The color chart 106 can be applied to either a reflection type or a self-luminous type. In the case of the reflective type, an illumination lamp is provided for illumination. In addition, you may make it illuminate the color chart 106 with the illumination part 3, without providing an illumination lamp separately. As a material for forming the color chart 106, a material such as paper that is rapidly deteriorated by heat is unsuitable. For example, a material resistant to heat such as a tile is used. In the case of the self-luminous type, LEDs of colors used for photographing calibration are arranged in a two-dimensional manner. The color chart 106 is preferably a self-luminous type from the viewpoints that illumination light is not necessary and the contents of the chart can be changed by the light emission pattern. When the self-luminous type color chart 106 is used, the characteristics of the illumination light of each LED are stored in the data storage unit 10.

  When shooting calibration is performed, for example, the image of the color chart 104 is shot by the camera 21 of the image shooting unit 2. The control unit 15 acquires gradation characteristics and color measurement parameters from an image obtained by photographing, and stores them in the parameter setting / management unit 9. Then, the control unit 15 appropriately corrects the image according to the characteristic change of the image capturing unit 2 based on the acquired gradation characteristics and color measurement parameters.

In addition, as shown in FIG. 8A, the drunk detection system 1 may include a distortion correction chart 107. In order to increase the accuracy of image analysis values such as personal authentication and pupils, it is necessary to obtain a front view image from the captured image in order to match the template and the captured image. In the distortion correction, when a captured image is not a front view image, the captured image is corrected to an image close to the front view.
In this case, as shown in FIG. 8B, the pattern of the chart 107 is changed to a lattice-shaped distortion correction pattern, and the image of the chart 107 is captured by the camera 21 of the image capturing unit 2, for example. The control unit 15 calculates a correction parameter for distortion correction from the pattern obtained by photographing, and stores it in the parameter setting / management unit 9. Then, when the captured image is not a front view image, the control unit 15 performs correction for correcting the captured image to an image close to the front view using the correction parameter.
Note that the distortion correction chart 107 may be a monochrome chart 107 as shown in FIG. In addition, for example, when a color chart 104 (see FIG. 7) used for obtaining parameters for performing correction according to the characteristic change of the image photographing unit 2 is provided with a self-luminous type, Instead of separately providing the distortion correction chart 107, the pattern of the color chart 104 (see FIG. 7) may be changed to a lattice-shaped distortion correction pattern to calculate a correction parameter for distortion correction.

  Note that the temperature in the vehicle is different between the time of shooting calibration and the time of determination of alcohol, and the shooting characteristics of the image shooting unit 2 may change accordingly. Therefore, it is preferable to acquire changes in the photographing characteristics of the image photographing unit 2 under a plurality of temperatures in advance and correct the difference in photographing characteristics of the image photographing unit 2 due to the difference in temperature in the vehicle. If there is no data of the same temperature, correction is performed using peripheral data close to that temperature.

  Note that the imaging calibration is preferably performed at night when there is no person in the vehicle. Moreover, it is preferable to carry out when the vehicle is placed in a dark place such as a parking lot. When taking an image for shooting calibration, the temperature in the vehicle at the time of shooting the image is measured, and the data is stored in the data storage unit 10 or the like.

Next, analysis calibration will be described.
The degree of alcohol is determined based on an evaluation function created in advance (or an evaluation table based on this), so if the conditions such as age and weight at the time of determination of alcohol are different from the conditions at the time of evaluation function creation, The relationship between the two changes. The analysis calibration corrects the influence of such age and weight, and creates an evaluation function (or an evaluation table based on this) that is close to the actual condition of the driver at the time of determination of drunkness.

For example, regarding the age, a mechanism is provided for experimentally examining the driver's biometric data for various levels of alcohol consumption or breath alcohol concentration at regular intervals of the driver's age and updating the evaluation function.
Specifically, for example, as shown in FIGS. 9 (a) and 9 (b), every time the driver's age reaches a certain level (for example, every five years of 30, 35, 40,...). Each alcohol consumption is caused by causing the driver to drink alcohol with various levels of alcohol consumption or breath alcohol concentration, or by applying an alcohol patch to the skin to observe the change corresponding to the prescribed alcohol consumption or breath alcohol concentration. Images of the driver's face, eyes, neck, etc. in the case of volume or breath alcohol concentration are taken, and biometric data such as face color, face temperature, and carotid pulse are extracted and generated based on this image. And based on the produced | generated biometric data, the evaluation function with respect to the drinking amount or breath alcohol concentration for every age is acquired.
The evaluation function for each age is stored in the data storage unit 10 together with the personal data and the experimental data at the time of creation, and when detecting the degree of alcohol, for each age stored in the data storage unit 10 The evaluation function is corrected according to the age of the driver (for example, 44 years in FIG. 9) at the time of detecting / determining the degree of alcohol consumption. Specifically, a weighting factor is derived from an evaluation function for the amount of alcohol consumption or breath alcohol concentration for each age, and the weighting factor is multiplied by the evaluation function to detect the degree of alcohol consumption as shown in FIG. -Create an evaluation function for age at the time of judgment.

The weight is measured by providing a weight sensor, a pressure sensor, etc. on the seat of the driver's seat, for example.
As with the age, the weight at the time of updating the evaluation function is measured and included in the personal data, and stored in the data storage unit 10 as with the age. When detecting the alcohol level, by referring to the evaluation function for the alcohol consumption or breath alcohol concentration for each body weight held in the data storage unit 10, the detection of the alcohol level is performed. The evaluation function is corrected according to the weight of the driver. Specifically, a weighting factor is derived from the evaluation function for the amount of alcohol consumed or the breath alcohol concentration for each body weight, and the weighting factor is multiplied by the evaluation function to create an evaluation function for the weight at the time of detecting the degree of alcohol consumption. .

  In addition, although not shown, a pressure sensor (a weight sensor, a pressure sensor, etc.) is provided in the seat portion or the backrest portion of the seat 106 of the driver's seat 101 of the vehicle in which the drunk detection system 1 is installed. The sensor outputs the detection result to the control unit 15. Thereby, the control part 15 can detect that the person sat in the driver's seat (boarded).

  Next, with reference to FIG. 10, a method for detecting alcohol consumption in the present embodiment will be described.

First, in the drunk detection system 1, the control unit 15 always determines from the detection result from the pressure sensor whether or not the driver gets on the driver's seat of the vehicle (step S1). If it is determined that the driver is not in the vehicle (step S1; NO), the determination in step S1 is repeated.
When it is determined that the driver has boarded the vehicle (step S1; YES), the control unit 15 operates the cameras 21 and 22 to perform color image photographing of the face portion, thermal image photographing of the face portion, and moving image of the eye portion. Image capturing, moving image capturing of the neck and jaws are performed, and each image is acquired (step S2). When the image is acquired, the data processing / analysis unit 8 analyzes each image, and extracts and generates biometric data (analysis target data) (step S3).

  When the biological data (analysis target data) is generated, the data processing / analysis unit 8 detects the difference between the analysis target data and the normal data and generates difference data (step S4). Then, the data processing / analyzing unit 8 analyzes the difference data while referring to the evaluation function or the evaluation table, and detects the alcohol level (step S5). Further, the data processing / analyzing unit 8 refers to a threshold value (upper limit value) for determining whether or not the driver is intoxicated, which is stored in the parameter setting / management unit 9, and determines whether or not the driver is intoxicated. (Step S6).

  The determination result by the data processing / analyzing unit 8 is sent to the control unit 15, and when it is determined that the state is drunken (step S 6; YES), the control unit 15 performs vehicle operation via the vehicle control interface unit 13. An instruction signal for vehicle control is output to the control unit 17 (step S7). On the other hand, when it is not determined that the person is drunken (step S6; NO), the data processing / analysis unit 8 has already been stored as normal data with the acquired analysis target data acquired this time. The normal data is averaged with the normal data to generate new normal data, and the generated normal data is stored in the data storage unit 10 (step S8). As a result, the normal data stored in the data storage unit 10 is updated to data including the analysis target data acquired this time.

  In the present embodiment, the timing of performing the above-mentioned alcoholic beverage detection process by the alcoholic beverage detection system 1 may be when the driver gets on the vehicle or after a certain period of time has elapsed after boarding. In addition, while the driver is in the vehicle, the drunk degree detection process may be repeatedly performed every certain period of time. The timing and the number of times of performing the alcohol level detection process may be set in advance, or may be arbitrarily set by the user.

  Note that the control unit 15 performs shooting calibration for the shooting characteristics of the image shooting unit 2 and analysis calibration for the evaluation function (or evaluation table) every time a fixed period or every time the degree of drunkness is detected. The timing for performing each calibration may be set in advance, or may be arbitrarily set by the user.

  As described above, according to the present embodiment, the image of the vehicle driver is taken, the biometric data obtained by analyzing this image is compared with the normal biometric data, and the biometric data and the vehicle driver are further compared. By referring to the evaluation function relating to the degree of alcohol, the degree of alcohol of the vehicle driver at the time of boarding can be detected in a non-contact manner. For this reason, it is possible to detect the alcohol level without bothering the driver himself / herself.

  Further, since the control unit 15 performs shooting calibration for the shooting characteristics of the image shooting unit, the image is corrected according to the shooting characteristics of the image shooting unit even when the shooting characteristics of the image shooting unit change over time. Thus, it is possible to stably acquire an image suitable for detecting the alcohol level.

  Further, since the control calibration is performed by the control unit 15 to correct the evaluation function or the evaluation table, even if the driver's age, weight, etc. have changed since the evaluation function or the evaluation table was created, the degree of drunkenness is adequate. Can be detected.

  Furthermore, when it is determined that the vehicle driver is in a drunken state, the vehicle operation control unit 17 that controls the functions and operations of the vehicle is informed via the vehicle control interface unit 13 and the situation is Since the instruction signal instructing to take a countermeasure according to the output is output, it is possible to prevent an accident caused by driving in a drunk state.

In the present embodiment, it is assumed that the boarding detection of whether or not the driver has boarded the driver's seat is performed by the pressure sensor provided on the seat 106 of the driver's seat 101, but the boarding detection method is limited to this. Not. For example, the driver's seat may be photographed constantly or at regular intervals by the camera 21 of the image photographing unit 2 and the photographed image may be analyzed to determine whether the driver is in the driver's seat.
Thus, when boarding detection is performed by the camera 21 of the image capturing unit 2, personal authentication of the boarded driver may be performed using head data after detection. The personal authentication methods include face authentication, iris authentication, voiceprint authentication, etc., but any method can be applied. As a result of personal authentication, when a driver is specified, driver information (for example, name, age, employee number, etc.) is attached to the captured image or biometric data obtained by analyzing the image, and a database, etc. You may make it manage with. In addition, driver information may be attached when reporting the driver's alcohol level to an external device.

Further, in the present embodiment, the face, eyes, and neck are all photographed simultaneously, but the face, eyes, and neck may be photographed separately or at two or more locations simultaneously.
Further, in the present embodiment, the order in which the analysis is performed after the imaging is performed has been described in the present embodiment, in which the images of all the detection sites are captured first, and then the image analysis of each detection site image is performed. The order is not limited to this, and the face part, the eye part, and the neck part may be separately photographed, and the image analysis may be sequentially performed in the photographed order.

  Further, in the present embodiment, the detection and determination of the degree of alcoholism are performed based on the image obtained by photographing the face, eyes, and neck and the biological data obtained by analyzing the image. The biometric data for detecting / determining the degree of alcohol is not limited to those exemplified here. For example, a biometric data sensor may be embedded in a position such as a handle, a seat, or a foot pedal that comes into contact with the living body of the vehicle driver, and the biometric data may be acquired using these sensors. Specifically, for example, a temperature sensor, a pulse sensor, and an oxygen saturation sensor may be provided on the handle. Further, a blood pressure sensor or the like may be provided on the seat belt or the seat.

  In this embodiment, the evaluation function or the evaluation table is created by experimentally acquiring the driver data. However, the data for creating the evaluation function or the evaluation table is the driver data. It is not limited. For example, an evaluation function (or an evaluation table based on this) may be created using experimental results for a plurality of general adults. In this way, when an evaluation function is created using experimental results for general adults, the driver is allowed to drink a specified amount of alcohol, the alcohol concentration in the breath is measured, and the driver is based on that value. Calibrate the evaluation function or evaluation table for the user.

  Further, in this embodiment, as biometric data extracted from an image, time series changes in the skin color (face color data) of the face of the living body, the skin surface temperature of the face (face temperature data), the size of the pupil and the center position However, other biological data may be extracted from the image.

  For example, in facial image analysis, thermal image data may be acquired as a moving image, the driver's respiratory rate may be extracted, and this may be used as biometric data for detecting / determining the degree of alcohol.

  In addition, the system according to the present embodiment is configured to determine whether or not the driving is in a drunken state, but the biometric data is obtained by capturing the driver's image and analyzing the captured image. Based on the method of determining whether or not the driver is in a normal state, for example, driving using drugs, driving after taking a drug, driving in an overworked state, driving asleep, sleeping or having a fever The same applies to the case of driving and the like.

  In the present embodiment, the case where the drunk detection system 1 is applied to an ordinary car has been described. However, vehicles to which the drunk detection system 1 can be applied are not limited to this, and for example, trains, trains, bicycles, and the like. It can also be applied to various vehicles.

  In addition, it is needless to say that the present invention is not limited to the above embodiment and can be modified as appropriate.

It is the principal part block diagram which showed schematic structure of one Embodiment of the liquor detection system which concerns on this invention. It is the figure which showed the example of mounting to the vehicle of the drunk detection system shown in FIG. FIG. 3A is a graph showing a time-series change in the size of the pupil as biological data. FIG. 3B is a diagram showing in more detail the time-series change of the pupil when the stimulus is added (b) in FIG. It is the graph which showed the time-sequential change of the center position of the pupil as biometric data. It is a conceptual diagram explaining the process which detects a drunkenness degree using analysis object data and normal time data. FIG. 6A is a graph showing the correlation between the surface temperature difference and the amount of alcohol consumed / expired alcohol concentration. FIG. 6B is a graph showing the correlation between the pulse rate difference and the alcohol consumption / expired alcohol concentration. FIG. 6C is a graph showing an evaluation function of the degree of alcohol. It is the figure which showed typically the installation position of the color chart used for imaging | photography calibration. FIG. 8A is a diagram schematically illustrating the installation position of a chart used for imaging calibration. FIG. 8B shows an example of a distortion correction pattern and distortion correction parameters. FIG. 9A is a graph showing the correlation between the difference in surface temperature and the amount of alcohol consumed / exhaled alcohol concentration. FIG. 9B is a graph showing the correlation between the pulse difference and the alcohol consumption / expired alcohol concentration. FIG. 9C is a graph showing an evaluation function of the degree of alcohol. It is a flowchart which shows the flow of the detection / judgment process of the alcoholic beverage degree in the alcoholic beverage detection system.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Alcohol detection system 2 Image photographing part 3 Illumination part 4 Light stimulation illumination part 5 User interface part 6 Display part 7 Memory part 8 Data processing / analysis part 9 Parameter setting / management part 10 Data storage part 11 I / O part 12 External communication unit 13 Vehicle control interface unit 15 Control unit

Claims (12)

An image capturing unit for capturing an image of the vehicle driver;
A normal data holding means for holding a normal image or normal biological data of a vehicle driver in a normal state that is not drunk;
Comparing and contrasting the image captured by the image capturing unit and the biometric data obtained by analyzing the image with the normal image or the normal biometric data stored in the normal data storage unit A data processing unit for detecting the degree of drunkness of the vehicle driver,
A drunkenness detection system characterized by comprising:   The alcohol image detection system according to claim 1, wherein the image capturing unit captures images of a head and neck of a vehicle driver. An association data holding means for holding an evaluation function or an evaluation table for relating the biometric data and the degree of drunkenness of the vehicle driver;
The said data processing part detects the drunkness degree of a vehicle driver, referring to the said evaluation function or the said evaluation table currently hold | maintained at the said correlation data holding means, The claim 1 characterized by the above-mentioned. The drunk detection system according to claim 2.   4. The alcohol detection according to claim 1, further comprising: a transmission unit that outputs an instruction signal relating to vehicle operation control based on a degree of alcohol consumption of the vehicle driver. 5. system.   The drunk detection system according to any one of claims 1 to 4, further comprising photographing calibration means for calibrating photographing characteristics of the image photographing unit.   6. The drunkenness detection system according to claim 3, further comprising an analysis calibration unit configured to calibrate the evaluation function or the evaluation table. An image capturing process for capturing an image of the vehicle driver;
By comparing and contrasting the captured image and the biometric data obtained by analyzing the image with the normal image or normal biometric data of the vehicle driver in a normal state that is not drunk, the driver's drunkness Data processing step to detect the degree,
A drunkenness detection method comprising:   The drunkenness detection method according to claim 7, wherein the image capturing step captures images of a head and a neck of a vehicle driver.   8. The data processing step detects the degree of drunkenness of the vehicle driver while referring to an evaluation function or an evaluation table relating the biometric data and the drunkenness of the vehicle driver. Item 9. The method for detecting drunkenness according to Item 8.   The detection of drunkness according to any one of claims 7 to 9, further comprising a transmission step of outputting an instruction signal related to vehicle operation control based on a degree of drunkness of the vehicle driver. Method.   11. The method for detecting drunkenness according to any one of claims 7 to 10, further comprising a photographing calibration step for performing calibration on photographing characteristics in image photographing.   12. The method for detecting drunkenness according to any one of claims 9 to 11, further comprising an analysis calibration step of performing calibration on the evaluation function or the evaluation table.

Images