JP2007025871A - Operator state evaluation device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately execute the evaluation of the load status of an operator by considering an environmental factor affecting face skin temperature distribution from the inside and outside of the compartment of a vehicle, and to accurately evaluate the load status of the operator by considering the change characteristics of the fractal dimension of each operator. <P>SOLUTION: A fractal dimension calculating circuit 5 calculates fractal dimensions by carrying out the fractal analysis of a temperature distribution image measured by a face temperature distribution measuring circuit 1. An operating load deciding circuit 6 decides the operation load of an operator based on the number of fractal dimensions calculated by the fractal dimension calculating circuit 5. Furthermore, when a traffic volume detected by a traffic volume detecting device 2 or the complicatedness level of road lineament detected by a road lineament complicatedness level detecting device 3 is changed, an operation load deciding circuit 6 changes the method of the decision of the operation load according to the changing amounts of the number of the calculated fractal dimensions. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a driver state evaluation apparatus that quantitatively or qualitatively evaluates a driver's driving load during driving of a vehicle.

  Conventionally, by analyzing information contained in so-called biological signals such as human heartbeat, pulse wave, respiration, blood, saliva, etc., the driving load given to the driver who is driving the vehicle etc. is quantitatively determined. Driver state devices that calculate and evaluate the state of the driver have been developed. In recent years, a non-contact type driver condition evaluation apparatus in which a sensor is not attached to a driver has been developed particularly for the purpose of mounting in a car.

For example, the fatigue characteristic evaluation device and the fatigue measurement device described in Patent Document 1 measure the surface temperature of the driver's face using an infrared thermal imaging device, and are given to the driver with the characteristics of changes in the measured temperature. By indicating the relationship of the driving load, the driver's load is estimated from the surface temperature of the face.
On the other hand, the distribution of facial surface temperature varies among individuals, but it is known that the distribution pattern basically exists, and in order to characterize the distribution pattern numerically, the facial skin temperature distribution is analyzed by fractal analysis. The technology for estimating the driver's load from the fractal analysis of the surface temperature of the face is known by calculating the fractal dimension using the above and showing the relationship between the calculated fractal dimension and the amount of driving load given to the driver It has become. With such a technology, it is possible to more accurately detect the state in which the driver is exposed to the driving load compared to the technology that estimates the driving load by directly observing the skin temperature of the face. became.
JP-A-9-28679

However, in evaluating the load state of the driver by performing fractal analysis of the facial skin temperature distribution, the evaluation should be made after eliminating the adverse effects of the environment in which the vehicle is placed.
Moreover, in evaluating the driving load based on the facial skin temperature distribution, it is necessary to eliminate the influence of sunlight entering the vehicle and the influence of the air conditioner provided in the vehicle.

Furthermore, when evaluating the driver's load state by fractal analysis of the facial skin temperature distribution, each driver has a different fractal dimension change characteristic according to the magnitude of the driving load, and the correct evaluation is performed. Therefore, an evaluation method that takes into account the change characteristics of the fractal dimension of each driver is required.
Accordingly, an object of the present invention was created in view of the above points, and in evaluating the driver's load state by fractal analysis of the facial skin temperature distribution, the facial skin temperature distribution from inside and outside of the vehicle. In addition to accurately evaluating the driver's load state by considering environmental factors that affect the driver, more accurate evaluation of the driver's load state by considering the fluctuating characteristics of each driver's fractal dimension It is in providing the driver state evaluation apparatus which can perform.

In order to solve the above problems and achieve the object of the present invention, the present invention has the following configuration.
That is, the present invention provides a facial skin temperature distribution measuring means for measuring a two-dimensional distribution of skin temperature on the face of a driver who drives a vehicle as a temperature distribution image, and surroundings of the vehicle when the vehicle travels. Traffic volume detecting means for detecting traffic volume, road linear complexity detecting means for detecting road linear complexity in front of the vehicle when the vehicle travels, and temperature distribution measured by the face temperature distribution measuring means Fractal dimension calculating means for calculating a fractal dimension by performing fractal analysis on an image, and driving load determining means for determining the driving load of the driver based on the number of dimensions of the fractal dimension calculated by the fractal dimension calculating means. The driving load determination means includes the traffic volume detected by the traffic volume detection means and the road linear complexity detected by the road linear complexity detection means. At least one of the case has changed, by an amount varied in the number of dimensions of the fractal dimension calculated by the fractal dimension calculation unit, a method of determining the operating load is adapted to be changed.

  According to the present invention, the load state of the driver can be evaluated by the pattern of the skin temperature distribution on the driver's face. During the evaluation, the traffic volume around the vehicle the driver is driving and the road ahead The driver's driving load for linear complexity can be evaluated from the change in fractal dimension.

Embodiments of the present invention will be described below with reference to the drawings.
(First embodiment)
FIG. 1 is a block diagram showing the configuration of the first embodiment of the present invention.
As shown in FIG. 1, the driver condition evaluation apparatus according to the first embodiment includes a facial skin temperature distribution measurement apparatus 1, a traffic volume detection apparatus 2, a road linear complexity detection apparatus 3, and an area division circuit 4. A fractal dimension calculation circuit 5, an operating load determination circuit 6, a process switching circuit 7, a determination area determination circuit 8, a divided area characteristic storage device 9, a load determination method change circuit 10, and a weight value assignment circuit 11 And an illuminance detection device 12 and a wind detection device 13.
Here, the facial skin temperature distribution measuring device 1, the traffic volume detecting device 2, the road linear complexity detecting device 3, and the area dividing circuit 4 are the facial skin temperature distribution measuring means, traffic volume detecting means, road linear complexity detecting means. , And area dividing means, respectively.

Further, the fractal dimension calculation circuit 5, the driving load determination circuit 6, the process switching circuit 7, and the determination area determination circuit 8 respectively correspond to the fractal dimension calculation means, the driving load determination means, the process switching means, and the determination area determination means. .
Further, the divided region characteristic storage device 9, the load determination method changing circuit 10, the weight value applying circuit 11, the illuminance detecting device 12, and the wind detecting device 13 are divided into a divided region characteristic storing unit, a load determining method changing unit, and a weight value applying unit. , Illuminance detection means, and wind detection means respectively.

Next, an overview of the functions of the components of the first embodiment shown in FIG. 1 will be described.
The facial skin temperature distribution measuring apparatus 1 measures a two-dimensional distribution of skin temperature on the face of a driver who drives a vehicle as a temperature distribution image. The traffic volume detection device 2 detects the traffic volume around the vehicle when the vehicle travels.
The road linearity complexity detection device 3 detects the complexity of the road linearity ahead of the vehicle when the vehicle travels. The area dividing circuit 4 divides the facial skin temperature image measured by the facial skin temperature distribution measuring apparatus 1 into divided areas having a predetermined area.

The fractal dimension calculation circuit 5 performs fractal analysis on each of the divided facial skin temperature images, and calculates the number of dimensions of the fractal dimension for each divided region. The driving load determination circuit 6 determines the driving load of the driver quantitatively or qualitatively based on the calculated number of dimensions of the fractal dimension, and the determination process is performed as described later.
The processing switching circuit 7 refers to the traffic volume detected by the traffic volume detection device 2 and the road linearity complexity detected by the road linearity complexity detection device 3, and switches processing contents under a predetermined condition as will be described later. It is. As will be described later, the determination area determination circuit 8 is such that neither the traffic detected by the traffic detection device 2 nor the complexity of the road alignment detected by the road linear complexity detection device 3 changes. When the driving load is detected, the driving load determination circuit 6 determines a determination region to be a target of the driving load determination.

The divided region characteristic storage device 9 stores the determination region determined by the determination region determination circuit 8. The load determination method changing circuit 10 is based on the condition that the driving load determination circuit 6 determines that there is no disturbance in determining the determination region, and the traffic volume detected by the traffic volume detection device 2 and the road linear complexity detection device 3. The weight to be given by the weight value assigning circuit 11 is calculated with reference to the detected complexity of the road alignment. The weight value giving circuit 11 gives the calculated weight.
The illuminance detection device 12 detects the illuminance of light directly irradiated on the driver's face. The wind detection device 13 detects the wind direction and the air volume in the vehicle interior.

Next, the detailed operation of each part of the first embodiment will be described with reference to the flowchart shown in FIG. This explanation is an operation example of each part when the driver is driving the vehicle, that is, when the vehicle is traveling.
In step S100, the facial skin temperature distribution measuring apparatus 1 measures the body temperature distribution in front of the driver's face as an image. As the facial skin temperature distribution measuring device 1, a so-called infrared thermal image device is used, which is disposed in a non-contact manner in front of the operator as shown in FIG. As a result, the infrared thermal imaging device detects infrared rays emitted from the driver's face and performs imaging that can be converted into a relative temperature distribution.

In step S101, the traffic volume detection device 2 detects the traffic volume around the vehicle. This traffic volume detection device 2 can be realized by detecting the presence or absence of a surrounding vehicle by recognizing a vehicle using an image taken by a camera, or by using a laser radar, an ultrasonic sensor, or the like.
In step S102, the road linear complexity detection device 3 detects the road linear complexity in front of the vehicle. The road linear complexity detection device 3 captures the front of the vehicle with a camera to acquire an image, extracts a road from the acquired image, and increases the complexity as the road alignment is more complicated. Alternatively, map information in front of the vehicle is taken from a so-called navigation device, and the complexity of the road information is detected from the map information.

In step S103, the area dividing circuit 4 performs processing for dividing the facial skin temperature distribution image measured in step S100 into areas of a predetermined size. Normally, the temperature distribution image of the facial skin temperature is composed of vertical N pixels and horizontal M pixels in units of two-dimensionally arranged pixels, and is expressed in grayscale shades and colors depending on the temperature of each pixel.
Therefore, in this example, the two-dimensional temperature distribution image is equally divided into p both vertically and horizontally (see FIG. 4). By this processing, the temperature distribution image is divided into a total of p ² regions, and each divided region is composed of N / p pixels in the vertical direction and M / p pixels in the horizontal direction. That is, each divided region can be represented by coordinates (Xi, Yj) where Y is vertical and X is horizontal. However, i = 0, 1,... P, j = 0, 1,.

In step S104, the fractal dimension calculation circuit 5 performs a fractal analysis on each of the p ² divided regions obtained in step S103, and calculates a fractal dimension. A known method can be used for fractal analysis of a two-dimensional image. When fractal analysis is performed on each divided region, the fractal dimension in each divided region has a value of 2 to 3. In other words, the fractal dimension Fij at the coordinates (Xi, Yj) of the divided area has a value of 2 to 3.
In step S105, the processing switching circuit 7 holds the traffic volume detected in step S101 and the road linearity complexity detected in step S102 for a predetermined time T, and at the predetermined time T, the traffic volume and the complexity are maintained. It is determined whether or not at least one change is a change (variation) within a predetermined range.

In this determination process, the traffic volume and the complexity are defined in advance in, for example, five levels, and if the level does not change for a predetermined time T, it is determined as “within a predetermined range”. If it changes, it is determined that “out of the predetermined range”. If the result of this determination is “within a predetermined range”, the process proceeds to step S106, and if “out of the predetermined range”, the process proceeds to step S110.
In step S106, the determination area determination circuit 8 determines whether at least one of the traffic volume detected in step S101 and the road linearity complexity detected in step S102 is low or low.

  In this individual processing, the traffic volume and the road linearity complexity are defined in advance, for example, in five stages, respectively. If the traffic volume is high or the complexity is high, 5 is assigned, and the traffic volume is low. Or 1 if the complexity is low. If the assigned value is 1 to 3, it is determined (determined) that the traffic volume or complexity is low, and if the assigned value is 4 or 5, the traffic volume or complexity is high. decide. As a result of this determination, if the traffic volume or complexity is low, the process proceeds to step S107, and if it is high, the process proceeds to step S108.

  In step S107, even when the determination area determination circuit 8 determines that the traffic volume or complexity is low in step S106 among the fractal dimensions in each divided area calculated in step S104, the fractal dimension shows a low value. If there is a corresponding divided area, the position of the searched divided area is a sensitive divided area whose fractal dimension changes even if it is determined that the traffic volume or complexity is low. This is stored in the divided area characteristic storage device 9.

  In step S108, when the determination area determination circuit 8 determines that the traffic volume or the complexity is high in step S106 among the fractal dimensions in each divided area calculated in step S104, the fractal dimension indicates a low value. When a segmented area is searched, and the corresponding segmented area exists, the segmented area is segmented as an insensitive segmented area whose fractal dimension changes even if it is determined that the traffic volume or complexity is high. This is stored in the area characteristic storage device 9.

  In step S109, the determination area determination circuit 8 determines and holds each position of the sensitive divided area and the insensitive divided area stored in the divided area characteristic storage device 9 as a determination area. In other words, the determination region determined in this way is a region where the fractal dimension has changed according to the traffic volume detected in step S101 or the road linearity complexity detected in step S102. This means that a segmented region where the fractal dimension changes with respect to changes in traffic volume and road line complexity is extracted.

In step S110, the driving load determination circuit 6 uses the detection information from the illuminance detection device 12 and the wind detection device 13 to determine whether there is a disturbance factor that affects the determination of the driving load of the driver inside and outside the vehicle. Determine whether or not.
Specifically, the illuminance detection device 12 uses a device capable of detecting at least the illuminance in the determination area determined in step S109, and the illuminance corresponding to the case where the determination area is directly irradiated with light from outside the vehicle. When it is detected, the driving load determination circuit 6 stops the determination process.

Further, the wind detection device 13 has a direct direction of the wind from the outside of the vehicle and the wind from the air conditioner with respect to at least the determination region determined in step S109 and affects the temperature of the face in the determination region. Is detected, and if the corresponding wind hits the determination area, the driving load determination circuit 6 stops the determination process.
On the other hand, when the driving load determination circuit 6 does not stop the determination processing of the driving load of the driver, the process proceeds to step S111. In step S111, the load determination method change circuit 10 determines whether at least one of the traffic volume detected in step S101 and the road linearity complexity detected in step S102 is low or low.

  In this individual processing, the traffic volume and the complexity are defined in advance, for example, in 5 stages, respectively. If the traffic volume is high or the complexity is high, 5 is assigned, and if the traffic volume is low, or If the complexity is low, 1 is assigned to each. In the case of the assigned values 1 to 3, it is determined (determined) that the traffic volume or the complexity is low, and in the case where the assigned value is 4 or 5, the traffic volume or the complexity is determined to be high. To do. As a result of this determination, if the traffic volume or complexity is low, the process proceeds to step S112, and if it is high, the process proceeds to step S113.

In step S112, the weighting circuit 11 performs weighting in the determination area determined in step S109. Usually, when the traffic volume or the complexity of the road alignment is low, the change in the fractal dimension obtained as a result of fractal analysis of the facial skin temperature image is very small. For this reason, many of the divided areas included in the determination area determined in step S109 cannot reflect changes in the driving load of the driver as changes in the fractal dimension.
Therefore, in order to reliably capture this minute change, in this example, the change in the fractal order in the sensitive divided region extracted in step S107 and stored in the divided region characteristic storage device 9 changes the driving load of the driver. A strong weight value is given to the sensitive divided region so as to surely affect the determination (see FIG. 5).

In step S113, the weighting circuit 11 performs weighting in the determination area determined in step S109. In general, when the traffic volume or the road alignment complexity is high, the influence of the fractal order obtained as a result of fractal analysis of the facial skin temperature image is large. For this reason, many of the divided regions included in the determination region determined in step S109 are saturated, and changes in the driving load of the driver cannot be reflected as changes in the fractal dimension.
Therefore, in order to reliably capture changes in the driver's driving load when the traffic volume or road linearity is high, in this example, the insensitivity extracted in step S108 and stored in the divided region characteristic storage device 9 is used. A strong weight value is given to the insensitive divided area so that the change in the fractal dimension in the divided area surely affects the determination that the driving load of the driver has changed (see FIG. 6).

In step S114, the driving load determination circuit 6 determines the driving load of the driver. In this determination, the fractal dimension in the divided region included in the determination region determined in step S109 is set as the determination target of the driver's driving load.
For example, when determining the total amount of operating load, it is preferable to determine the sum of the fractal dimensions of each divided region included in the determination region, and to determine that the operating load is larger as the calculated sum is smaller. In this way, it is possible to evaluate the state of the driver in which the influence of the region where the fractal dimension changes with respect to factors other than the driving load is eliminated as much as possible.

  Further, for example, when detecting a change in the driving load, the fractal value obtained by multiplying each fractal dimension by the weight value assigned to each divided region given in step S112 and step S113, and further multiplying the weight in each divided region. The total sum of dimensions may be obtained, and it may be determined that the change in the driving load of the driver is larger as the change in the calculated sum value becomes more prominent.

  As described above, according to the first embodiment, by providing the region dividing circuit 4, the facial skin temperature distribution measured by the facial skin temperature distribution measuring apparatus 1 is divided, and the fractal dimension calculating circuit 5 is further provided. Since the fractal dimension is calculated for the area divided by the area dividing circuit 4 by providing, the characteristics of each divided area, that is, the area where a strong driving load and a weak driving load can be evaluated In addition, it is possible to evaluate the driver's condition based on an area where the influence of factors from the external environment can be detected.

In the first embodiment, since the traffic volume inspection device 2 and the road linearity complexity detection device 3 are provided, the factors affecting the driver's load can be detected.
In the first embodiment, since the process switching circuit 7 and the determination area determination circuit 8 are provided, the relationship between the detected operating load factor and the change in the fractal dimension in each divided area is considered. Even if the driver changes or the driver's own characteristics change, the driver's condition can be continuously evaluated.

In the first embodiment, the determination area determination circuit 8 and the divided area characteristic storage device 9 are provided. The determination area determination circuit 8 determines a sensitive divided area and an insensitive divided area, and divides both the determined divided areas. Since it is stored in the area characteristic storage device 9, it is possible to evaluate the driver's state more accurately regardless of the driving load.
In the first embodiment, since the weight value assigning circuit 11 that performs weighting in the determination region is provided, the driving load determination circuit 6 can perform determination considering the weighting when determining the driving load. Therefore, the driver's condition can be more accurately evaluated regardless of the driving load.

  Furthermore, in the first embodiment, since the illuminance detection device 12 and the wind detection device 13 are provided, the factors from the external environment can be detected by these, and the detected factors are detected by the determination region determination circuit 8 by the driver. When there is a possibility of affecting the region for determining the state of the driver, the driver's state evaluation process is stopped, so that the driver's state can be evaluated more accurately.

(Second Embodiment)
FIG. 7 is a block diagram showing the configuration of the second exemplary embodiment of the present invention.
As shown in FIG. 7, the driver condition evaluation apparatus according to the second embodiment includes a facial skin temperature distribution measurement apparatus 1, a traffic volume detection apparatus 2, a road linear complexity detection apparatus 3, and a region dividing circuit 21. A fractal dimension calculation circuit 5, a driving load determination circuit 22, a switch 23, a process switching circuit 24, a determination area determination circuit 25, and a non-determination area determination circuit 26.
Here, the facial skin temperature distribution measuring device 1, the traffic volume detecting device 2, the road linear complexity detecting device 3, and the area dividing circuit 21 are a facial skin temperature distribution measuring means, a traffic volume detecting means, a road linear complexity detecting means. , And area dividing means, respectively.

Further, the fractal dimension calculation circuit 5, the driving load determination circuit 22, the switch 23, the processing switching circuit 24, the determination area determination circuit 25, and the non-determination area determination circuit 26 include a fractal dimension calculation means, an operation load determination means, an instruction means, It corresponds to a process switching means, a determination area determination means, and a non-determination area determination means, respectively.
The second embodiment shown in FIG. 7 includes the same constituent elements as those in the first embodiment shown in FIG. 1. Therefore, the same constituent elements are denoted by the same reference numerals, and the description thereof is omitted as much as possible.

Next, an overview of the functions of the components of the second embodiment shown in FIG. 7 will be described.
The area dividing circuit 21 converts the facial skin temperature image measured by the facial skin temperature distribution measuring device 1 into the traffic volume detected by the traffic volume detection device 2 and the road linearity complexity detected by the road linearity complexity detection device 3. Based on this, it is divided into regions of a predetermined area as will be described later.
The driving load determination circuit 22 determines the driving load of the driver quantitatively or qualitatively based on the number of dimensions of the fractal dimension calculated by the fractal dimension calculation circuit 5, and is determined as described later.
The switch 23 is used by the driver to input processing switching to the processing switching circuit 24. The process switching circuit 24 switches between a series of processes for determining the driving load and a series of processes for determining the determination processing area in accordance with a switching instruction from the switch 23, and instructs each unit to switch the processes. Yes.

In accordance with an instruction from the processing switching circuit 24, the determination area determination circuit 25 is predetermined for at least one of the traffic volume detected by the traffic volume detection device 2 and the road linearity complexity detected by the road linearity complexity detection device 3. When there is a change, the predetermined divided area is determined as the determination area.
In accordance with an instruction from the processing switching circuit 24, the non-determination area determination circuit 26 sets at least one of the traffic volume detected by the traffic volume detection device 2 and the road linearity complexity detected by the road linearity complexity detection device 3. When there is a predetermined change, the predetermined divided area is determined as a non-determination area.

Next, the detailed operation of each part of the second embodiment will be described with reference to the flowchart shown in FIG.
In step S200, the facial skin temperature distribution measuring apparatus 1 measures the body temperature distribution in front of the driver's face as an image. In step S201, the traffic volume detection device 2 detects the traffic volume of vehicles around the running vehicle. In step S202, the road linear complexity detection device 3 detects the road linear complexity ahead of the traveling vehicle.
In step S203, the area dividing circuit 21 determines an area for dividing the facial skin temperature distribution image measured in step S200 into predetermined areas in the next step S204. This process is performed based on the traffic volume detected in step S201 and the road alignment complexity detected in step S202.

  That is, the area dividing circuit 21 is configured such that the smaller the detected traffic volume and the smaller the detected road alignment complexity, the smaller the area for dividing. Specifically, the traffic volume and the complexity of the road alignment are defined in advance in, for example, five levels, and 5 is assigned if the traffic volume is high or the complexity is high, respectively, and the traffic volume is low, or If the complexity is low, 1 is assigned. If the traffic or complexity level is 5, then the density distribution image is divided into q equal parts both vertically and horizontally, if it is 4, it is divided into q × 2 and if it is 3, it is divided into q × 4. It is decided to divide q × 8 equally if 2 and q × 16 equally if 1 (see FIG. 9).

  In step S204, the region dividing circuit 21 performs a process of dividing the facial skin temperature distribution image obtained by the measurement in step S200 into the area determined in step S203. Normally, the temperature distribution image of the facial skin temperature is composed of vertical N pixels and horizontal M pixels in units of two-dimensionally arranged pixels, and is expressed in grayscale shades and colors depending on the temperature of each pixel. Therefore, in this example, the two-dimensional temperature distribution image is equally divided into the size determined in step S203 in both vertical and horizontal directions.

In step S205, the fractal dimension calculation circuit 5 performs fractal analysis on each of the divided areas obtained in step S204, and calculates each of the fractal dimensions.
In step S <b> 206, the process switching circuit 24 determines whether or not there is a switching input (instruction) from the switch 23. The driver uses the switch 23 to instruct one of a process for determining the driving load and a process for determining a determination processing region in the temperature distribution image of the facial skin temperature for determining (detecting) the driving load. Can be selected.

  The process switching circuit 24 switches between a series of processes for determining an operating load and a series of processes for determining a determination process area in accordance with an input instruction from the switch 23. That is, when there is an input from the switch 23, the process proceeds to step S207 for determining the determination processing region, and when there is no input, the process proceeds to step S210 for determining the driving load.

  In step S207, the determination area determination circuit 25 determines whether there is a change in the traffic volume detected in step S201, or whether there is a change in the road linearity complexity detected in step S202. In this determination, for example, the traffic volume and the road alignment complexity are defined in advance in five stages according to the degree, and if the traffic volume is high or the road alignment complexity is high, 5 is given, If the traffic volume is low or the road alignment complexity is low, 1 is assigned. Then, when the traffic volume or the complexity of the road alignment changes to another stage, it is determined that it has changed. As a result of this determination, if there is a change, the process proceeds to step S208, and if there is no change, the process ends.

  In step S208, the determination area determination circuit 25 determines a divided area that is a target for determining the driver's driving load based on the fractal dimension. That is, the determination area determination circuit 25 performs the fractal dimension calculation circuit 5 in each divided area divided by the area dividing circuit 21 when the traffic volume or the road linearity complexity changes to another stage in step S207. The region where the fractal dimension calculated in (1) is changed is determined as the determination region, and the determined determination region is held.

  In step S209, the non-determination area determination circuit 26 determines a divided area that is not a target for determining the driving load of the driver based on the fractal dimension. That is, the non-determination area determination circuit 26 performs the fractal dimension in each divided area divided by the area dividing circuit 21 even though the traffic volume or road linearity complexity is changed to another stage in step S207. An area where the fractal dimension calculated by the calculation circuit 5 has not changed is determined as a non-determination area, and the determined non-determination area is held.

In step S210, the driving load determination circuit 22 determines whether there is a change in the fractal dimension calculated by the fractal dimension calculation circuit 5 in the divided area determined by the non-determination area determination circuit 26 as the non-determination area in step S209. To do.
Here, when the fractal dimension changes in the non-judgment region, factors other than the driver's driving load, that is, factors that affect the facial skin temperature, such as sunlight from outside and air conditioners, Since there is a possibility that it has an influence, the processing is terminated without determining the driving load of the driver. On the other hand, if the fractal dimension has not changed in the non-determination area, the process proceeds to step S211.

  In step S <b> 211, the driving load determination circuit 22 determines the driving load of the driver using each fractal dimension of the divided area calculated by the fractal dimension calculation circuit 5. At this time, the driving load determination circuit 22 sets the fractal dimension in the divided region included in the determination region extracted in step S208 as the determination target of the driving load. For example, when determining the total amount of operating load, the sum of the fractal dimensions of each divided region included in the determination region is obtained, and it is determined that the operating load is higher as the obtained sum is smaller.

  As described above, according to the second embodiment, in the area dividing circuit 21, factors affecting the driving load of the driver detected by the traffic volume detection device 2 and the road linear complexity detection device 3 are When the influence is expected to be small, the area to be divided is reduced, and when the influence is expected to be large, the area of the divided region is increased. For this reason, in addition to the effect of 1st Embodiment, it is possible to evaluate more accurately the influence which the factor of driving load has on a driver.

In the second embodiment, since the non-determination area determination circuit 26 is provided, it is possible to detect a state in which factors other than the driving load affect the driver's situation evaluation by fractal analysis. It is possible to stop the evaluation by detecting a state where the state cannot be accurately evaluated.
Furthermore, since the switch 23 and the process switching circuit 24 are provided in the second embodiment, it is possible to switch between the process for determining the determination region and the process for determining the driving load according to the driver's intention.

(Third embodiment)
FIG. 10 is a block diagram showing the configuration of the third exemplary embodiment of the present invention.
As shown in FIG. 10, the driver condition evaluation apparatus according to the third embodiment includes a facial skin temperature distribution measurement apparatus 1, a traffic environment amount detection apparatus 31, a region division circuit 32, and a nose extraction circuit 33. , A fractal dimension calculation circuit 5, an operation load determination circuit 34, a process switching circuit 35, and a determination region determination circuit 36.
Here, the facial skin temperature distribution measuring device 1, the traffic environment quantity detecting device 31, the area dividing circuit 32, and the nose extraction circuit 33 are the facial skin temperature distribution measuring means, the traffic environment quantity detecting means, the area dividing means, and the nose. Each corresponds to a part extraction means.

The fractal dimension calculation circuit 5, the driving load determination circuit 34, the process switching circuit 35, and the determination area determination circuit 36 correspond to the fractal dimension calculation means, the driving load determination means, the process switching means, and the determination area determination circuit means, respectively. To do.
In addition, since this 3rd Embodiment contains the same component as 1st Embodiment shown in FIG. 1, the same code | symbol is attached | subjected about the same component and the description is abbreviate | omitted as much as possible.

Next, an overview of the functions of the components in the third embodiment shown in FIG. 10 will be described.
The traffic environment detection device 31 detects the complexity of the traffic environment while the vehicle is traveling. The area dividing circuit 32 divides the facial skin temperature image measured by the facial skin temperature distribution measuring apparatus 1 into areas of a predetermined area with reference to the complexity of the traffic environment detected by the traffic environment detecting apparatus 31. is there.

The nose extraction circuit 33 extracts an area including the driver's nose from the divided areas of the skin temperature distribution image measured by the facial skin temperature distribution measuring apparatus 1. The driving load determination circuit 34 determines the driving load of the driver quantitatively or qualitatively based on the number of dimensions of the fractal dimension calculated by the fractal dimension calculation circuit 5, and is determined as described later.
The process switching circuit 35 switches between a series of processes for determining the determination area and a series of processes for determining the driving load in accordance with a change in the complexity of the traffic environment detected by the traffic environment detection device 31. The determination region determination circuit 36 corresponds to a nose as a determination region when the driving load determination circuit 34 determines a driving load when a series of processing instructions for determining a determination region is received from the processing switching circuit 35. Is specified.

Next, the detailed operation of each part of the third embodiment will be described with reference to the flowchart shown in FIG.
In step S300, the facial skin temperature distribution measuring apparatus 1 measures the body temperature distribution in front of the driver's face as an image.
In step S301, the traffic environment detection device 31 detects the complexity of the traffic environment. The traffic environment detection device 31 in this example includes, for example, a traffic environment complexity detection device using a known chaos analysis.

  This complexity detection device includes a camera that captures the front of the vehicle, and a chaos analysis circuit that performs a chaos analysis on a front image acquired by the camera and calculates a Lyapunov exponent. The Lyapunov exponent tends to show a larger value as the information contained in the image is more complicated. Therefore, the complexity of the traffic environment ahead of the vehicle can be quantitatively quantified with the Lyapunov exponent.

In step S302, the region dividing circuit 32 performs processing for dividing the facial skin temperature distribution image measured in step S300 into regions of a predetermined size.
In step S303, the nose extraction circuit 33 extracts an area including the driver's nose from the divided areas of the distribution image. In general, it can be said that the influence of the driving load is easily reflected in the distribution image of the facial skin temperature, particularly in the nose. Therefore, in this example, a camera is placed at the same position as the facial skin temperature distribution region device 1, the face of the driver is photographed with the camera, and a predetermined image recognition process is performed to identify the nose. . The identified position of the nose is held in the nose extraction circuit 33 for use as coordinates on the screen.

In step S304, the fractal dimension calculation circuit 5 performs a fractal analysis on each of the divided areas obtained in step S302, and calculates a fractal dimension for each divided area.
In step S305, the process switching circuit 35 determines whether or not the change in the complexity of the traffic environment detected by the traffic environment detection device 31 in step S301 is within a predetermined range, and a determination is made according to the determination result. Switching between a series of processes for determining a region and a series of processes for determining an operation load.

In this determination, the complexity of the traffic environment is maintained for a predetermined time T, and it is determined whether or not the change in the complexity of the traffic environment is within a predetermined range at the predetermined time T. For example, the complexity is defined in advance in five levels, and if the level does not change during a predetermined time T, it is determined as “in range”, and if the level changes, it is determined as “out of range”. I tried to do it.
If the result of this determination is “within range”, the determination area determination circuit 36 proceeds to step S306 for performing a series of processes for determining the determination area. If it is “out of range”, the driving load The process proceeds to step S308 where the determination circuit 34 performs a process of determining the driving load.

  In step S306, the determination area determination circuit 36 determines whether the complexity of the traffic environment detected by the traffic environment detection device 31 is high or low. In this determination, for example, the complexity of the traffic environment is defined in advance in five stages according to the degree, 5 is assigned if the complexity of the traffic environment is high, and 1 is assigned if the complexity of the traffic environment is low. Give. Then, when the level is 1, it is determined that the complexity of the traffic environment is low, and when the level is 2 to 5, it is determined that the complexity of the traffic environment is high. As a result of this determination, if it is determined that the complexity of the traffic environment is high, the process proceeds to step S307, and if it is determined that the complexity of the traffic environment is low, the process ends.

In step S307, when the determination area determination circuit 36 has a high degree of complexity of the traffic environment determined in step S306, among the divided areas where the fractal dimension value is small, the driving extracted particularly in step S303. A divided region corresponding to the person's nose is specified as a nose determination region.
In step S308, the driving load determination circuit 34 determines the driving load of the driver based on each fractal dimension of the divided area calculated by the fractal dimension calculation circuit 5. At this time, the driving load determination circuit 34 sets the fractal dimension in the divided region included in the nose determination region extracted in step S307 as the determination target of the driving load. For example, when determining the total amount of driving load, the sum of the fractal dimensions of each divided region included in the nose determination region is obtained, and it is determined that the driving load is higher as the obtained sum is smaller.
As described above, according to the third embodiment, the nose part extraction circuit 33 is provided, and the nose part that is most susceptible to the influence of the driving load on the driver's facial temperature distribution is specified. In addition to the effects of the first and second embodiments, the driver's condition can be more accurately evaluated.

(Other embodiments)
In the above first to third embodiments, as shown in FIG. 1, FIG. 7, and FIG. However, as another embodiment of the present invention, these circuits and the like may be realized using a computer, and arithmetic processing as shown in FIGS. 2, 8, and 11 may be performed.
In this case, for example, in FIG. 1, an area division circuit, a fractal dimension calculation circuit, an operation load determination circuit, a process switching circuit, a determination area determination circuit, a load determination method change circuit, a weight value assignment circuit, etc. ) Or memory.

It is a block diagram which shows the structure of 1st Embodiment of this invention. It is a flowchart explaining operation | movement of 1st Embodiment. It is explanatory drawing explaining arrangement | positioning of the facial skin temperature distribution measuring apparatus. It is explanatory drawing explaining the determination method of the division area in 1st Embodiment. It is explanatory drawing explaining the determination method of weight value provision in 1st Embodiment. It is explanatory drawing explaining the determination method of weight value provision in 1st Embodiment. It is a block diagram which shows the structure of 2nd Embodiment of this invention. It is a flowchart explaining operation | movement of 2nd Embodiment. It is explanatory drawing explaining the determination method of the division area in 2nd Embodiment. It is a block diagram which shows the structure of 3rd Embodiment of this invention. It is a flowchart explaining operation | movement of 3rd Embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Facial skin temperature distribution measuring device 2 Traffic detection device 3 Road linear complexity detection device 4, 21, 32 Area division circuit 5 Fractal dimension calculation circuit 6, 22, 34 Driving load judgment circuit 7, 24, 35 Processing switching circuit 8 , 25, 36 Determination region determination circuit 9 Division region characteristic storage device 10 Load determination method change circuit 11 Weight value assignment circuit 12 Illuminance detection device 13 Wind detection device 23 Switch 26 Non-determination region determination circuit 31 Traffic environment detection device 33 Nose extraction circuit

Claims (16)

Facial skin temperature distribution measuring means for measuring a two-dimensional distribution of skin temperature on the face of the driver who drives the vehicle as a temperature distribution image;
Traffic volume detection means for detecting traffic volume around the vehicle when the vehicle travels;
Road linear complexity detection means for detecting the road linear complexity in front of the vehicle when the vehicle travels;
Fractal dimension calculating means for calculating a fractal dimension by fractal analysis of the temperature distribution image measured by the facial temperature distribution measuring means;
Driving load determination means for determining the driving load of the driver based on the number of dimensions of the fractal dimension calculated by the fractal dimension calculation means,
The driving load determination means is a fractal dimension calculation means when at least one of the traffic volume detected by the traffic volume detection means and the road linearity complexity detected by the road linearity complexity detection means changes. The driver state evaluation apparatus, wherein the driving load determination method is changed according to the calculated amount of change in the number of dimensions of the fractal dimension. A region dividing unit that divides the temperature distribution image measured by the facial skin temperature distribution measuring unit into regions of a predetermined area;
The fractal dimension calculating means calculates a fractal dimension for each area divided by the area dividing means,
2. The driver state according to claim 1, wherein the driving load determination unit determines the degree of driving load on the driver based on a correlation between fractal dimensions in each region divided by the region dividing unit. Evaluation device.   The fractal in each region divided by the region dividing unit when at least one of the traffic detected by the traffic detecting unit and the road linear complexity detected by the road linear complexity detecting unit changes. 3. The driver state according to claim 2, further comprising load determination change means for changing the determination of the driving load in the driving load determination means according to the amount of change in the fractal dimension calculated by the dimension calculation means. Evaluation device.   In a predetermined time interval, when the traffic volume detected by the traffic volume detection unit and the road linearity complexity detected by the road linearity complexity detection unit do not change, the fractal is determined for the area divided by the area dividing unit. The determination area determination means for determining the area where the fractal dimension calculated by the dimension calculation means changes as the target of determination of the driving load by the driving load determination means is further provided. The driver condition evaluation device described. Further comprising divided region characteristic storage means,
The determination area determination means, when determining the determination area,
In the traffic volume detected by the traffic volume detecting means and the road linearity complexity detected by the road linear complexity detecting means, the fractal dimension calculating means calculates when at least one of the changes is smaller than a predetermined value. The segmented region where the fractal dimension value changes is determined as the sensitive segmented region,
When the change in at least one of the detected traffic volume and the detected road linearity complexity is larger than a predetermined value, the insensitive division is performed on the divided area in which the fractal dimension value calculated by the fractal dimension calculating unit changes. Determined as an area,
The driver state evaluation apparatus according to claim 4, wherein the divided area characteristic storage means stores the determined sensitive divided area and the insensitive divided area separately. When at least one of the traffic volume detected by the traffic volume detection means and the road linearity complexity detected by the road linear complexity detection means changes, the fractal dimension for the area divided by the area dividing means is changed. A determination region determination unit that determines a region in which the fractal dimension calculated by the calculation unit is changed as a target of determination of the driving load by the driving load determination unit;
When at least one of the traffic volume detected by the traffic volume detection means and the road linearity complexity detected by the road linear complexity detection means changes, the fractal dimension for the area divided by the area dividing means is changed. A non-determination area determining means for determining an area where the fractal dimension calculated by the calculating means is not changed by the driving load determining means as being excluded from the determination of the driving load;
The driver state evaluation apparatus according to claim 2, further comprising:   The driving load determination means targets each divided area divided by the divided area means included in the determination area determined by the determination area determination means when determining the driving load. 7. The sum of the fractal dimensions calculated by the fractal dimension calculating means is obtained for each divided region, and it is determined that the driving load on the driver is larger as the obtained sum is smaller. The driver state evaluation apparatus according to any one of the above. Processing for determining the determination region by the determination region determination means;
For the determination region determined by the determination region determination means, the fractal dimension calculation means calculates the fractal dimension so that the driving load determination means determines the driving load. The driver state evaluation apparatus according to any one of claims 4 to 7, further comprising a process switching means for selectively switching both processes. The process switching means includes
If at least one of the traffic volume detected by the traffic volume detection means and the road linearity complexity detected by the road linearity complexity detection means is within a predetermined range within a preset time, the determination Let the area determination means perform the process of determining the determination area,
When at least one of the detected traffic volume and the detected change in complexity is not within a predetermined range, the driving load determination means performs processing for determining the driving load. The driver state evaluation apparatus according to claim 8. Further comprising instruction means for a driver to instruct the switching operation of the process switching means;
The driver state evaluation apparatus according to claim 8, wherein the switching unit is configured to perform a switching operation in accordance with an instruction from the instruction unit. Weight division means for assigning a weight value to each divided area divided by the area division means according to its importance,
The weighting means is
When at least one of the traffic volume detected by the traffic volume detecting means and the road linear complexity detected by the road linear complexity detecting means is small or small, the divided area characteristic storage means sets the sensitive divided areas as the sensitive divided areas. Giving a large weight to the stored divided areas,
When at least one of the traffic volume detected by the traffic volume detection unit and the road linearity complexity detected by the road linearity complexity detection unit is large or large, the divisional region characteristic storage unit defines the insensitive divisional region. Giving a large weight to the stored divided areas,
Furthermore, the determination of the driving load in the driving load determination means is performed by taking into account the weight value assigned to each of the divided areas by the weight applying means. The driver state evaluation apparatus in any one of. The area of the divided region divided by the region dividing means is
When at least one of the traffic volume detected by the traffic volume detection means and the road linearity complexity detected by the road linear complexity detection means is small or small, it is divided so as to increase.
12. The method according to any one of claims 2 to 11, wherein at least one of the traffic volume to be detected and the complexity of the road alignment to be detected is large or large so that the traffic volume is small. The driver condition evaluation device described. A nose part extracting means for extracting an area including the driver's nose from among the divided areas divided by the area dividing means,
The determination area determined by the determination area determination means is the driver's nose extracted by the nose extraction means,
The determination of the driving load of the driver by the driving load determination means determines that the driving load of the driver is larger as the fractal dimension of the determination region decreases. The driver state evaluation apparatus according to any one of the above.   The driving load determining means determines that the driving load when the fractal dimension value calculated by the fractal dimension calculating means has changed by a predetermined value or more in the divided area determined as non-determined by the non-determining area determining means. The driver state evaluation apparatus according to any one of claims 5 to 13, wherein the process for determining the vehicle is not performed. It further comprises illuminance detection means for detecting the illuminance of the light directly irradiated on the driver's face,
When the illuminance detected by the illuminance detection means is equal to or greater than a threshold value that gives an error in the determination of the determination area determined by the determination area determination means, at least the driving load determination means does not perform a process of determining the driving load. The driver state evaluation apparatus according to any one of claims 4 to 14, wherein the driver state evaluation apparatus is configured as described above. Wind detection means for detecting the wind direction and air volume in the vehicle interior;
When the wind direction and the air volume detected by the wind detection means give an error to the determination of the determination area determined by the determination area determination means, at least the operation load determination means does not perform the process of determining the operation load. The driver state evaluation device according to any one of claims 4 to 15, wherein the driver state evaluation device is configured as follows.

Images