JP2001331900A - On-vehicle danger forecast alarm system and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To issue a proper alarm to the driver of a vehicle for preventing a traffic accident or a dangerous state which can happen from a present situation during driving the vehicle. SOLUTION: Traffic situations which can cause the traffic accident or the dangerous state are stored as reference picture recognition information in advance. A picture recognition means 2 catches visual information in the periphery of the vehicle to identify and recognize a road, people, vehicles buildings, etc., to make picture recognition information. Picture recognition information is compared with reference picture recognition information by a picture recognition information comparing means 5. When they are matched with each other, an alarm means 4 calls the attention of the driver. Thus, the proper alarm for preventing the traffic accident or the dangerous state can be issued.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an in-vehicle danger prediction / warning apparatus, and more particularly to an in-vehicle danger prediction / warning system which issues a warning based on image recognition when a driver of a vehicle may have a traffic accident or a dangerous state. It relates to an alarm device.

[0002]

2. Description of the Related Art As a conventional in-vehicle danger prediction / warning system, there is an in-vehicle speed warning system which issues a warning when a traveling speed exceeds a predetermined value. There is also an inter-vehicle distance alarm system that measures the inter-vehicle distance to a vehicle in front with a millimeter-wave radar or the like and issues an alarm when the distance falls below a predetermined value.
With these in-vehicle danger prediction warning systems, the driver can reduce the speed and increase the distance between vehicles,
Dangerous situations can be avoided beforehand.

[0003]

However, the conventional on-vehicle danger prediction / warning system does not consider conditions other than the traveling speed and the inter-vehicle distance. Even when it is determined that there is a possibility that a traffic accident or a dangerous state may occur, a warning cannot be issued, and there is a problem that it is insufficient to prevent an accident or a danger in advance.

In a conventional vehicle speed warning system,
A warning is issued when the traveling speed exceeds a predetermined value set in advance, so even if you drive continuously on a road with a different speed limit, the warning will be issued at a uniform overspeed, and the flexibility of the warning will be inflexible. was there.

[0005] The present invention solves the above-mentioned conventional problems, and when it is determined that there is a possibility that a traffic accident or a dangerous state may occur from the surroundings of the own vehicle such as a road, a pedestrian or another vehicle,
Provide an in-vehicle traveling speed excess warning system that can issue an alarm, and even when traveling continuously on roads with different speed limits, accurately notify the driver that the traveling speed has exceeded the speed limit and issue an alarm. The purpose is to do.

[0006]

In order to solve the above-mentioned problems, according to the present invention, an in-vehicle danger prediction / warning apparatus is provided which is capable of recognizing a traffic situation image around a vehicle and outputting image recognition information relating to the traffic situation. Recognition means, reference image recognition information storage means for storing reference image recognition information relating to dangerous traffic conditions in advance, warning means for alerting the driver, and a comparison result between the image recognition information and the reference image recognition information. And an image recognition information comparison means for issuing an alarm based on the alarm means based on the information.

[0007] With this configuration, when it is determined that a traffic accident or a dangerous state may occur based on the surroundings of the vehicle such as a road, a pedestrian, and another vehicle, the driver is warned. Can be issued.

Further, the image recognition information comparing means is provided with a means for changing a warning method according to the degree of similarity between the image recognition information and the reference image recognition information. With this configuration, it is determined that the higher the degree of similarity, the higher the possibility that a traffic accident or a dangerous state will occur, and a larger alarm can be issued to the driver.

Further, a means is provided for acquiring vehicle traveling information relating to the traveling position, traveling speed, time and weather of the vehicle, and a means for changing an alarm method based on the vehicle traveling information is provided for the image recognition information comparing means. With this configuration, it is possible to issue an alarm to the driver based on the possibility of a traffic accident or a dangerous state that differs depending on the current position, speed, season, time, weather, and temperature.

[0010] Further, means for acquiring driver information relating to the driver's skill and experience is provided, and the image recognition information comparing means includes:
A means for changing the warning method based on the driver information is provided. With this configuration, it is determined that the greater the degree of immaturity of the driving technique, the degree of fatigue, and the degree of unfamiliarity of the driving route, the higher the possibility that a traffic accident or a dangerous state will occur. A big alarm can be issued.

[0011] Further, an external communication means for communicating with the outside to acquire new reference image recognition information is provided, and the reference image recognition information storage means is provided with means for storing new reference image recognition information. With this configuration, it is possible to predict a traffic accident or a dangerous state in more traffic events, and it is possible to issue more useful warnings to the driver.

Further, the reference image recognition information storage means is provided with means for storing information relating to a traffic sign on the road or a traffic sign along the road as reference image recognition information. With such a configuration, it is possible to warn the driver of the speed limit of the road on which the vehicle is traveling, the presence of a pedestrian crossing ahead, and the like.

The image recognition information comparing means includes means for acquiring the speed limit of the road on which the vehicle is traveling based on the comparison result between the reference image recognition information and the image recognition information; and comparing the speed limit with the vehicle speed. Means for alarming based on the result is provided. With this configuration, the driver can be alerted to drive at a safe traveling speed.

Further, the alarm means is provided with means for displaying the speed limit, the traveling speed, and the excess of the speed limit. With this configuration, it is possible to accurately inform the driver how much the traveling speed exceeds the speed limit of the road on which the vehicle is traveling.

[0015] The image recognition information comparing means may include means for acquiring a travelable direction of a traveling road based on a comparison result between the reference image recognition information and the image recognition information, and means for acquiring a traveling direction of the vehicle. Means for issuing an alarm based on the result of comparison between the direction in which the vehicle can travel and the traveling direction. With this configuration, it is possible to inform the driver whether the right / left turning direction of the vehicle is correct with respect to the right / left turning direction of the road on which the vehicle is traveling.

Also, there is provided means for presetting alarm control information comprising at least one of an alarm reference such as a current position and a traveling speed, an alarm rank for each type of alarm, and alarm selection information for selecting the type of alarm. The image recognition information comparing means is provided with a means for changing a warning method based on warning control information. With this configuration, an alarm can be issued using only the alarm control information desired by the driver from the plurality of pieces of alarm control information, and the flexibility can be increased.

[0017]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below in detail with reference to FIGS.

(First Embodiment) In a first embodiment of the present invention, a traffic situation in which a traffic accident or a dangerous state may occur is stored as reference image recognition information in advance. This is an in-vehicle danger prediction warning system that generates image recognition information on traffic conditions by recognizing a traffic situation image around a vehicle and generates an alarm based on a result of comparing the image recognition information with reference image recognition information.

FIG. 1 is a functional block diagram of an in-vehicle danger prediction warning system according to a first embodiment of the present invention.
In FIG. 1, an in-vehicle danger prediction / warning system 1 is an in-vehicle device that issues a warning to a driver of a vehicle based on image recognition that a traffic accident or a dangerous state may occur. The image recognizing means 2 is a means for recognizing and processing a traffic situation image around the vehicle to identify and recognize a road, a person, a car, a building, and the like. The reference image recognition information accumulating means 3 is means for accumulating traffic conditions in which a traffic accident or a dangerous state may occur as reference image recognition information in advance. The warning unit 4 is a unit that alerts the driver of the car when a traffic accident or a dangerous state may occur. The image recognition information comparison unit 5 is a unit that compares the image recognition information from the image recognition unit 2 with the reference image recognition information from the reference image recognition information storage unit 3.

The position detecting means 6 is a means for detecting the current position of the vehicle. The speed detecting means 7 is a means for detecting the running speed of the vehicle. The date and time information acquiring means 8 is means for acquiring date and time information such as what month, what day, what time, and what minute. The weather information detecting means 9 is a means for detecting weather information near the traveling position of the car. The driver information setting means 10 is a means for setting information such as a driving skill, a visual acuity, and a degree of fatigue of a driver of the car. The route familiarity setting means 11 is a means for setting the familiarity information of the driver of the vehicle with respect to the traveling route to the destination. The left / right turn direction setting means 12 is means for setting the right / left turn direction of the vehicle.

The alarm control setting means 13 is a means for selecting and specifying a necessary alarm control standard from a plurality of alarm control standards. The alarm rank setting means 14 is a means for setting an alarm rank for each type of alarm. The alarm type setting means 15 is a means for selecting and specifying a necessary alarm type from a plurality of alarm types. The out-of-vehicle communication means 16 is means for taking in new reference image recognition information into the reference image recognition information storage means 3 from outside the vehicle.

FIG. 2 is a diagram showing a traffic situation when traveling on a right curve. FIG. 2A shows a traffic situation corresponding to the accumulated reference image recognition information, and FIG. Is a traffic condition corresponding to the reference image recognition information accumulated in association with the traffic condition predicted from the traffic condition in FIG.

The on-vehicle danger prediction warning system according to the first embodiment of the present invention configured as described above,
The operation will be described with reference to FIGS. Car, Fig. 2
As shown in (a), the vehicle is traveling on a road. There is a drive-in with a parking lot near the front right curve. From the drive-in parking lot, another vehicle is about to enter the oncoming lane.

The image recognizing means 2 shown in FIG. 1 captures this situation, and recognizes that a right curve is ahead of the road and that another vehicle is heading from the parking lot to the road near the curve. These are sent to the image recognition information comparison means 5 as image recognition information.

On the other hand, the reference image recognition information storage means 3
A traffic situation in which a traffic accident or a dangerous state may occur is stored in advance as reference image recognition information that identifies and recognizes a road, a person, a car, a building, and the like. In this case, the reference image recognition information as shown in FIG. 2A and the reference image recognition information as shown in FIG. 2B in which a traffic accident or a dangerous state is predicted are stored in association with each other. .

The image recognition information comparing means 5 stores the image recognition information from the image recognition means 2 and the reference image recognition information storage means 3
Is compared with the reference image recognition information. If they match, a warning is issued to the driver by the warning means 4. For example, a warning is issued by voice synthesis or the like, "Be careful of the approach of another vehicle before the right curve, and be careful of the right turn of the oncoming vehicle after the curve." As a result, the driver is distracted by the approach of another vehicle in front of the right curve that looks ahead, and after turning the curve, turns right on another vehicle on the opposite lane that is waiting.
It can detect dangerous situations that are difficult to predict in advance,
It is possible to prevent an unexpected situation from occurring. In particular, it is very difficult for a driver with little driving experience to detect such a dangerous state in advance, which is very effective in preventing a traffic accident.

In such a traffic situation, if no other vehicle is found in the drive-in parking lot, the image recognition information comparison means 5 refers to the image recognition information from the image recognition means 2 to refer to the image recognition information. The comparison with the reference image recognition information from the image recognition information storage means 3 is performed, and they do not match, but it is determined that the similarity is high. And warn them differently. As described above, even if the traffic condition is not necessarily according to the traffic condition previously stored in the reference image recognition information storage unit 3, the warning according to the magnitude of the similarity is issued to the driver, so that the limited reference image recognition information can be obtained. Therefore, it is possible to issue more useful alarms to the driver.

Further, new reference image recognition information is acquired from outside by the external communication means 16 and stored in the reference image recognition information storage means 3. The external communication means 16 may be a wireless communication device such as a mobile phone or a PHS, or a CD-RO.
M and an information recording medium such as a semiconductor memory. As a result, it is possible to predict a traffic accident or a dangerous state in more traffic events, and it is possible to issue more useful warnings to the driver.

As described above, in the first embodiment of the present invention, the in-vehicle danger prediction / warning system stores in advance a traffic situation in which a traffic accident or a dangerous state may occur as reference image recognition information. In addition, capturing visual information around the driving vehicle, identifying and recognizing roads, people, cars, buildings, etc., generating image recognition information, and comparing the image recognition information with the reference image recognition information A warning is issued based on the road, pedestrians, other vehicles, and other surroundings of the vehicle. Can be emitted.

(Second Embodiment) In a second embodiment of the present invention, a traffic sign on the road or a traffic sign along the road is stored in advance as reference image recognition information, and the road on which the vehicle is traveling is stored. This is an in-vehicle danger prediction warning system that issues a warning based on the result of comparison between the speed limit of the vehicle and the traveling speed of the vehicle.

The basic configuration of the vehicle-mounted danger prediction / warning system according to the second embodiment of the present invention is the same as that of the first embodiment shown in FIG. The second embodiment is the first
The difference from this embodiment is that reference image recognition information such as a traffic sign is stored in the reference image recognition information storage means 3 in advance. FIG. 3A is a diagram illustrating a traffic situation when a road sign is image-recognized, and FIG. 3B is a diagram illustrating a traffic situation when a road sign is image-recognized. FIGS. 4A and 4B show a traffic situation when traveling on a road having a pedestrian crossing ahead and a traffic situation when predicted.

The operation of the vehicle-mounted danger prediction / warning system according to the second embodiment of the present invention configured as described above will be described. A traffic sign on the road or a traffic sign along the road is stored in advance in the reference image recognition information storage means 3 as reference image recognition information.

As shown in FIG. 3 (a), a car is traveling in front of a traffic sign "40" on the road. Alternatively, FIG.
As shown in (b), a car is running in front of a traffic sign “30” along the road. In such a case, the image recognition means 2 of FIG. 1 captures the traffic sign or the traffic sign and outputs image recognition information. The image recognition information comparing means 5 compares the image recognition information with the reference image recognition information, and recognizes that the speed limit of the traveling road is 40 km / h or 30 km / h. The warning means 4 warns the driver that the speed limit here is 40 km / h (30 km / h). This can compensate for the driver's lack of recognition of the speed limit due to a traffic sign or a traffic sign being overlooked.

As shown in FIG. 4A, when a car is running in front of a traffic sign "標" on the road, the image recognizing means 2 of FIG. 1 catches the traffic sign and recognizes them. The information is sent to the image recognition information comparison means 5 as information. On the other hand, the reference image recognition information storage means 3 stores in advance traffic conditions in which a traffic accident or a dangerous state may occur as reference image recognition information that identifies and recognizes a road, a person, a car, a building, and the like. . In this case, the reference image recognition information as shown in FIG. 4B and a traffic accident or a dangerous state are predicted in FIG.
The reference image recognition information as shown in FIG.

The image recognition information comparison means 5 in FIG. 1 compares the image recognition information from the image recognition means 2 with the reference image recognition information from the reference image recognition information storage means 3,
Recognizing that there is a pedestrian crossing several tens of meters ahead of this running road, the warning means 4 warns the driver that there is a pedestrian crossing ahead. As a result, the driver can be informed in advance of a traffic situation that may cause a traffic accident, such as a pedestrian crossing ahead and a person crossing or may be crossing. .

Although the road display and the road sign have been described by taking the speed limit and the pedestrian crossing as an example, other objects such as "be careful of a cliff ahead" can be similarly implemented. Thereby, the driver can be urged in advance to receive more useful warnings and cautions.

When the image recognition information comparing means 5 compares the image recognition information with the reference image recognition information and controls the warning means 4, an alarm is issued in accordance with the right / left turn direction information from the right / left turn direction setting means 12. Change the way. The left / right turn direction setting means 12 includes a direction indicator of a car. For example, if the driver is trying to make a U-turn and has set the right / left-turn direction setting means 12 to a right-turn while recognizing that the front intersection store on the traveling road is prohibited from making a U-turn, the message “Here is a U-turn Turn is prohibited. " Alternatively, if the driver does not set the right / left turn direction setting means 12 to a right turn while recognizing that the driving lane of the road is dedicated to a right turn, the driver is told that "this is a right turn only lane". An alarm can be issued.

As described above, in the second embodiment of the present invention, the in-vehicle danger prediction / warning system stores traffic signs on the road and traffic signs along the road in advance as reference image recognition information. A warning is issued based on the result of comparison between the speed limit of the traveling road and the traveling speed of the car, so that the driver is warned that the speed limit of the traveling road or that there is a pedestrian crossing ahead. be able to. Further, it is possible to accurately inform the driver how much the traveling speed exceeds the speed limit of the road on which the vehicle is traveling. Further, it is possible to issue a warning that assists the driver in driving such as exceeding the speed limit or prohibiting the U-turn.

(Third Embodiment) A third embodiment of the present invention is an in-vehicle danger prediction warning system that changes a warning method according to the current position and traveling speed of a vehicle.

The basic configuration of the vehicle-mounted danger prediction warning system according to the third embodiment of the present invention is the same as that of the first embodiment shown in FIG. The third embodiment is the first
The difference from this embodiment is that the image recognition information comparing means 5 is provided with a means for acquiring the speed limit of the traveling road, and the warning means 4 is provided with a speed meter for displaying the speed limit, the running speed and the excess speed. It is a point. FIG. 5 is a diagram showing a speedometer with a speed limit display as a visual warning means.

The operation of the vehicle-mounted danger prediction / warning system according to the third embodiment of the present invention configured as described above will be described. When the image recognition information comparing means 5 in FIG. 1 compares the image recognition information with the reference image recognition information and controls the warning means 4, an alarm is issued according to the current position of the vehicle detected by the position detection means 6. Change the way. As the position detecting means 6,
There are a GPS and a car navigation device using the GPS.

For example, when traveling near an urban area, there is a high possibility that pedestrians, bicycles and other vehicles are present, and the risk of causing a traffic accident increases. Even with such a relatively low alarm level, an alarm is always issued to the driver. If you are traveling near the suburbs, it is unlikely that pedestrians, bicycles or other vehicles will be present, and the risk of causing a traffic accident is small, so a warning such as `` There is a pedestrian crossing ahead '' is issued. do not do.

When the image recognition information comparison means 5 compares the image recognition information with the reference image recognition information and controls the warning means 4, the warning method is determined according to the traveling speed detected by the speed detection means 7. Change. Examples of the speed detecting means 7 include a detector using a vehicle speed pulse.

For example, when driving at a high speed, there is a high possibility that the road sign will be overlooked and the risk of causing a traffic accident increases, so that a relatively warning such as “There is a pedestrian crossing ahead” is issued. Even if the warning level is low, a warning is always issued to the driver. If the vehicle is traveling at a reduced speed, it is unlikely that a road sign will be overlooked and the risk of causing a traffic accident is small, so that a warning such as “a pedestrian crossing ahead” is not issued.

As shown in FIG. 3B, when the vehicle is running just before the traffic sign "30" along the road, the image recognition information comparing means 5 of FIG. By comparing the recognition information, it is recognized that the speed limit of the traveling road is 30 km / h. Compared with the speed information from the speed detecting means 7, the warning means 4 gives the driver "30 km / h speed limit, slow down" or "20 km speed exceeded, slow down. Please ". Thus, the driver can be alerted to drive at a safe traveling speed within the speed limit. Moreover, unlike the conventional vehicle speed warning method that alerts the driver that the vehicle is overspeed only when the vehicle speed exceeds the preset speed, the speed limit of the road on which the vehicle is traveling is automatically recognized by image recognition. In addition, it is possible to accurately inform the driver how much the speed limit differs depending on the road.

The warning means 4 for notifying the driver of the traveling speed of the vehicle with respect to the speed limit is as follows.
It may be done visually rather than aurally. This example is shown in FIG.
Shown in FIG. 5A is a mechanical speed meter, and FIG.
Is an electronic speed meter with a speed limit of 30k / h
m and a running speed of 50 km. FIG.
In the mechanical speed meter of (a), the short hand indicates the speed limit, the long hand indicates the running speed, and the difference between the short hand and the long hand indicates the excess of the running speed with respect to the speed limit. In the electronic speed meter shown in FIG. 5B, the boundary between blue and red indicates the speed limit, the rightmost portion of red and the lower number indicate the running speed, and the red portion indicates the excess of the running speed with respect to the speed limit. Represents. When the speed limit of the traveling road changes, the short hand or the boundary between blue and red moves. This allows the driver to visually know at a glance the speed limit of the road on which the vehicle is currently running, the running speed, and the excess of the running speed with respect to the speed limit, and encourages the driver to drive at a safe running speed. Can be.

As described above, in the third embodiment of the present invention, the on-vehicle danger prediction warning system is configured to change the warning method according to the current position and the traveling speed of the vehicle. An alarm can be issued according to the traveling position and speed of the vehicle. Judging that the higher the speed of the car, the more likely it is that a traffic accident or dangerous situation will occur,
A larger alarm can be issued.

(Fourth Embodiment) The fourth embodiment of the present invention is an in-vehicle danger prediction warning system that changes a warning method according to date and time information and weather information.

The basic configuration of the vehicle-mounted danger prediction / warning system according to the fourth embodiment of the present invention is the same as that of the first embodiment shown in FIG. The fourth embodiment is the first
The difference from this embodiment is that risk information corresponding to date and time and weather is set in the image recognition information comparison means 5.

The operation of the vehicle-mounted danger prediction / warning system according to the fourth embodiment of the present invention configured as described above will be described. When the image recognition information comparing means 5 in FIG. 1 compares the image recognition information with the reference image recognition information and controls the warning means 4, according to the date and time information from the date and time information acquiring means 8,
Change the way of alerting.

For example, when traveling during the time of commuting to work, there is a high possibility that pedestrians, bicycles and other vehicles are present, and the risk of causing a traffic accident increases. Even if the alarm is relatively low, such as "Yes", the driver will always be warned. If you are traveling early in the morning or late at night, it is unlikely that pedestrians, bicycles, or other vehicles will be present, and there is less risk of a traffic accident. No such alarm is issued.

When the image recognition information comparison means 5 compares the image recognition information with the reference image recognition information and controls the warning means 4, the warning method is changed according to the weather information from the weather information detection means 9. . Examples of the weather information detecting means 9 include a temperature sensor, a humidity sensor, and a raindrop sensor.

For example, when traveling in rainy weather,
"There is a pedestrian crossing ahead" as it is more likely to overlook road markings and increase the risk of a traffic accident
Even if the warning level is relatively low,
Always warn the driver. When driving in fine weather, there is a low possibility of overlooking the road sign and the danger of causing a traffic accident is small, so that a warning such as "a pedestrian crossing ahead" is not issued.

As described above, in the fourth embodiment of the present invention, the in-vehicle danger prediction warning system is configured to change the warning method according to the date and time and the weather. The driver can be alerted to the possibility of a traffic accident or a dangerous state depending on the time, weather, or temperature.

(Fifth Embodiment) The fifth embodiment of the present invention is an in-vehicle danger prediction warning system that changes a warning method according to driver information.

The basic configuration of the vehicle-mounted danger prediction / warning system according to the fifth embodiment of the present invention is the same as that of the first embodiment shown in FIG. The fifth embodiment is the first
The difference from this embodiment is that the risk information corresponding to the driver information is set in the image recognition information comparing means 5.

The operation of the vehicle-mounted danger prediction / warning system according to the fifth embodiment of the present invention configured as described above will be described. When the image recognition information comparing means 5 compares the image recognition information with the reference image recognition information and controls the warning means 4, when the driving information from the driver information setting means 10 such as the driving skill and the degree of fatigue of the driver is used. Change the warning method according to the person information. The driver information setting means 10 includes a method using a display monitor and a remote controller.

For example, when the driving skill of the driver is inexperienced or the degree of fatigue is large, there is a high possibility that the road sign will be overlooked and the risk of causing a traffic accident increases.
Even if the warning level is relatively low, such as "There is a pedestrian crossing ahead," the driver always be warned. If the driving skills are mature and the degree of fatigue is small, there is no possibility of overlooking the road sign and the risk of causing a traffic accident is small, so there is no warning such as `` There is a pedestrian crossing ahead '' . The driver information is not limited to the driving technique and the degree of fatigue, but may be other information such as visual acuity or frequency of driving such as whether driving is daily or after a long time.

When the image recognition information comparing means 5 compares the image recognition information with the reference image recognition information and controls the warning means 4, an alarm is issued according to the route familiarity information from the route familiarity setting means 11. Change the way. Examples of the route familiarity setting means 11 include a method using a display monitor and a remote controller.

For example, when a driver has never experienced a driving route for the first time or has little experience, the possibility of overlooking a road sign and the danger of causing a traffic accident increase, so that "a U-turn ahead" Even if the warning level is relatively low, such as "Prohibition," the driver will always be warned. If the driver is well experienced in the driving route of the car, the warning such as “U-turn is prohibited in the future” is not issued because the possibility of overlooking the road sign is low and the risk of causing a traffic accident is low.

As described above, in the fifth embodiment of the present invention, the in-vehicle danger prediction / warning system is configured to change the warning method according to the driver information. The greater the degree of fatigue and the degree of unfamiliarity of the traveling route, the higher the possibility that a traffic accident or a dangerous state will occur, and can issue a larger warning to the driver.

(Sixth Embodiment) A sixth embodiment of the present invention is an in-vehicle danger prediction warning system that changes a warning method according to set warning control information.

The basic configuration of the vehicle-mounted danger prediction / warning system according to the sixth embodiment of the present invention is the same as that of the first embodiment shown in FIG. The sixth embodiment is the first
The difference from the embodiment is that the image recognition information comparing means 5 is provided with a means for controlling the warning means 4 according to the warning control information.

FIG. 6 is a diagram showing an alarm control selection screen for selecting alarm control information in advance. FIG.
It is the figure which showed the alert rank selection screen for selecting the alert rank according to the kind of alert in advance. FIG. 8 is a diagram showing an alarm type selection screen for selecting an alarm type in advance.

The operation of the vehicle-mounted danger prediction / warning system according to the sixth embodiment of the present invention configured as described above will be described. The alarm control setting means 13 sets in advance the alarm control information desired by the driver from the plurality of pieces of alarm control information. When the image recognition information comparing means 5 compares the image recognition information with the reference image recognition information and controls the alarm means 4, an alarm method is performed in accordance with only the alarm control information set by the alarm control setting means 13. change. Alarm control setting means
As the method 13, there is a method using a display monitor and a remote controller.

For example, as shown in FIG. 6, if only the traveling speed control is set to ON, a warning is issued to the driver in response to only the traveling speed of the vehicle. When a plurality of drivers alternately drive the same car, an alarm can be issued only with the alarm control information desired by the driver by resetting the settings for each driver. In addition, even the same driver can change the time of driving the car, and the flexibility can be increased.

The alarm rank setting means 14 presets an alarm rank desired by the driver for each of a plurality of types of alarms. When the image recognition information comparing means 5 compares the image recognition information with the reference image recognition information and controls the alarm means 4, an alarm method according to the type of alarm rank from the alarm rank setting means 14 is used. change. Examples of the alarm rank setting means 14 include a method using a display monitor and a remote controller.

For example, as shown in FIG. 7, oncoming vehicle attention, speed limit attention, and U-turn attention are set to a high warning rank, and left road attention and one-way attention are set to a low warning rank.
When traveling speed is fast, all of oncoming vehicle attention, speed limit attention, U-turn attention, left road attention, and one-way attention,
Alerts you when traffic conditions are appropriate. If the traveling speed is slow, an alert with only oncoming traffic, speed limit warning, and U-turn warning is issued in appropriate traffic conditions.

The alarm type setting means 15 presets the type of alarm desired by the driver from a plurality of alarm types. When the image recognition information comparing means 5 compares the image recognition information with the reference image recognition information and controls the warning means 4, when the warning type is set by the alarm type setting means 15, the warning method is used. change. Examples of the alarm type setting means 15 include a method using a display monitor and a remote controller.

For example, as shown in FIG. 8, only the oncoming vehicle caution and the U-turn caution are set to ON. Even in a traffic situation where there is a pedestrian crossing ahead, there should be no warning that there is a pedestrian crossing ahead, but only an oncoming vehicle warning and a U-turn warning. Emitted during traffic conditions.

Although the types of alarms and the alarm ranks have been described with reference to specific examples, other types of alarms necessary for safely driving a car may be added, or the alarm ranks may be set in more stages. . Thereby, more safety and flexibility can be given to the driver.

As described above, in the sixth embodiment of the present invention, the in-vehicle danger prediction warning system is configured to change the warning method according to the set warning control information.
An alarm can be issued according to the driver's request, and flexibility can be increased.

[0073]

As is apparent from the above description, according to the present invention, an in-vehicle danger prediction / warning device is provided with image recognition means for recognizing a traffic situation image around a vehicle and outputting image recognition information relating to the traffic situation. Reference image recognition information storage means for storing reference image recognition information relating to dangerous traffic conditions in advance, warning means for calling a driver's attention, and warning means based on a comparison result between the image recognition information and the reference image recognition information It is determined that there is a possibility that a traffic accident or a dangerous state may occur from the surrounding conditions of the own vehicle such as a road, a pedestrian, and another vehicle because the configuration includes the image recognition information comparing means that issues an alarm according to In this case, an effect is obtained that an alarm can be issued to the driver.

Further, since the image recognition information comparing means is provided with a means for changing a warning method in accordance with the degree of similarity between the image recognition information and the reference image recognition information, the higher the similarity, the more the traffic accident or the like. By judging that there is a high possibility that a dangerous state will occur, it is possible to obtain an effect that a larger alarm can be issued to the driver.

Further, a means for acquiring vehicle traveling information relating to the traveling position, traveling speed, time and weather of the vehicle is provided, and a means for changing an alarm method based on the vehicle traveling information is provided in the image recognition information comparing means. An effect is obtained that an alarm can be issued to the driver based on the magnitude of the possibility of a traffic accident or a dangerous state that differs depending on the current position, speed, season, time, weather, and temperature.

A means for acquiring driver information relating to the driver's skill and experience is provided.
Since the means of changing the warning method based on the driver information is provided, the greater the immaturity of the driving technique, the degree of fatigue, and the greater the degree of unfamiliarity with the driving route, the higher the possibility that a traffic accident or a dangerous state will occur. By judging, it is possible to obtain an effect that a larger alarm can be issued to the driver.

Further, the vehicle communication means for obtaining new reference image recognition information by communicating with the outside is provided, and the reference image recognition information storage means is provided with means for storing new reference image recognition information. In this case, it is possible to predict a traffic accident or a dangerous state in a traffic event, and an effect that more useful warnings can be issued to the driver can be obtained.

Further, the reference image recognition information storage means is provided with means for storing information on traffic signs on the road or traffic signs along the road as reference image recognition information.
It is possible to obtain an effect that a driver can be warned of a speed limit of a running road or a pedestrian crossing ahead.

Further, the image recognition information comparison means obtains the speed limit of the road on which the vehicle is traveling based on the comparison result between the reference image recognition information and the image recognition information, and compares the speed limit with the vehicle speed. Since the means for giving an alarm based on the result is provided, an effect is obtained that the driver can be alerted to drive at a safe traveling speed.

Further, since the warning means is provided with means for displaying the speed limit, the running speed and the excess speed limit, it is possible to determine how much the running speed exceeds the speed limit of the road on which the vehicle is running. The effect of being able to accurately inform the driver is obtained.

Further, the image recognition information comparing means includes means for acquiring a travelable direction of the traveling road based on a comparison result between the reference image recognition information and the image recognition information, and means for acquiring the traveling direction of the vehicle. A means for giving an alarm based on the result of comparison between the direction in which the vehicle can travel and the traveling direction is provided. Is obtained.

A means for presetting alarm control information comprising at least one of an alarm criterion such as a current position and a traveling speed, an alarm rank for each type of alarm, and alarm selection information for selecting an alarm type is provided. Since the image recognition information comparing means is provided with means for changing the method of alarming based on the alarm control information, it is possible to issue an alarm using only the alarm control information desired by the driver from among a plurality of alarm control information. Thus, the effect of increasing the flexibility of the alarm can be obtained.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of an in-vehicle danger prediction warning system according to a first embodiment of the present invention;

FIGS. 2A and 2B show a traffic condition corresponding to reference image recognition information stored in a vehicle-mounted danger prediction and warning system according to the first embodiment of the present invention, and a traffic condition diagram showing a predicted traffic condition; ),

FIGS. 3A and 3B are diagrams illustrating a state where a traffic sign on a road is image-recognized and a diagram illustrating a state where a road sign is image-recognized in the vehicle-mounted danger prediction warning system according to the second embodiment of the present invention; b),

FIGS. 4A and 4B are diagrams illustrating a state in which an in-vehicle danger prediction warning system according to a second embodiment of the present invention recognizes an image of a notice road sign of a pedestrian crossing, and a diagram illustrating a predicted traffic condition; ),

FIGS. 5A and 5B show a speed meter with a mechanical speed limit display and a speed meter with an electronic speed limit display in a vehicle-mounted danger prediction warning system according to a third embodiment of the present invention; FIGS. ,

FIG. 6 is a diagram showing an alarm control selection screen of a vehicle-mounted danger prediction alarm system according to a sixth embodiment of the present invention;

FIG. 7 is a diagram of an alarm rank selection screen of a vehicle-mounted danger prediction alarm system according to a sixth embodiment of the present invention;

FIG. 8 is a diagram of an alarm type selection screen of a vehicle-mounted danger prediction alarm system according to a sixth embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 In-vehicle danger prediction warning system 2 Image recognition means 3 Reference image recognition information storage means 4 Warning means 5 Image recognition information comparison means 6 Position detection means 7 Speed detection means 8 Date and time information acquisition means 9 Weather information detection means 10 Driver information setting means 11 Route familiarity setting means 12 Right / left turn direction setting means 13 Alarm control setting means 14 Alarm rank setting means 15 Alarm type setting means 16 Outside vehicle communication means

Claims (20)

[Claims] An image recognizing means for recognizing a traffic situation image around a vehicle and outputting image recognition information on a traffic situation, and accumulating reference image recognition information on a dangerous traffic situation in advance. Means, warning means for calling a driver's attention, and image recognition information comparison means for issuing a warning by the warning means based on a comparison result between the image recognition information and the reference image recognition information. In-vehicle danger prediction warning device. 2. The apparatus according to claim 1, wherein said image recognition information comparing means includes means for changing a warning method according to the degree of similarity between said image recognition information and said reference image recognition information. In-vehicle danger prediction warning device according to the description. 3. A means for acquiring vehicle traveling information relating to a current position, a traveling speed, a time, and weather of the vehicle, and a means for changing a warning method based on the vehicle traveling information is provided in the image recognition information comparing means. 2. The method according to claim 1, wherein
In-vehicle danger prediction warning device according to the description. 4. A means for acquiring driver information relating to a driver's skill and experience, wherein said image recognition information comparing means includes:
2. The in-vehicle danger prediction warning device according to claim 1, further comprising means for changing a warning method based on the driver information. 5. An external communication means for acquiring new reference image recognition information by communicating with the outside, wherein said reference image recognition information storage means includes means for storing said new reference image recognition information. The in-vehicle danger prediction warning device according to claim 1, wherein: 6. The on-vehicle vehicle according to claim 1, wherein said reference image recognition information storage means includes means for storing information relating to a traffic sign on a road or a traffic sign along the road as reference image recognition information. Danger prediction warning device. 7. A means for acquiring a speed limit of a traveling road on the basis of a comparison result between the reference image recognition information and the image recognition information; 7. An in-vehicle traveling speed excess warning device according to claim 6, further comprising means for issuing an alarm based on a result of comparison with the speed. 8. The in-vehicle traveling speed excess warning device according to claim 7, wherein said warning means is provided with a means for displaying said speed limit, said traveling speed and an excess of the speed limit. 9. The image recognition information comparison means for obtaining a traveling direction of a traveling road based on a comparison result between the reference image recognition information and the image recognition information, and acquiring a traveling direction of a vehicle. 7. The on-vehicle danger prediction warning device according to claim 6, further comprising: means for performing a warning, and means for warning based on a comparison result between the possible traveling direction and the traveling direction. 10. A means for presetting alarm control information comprising at least one of alarm criteria such as a current position of a vehicle and a traveling speed, an alarm rank for each type of alarm, and alarm selection information for selecting an alarm type. 2. The in-vehicle danger prediction warning device according to claim 1, wherein the image recognition information comparison unit includes a unit that changes a warning method based on the warning control information. 11. Reference image recognition information relating to a dangerous traffic situation is stored in advance, and a traffic situation image around a vehicle is subjected to recognition processing to generate image recognition information relating to a traffic situation, and the image recognition information and the reference image are generated. A danger prediction warning method characterized by issuing a warning based on a result of comparison with recognition information. 12. The danger prediction warning method according to claim 11, wherein a warning method is changed according to a degree of similarity between the image recognition information and the reference image recognition information. 13. The method according to claim 1, further comprising acquiring vehicle traveling information relating to a current position, a traveling speed, a time, and weather of the vehicle, and changing a warning method based on the vehicle traveling information.
1. The danger prediction warning method according to 1. 14. The danger prediction warning method according to claim 11, wherein driver information relating to the driver's skill and experience is acquired, and a warning is issued by changing a warning method based on the driver information. 15. The danger prediction warning method according to claim 11, wherein new reference image recognition information is acquired from outside and stored. 16. The danger prediction warning method according to claim 11, wherein a traffic sign on the road or a traffic sign along the road is stored in advance as reference image recognition information. 17. A speed limit for a traveling road is acquired based on a comparison result between the reference image recognition information and the image recognition information, and an alarm is issued based on a comparison result between the speed limit and a vehicle speed. 17. The danger prediction warning method according to claim 16, wherein: 18. The danger prediction warning method according to claim 17, wherein the speed limit, the traveling speed, and the speed limit excess are displayed. 19. A traveling direction of a traveling road is acquired based on a comparison result between the reference image recognition information and the image recognition information, a traveling direction of a vehicle is acquired, and the traveling direction and the traveling direction are acquired. 17. The danger prediction warning method according to claim 16, wherein a warning is issued based on a result of comparison with. 20. Alarm control information comprising at least one of alarm criteria such as a traveling position and a traveling speed, an alarm rank for each type of alarm, and alarm selection information for selecting an alarm type, is set in advance. The danger prediction warning method according to claim 11, wherein a warning method is changed based on the warning control information.