JP2007263737A - Navigation apparatus, method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a reliable technique for outputting an alarm related to a signal while placing a less burden thereon. <P>SOLUTION: The technique comprises: a means for recognizing a traffic light based on image data photographed by a camera and distinguishing display colors; a means for calculating/estimating a vehicle stopping distance based on a vehicle speed; a means for calculating a distance from the traffic light based on navigation-use data of a road map and an own vehicle position; and a means for outputting the alarm based on a combination of the distinguished display color and a magnitude relation between the vehicle stopping distance and the distance from the traffic light. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an improvement in navigation technology.

  In recent years, with the spread of technology in information processing and communication, navigation devices are also widely used in automobiles. The navigation device uses pre-prepared road map data, displays the position of the vehicle that is sequentially calculated by GPS or the like on a surrounding map, or calculates and guides a route to a destination specified by a map or facility search. Is.

On the other hand, in recent years, the safety aspect of vehicles has also been regarded as important, and a vehicle collision prevention system based on distance detection using an ultrasonic sensor, a rear confirmation system using a camera to prevent collisions during back travel, etc. have been developed. Is mounted on the vehicle.
JP 2004-199148 A

  However, in terms of safety, it is still not enough to avoid or prevent danger in every situation / situation where the vehicle travels, and there is plenty of room to deal with. Rather, it is no exaggeration to say that a vehicle safety countermeasure system is installed in a vehicle in order to avoid or prevent danger under certain limited conditions, but safety measures are taken.

  Now, one of the traffic accidents is a collision caused by an erroneous approach of a vehicle due to an oversight of a traffic light at an intersection. There is a collision between vehicles, and there is also a collision with a person walking on a pedestrian crossing. If you check the traffic light and take action based on the appropriate situation judgment (stop because the traffic light is red, etc.), it will not result in an accident, but it will result in an accident due to the so-called accident. Furthermore, because the traffic light changes (from yellow to red), an accident may be induced by trying to pass through the intersection forcibly. Surprisingly, at an intersection with no traffic lights, the driver pays attention and slows down the vehicle speed, so even if an accident occurs, it is unlikely that it will be a major accident, but an intersection with a traffic light may sometimes cause catastrophe .

  In recent navigation systems, utilizing integrated integration of navigation, audio, and visual functions to prevent collisions at intersections, the driver is warned by the distance and display color of the traffic lights recognized by the stereo camera installed in the vehicle. Although there are proposals to output (see Patent Document 1), a plurality of camera images are indispensable, and it is difficult to increase the cost and adjust due to the complexity of the structure, the complexity of the parallax calculation by the left and right image matching and the calculation load, etc. In addition, there was a problem such as a decrease in reliability.

  The present invention solves the above-described problems, and an object of the present invention is to provide a low-load and high-reliability technique for outputting an alarm related to a signal.

  Based on the above object, one aspect of the navigation device according to the present invention includes means for recognizing a traffic light from image data captured by a camera and determining its display color, means for estimating and calculating a vehicle stop distance from vehicle speed, Based on the combination of the means for calculating the distance to the traffic light from the vehicle position and road map data, the magnitude relationship between the vehicle stop distance and the distance to the traffic light, and the determined display color And means for performing.

  Another aspect of the present invention is a navigation device that predicts a danger at an intersection and promotes safety to a driver, and stores road map data storing road map data including a road position and a traffic light or an intersection position. An image pickup means attached to the vehicle for picking up an image of the front of the vehicle, an image data conversion means for converting an image picked up by the image pickup means into image data, and a traffic light from the image data converted by the image data conversion means. A traffic light recognition means for recognizing, a display color judgment means for judging the display color of the traffic light recognized by the traffic light recognition means, a speed detection means for detecting the traveling speed of the vehicle, and the speed detection means detected by the speed detection means. Vehicle stop distance calculating means for calculating the vehicle stop distance from the traveling speed, own vehicle position estimating means for estimating the current own vehicle position, and the road map data storage means Based on the stored road map data, the traffic signal position acquisition means for acquiring the position of the traffic signal recognized by the traffic signal recognition means, the position of the traffic signal acquired by the traffic signal position acquisition means, and the vehicle position estimation A traffic light distance calculating means for calculating a distance from the own vehicle to the traffic light based on the host vehicle position estimated by means; a distance to the traffic light calculated by the traffic light distance calculating means; and the vehicle stop A difference calculating means for calculating a difference from the vehicle stop distance calculated by the distance calculating means; and the difference calculated by the difference calculating means is equal to or less than a set value and the display of the traffic light by the display color determining means. When the color is determined to be red or yellow, the arithmetic control unit of the computer or the electronic circuit realizes an alarm control means for outputting an alarm to alert the driver. That.

  Like these, when the vehicle approaches the intersection, the traffic light is recognized by the image of a single camera provided on the vehicle, the display color is also judged, the color is red or yellow, and the stop distance calculated from the vehicle speed is navigation By outputting a warning when the distance to the traffic signal calculated from the vehicle position and road map data is exceeded, the structure and control calculations are simplified, and high reliability is realized with less burden.

Another aspect of the present invention further includes means for detecting each display color and position of the traffic light from the image data, and the traffic light recognition means detects the traffic light based on the detected arrangement of the display colors. It is characterized by recognition.
Thus, by detecting the display colors of blue, yellow, and red and their positions from the image data and using the arrangement as a reference, it is possible to quickly and reliably recognize the traffic signal with a simple algorithm.

Another aspect of the present invention further includes signal luminance calculating means for calculating each luminance of each display color of the traffic light from the image data, and the display color determining means is configured to calculate the luminance calculated by the luminance calculating means. The display color is determined based on the relative level of each luminance of the display color.
In this way, it is possible to quickly and reliably determine the display color of a traffic light with a simple algorithm by calculating each luminance of blue, yellow, and red, which are the display colors of the traffic light, from the image data and using the relative height as a reference. It becomes.

In another aspect of the present invention, the traffic signal recognition means selects the closest traffic signal based on the size of the position range of each display color when a plurality of traffic signals are recognized by the image data. It is characterized by.
In this way, when there are a plurality of traffic lights in the image data, by selecting the closest one from the size, the possibility of the most pressing danger can be processed with a simple algorithm.

According to another aspect of the present invention, the traffic signal position acquisition means is a road map data stored in the road map data storage means at an intersection in the traveling direction at a link of the own vehicle position obtained by map matching. The position of the corresponding node is acquired as the position of the traffic light.
In this way, by considering the position of the intersection as the position of the traffic signal, even if there is no individual traffic signal position data or there are multiple traffic signal position data, there is a simple algorithm for the position of the traffic signal that can be used in practice. It can be acquired.

  In addition, about each item above, it is the same also about a method and a program.

  Next, the best mode for carrying out the present invention will be described with reference to the drawings. Note that assumptions common to the explanation in the background art and problems are omitted as appropriate.

〔Constitution〕
The present embodiment relates to a navigation device (referred to as “this device”) that predicts a danger at an intersection and promotes safety to a driver. This device recognizes a traffic light from image data captured by a camera and displays the signal. Means for determining the color, means for estimating and calculating the vehicle stop distance from the vehicle speed, means for calculating the distance to the traffic light from the navigation vehicle position and road map data, and the vehicle stop distance and the traffic light And a means for outputting an alarm based on a combination of the magnitude relationship of the distance and the determined display color.

  Specifically, as shown in the configuration diagram of FIG. 1, the present apparatus includes a main unit 2, a GPS receiver 4, a beacon receiver 6, an FM multiplex receiver 8, a drive device 10, a display 12, a speaker 14, and the like. Have.

  Among these, the GPS receiver 4 receives radio waves from each satellite and measures its own position. In recent years, a differential GPS (D-GPS) service for correcting a positioning error due to a radio wave reception state from a GPS satellite is provided by FM multiplex broadcasting, and an FM multiplex receiver 8 is used for that purpose. The FM multiplex receiver 8 is also used in a road traffic information communication system (VICS) described later.

  The beacon receiver 6 receives traffic information such as traffic congestion and traffic regulations provided by the Japan Road Traffic Information Center in the above-mentioned VICS and uses it as information for traveling.

  The main unit 2 includes a CPU (arithmetic control unit) which is the center of arithmetic processing in a computer, various sensors 16 (gyro sensors such as a vibration gyro for detecting an angle and a 3D gyro for detecting an inclination angle and acceleration, and a traveling speed of a vehicle. And a storage medium drive device 10 (CD, DVD, hard disk drive, etc.) for storing a map, geographical information, regional information, and various data related thereto.

  The main unit 2 also includes a decoder 18 that decodes data from the drive device 10, an audio output device 20 that outputs voice guidance to the speaker 14, a display 12 that displays information to the driver and passengers, and a drawing. And a storage device (main memory) 24 such as DRAM used for internal calculation.

  The display 12 has recently been used as a touch panel and has become one of the user interfaces 13. However, the present embodiment includes the case where a conventional infrared remote controller is used.

  The drive device 10 also serves as road map data storage means for storing road map data including the position of the road and the position of the traffic light or the intersection, and the apparatus further includes a camera 26 and an image calculation processing unit 28. Of these, the camera 26 is an imaging unit such as a CCD camera unit that is attached to the vehicle and images the front of the vehicle, and is typically installed in front of the vehicle as shown in FIG. If possible, the installation location is not limited.

  The image calculation processing unit 28 is an electronic circuit that implements the following means (reference numeral 30 and later) shown in FIG. 3, and all or a part of each of these means is an arithmetic control unit of a computer. A certain CPU 29 can be realized using a storage unit such as a memory according to a control program stored in advance. The CPU 29 serves both as a vehicle position estimation means for successively estimating and calculating the vehicle position based on the outputs from the various sensors 16 and a speed detection means for detecting the vehicle speed with vehicle speed pulses.

The image calculation processing unit 28 includes a traffic signal recognition means 30 for recognizing a traffic signal based on image data obtained by digitally converting an image signal from a camera, and luminances of blue, yellow, and red lights of the traffic signal from the image data. Based on the signal brightness calculation means 32 for calculating the distance, the traffic light distance calculation means 34 for calculating the distance from the traffic light based on the own vehicle position sequentially calculated in the navigation, and the results and the vehicle speed. In addition to the alarm control means 36 that outputs an alarm for safety, the following means shown in FIG. 3 are provided.
The procedure of processing by each of these means is shown in the flowchart of FIG.

[Same points as before]
First, in this apparatus, as in the prior art, the CPU 29 that also serves as the vehicle position estimation means estimates the current vehicle position (vehicle position estimation process). This process is based on a combination of other navigation that measures the position of the vehicle by the GPS receiver 4 and autonomous navigation that calculates the position of the vehicle by various sensors 16, based on the position of the vehicle that is calculated from them. Map matching is performed to display an appropriate vehicle position.

  That is, as shown in FIG. 5, the road map data in the prior art and the present embodiment is generally called a node or a link with elements of intersections (may be end points) and lines connecting the intersections. It was created by associating various elements with coordinates. Thus, nodes mainly correspond to intersections (including branches, T-junctions and inflection points), and links correspond to roads and road sections.

  The map matching is based on the vehicle position A (dead reckoning position coordinates) that is calculated by the sensor and generally includes an error, even if it is slightly deviated from the actual road position. It is to obtain the nearest neighbor link and to determine the coordinate B (map matching position) of the vehicle position at a suitable place on the link.

  That is, even in this embodiment, the positions A and B of the own vehicle are calculated. The calculation update cycle is the same as or faster than the map matching timing. Some apparatuses perform map matching at a period of 0.1 seconds or the like. These are common points.

[Signal recognition]
On the other hand, FIG. 4 shows a flowchart of operations representing the features of the present apparatus which are different from the conventional one. That is, as long as the functions of the present invention and this embodiment are not turned off (step S11), the image data conversion means 38 (FIG. 3) such as an A / D converter continuously converts the image captured by the camera. Thus, the image data is acquired (step S12, image data conversion process), and the traffic signal recognition means 30 tries to recognize the traffic signal based on the acquired image data (steps S13 and S14. Traffic signal recognition process). .

  In other words, the camera captures the traveling direction of the vehicle and recognizes whether there is a traffic light ahead based on the image data. The recognition of such a traffic light is based on the blue, which is the lighting location of the traffic light, from the image data. It can be considered to extract the arrangement of yellow and red colors.

  The color extraction method in this case can be freely selected from conventionally known ones in each direction. As an example, there are three basic colors by performing color separation using image processing data from image data. Blue, green, and red are extracted, and the color is obtained from image information that has been subjected to noise removal and color density processing from the plain image of each color. As a result, since the digitized color is obtained, it is possible to perform numerical calculation processing by a microcomputer or the like, and it is possible to find the arrangement of blue, yellow, and red colors, which are the lighting positions of the traffic light, and determine the traffic light.

  Specifically, in this apparatus, the signal luminance calculation means 32 performs pre-processing for detecting each display color (blue, yellow, red), each position and range, and luminance from the image data. The traffic signal recognition means 30 determines the color arrangement from the information of each display color and recognizes the traffic signal based on this. Thus, by detecting blue, yellow, and red, which are display colors of traffic lights, and their positions from the image data and using the arrangement as a reference, the traffic lights can be quickly and reliably recognized with a simple algorithm.

[When there are multiple traffic lights]
Further, when the traffic light recognition means 30 recognizes a plurality of traffic lights existing in front from the image data (step S15), the traffic light recognition means 30 discriminates the perspective based on the size of the light of the traffic light (position range of each display color) from the data. Accordingly, the nearest traffic signal is selected (step S16). In this way, when there are a plurality of traffic lights in the image data, by selecting the closest one from the size, the possibility of the most pressing danger can be processed with a simple algorithm.

[Judgment of color]
Further, the display color determination unit 40 determines the display color of the traffic signal recognized by the traffic signal recognition unit 30 (step S17). That is, when the traffic light is confirmed, the signal brightness calculation means 32 described above detects or calculates the brightness of the detected blue, yellow, and red lights of the traffic light.

  The signal luminance calculation means may be provided as a separate system from the processing related to traffic signal recognition, but in either case, for example, the luminance calculation is performed for each map matching period of the navigation device. Specifically, the luminance may be calculated from the calculated color intensity, or the measurement principle of a luminance meter of an optical device may be used, for example. Then, the color having the highest luminance is selected from the three colors thus obtained (step S17).

  In this case, instead of calculating the absolute value of the luminance of each color, it is only necessary to determine the relative luminance of the three luminances, so there is no need for particularly high-precision luminance calculation, and which color of the three lights is lit It is enough to be able to judge whether or not

  When such relative determination is used, the signal luminance calculation means 32 calculates each luminance of each display color of the traffic light from the image data, and determines the display color based on the relative level of each luminance of each display color. The means 40 determines the display color. In this way, it is possible to quickly and reliably determine the display color of a traffic light with a simple algorithm by calculating each luminance of blue, yellow, and red, which are the display colors of the traffic light, from the image data and using the relative height as a reference. It becomes.

[Calculation of position and distance]
Subsequently, the traffic signal position acquisition means 42 acquires the position of the recognized traffic signal based on the road map data stored in the road map data storage means (signal position acquisition processing), and the position of the traffic signal thus acquired. Then, based on the host vehicle position estimated by the CPU 29, the traffic light distance calculation means 34 calculates a distance L (FIG. 2) from the host vehicle to the traffic signal (signal light distance calculation processing).

  At this time, the node corresponding to the intersection ahead of the vehicle position B (FIG. 5) obtained by map matching is determined to be an intersection with a traffic signal, and the distance L from the vehicle position to the intersection is determined as a road map. It can be calculated from the data. That is, the distance L is obtained by calculating the inter-coordinate distance (L) between the traffic signal location (coordinates) of the map data and the vehicle position coordinate A, and the coordinates for each traffic signal are also recorded. Is ideal, but in order to simplify the process and avoid the influence of errors, it is also realistic to calculate from the difference between the coordinates of the node that is the intersection ahead and the coordinates on the link that is the vehicle position B It is considered correspondence.

  In this case, the traffic signal position acquisition means 42 indicates the position of the node corresponding to the intersection in the traveling direction at the link of the own vehicle position obtained by map matching among the road map data stored in the road map data storage means. Get as the position. In this way, by considering the position of the intersection as the position of the traffic signal, even if there is no individual traffic signal position data or there are multiple traffic signal position data, there is a simple algorithm for the position of the traffic signal that can be used in practice. It can be acquired. Further, not only the dead reckoning position coordinates but also the map matching position is used as the own vehicle position, and the influence of errors can be avoided by an easy process.

[Calculation of vehicle stop distance]
Further, at an appropriate timing, preferably for each map matching or sequentially in real time, the speed detecting means acquires the traveling speed (vehicle speed) of the vehicle from a vehicle speed sensor (not shown) included in the various sensors 16 (step S22. Speed detection). Processing) From this vehicle speed, the vehicle stop distance calculating means calculates a vehicle stop distance Lstop, which is an estimated vehicle stop distance (step S23, vehicle stop distance calculating process).

As a reference example for this calculation, according to Norio Takagi “Authorization of negligence in the case of automobile (heavy) negligence and injury”, Judicial Research Report 21-2, p289,

◎ Stopping distance =
Running distance (braking distance)

For example,

Free running distance = speed km / 3600 x 0.75
Braking distance = speed ^ 2 / (2 x 9.8 x friction coefficient)

Examples can be given.

  In other words, in the calculation of distance, the reaction time (idle time) from the time when the driver feels dangerous and sudden braking is necessary to the time when the brake pedal is depressed and the brakes start to be effective varies (individual difference) The average reaction time of a normal person was calculated as 0.75 seconds. FIG. 6 shows the friction coefficients of tires for various road surfaces used in the calculation of the braking distance. In this case, it is most suitable to vary the friction coefficient in accordance with the road surface condition. However, since the configuration becomes complicated, this is set to a fixed value of about 0.5 assuming a general environment.

  As described above, by calculating the stop distance Lstop, it can be seen whether the vehicle can sufficiently stop from the distance (L) to the intersection. Therefore, it is possible to ensure a safe driving state for the driver by issuing a warning for preventing a rear-end collision according to the state of the front traffic light at that time.

[Alarm condition judgment]
That is, as a result of the above processing, for each map matching, three items of information are obtained: the vehicle stop distance Lstop, the distance L to the traffic light, and the display color state of the traffic light (either blue, yellow, or red). Based on these information, the alarm control means 36 performs the following processing.

  That is, as a method of activating an alarm, an alarm output is synthesized from the audio output device 20 to the speaker 14 of the navigation device by a combination of the current relationship between the distance L to the traffic signal and the stop distance Lstop and the state of the traffic signal. Output with audio.

  Specifically, the difference calculation means 46 calculates the difference between the distance L to the traffic light calculated by the traffic light distance calculation means 34 and the vehicle stop distance Lstop calculated by the vehicle stop distance calculation means 44 (difference calculation). Processing), the alarm control means 36 determines that the difference calculated in this way is equal to or less than a set value (for example, zero) (step S20), and the display color determination means 40 determines that the display color of the traffic light is red or yellow. In such a case (steps S21 and S24), an alarm for alerting the driver is output according to the display color (steps S25 and S26, alarm control processing). This combination is shown in FIG. Thereby, the function for supporting the safe driving | running | working driving | operation to a driver | operator can be provided via a navigation apparatus.

  As the types of alarm output, for example, in the case of a speaker, the alarm output 1 is provided as a sound (for example, the alarm output 1 is a slow intermittent alarm sound: Peep, the alarm output 2 is a fast intermittent alarm sound: a beep), Alternatively, the types and combinations for notifying the driver are free, such as displaying the warning outputs 1 and 2 visually by voice guidance and the display screen.

  Since the above processing is always performed unless the function in the present invention or the present embodiment is turned OFF by setting or the like, for example, after an alarm output is issued from the alarm control means 36, the vehicle speed is decelerated, and the condition of FIG. If α> 0 is established (step S20), the alarm is stopped (step S30). Further, when the vehicle stops (step S27) or when the vehicle passes a traffic light (intersection) (step S28), the alarm output is canceled (step S29). Thereafter, monitoring is performed again to realize that this function always operates in a situation where the vehicle is traveling.

  It should be noted that the navigation device can be customized according to the application by allowing the driver to select whether or not this function can be used from the navigation device menu.

[Main effects]
As described above, according to the present embodiment, when the vehicle approaches the intersection, the traffic light is recognized from the image of a single camera provided on the vehicle, the display color is also determined, the color is red or yellow, and the vehicle speed is By outputting a warning when the stop distance to be calculated exceeds the distance to the traffic signal calculated from the position of the vehicle for navigation and the road map data, the structure and control calculations are simplified, and high reliability is achieved with less burden Is realized. In addition, by providing an in-vehicle navigation device for supporting safe vehicle travel, it is possible to support safe vehicle travel at intersections and the like, which can contribute to reduction of accidents.

The figure which shows the structure of the navigation apparatus in embodiment of this invention. The conceptual diagram which shows the distance L of the camera installed in the vehicle and the signal apparatus in embodiment of this invention. The figure which shows the specific structure of the image calculation process part in embodiment of this invention. The flowchart which shows the process sequence in embodiment of this invention. The figure which shows the relationship between the node and link contained in road map data in embodiment of this invention. The figure which illustrates the friction coefficient used by embodiment of this invention. The figure which shows the conditions of the alarm in embodiment of this invention.

Explanation of symbols

2 ... main unit 4 ... GPS receiver 6 ... beacon receiver 8 ... FM multiplex receiver 10 ... drive device 12 ... display 13 ... user interface 14 ... speaker 16 ... various sensors 18 ... decoder 20 ... audio output device 22 ... graphic renderer 26 ... Camera 28 ... Image arithmetic processing unit 30 ... Signal recognition means 32 ... Signal luminance calculation means 34 ... Signal distance calculation means 36 ... Alarm control means 38 ... Image data conversion means 40 ... Display color determination means 42 ... Signal position acquisition means 44 ... vehicle stop distance calculation means 46 ... difference calculation means A ... own vehicle position coordinates B by dead reckoning ... own vehicle position L by map matching ... distance Lstop ... vehicle stop distance S ~ ... each step of the procedure

Claims (14)

Means for recognizing a traffic light from image data captured by a camera and determining its display color;
Means for estimating and calculating the vehicle stop distance from the vehicle speed;
Means for calculating the distance to the traffic signal from the navigation vehicle position and road map data;
Means for outputting an alarm based on the combination of the magnitude relationship between the vehicle stop distance and the distance to the traffic light and the determined display color;
A navigation device comprising: A navigation device that predicts danger at intersections and promotes safety for drivers,
Road map data storage means for storing road map data including the position of the road and the position of the traffic light or the intersection;
An imaging means attached to the vehicle for imaging the front of the vehicle;
Image data conversion means for converting an image captured by the imaging means into image data;
A traffic light recognition means for recognizing a traffic light from the image data converted by the image data conversion means;
Display color determination means for determining the display color of the traffic signal recognized by the traffic signal recognition means;
Speed detecting means for detecting the traveling speed of the vehicle;
Vehicle stop distance calculating means for calculating a vehicle stop distance from the travel speed detected by the speed detecting means;
Vehicle position estimation means for estimating the current vehicle position;
Traffic signal position acquisition means for acquiring the position of the traffic signal recognized by the traffic signal recognition means based on the road map data stored in the road map data storage means;
A traffic light distance calculating means for calculating a distance from the own vehicle to the traffic light based on the position of the traffic light acquired by the traffic light position acquiring means and the own vehicle position estimated by the own vehicle position estimating means;
Difference calculating means for calculating a difference between the distance to the traffic light calculated by the traffic light distance calculating means and the vehicle stop distance calculated by the vehicle stop distance calculating means;
When the difference calculated by the difference calculation means is less than or equal to a set value and the display color determination means determines that the display color of the traffic light is red or yellow, an alarm for alerting the driver is output. Alarm control means;
Is realized by an arithmetic control unit or an electronic circuit of a computer. Means for detecting each display color and position of the traffic light from the image data;
The navigation apparatus according to claim 2, wherein the traffic light recognition unit recognizes a traffic light based on the detected arrangement of the display colors. Signal luminance calculation means for calculating the luminance of each display color of the traffic light from the image data,
The navigation apparatus according to claim 2 or 3, wherein the display color determination unit determines the display color based on a relative level of each luminance of each display color calculated by the luminance calculation unit.   5. The traffic signal recognizing unit, when recognizing a plurality of traffic signals based on the image data, selects the closest traffic signal based on a size of a position range of each display color. The navigation device according to any one of the above.   The traffic signal position acquisition means is configured to determine a position of a node corresponding to an intersection in a traveling direction at a link of the vehicle position obtained by map matching among road map data stored in the road map data storage means. The navigation device according to claim 2, wherein the navigation device is acquired as a position. Processing for recognizing a traffic light from image data captured by the camera and determining its display color;
Processing to estimate the vehicle stop distance from the vehicle speed;
A process for calculating the distance to the traffic signal from the navigation vehicle position and road map data;
A process of outputting an alarm based on a combination of the relationship between the vehicle stop distance and the distance to the traffic light and the determined display color;
A navigation method comprising: Road map data storage means for storing road map data including the position of the road and the position of the traffic light or the intersection;
An imaging means attached to the vehicle for imaging the front of the vehicle;
Speed detecting means for detecting the traveling speed of the vehicle;
Have
A navigation method using a navigation device that predicts danger at an intersection and promotes safety to the driver,
Image data conversion processing for converting an image captured by the imaging means into image data;
A traffic light recognition process for recognizing a traffic light from the image data converted by the image data conversion process;
Display color determination processing for determining the display color of the traffic signal recognized in the traffic signal recognition processing;
A vehicle stop distance calculation process for calculating a vehicle stop distance from the travel speed detected by the speed detection means;
A vehicle position estimation process for estimating the current vehicle position;
Based on the road map data stored in the road map data storage means, a traffic signal position acquisition process for acquiring the position of the traffic signal recognized in the traffic signal recognition process;
A traffic light distance calculation process for calculating a distance from the own vehicle to the traffic light based on the position of the traffic signal acquired in the traffic light position acquisition process and the own vehicle position estimated in the own vehicle position estimation process;
A difference calculation process for calculating a difference between the distance to the traffic light calculated in the traffic light distance calculation process and the vehicle stop distance calculated in the vehicle stop distance calculation process;
When the difference calculated in the difference calculation process is equal to or less than a set value and the display color determination process determines that the display color of the traffic light is red or yellow, an alarm for alerting the driver is output. Alarm control processing;
Is executed by an arithmetic control unit or an electronic circuit of a computer. From the image data, including processing for detecting each display color and position of the traffic light,
The navigation method according to claim 8, wherein the traffic light recognition process recognizes a traffic light based on the arrangement of the display colors detected by the luminance calculation process. A signal luminance calculation process for calculating the luminance of each display color of the traffic light from the image data,
The navigation method according to claim 8 or 9, wherein the display color determination process determines the display color based on a relative level of each luminance of each display color calculated by the luminance calculation process.   11. The traffic signal recognition process according to claim 8, wherein when a plurality of traffic signals are recognized based on the image data, the closest traffic signal is selected based on the size of the position range of each display color. The navigation method according to any one of the above.   In the traffic signal position acquisition process, among the road map data stored in the road map data storage means, the position of the node corresponding to the intersection in the traveling direction at the link of the own vehicle position obtained by map matching is calculated. It acquires as a position, The navigation method as described in any one of Claims 8-11 characterized by the above-mentioned. By controlling the calculation control part of the computer,
Processing for recognizing a traffic light from image data captured by the camera and determining its display color;
Processing to estimate the vehicle stop distance from the vehicle speed;
A process for calculating the distance to the traffic signal from the navigation vehicle position and road map data;
A process of outputting an alarm based on a combination of the relationship between the vehicle stop distance and the distance to the traffic light and the determined display color;
A navigation program characterized by causing Road map data storage means for storing road map data including the position of the road and the position of the traffic light or the intersection;
An imaging means attached to the vehicle for imaging the front of the vehicle;
Speed detecting means for detecting the traveling speed of the vehicle;
A computer control unit;
Have
A navigation program for predicting danger at an intersection and controlling a navigation device that promotes safety to the driver,
The program controls the arithmetic control unit,
Image data conversion processing for converting an image captured by the imaging means into image data;
A traffic light recognition process for recognizing a traffic light from the image data converted by the image data conversion process;
Display color determination processing for determining the display color of the traffic signal recognized in the traffic signal recognition processing;
A vehicle stop distance calculation process for calculating a vehicle stop distance from the travel speed detected by the speed detection means;
A vehicle position estimation process for estimating the current vehicle position;
Based on the road map data stored in the road map data storage means, a traffic signal position acquisition process for acquiring the position of the traffic signal recognized in the traffic signal recognition process;
A traffic light distance calculation process for calculating a distance from the own vehicle to the traffic light based on the position of the traffic signal acquired in the traffic light position acquisition process and the own vehicle position estimated in the own vehicle position estimation process;
A difference calculation process for calculating a difference between the distance to the traffic light calculated in the traffic light distance calculation process and the vehicle stop distance calculated in the vehicle stop distance calculation process;
When the difference calculated in the difference calculation process is equal to or less than a set value and the display color determination process determines that the display color of the traffic light is red or yellow, an alarm for alerting the driver is output. Alarm control processing;
A navigation program characterized by causing

Images