JP5001911B2 - Stop duty point learning device, stop duty point learning program, and navigation device using the same - Google Patents

Description

  The present invention relates to a stop duty point learning device, a stop duty point learning program that generates stop duty information indicating that there is a temporary stop duty, and stores the stop duty information in association with an intersection with a temporary stop duty, and a stop duty point learning program The present invention relates to a navigation device using the.

  In recent years, various types of information have been provided to assist the driving of a vehicle by providing information on the road on which the vehicle is being driven or assisting or intervening in the operation control of the vehicle based on such information. Technology has been proposed. For example, when there is an intersection that is obligated to pause in front of the road on which the vehicle is traveling, it is recognized and notified to the driver of the vehicle, or the engine speed, braking operation, etc. before the intersection There is known a driving support device that can safely and surely execute a temporary stop by controlling (see, for example, Patent Document 1 below).

  The driving support device associates stop obligation information with own vehicle position information based on image recognition means for recognizing a stop sign included in image information around the own vehicle position and the image recognition result of the stop sign. And storing means for storing. When the image recognition unit succeeds in recognizing the stop sign image and there is no corresponding information in the storage unit, new stop duty information is added to update the information. And a driving assistance device performs driving assistance based on the storage contents after updating of a storage means.

JP 2006-275690 A

  In the driving support device described in Patent Literature 1, the stop duty information based on the image recognition result of the stop sign is stored in association with the own vehicle position information, thereby learning the intersection with the temporary stop duty. Here, the vehicle position information can be obtained by analyzing a signal received from a GPS satellite via a GPS receiver. The vehicle position information acquired in this way generally has an error of several meters to several tens of meters. Therefore, it is necessary to set the range for performing the image recognition processing of the stop sign corresponding to the intersection having the obligation to stop temporarily in consideration of the magnitude of the error included in the vehicle position information. Specifically, for example, it is necessary to set the image recognition range with a front-rear width of a certain distance with reference to the position of the stop sign.

  However, when setting the image recognition range in this way, when a plurality of intersections are close to each other, the image recognition ranges for performing the image recognition processing of the stop sign corresponding to each intersection are mutually exclusive. It will be duplicated. An example in which such a situation occurs is shown in FIG. In the example of FIG. 11, the intersection U1 and the intersection U2 are relatively close to each other, and the image recognition range X1 of the stop sign corresponding to the intersection U1 and the image recognition range X2 of the stop sign corresponding to the intersection U2 , Partly overlaps in the range Y. Originally, when the image recognition of the stop sign is successful in the image recognition range X1, it is determined that the intersection U1 has the obligation to stop, and similarly, the image recognition of the stop sign succeeds in the image recognition range X2. In such a case, it should be determined that the intersection U2 is obliged to pause. However, if the image recognition of the stop sign is successful in the range Y, it is not possible to uniquely determine which stop duty information should be stored corresponding to the intersection U1 or the intersection U2.

  Therefore, when image recognition of a stop sign is successful in a range where multiple image recognition ranges overlap each other, it is known that there is at least an intersection with a stop obligation, but the corresponding intersection cannot be determined. There was a problem that the learning efficiency was reduced. On the other hand, according to a predetermined rule such as dividing the overlapping range into two and allocating to two intersections, when storing the stop obligation information corresponding to any intersection, store it corresponding to the intersection different from the actual In some cases, driving assistance that matches the actual situation may not be possible.

  The present invention has been made in view of the above problems, and a stop duty point learning device and a stop duty point that can appropriately determine an intersection that associates stop duty information while minimizing a decrease in learning efficiency. It is an object to provide a learning program and a navigation device using the learning program.

  In order to achieve this object, a stop duty information learning device that generates stop duty information indicating that there is a temporary stop duty according to the present invention and stores the stop duty information in association with an intersection with a temporary stop duty is stored. The feature configuration is included in recognition range setting means for setting an image recognition range for performing image recognition of a stop duty index indicating that there is a duty to stop temporarily, based on each intersection, and image information around the vehicle position, Image recognition means for performing image recognition of the stop duty index, stop duty information generation means for generating the stop duty information based on the image recognition result of the stop duty index, and the image recognition range is N (N is 2 or more) Natural number), the N image recognition ranges are defined as a sum range that is one continuous range, and the stop that has been successfully recognized in the image is included in the sum range. Based on the number of compulsory indicators, in that and a stop duty point determining means for determining the intersection associating said stop duty information.

  According to the above characteristic configuration, by generating and storing the stop duty information based on the image recognition result of the stop duty index indicating that there is a stop duty included in the image information around the vehicle position, Information on intersections that are obliged to stop can be collected. At this time, it is determined that there is a suspension obligation based on the number of suspension obligation indicators that have been successfully recognized by the stop duty point determination means included in the sum range of a plurality of image recognition ranges set with reference to each intersection. An intersection is determined that associates the stop obligation information that it represents. In this case, even if image recognition of the stop obligation index succeeds in a range where a plurality of image recognition ranges overlap with each other, the stop obligation information may be changed depending on the number of stop obligation indexes that succeeded in image recognition within the sum range. The associated intersection can be established. Therefore, it is possible to provide a stop duty point learning device that can appropriately determine an intersection associated with stop duty information while minimizing a decrease in learning efficiency.

  Here, when the stop obligation point determination means has succeeded in image recognition of the N stop obligation indicators in the sum range, the stop duty point determination means determines the stop at the intersection corresponding to each image recognition range constituting the sum range. It is preferable that the configuration is such that the duty information is determined to be associated.

  According to this configuration, when the number of image recognition ranges that overlap sequentially and the number of stop duty indicators that succeeded in image recognition coincide with each other, the stop duty point determination unit determines the position at which image recognition of the stop duty indicator succeeded. Regardless of this, it is possible to determine to associate the stop duty information with the intersection corresponding to each image recognition range. Therefore, even if the image recognition of any stop duty index is successful in the overlapping range of the plurality of image recognition ranges, the stop duty information can be reliably associated with all the intersections.

  Moreover, the said stop duty point determination means is an image recognition range which does not overlap with an adjacent image recognition range when the image recognition of the said stop duty index of 1 or more and less than N is successful in the said sum range. When the image recognition of the stop duty index is successful in the non-overlapping range, it is determined to associate the stop duty information with the intersection corresponding to the non-overlapping range, and the image recognition ranges adjacent to each other are overlapped with each other. When image recognition of the stop duty index is successful in a certain overlapping range, it is preferable to determine that the stop duty information is not associated with intersections corresponding to the plurality of image recognition ranges including the overlapping range. is there.

  According to this configuration, the stop duty point determination means is the case where the image recognition of at least one stop duty index is successful in the sum range, and the number of the overlapping image recognition ranges and the stop that succeeds in the image recognition. When the number of duty indicators does not match, the intersection with which the stop duty information is associated can be appropriately determined according to the position where the image recognition of the stop duty indicator is successful. In other words, the position that succeeded in the image recognition of the stop duty index is determined by determining the intersection that associates the stop duty information only when the position that succeeded in the image recognition of the stop duty index is a non-overlapping range where there is no possibility of erroneous learning. If it is an overlapping range that is likely to be mislearned, it is determined that the stop obligation information should not be associated with intersections corresponding to a plurality of image recognition ranges including the overlapping range, thereby suppressing mislearning and Depending on the position where image recognition of the index is successful, an intersection that associates stop obligation information can be determined.

  Further, if the stop duty point determining means does not succeed in image recognition of the stop duty index in the sum range, the stop duty point is determined at the intersection corresponding to each image recognition range constituting the sum range. A configuration in which it is determined that information is not associated is preferable.

  According to this configuration, when the image recognition of the stop obligation index is not successful from the image information, it is appropriate that the intersection corresponding to each image recognition range that constitutes the sum range has no pause obligation. Can be determined.

  The overlapping range number acquisition means for acquiring the number of overlapping ranges indicating the number of the image recognition ranges included in the sum range, and the number of stop obligation points indicating the number of the stop obligation indicators that have succeeded in image recognition in the sum range. It is preferable that the stop duty point determination unit is configured to determine an intersection that associates the stop duty information based on the number of overlapping ranges and the number of stop duty points. It is.

  According to this configuration, by referring to the number of overlapping ranges acquired by the overlapping range number acquiring means and the number of stop obligation points measured by the stop duty point number measuring means, the stop duty information by the stop duty point determining means It is possible to appropriately execute the process of determining the intersection that associates.

  Further, the information processing apparatus includes map information storage means for storing map information including road network information representing road connection relations by nodes corresponding to intersections and links corresponding to roads connecting the intersections, and the stop obligation information includes the nodes or It is preferable that the information is stored in association with one end of the link.

  According to this configuration, it is possible to store the stop duty information appropriately associated with the intersection having the temporary stop duty by using the information regarding the node or the link stored in the map information storage means. Therefore, for example, it is possible to appropriately use for various uses such as alerting the driver of the vehicle to urge the suspension and vehicle control.

  Further, the stop obligation index is a combination of a stop line provided on the road surface and at least a part of characters in a character string of “to rare” provided adjacent to the stop line; It is preferable.

  In general, a stop line is provided in front of an intersection that is obliged to temporarily stop, and a character string such as “Tare” is painted on the road surface. Therefore, by adopting the above configuration, it is possible to appropriately select an intersection having a suspension obligation. In addition, since it is determined that the image recognition of the stop duty index is successful only when the image recognition of both the stop line and the character string “Tore rare” is successful, it is possible to suppress erroneous recognition. Furthermore, the “to rare” character string is configured to perform image recognition of only some of the characters, thereby reducing the image recognition processing burden.

  The characteristic configuration of the navigation device according to the present invention includes the stop duty point learning device that has been described so far, and guidance means that guides the intersection having the suspension obligation based on the stop duty information. It is in.

  According to this characteristic configuration, the navigation device refers to the stop duty information collected using the stop duty point learning device described above, and guides the intersection having the temporary stop duty. For intersections that have not been stored and that have a duty to stop, after generating and storing information on the obligation to stop, for example, the driver of the vehicle is informed that there is a duty to stop and is urged to suspend Arousing can be performed.

  Features of a stop duty point learning program according to the present invention for generating stop duty information indicating that there is a stop duty and causing a computer to execute processing for associating and storing the stop duty information in association with an intersection with a stop duty The configuration includes a recognition range setting step for setting an image recognition range for performing image recognition of a stop obligation index indicating that there is a temporary stop obligation, with each intersection as a reference, and the stop included in the image information around the vehicle position. An image recognition step for performing image recognition of the duty index, a stop duty information generation step for generating the stop duty information based on an image recognition result of the stop duty index, and the image recognition range is N (N is a natural number of 2 or more) ) When there are sequential overlaps, the consecutive N image recognition ranges are defined as a sum range that is one continuous range, and are included in the sum range. Is the succeeded in image recognition based on the number of stop duty indicator, there are a stop duty point determining step of determining an intersection associating said stop duty information, to a point for causing a computer to execute.

  According to this feature configuration, the computer executes processing for generating and storing stop duty information based on the image recognition result of the stop duty index indicating that there is a duty to stop included in the image information around the vehicle position. By doing so, it is possible to collect information on intersections that are obligated to pause. At this time, in the stop duty point determination step, it is determined that there is a stop duty based on the number of stop duty indices that have been successfully recognized in the image, which are included in the sum of the plurality of image recognition ranges set with reference to each intersection. An intersection is determined that associates the stop obligation information that it represents. In this case, even if image recognition of the stop obligation index succeeds in a range where a plurality of image recognition ranges overlap with each other, the stop obligation information may be changed depending on the number of stop obligation indexes that succeeded in image recognition within the sum range. The associated intersection can be established. Therefore, it is possible to provide a stop duty point learning program that can appropriately determine an intersection associated with stop duty information while minimizing a decrease in learning efficiency.

  Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a schematic configuration of a navigation device 1 according to the present embodiment. FIG. 2 is a diagram showing an example of the configuration of map information M stored in the map database DB1, and FIG. 3 shows an example of feature information F such as road markings stored in the feature database DB2. FIG. FIG. 4 is a diagram illustrating an example of an arrangement configuration of the imaging device 18 on the host vehicle. FIG. 5 is an explanatory diagram for explaining an example of a method for setting the image recognition range C, and FIGS. 6 and 7 illustrate a method for determining an intersection associated with the stop duty information s by the stop duty point determination unit 9. It is explanatory drawing for.

  The stop duty point learning device 2 included in the navigation device 1 according to the present embodiment performs image recognition of the stop duty index I that is included in the image information G around the vehicle position and indicates that there is a duty to pause, and performs image recognition. Stop obligation information s is generated based on the result. Then, the stop duty point learning device 2 stores the stop duty information s in association with the intersection having the temporary stop duty. Further, the navigation device 1 performs normal processing such as route guidance to the destination, and refers to the stop duty information s that is learned and stored by the stop duty point learning device 2, and about the intersection having the temporary stop duty. I will guide you.

  Each functional unit of the spot information collection device 2 shown in FIG. 1, specifically, the own vehicle position information acquisition unit 3, the image information acquisition unit 4, the recognition range setting unit 5, the overlapping range number acquisition unit 6, and the image recognition unit 7. The stop duty point number measurement unit 8, the stop duty point determination unit 9, the stop duty information generation unit 10, and the navigation calculation unit 11 are input using a common processing unit such as a CPU as a core member. A functional unit for performing various processes on the data is implemented by hardware or software (program) or both. Each of these functional units is configured to be able to exchange information with each other via a communication line such as a digital transfer bus. Further, the map database DB1 and the feature database DB2 are, for example, a recording medium capable of storing information, such as a hard disk drive, a DVD drive equipped with a DVD-ROM, a CD drive equipped with a CD-ROM, and a driving means thereof. The learning database DB3 includes a recording medium on which information can be rewritten, such as a hard disk drive, a DVD drive equipped with a DVD-RAM, and a driving means thereof. The apparatus which has is provided as a hardware configuration. Hereinafter, the configuration of each part of the navigation device 1 according to the present embodiment will be described in detail.

1. Map Database The map database DB1 is a database in which map information M divided for each rectangular area called a section divided by predetermined longitude and latitude is stored. FIG. 2 is a diagram illustrating an example of the configuration of the map information M stored in the map database DB1. As shown in this figure, the map information M has road information Ra representing a road network by connection relations between a large number of nodes n corresponding to intersections and links k corresponding to roads connecting the intersections. . Each node n has information on the position (coordinates) on the map expressed by latitude and longitude. Each link k is connected via the node n, and has attribute information such as road type, area type, link length, road width, and shape interpolation point for expressing the link shape. . In FIG. 2, only the road information Ra of one section is illustrated, and the road information Ra of other sections is omitted.

2. Feature database The feature database DB2 is a database for storing feature information provided on the road surface of the road, that is, feature information F. FIG. 3 is a diagram illustrating an example of the feature information F of the road marking stored in the feature database DB2. In the present embodiment, as the feature information F stored in the feature database DB2, for example, a stop line provided on the road surface of the road, or a character “to rare” provided adjacent to the stop line. Column etc. are mentioned. In addition, for example, a pedestrian crossing, a stop line, a speed sign indicating a maximum speed, a zebra zone, a lane line that divides a lane along the road (including various lane lines such as a solid line, a broken line, and a double line), It may include feature information F relating to a traveling direction marking (designating an arrow marking, for example, a straight arrow, a right turn arrow, etc.) specifying the traveling direction of each lane. In addition to such road markings, the features in which the feature information F is stored can include various features such as traffic lights, signs (for example, temporary stop signs), overpasses, and tunnels. The feature information F includes position information of each feature and feature attribute information associated therewith as its contents. Here, the position information includes information on the position (coordinates) on the map of the representative point of each feature associated with the link k or the node n constituting the road information Ra, and the direction of each feature. The feature attribute information includes type information indicating the type of each feature, or feature form information such as the shape, size, and color of the feature. Here, the feature type specifically indicates the type of feature having basically the same form such as “pedestrian crossing”, “stop line”, “speed indication (30 km / hour)”, and the like. Information. Although not described in detail, the feature information F can be used for correction of the vehicle position information P, vehicle control, and the like, together with image recognition processing of actual features on the road by the image recognition unit 7. .

3. Learning Database The learning database DB3 is a database that stores stop duty information s generated by the stop duty information generation unit 10 described later, and functions as a stop duty information storage unit. In this learning database DB3, stop duty information s generated based on an image recognition result of a stop duty index I (details will be described later) by the image recognition unit 7 is stored in association with an intersection having a temporary stop duty.

4). Own vehicle position information acquisition unit The own vehicle position information acquisition unit 3 functions as own vehicle position information acquisition means for acquiring own vehicle position information P indicating the current position of the own vehicle. Here, the vehicle position information acquisition unit 3 is connected to the GPS receiver 15, the direction sensor 16, and the distance sensor 17. Here, the GPS receiver 15 is a device that receives a GPS signal from a GPS (Global Positioning System) satellite. This GPS signal is normally received every second and output to the vehicle position information acquisition unit 3. The own vehicle position information acquisition unit 3 can analyze a signal from a GPS satellite received by the GPS receiver 15 and acquire information such as the current position (latitude and longitude), traveling direction, and moving speed of the own vehicle. it can. The direction sensor 16 is a sensor that detects a traveling direction of the host vehicle or a change in the traveling direction. The azimuth sensor 16 includes, for example, a gyro sensor, a geomagnetic sensor, an optical rotation sensor attached to the rotating part of the handle, a rotary resistance volume, and an angle sensor attached to the wheel part. Then, the direction sensor 16 outputs the detection result to the vehicle position information acquisition unit 3. The distance sensor 17 is a sensor that detects a vehicle speed and a moving distance of the host vehicle. The distance sensor 17 integrates, for example, a vehicle speed pulse sensor that outputs a pulse signal every time a drive shaft or wheel of the vehicle rotates by a certain amount, a yaw / G sensor that detects the acceleration of the host vehicle, and the detected acceleration. It is composed of a circuit or the like. Then, the distance sensor 17 outputs information on the vehicle speed and movement distance as the detection result to the own vehicle position information acquisition unit 3.

  And the own vehicle position information acquisition part 3 performs the calculation which pinpoints an own vehicle position by a well-known method based on the output from these GPS receiver 15, the direction sensor 16, and the distance sensor 17. FIG. In addition, the vehicle position information acquisition unit 3 acquires map information M around the vehicle position from the map database DB1 and performs known map matching based on the map information M on the road indicated by the map information M. Correction to match is also performed. In this way, the host vehicle position information acquisition unit 3 acquires host vehicle position information P including information on the current position of the host vehicle represented by latitude and longitude and information on the traveling direction of the host vehicle. However, the signal received from the GPS satellite by the GPS receiver 15 generally has an error of several meters to several tens of meters. Therefore, the own vehicle position information P acquired in this way also has an error of several meters to several tens of meters. The vehicle position information P obtained by the vehicle position information acquisition unit 3 is output to the recognition range setting unit 5 and the spot information generation unit 9.

5. Image Information Acquisition Unit The image information acquisition unit 4 functions as an image information acquisition unit that acquires image information G around the host vehicle captured by the imaging device 18. Here, the imaging device 18 is an in-vehicle camera or the like provided with an imaging element, and is provided at a position where at least a road surface of a road around the host vehicle can be imaged. As such an imaging device 18, for example, a back camera that images the road surface behind the host vehicle as shown in FIG. 4 is preferably used. Below, the example which acquires the image information G around the own vehicle using such a back camera is demonstrated. The image information acquisition unit 4 captures image information G captured by the imaging device 18 at predetermined time intervals via a frame memory (not shown) or the like. The time interval for capturing the image information G at this time can be set to, for example, about 10 to 50 ms. Thereby, the image information acquisition unit 4 can continuously acquire the image information G of a plurality of frames captured by the imaging device 18. The acquired image information G is output to the image recognition unit 7.

6). Recognition range setting unit The recognition range setting unit 5 sets an image recognition range C for performing image recognition of a stop obligation index I (details will be described later) indicating that there is a temporary stop obligation, with each intersection T as a reference. Functions as setting means. When setting the image recognition range C, as shown in FIG. 5, the recognition range setting unit 5 first determines a predetermined distance D ahead of the vehicle position in the traveling direction based on the vehicle position information P and the map information M. The node n1 corresponding to the nearest intersection T1 within is extracted. Next, using the position of the node n1 corresponding to the intersection T1 as the first reference point B1, a position offset by a predetermined distance W toward the own vehicle position along the link k (vehicle traveling direction) is set as the second reference point. Obtained as B2. Here, the predetermined distance W to be offset from the position of the node n is set by adding a predetermined distance to a half of the road width of the road intersecting with the road where the host vehicle is traveling at the intersection T1. It can be a variable value. The fixed distance set in advance is preferably a fixed value (for example, 1 meter) set based on the average value of the distance between the stop duty index I and the entrance of the intersection. In general, since the position of the node n is set at the center point of the intersection T, by setting the distance as described above as the predetermined distance W, the position immediately before the second reference point B2 enters the intersection T. Can be set at substantially the same position as the position of the stop duty index I, which is often provided at. Information regarding the road width of the road intersecting with the road on which the host vehicle is traveling at the intersection T can be acquired by referring to attribute information of the link k corresponding to the road.

  Next, the recognition range setting unit 5 sets, as the image recognition range C1, a range that is expanded by a predetermined distance E on both sides of the second reference point B2 along the link k (vehicle traveling direction). Here, the predetermined distance E can be a fixed value (for example, 50 meters) set in accordance with the maximum value of the error of the vehicle position information P described above. By setting such a distance as the predetermined distance E, even if the vehicle position information P has a predetermined error, if the stop obligation index I exists at the position of the second reference point B2, In any position within the image recognition range C where the total length is 2E, the image recognition unit 7 described later can perform image recognition.

  In this embodiment, in addition to the previously extracted intersection T1, the recognition range setting unit 5 also includes a node n2 corresponding to the next intersection T2 existing further ahead of the vehicle position in the traveling direction than the intersection T1. Extract. Then, the image recognition range C2 is set for the intersection T2 in the same manner as described above (see FIG. 6). Next, the recognition range setting unit 5 determines whether or not the image recognition range C1 and the image recognition range C2 partially overlap. When the image recognition range C1 and the image recognition range C2 do not overlap, the recognition range setting unit 5 sets only the image recognition range C1 as the image recognition range C. On the other hand, when the image recognition range C1 and the image recognition range C2 overlap, the recognition range setting unit 5 adds to the vehicle further than the intersections T1 and T2 in addition to the intersections T1 and T2 extracted earlier. A node n3 corresponding to the next intersection T3 existing on the front side in the traveling direction of the position is also extracted. The image recognition range C3 is set for the intersection T3 in the same manner as described above. Next, the recognition range setting unit 5 determines whether or not the image recognition range C2 and the image recognition range C3 partially overlap. When the image recognition range C2 and the image recognition range C3 do not overlap, the recognition range setting unit 5 sets the image recognition range C1 and the image recognition range C2 as the image recognition range C. On the other hand, when the image recognition range C2 and the image recognition range C3 overlap, the recognition range setting unit 5 uses the intersections T1, T2, and T3 in addition to the previously extracted intersections T1, T2, and T3. Further, a node n4 corresponding to the next intersection T4 existing on the front side in the traveling direction of the vehicle position is also extracted. In such a process, the image recognition range CN corresponding to the Nth (N is a natural number) intersection TN and the image recognition range C (N + 1) corresponding to the (N + 1) th intersection T (N + 1) do not overlap. Repeat until it is determined. In this way, the recognition range setting unit 5 uses the N image recognition ranges C1, C2,..., CN, which are sequentially overlapped, with the intersections T1, T2,. Set sequentially.

In the example shown in FIG. 6, the image recognition range C1 and the image recognition range C2 partially overlap each other, but the image recognition range C2 and the image recognition range C3 are completely separated without overlapping. . Therefore, in the example of FIG. 6, the recognition range setting unit 5 sequentially sets two image recognition ranges C1 and C2 that are sequentially overlapped with respect to the intersections T1 and T2, respectively. At this time, the recognition range setting unit 5 also sets an overlapping range L that is a range in which image recognition ranges C adjacent to each other overlap and a non-overlapping range R that is an image recognition range that does not overlap with the adjacent image recognition range C. Configured to do. In the illustrated example, the image recognition range C1 is divided into a non-overlapping range R ₁ and overlapping range L _12, the image recognition range C2 is divided into the overlapping range L ₁₂ and a non-overlapping range R _2. And the range which caught the image recognition range C1 and the image recognition range C2 as one continuous range is made into the "sum range A" in this invention. That is, the sum range A includes a non-overlapping range R ₁ of the image recognition range C 1, an overlapping range L ₁₂ of the image recognition range C ₁ and the image recognition range C _2, and a non-overlapping range R ₂ of the image recognition range C _2. It becomes. In other words, the sum range A is from the rear end portion of the vehicle along the link k in the image recognition range C1 to the front end portion of the vehicle along the link k in the image recognition range C2. It becomes a range.

7). Overlapping Range Number Acquisition Unit The overlapping range number acquisition unit 6 functions as overlapping range number acquisition means for acquiring the overlapping range number Nl (Nl is a natural number) representing the number of image recognition ranges C included in the sum range A. As described above, the recognition range setting unit 5 extracts the intersections TN and T (N + 1) existing on the front side in the traveling direction of the vehicle position, and the image recognition ranges CN and C (C (C) respectively corresponding to the intersections T. N + 1) is set, and it is determined whether or not the image recognition range CN and the image recognition range C (N + 1) partially overlap. Such processing is repeated until it is determined that the image recognition range CN and the image recognition range C (N + 1) do not overlap. In the present embodiment, the overlapping range number acquisition unit 6 acquires the value of N when it is determined that the image recognition range CN and the image recognition range C (N + 1) do not overlap as the overlapping range number Nl. To do. The acquired overlapping range number Nl is output to the stop duty point determination unit 9.

8). Image Recognizing Unit The image recognizing unit 7 functions as an image recognizing unit that performs image recognition of the stop duty index I included in the image information G around the vehicle position. The image recognition unit 7 performs image recognition of the stop obligation index I from the image information G in the image recognition range C set by the recognition range setting unit 5. Here, the stop duty index I is an index that is provided corresponding to an intersection with a temporary stop obligation and notifies the vehicle driver that there is a temporary stop obligation before the intersection. For example, there are temporary stop signs ("To rare" signs) installed on the shoulders of roads, "To rare" signs provided adjacent to a stop line provided on the road surface, and the like. In the case of using the “tore rare” sign, the entire “tore rare” sign may be used as the stop obligation index I, or at least some characters in the “to rare” character string may be used as the stop duty index I. Further, a combination of these and the stop line may be used as the stop obligation index I. In the present embodiment, the combination of the stop line provided on the road surface and the characters “MA” and “RE” in the character string “TO rare” provided adjacent to the stop line is an obligation to stop. It is set as index I.

  Therefore, in the present embodiment, the image recognition unit 7 uses the image “G”, “Re”, and stop line images as the stop obligation index I from the image information G in the image recognition range C included in the sum range A. Perform recognition processing. Here, of the character string “Tare rare” included in the stop obligation index I, the characters “MA” and “RE” are subjected to image recognition processing only on the road. Among the various features that have been created, there are relatively many edges, complex shapes, and inherently character-like features. This is because it can be considered that the entire image recognition is successful. When recognizing the characters “ma”, “re”, and the stop line, binarization processing, edge detection processing, etc. are performed on the image information G, and the contour information of the feature included in the image information G is obtained. Extract. The image recognizing unit 7 performs pattern matching between the extracted feature outline information and the stop duty index I, which are the characters “ma”, “re”, and the feature amount of the stop line form, and is included in the image information G. The images of the characters “MA”, “RE” and the stop line are extracted. As a result, when the images of the characters “MA”, “RE”, and the stop line can be extracted from the image information G, the image recognition unit 7 determines that the image recognition of the stop obligation index I is successful, The result is output to the stop duty point number measuring unit 8 and the stop duty information generating unit 10.

9. Stop duty point number measurement unit Stop duty point number measurement unit 8 measures the number of stop duty points Ns (Ns is a natural number) that represents the number of stop duty indicators I that succeeded in image recognition in the sum range A. Functions as a means. The stop duty point number measuring unit 8 has “0” as an initial value of the stop duty point number Ns, and in the image recognition range C included in the sum range A, as the stop duty index I from the image information G. Each time the image recognition process of the characters “MA”, “RE” and the stop line is successfully performed, a process of adding “1” to the stop duty number Ns is performed (Ns = Ns + 1). The stop duty point number Ns when the image recognition processing for all the image information G in the sum range A is finished is output to the stop duty point determination unit 9.

10. Stop Obligation Point Determination Unit The stop duty point determination unit 9 functions as a stop duty point determination unit that determines an intersection T that associates stop duty information s described later. In this embodiment, the stop duty point determination unit 9 stops based on the overlap range number Nl output from the overlap range number acquisition unit 6 and the stop duty point number Ns output from the stop duty point number measurement unit 8. The intersection T with which the duty information s is associated is determined. Below, the determination method of the intersection T which associates the stop duty information s by the stop duty point determination part 9 is demonstrated.

First, when the overlapping range number Nl is 2 or more, at least the image recognition range C1 and the image recognition range C2 overlap, and the sum range A is defined as one continuous range from the image recognition range C1 to the image recognition range CNl. Matches the captured range. In the example shown in FIG. 6, since only the image recognition range C1 and the image recognition range C2 overlap, and the image recognition range C2 and the image recognition range C3 do not overlap, the number of overlap ranges Nl is “2”. At this time, the sum range A includes a non-overlapping range R ₁ of the image recognition range C 1, an overlapping range L ₁₂ of the image recognition range C 1 and the image recognition range C _2, and a non-overlapping range R ₂ of the image recognition range C _2. included. Here, when image recognition of the stop obligation index I is successfully performed from the image information G in the non-overlapping range R ₁ of the image recognition range C1, the intersection T1 corresponding to the image recognition range C1 has a temporary stop obligation. It can be determined with certainty. Similarly, if from the image information G in the non-overlapping range R ₂ of the image recognition range C2 succeeded in image recognition of the stop duty index I is, if there is a pause obligation intersection T2 corresponding to the image recognition range C2 It can be determined with certainty. On the other hand, if from the image information G in the image recognition range C1 and the overlap region L ₁₂ between the image recognition range C2 succeeded in image recognition of the stop duty index I is an intersection T1 and image recognition corresponding to the image recognition range C1 It cannot be reliably determined which of the intersections T2 corresponding to the range C2 is obliged to pause. Therefore, when the image recognition of the stop obligation index I is successful in the overlapping range L ₁₂ , it is known that there is at least an intersection having a stop obligation, but the corresponding intersection cannot be reliably determined. Learning efficiency is reduced. Therefore, the stop duty point determination unit 9 determines the intersection T that associates the stop duty information s according to the following determination pattern based on the relationship between the overlap range number Nl and the stop duty point number Ns, thereby improving the learning efficiency. It is possible to minimize the decline.

Decision pattern 1
When the stop obligation point determination unit 9 succeeds in the image recognition of the number of stop obligation indicators I equal to the overlap range number Nl from the image information G in the sum range A, that is, when it is determined that Ns = Nl, The stop duty information s is determined to be associated with the intersection T corresponding to each image recognition range C constituting the sum range A.
A method of determining the intersection T that associates the stop obligation information s according to the pattern 1 will be described with reference to FIG. FIG. 6A shows characters as two sets of stop obligation indicators I on the road on which the host vehicle is traveling within the sum range A where the two image recognition ranges C1 and C2 are regarded as one continuous range. The state where “MA”, “RE”, and a stop line exist is shown. 6 (b) to 6 (d) correspond to FIG. 6 (a), an intersection T on the ongoing road, a second reference point B2 (corresponding to the stop duty index I), and an actual image. The position of the point indicated by the own vehicle position information P at the time of successful recognition is shown in a simplified manner. 6B to 6D, the intersection T is “●”, the second reference point B2 is “Δ”, and the point indicated by the vehicle position information P when the image recognition is actually successful is shown. It is indicated by “☆” (the same applies to FIG. 7). Here, the distance between the second reference point B2 substantially corresponding to the stop duty index I and the point indicated by the vehicle position information P at the time when the image recognition is actually successful is the vehicle position described above. This corresponds to the error of the information P. In this example, since Ns = Nl = 2, in the determination pattern 1, in any of these cases, the stop duty point determination unit 9 performs intersections T1 and T2 corresponding to the respective image recognition ranges C1 and C2. Is determined to be associated with the stop obligation information s. According to such a determination pattern 1, for example, as shown in FIGS. 6B and 6D, the vehicle position information P has a relatively large error on the rear side or the front side with respect to the traveling direction of the vehicle. in that state when (a relatively large displacement position relative to the actual vehicle position in the traveling direction of the rear side or front side of the vehicle recognized by that state as the vehicle position), either in the overlap region L ₁₂ Even if image recognition of one stop duty index I is to be performed, the stop duty information s can be reliably associated with both the intersections T1 and T2. Therefore, it is possible to minimize a decrease in learning efficiency.

Decision pattern 2
If the stop obligation point determination unit 9 succeeds in image recognition of one or more stop obligation indicators I from the image information G in the sum range A, that is, if it is determined that 1 ≦ Ns <Nl. The intersection T to which the stop duty information s is associated is variably determined according to the position where the image recognition of the stop duty index I is successful. Specifically, when the image recognition unit 7 succeeds in the image recognition of the stop duty index I in the non-overlapping range R, the stop duty point determining unit 9 is obliged to stop at the intersection T corresponding to the non-overlapping range R. Decide to associate information s. On the other hand, when the image recognition unit 7 succeeds in image recognition of the stop duty index I in the overlap range L, the stop duty point determination unit 9 determines the intersection T corresponding to the plurality of image recognition ranges C including the overlap range L. Is determined not to be associated with the stop obligation information s.
A method of determining the intersection T that associates the stop obligation information s according to the pattern 2 will be described with reference to FIG. FIG. 7A shows characters as a set of stop obligation indicators I on the road on which the host vehicle is traveling within the sum range A where the two image recognition ranges C1 and C2 are regarded as one continuous range. The state where “MA”, “RE”, and a stop line exist is shown. In the determination pattern 2, the processing content may differ depending on the magnitude and direction of the error included in the vehicle position information P. That is, for example, as shown in FIG. 7B, the vehicle position information P has a relatively large error on the rear side with respect to the traveling direction of the vehicle (the vehicle traveling relative to the actual vehicle position). in state) recognizes the relatively large displacement position in the direction of the rear side as a vehicle position, so that a successful image recognition stop duty index I in the overlap range L _12. Therefore, stop duty point determining unit 9, the intersection T1 and T2 corresponding to the image recognition range C1 and C2 containing overlapping range L ₁₂ determines not associate stop duty information s. In this way, it is possible to prevent the intersection T associated with the stop obligation information s from being erroneously determined. On the other hand, as shown in FIG. 7C, for example, the vehicle position information P has a relatively large error on the front side with respect to the traveling direction of the vehicle (the vehicle travels relative to the actual vehicle position). in state) recognizes the relatively large displacement position in the forward side as the vehicle position, so that in the non-overlapping ranges R ₂ successfully image recognition stop duties indicators I. Therefore, stop duty point determining unit 9 determines to associate the stop duty information s the intersection T2 corresponding to the non-overlapping range R _2. In this way, when there is no possibility of erroneously determining the intersection T associated with the stop duty information s, the intersection T associated with the stop duty information s can be reliably determined, and a decrease in learning efficiency can be suppressed. .

Decision pattern 3
If the stop obligation indicator I is not successfully recognized from the image information G in the sum range A, that is, if it is determined that Ns = 0, the stop duty point determination unit 9 configures the sum range A. It is determined that the stop duty information s is not associated with the intersection T corresponding to each image recognition range C. In this way, if the image recognition G does not succeed in recognizing the stop obligation index I from the image information G, the intersection T corresponding to each image recognition range C constituting the sum range A is obliged to pause. It can be determined appropriately that there is no.

  When the overlapping range number Nl is 1, the image recognition range C1 and the image recognition range C2 do not overlap, so the sum range A matches the image recognition range C1. In this case, if the image recognition unit 7 succeeds in image recognition of the characters “MA”, “RE”, and the stop line as the stop obligation index I in the image recognition range C1, the stop duty point determination unit 9 performs image recognition. It is determined that the stop obligation information s is associated with the intersection T1 corresponding to the range C1. That is, the process performed when Nl = 1 is the same as the normal process, and in this case, the stop duty point determination unit 9 does not perform any process.

11. Stop Duty Information Generation Unit The stop duty information generation unit 10 functions as a stop duty information generation unit that generates the stop duty information s based on the image recognition result of the stop duty index I. In the present embodiment, the stop duty information s is generated as attribute information indicating that there is a temporary stop duty. The generated stop obligation information s is output to the learning database DB3 and stored in the learning database DB3. At this time, each stop duty information s is stored in association with the map information M corresponding to the intersection T to be associated according to the result determined by the stop duty point determination unit 9. In the present embodiment, each stop obligation information s is stored in association with the link k information included in the map information M. Specifically, the stop duty information s is associated with the link end on the node n side in the link k connected to the node n corresponding to the intersection T that associates the stop duty information s determined by the stop duty point determination unit 9. Is remembered. Such a position corresponds to the position of the road edge that is actually connected to the intersection that is obligated to temporarily stop, and therefore, the stop obligatory information s is used to appropriately guide the intersection that is obliged to temporarily stop. It becomes possible.

12 Navigation Calculation Unit The navigation calculation unit 11 operates according to an application program to execute navigation functions such as display of the vehicle position, calculation of a route from the departure point to the destination, route guidance to the destination, and destination search. Arithmetic processing means. For example, the navigation calculation unit 11 acquires the map information M around the host vehicle from the map database DB1, displays a map image on the display input device 19, and displays the map position information P on the map image. Based on this, the process of displaying the self-position mark in a superimposed manner is performed. In addition, the navigation calculation unit 11 uses one or both of the display input device 19 and the audio output device 20 based on the route from the departure point to the destination calculated by a known method and the vehicle position information. Provide route guidance. At this time, the navigation calculation unit 11 refers to the stop duty information s stored in the learning database DB3 and provides guidance for the intersection T having the temporary stop duty. Therefore, in the present embodiment, the navigation calculation unit 11 functions as a guidance unit that performs guidance for the intersection T having a suspension obligation based on the suspension obligation information s. In addition, the navigation calculation unit 11 is connected to various known configurations necessary for the navigation device 1 such as a remote controller and a user interface such as a touch panel provided integrally with the display input device 19. .

13. Next, the procedure of the stop duty point learning process executed in the stop duty point learning device 2 applied to the navigation device 1 according to the present embodiment will be described. FIG. 8 is a flowchart showing the processing order of the stop duty point learning program according to this embodiment, and FIG. 9 is a flowchart showing the processing procedure of the stop duty point determination process. The procedure of the stop duty point learning process described below is executed by hardware, software (program), or a combination of both constituting each functional unit of the stop duty point learning device 2 described above. When each function part of the stop duty point learning device 2 is configured by a program, the arithmetic processing unit included in the stop duty point learning device 2 is a computer that executes the stop duty point learning program that constitutes each of the above function units. Works as.

  First, the overall procedure of the stop duty point learning process will be described. As shown in FIG. 8, first, the vehicle position information acquisition unit 3 acquires the vehicle position information P based on the outputs from the GPS receiver 15, the direction sensor 16, and the distance sensor 17 (step # 01). Next, the recognition range setting unit 5 extracts the node n corresponding to the nearest intersection T within the predetermined distance D ahead of the own vehicle position in the traveling direction based on the own vehicle position information P and the map information M ( In step # 02), an image recognition range C for performing image recognition of the stop duty index I is set with the intersection T as a reference (step # 03). Next, the recognition range setting unit 5 extracts a node n ′ corresponding to the next intersection T ′ existing further forward than the intersection T in the traveling direction of the host vehicle position (step # 04), and determines the intersection T ′. As a reference, an image recognition range C ′ for performing image recognition of the stop obligation index I is set (step # 05). Next, the recognition range setting unit 5 determines whether or not the image recognition range C and the image recognition range C ′ overlap each other (step # 06). If it is determined that the image recognition range C and the image recognition range C ′ overlap each other (step # 06: Yes), the processing from step # 04 to step # 06 is repeated again. On the other hand, when it is determined that the image recognition range C and the image recognition range C ′ do not overlap each other (step # 06: No), the overlapping range number acquisition unit 6 determines that the images included in the sum range A The overlapping range number Nl representing the number of recognition ranges C is acquired (step # 07), and it is determined whether the overlapping range number Nl is 2 or more (step # 08). When it is determined that the overlapping range number Nl is less than 2, that is, “1” (step # 08: No), the image recognition unit 7 includes the image recognition range C in the image information G around the vehicle position. Is recognized (step # 13), and then the process proceeds to step # 14. On the other hand, when it is determined that the overlapping range number Nl is 2 or more (step # 08: Yes), the recognition range setting unit 5 sets the overlapping range L and the non-overlapping range R for each image recognition range C. Setting is made (step # 09).

  Next, the image recognition unit 7 performs image recognition of the stop obligation index I included in the image information G around the vehicle position in the image recognition range C (step # 10). Then, the stop duty point number measuring unit 8 measures the stop duty point number Ns indicating the number of the stop duty index I that has succeeded in image recognition in the sum range A (step # 11). Next, the stop duty point determination unit 9 executes a stop duty point determination process for determining the intersection T with which the stop duty information s is associated based on the overlapping range number Nl and the stop duty point number Ns (step # 12). The procedure of the stop duty point determination process will be described later. Finally, the stop duty information generating unit 10 generates stop duty information s, and each stop duty information s is stored in association with link information corresponding to the intersection T to be associated (step # 14). At this time, if it is determined in step # 08 that the overlapping range number Nl is 2 or more, according to the result determined by the stop duty point determination process, each stop duty information s should be associated. It is stored in association with the link information corresponding to T. Thus, the stop duty point learning process is completed.

14 Next, details of the stop duty point determination process in step # 12 will be described. As shown in FIG. 9, the stop duty point determination unit 9 first determines whether or not the stop duty point number Ns and the overlapping range number Nl are equal (step # 21). When it is determined that the number Ns of stop duty points and the number Nl of overlapping ranges are equal (step # 21: Yes), the stop duty point determination unit 9 corresponds to each image recognition range C constituting the sum range A. It is determined to associate the stop obligation information s with the intersection T to be performed (step # 22). On the other hand, when it is determined that the number Ns of stop duty points and the overlap range number Nl are not equal (step # 21: No), the stop duty point determination unit 9 then sets the stop duty point number Ns to “0”. Is determined (step # 23). When it is determined that the stop duty point number Ns is “0” (step # 23: Yes), the stop duty point determination unit 9 performs intersections corresponding to the respective image recognition ranges C constituting the sum range A. It is determined that the stop obligation information s is not associated with T (step # 24). On the other hand, if it is determined that the number Ns of stop duty points is not “0”, that is, the number Ns of stop duty points is 1 or more and less than the overlapping range number Nl (step # 23: No), then the stop duty point The determination unit 9 determines whether or not the position at which the image recognition unit 7 has successfully recognized the stop obligation index I is within the non-overlapping range R (step # 25). When it is determined that it is within the non-overlapping range R (step # 25: Yes), the stop duty point determination unit 9 determines to associate the stop duty information s with the intersection T corresponding to the non-overlapping range R. (Step # 26). On the other hand, when it is determined that it is not within the non-overlapping range R, that is, within the overlapping range L (step # 25: No), the stop duty point determination unit 9 recognizes a plurality of images including the overlapping range L. It is determined that the stop duty information s is not associated with the intersection T corresponding to the range C (step # 27). The stop duty point determination process is thus completed.

[Other Embodiments]
(1) In the above embodiment, the image recognition range C1 corresponding to the intersection T1 and the image recognition range C2 corresponding to the intersection T2 overlap, and the image recognition range C2 corresponding to the intersection T2 and the image corresponding to the intersection T3. The case where there is no overlap with the recognition range C3, that is, the case where the overlap range number Nl is “2” has been described as an example. However, the embodiment of the present invention is not limited to this. That is, the present invention can naturally be applied to the case where the overlapping range number Nl is 3 or more. FIG. 10 shows an example in which the overlapping range number Nl is “4”. In the illustrated example, an image recognition range C1 corresponding to the intersection T1, an image recognition range C2 corresponding to the intersection T2, an image recognition range C3 corresponding to the intersection T3, and an image recognition range C4 corresponding to the intersection T4 are sequentially provided. Duplicate. Although not shown, the image recognition range C4 corresponding to the intersection T4 does not overlap with the image recognition range C5 corresponding to the intersection T5 appearing next to the intersection T4. In this case, a range obtained as a continuous range from the image recognition range C1 to the image recognition range C4 is set as a “sum range A”. Then, the image recognition range C1 is divided into a non-overlapping range R ₁ and overlapping range L _12, the image recognition range C2 is divided into the overlapping range L ₁₂ and the non-overlapping range R ₂ and overlapping range L _23. Further, the image recognition range C3 is divided into a duplicated range L ₂₃ and the non-overlapping range R ₃ and overlapping range L _34, the image recognition range C4 is divided into an overlapping range L ₃₄ and a non-overlapping range R _4.

In this example, when the stop duty point number Ns representing the number of stop duty indices I that have been successfully recognized in the sum range A is “4”, the stop duty point determination unit 9 sets the sum range A to It is determined that the stop duty information s is associated with the intersections T1 to T4 corresponding to the respective image recognition ranges C1 to C4. When the stop duty point number Ns representing the number of stop duty indicators I that have been successfully recognized in the sum range A is “0”, the stop duty point determination unit 9 configures the sum range A. It is determined that the stop duty information s is not associated with the intersections T1 to T4 corresponding to the image recognition ranges C1 to C4. When the number Ns of stop duty points indicating the number of stop duty indices I that have been successfully recognized in the sum range A is any one of “1” to “3”, the image recognition unit 7 When image recognition of the stop duty index I is successful in any of R ₁ , R ₂ , R ₃ , and R ₄ , the stop duty point determination unit 9 determines the intersections T1 and T2 corresponding to the non-overlapping range R, respectively. , T3, and T4 are determined to be associated with the stop obligation information s. On the other hand, when the image recognition unit 7 succeeds in the image recognition of the stop duty index I in any of the overlapping ranges L ₁₂ , L ₂₃ , and L ₃₄ , the stop duty point determining unit 9 includes a plurality of the overlapping ranges L including the overlapping range L. It is determined that the stop duty information s is not associated with the intersection T corresponding to the image recognition range C.

(2) In the above embodiment, the position of the node n corresponding to the intersection T is set to the first reference point for the purpose of enabling the image recognition range C to be set around the position where the stop duty index I exists. B1 and a position offset by a predetermined distance W toward the vehicle position along the link k (vehicle traveling direction) as a second reference point B2, and the second along the link k (vehicle traveling direction). An example has been described in which a range expanded by a predetermined distance E on both sides of the reference point B2 is set as the image recognition range C. However, the embodiment of the present invention is not limited to this. That is, a range expanded by a predetermined distance E on both sides of the position of the node n corresponding to the intersection T may be set as the image recognition range C without setting the second reference point B2. In this case, there may occur a situation in which the position where the stop obligation index I exists does not substantially coincide with the center of the image recognition range C. However, the method for setting the image recognition range C can be simplified, and the processing load therefor can be reduced. .

(3) In the above embodiment, the predetermined distance W is set by adding a predetermined distance to a half of the road width of the road intersecting with the road where the host vehicle is traveling at the intersection T1. As an example, a case has been described in which the predetermined distance E is a fixed value set in accordance with the maximum value of the error included in the vehicle position information P. However, the settings of the predetermined distance W and the predetermined distance E are arbitrary and are not limited to these. For example, the predetermined distance W may be a fixed value set in advance, or the predetermined distance E may be a variable value set according to the travel distance after the map matching process.

(4) In the above embodiment, the characters “MA” and “RE” in the stop line provided on the road surface of the road and the character string “TORAARE” provided adjacent to the stop line The case where the combination is set as the stop duty index I has been described as an example. However, the embodiment of the present invention is not limited to this. In other words, for example, the stop line provided on the road surface of the road and the entire character string of “to rare” provided adjacent to the stop line can be set as the stop duty index I. In addition to this, a stop sign, a combination of a stop sign and a stop line, or the like can be set as the stop duty index I as long as it can serve as an index indicating that there is a duty to stop.

(5) In the above embodiment, the case where the generated stop duty information s is stored in association with the link information corresponding to the intersection T to be associated with has been described as an example. However, the embodiment of the present invention is not limited to this. That is, the generated stop obligation information s may be stored in association with node information corresponding to the intersection T to be associated. In the case of this configuration, since the node n itself constituting the map information M corresponds to the intersection, there is an advantage that it is easy to correspond to the intersection having the suspension obligation to which the suspension obligation information s should be associated. In this case, it is preferable that information indicating which link k of the link k connected to the node n has a suspension obligation is stored together.

(6) In the above embodiment, the case where the generated stop obligation information s is stored in the learning database DB3 has been described as an example. However, the embodiment of the present invention is not limited to this. That is, for example, without providing the learning database DB3, the generated stop obligation information s is associated with the road network information composed of the node n and the link k and stored directly in the map database DB1. May be. In this case, the map database DB1 has a hardware configuration of an apparatus having a recording medium capable of rewriting information and a driving means thereof, such as a hard disk drive, a DVD drive equipped with a DVD-RAM, and the like. It is set as the structure prepared as.

(7) In the above embodiment, an example in which the image information G around the host vehicle is acquired using the back camera that captures the road surface behind the host vehicle has been described. However, the embodiment of the present invention is not limited to this. That is, for example, the configuration may be such that image information G around the host vehicle is acquired using a front camera that captures the situation ahead of the host vehicle including the road surface. In general, the front camera is often provided to acquire a relatively wide range of image information G in front of the host vehicle. Therefore, for example, as a configuration of the imaging device 18 when a stop sign is set as the stop obligation index I, Is suitable.

(8) In each of the embodiments described above, the case where all the configurations of the stop duty point learning device 2 and the navigation device 1 are loaded on the vehicle has been described. However, the embodiment of the present invention is not limited to this. That is, a part of the stop obligation point learning device 2 and the navigation device 1 may be provided outside the vehicle via a communication network. For example, a part or all of each functional unit and each database may be provided in an external server, and the stop duty point learning process and the navigation process may be performed via a communication network.

  The present invention relates to a stop duty point learning device, a stop duty point learning program that generates stop duty information indicating that there is a temporary stop duty, and stores the stop duty information in association with an intersection with a temporary stop duty, and a stop duty point learning program It can utilize suitably for the navigation apparatus using.

1 is a block diagram showing a schematic configuration of a navigation device according to the present embodiment. The figure which shows the example of a structure of the map information memorize | stored in the map database The figure which shows the example of the feature information of the road marking memorize | stored in the feature database The figure which shows an example of the arrangement configuration of the imaging device to the own vehicle Explanatory drawing for demonstrating an example of the setting method of an image recognition range Explanatory drawing for demonstrating the determination method of the intersection which links the stop duty information by the stop duty point determination means Explanatory drawing for demonstrating the determination method of the intersection which links the stop duty information by the stop duty point determination means The flowchart which shows the processing order of the stop duty point learning program which concerns on this embodiment Flowchart showing the processing procedure of stop duty point determination processing Explanatory drawing for demonstrating the determination method of the intersection which links the stop duty information by the stop duty point determination means which concerns on another embodiment Explanatory drawing for demonstrating the subject of a prior art

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Navigation apparatus 2 Stop stop point learning apparatus 5 Recognition range setting part (recognition range setting means)
6 Overlapping range number acquisition unit (overlapping range number acquisition means)
7 Image recognition unit (image recognition means)
8 stop duty point number measurement part (stop duty point number measuring means)
9 Stop duty point determination section (stop duty point determination means)
10 Stop obligation information generator (stop duty information generator)
11 Calculation unit DB1 for navigation Map database (map information storage means)
G Image information M Map information n Node k Links s Stop obligation information I Stop obligation index C Image recognition range A Sum range L Overlapping range R Non-overlapping range Nl Overlapping range number Ns Stopping duty point number T Intersection

Claims (9)

A stop duty point learning device that generates stop duty information indicating that there is a temporary stop duty and stores the stop duty information in association with an intersection with a temporary stop duty,
A recognition range setting means for setting an image recognition range for performing image recognition of a stop duty index indicating that there is a temporary stop duty, based on each intersection;
Image recognition means for performing image recognition of the stop duty index included in the image information around the vehicle position;
Stop duty information generating means for generating the stop duty information based on the image recognition result of the stop duty index;
When the image recognition ranges are sequentially overlapped over N (N is a natural number of 2 or more), the N image recognition ranges are set as a continuous range, and the sum range is included in the sum range. A stop duty point determining means for determining an intersection to be associated with the stop duty information based on the number of the stop duty indicators that have been successfully recognized in the image;
Stop duty point learning device comprising.   When the stop duty point determining means succeeds in image recognition of the N stop duty indicators in the sum range, the stop duty point determination means sets the stop duty information at intersections corresponding to the image recognition ranges constituting the sum range. The stop duty point learning device according to claim 1, wherein the stop duty point learning device is determined to be associated. If the stop obligation point determination means succeeds in image recognition of the stop obligation index of 1 or more and less than N in the sum range,
When image recognition of the stop duty index is successful in a non-overlapping range that is an image recognition range that does not overlap with an adjacent image recognition range, the decision to associate the stop duty information with an intersection corresponding to the non-overlapping range And
When image recognition of the stop duty index is successful in an overlapping range in which image recognition ranges adjacent to each other overlap, the stop duty is given to intersections corresponding to a plurality of the image recognition ranges including the overlapping range. The stop duty point learning device according to claim 1, wherein it is determined that the information is not associated.   When the stop duty point determining means does not succeed in image recognition of the stop duty index in the sum range, the stop duty information is provided at the intersection corresponding to each image recognition range constituting the sum range. The stop duty point learning device according to any one of claims 1 to 3, wherein the stop obligatory point learning device is determined not to be associated. Overlapping range number acquisition means for acquiring the number of overlapping ranges representing the number of the image recognition ranges included in the sum range;
A stop duty point number measuring means for measuring the number of stop duty points indicating the number of the stop duty index that has been successfully image recognition in the sum range, and
The stop duty point learning device according to any one of claims 1 to 4, wherein the stop duty point determination unit determines an intersection that associates the stop duty information based on the number of overlapping ranges and the number of stop duty points. . Comprising map information storage means for storing map information including road network information representing connection relations of roads by nodes corresponding to intersections and links corresponding to roads connecting the intersections;
The stop duty point learning device according to any one of claims 1 to 5, wherein the stop duty information is stored in association with one end of the node or the link.   The stop duty index is a combination of a stop line provided on a road surface of a road and at least a part of characters in a character string of “to rare” provided adjacent to the stop line. The stop duty point learning device according to any one of claims 6 to 6. Stop duty point learning device according to any one of claims 1 to 7,
Based on the stop duty information, guide means for guiding the intersection having the temporary stop duty;
A navigation device comprising: A stop duty point learning program that generates stop duty information indicating that there is a temporary stop duty, and causes the computer to execute a process of storing the stop duty information in association with an intersection with the temporary stop duty,
A recognition range setting step for setting an image recognition range for performing image recognition of a stop duty index indicating that there is a temporary stop duty, based on each intersection;
An image recognition step for performing image recognition of the stop duty index included in the image information around the vehicle position;
A stop duty information generating step for generating the stop duty information based on an image recognition result of the stop duty index;
When the image recognition ranges are sequentially overlapped over N (N is a natural number equal to or greater than 2), the N consecutive image recognition ranges are set as one continuous range, and the sum A stop duty point determination step for determining an intersection to be associated with the stop duty information based on the number of the stop duty indices that have been successfully recognized in the image included in the range;
Stop point learning program to make the computer execute.

Images