JP2008305042A - Car navigation device, road sign recognition method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a car navigation device, a road sign recognition method and a program for shortening a time until a road signa is recognized. <P>SOLUTION: In a processing part 20, a road circumstance judgment part 21 judges the road circumstances of a road on which a vehicle 2 is traveling. Then, a recognition sequence for preferentially recognizing a road signal which is highly likely to exist according to the road circumstances is stored for each road circumstance into recognition sequence information 12. Thereafter, a recognition sequence determination part 22 determines the proper recognition sequence of the road sign according to the judged road circumstances by referring to the recognition sequence information 12, and an image recognition processing part 23 recognizes the road sign according to the determined recognition sequence. The image recognition processing part 23 therefore preferentially recognizes the road sign which is highly likely to exist according to the road circumstances of the road on which the vehicle 2 is traveling. Thus, it is possible to shorten a time until the road sign is recognized. Also, a warning output part 28 outputs the warning of the road sign as necessary according to the traveling circumstances of the vehicle 2 judged by the traveling circumstance judgment part 24. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a car navigation device that guides a vehicle by displaying a guidance route, and a road sign recognition method and program in the car navigation device.

  In recent years, various functions have been added to car navigation devices in addition to functions that support navigation. For example, Patent Document 1 discloses an example of a car navigation device that includes a reference camera and recognizes a road sign from an image in front of a vehicle photographed by the camera. According to Patent Document 1, the car navigation device recognizes a road sign from a captured image, identifies its type, accumulates information on the identified road sign type and position in association with map information, and routes It is assumed that the information of the accumulated road sign is used for guidance.

  In addition, as a recognition method used for recognition of such a road sign, for example, as disclosed in Patent Document 2, a method of collating a photographed image with a pre-registered road sign template is generally used. Is. However, in this type of recognition method, since it is necessary to collate a captured image with a plurality of template images one by one, the recognition processing time tends to be long. On the other hand, recognition of road signs must be performed in real time for moving images acquired when the vehicle is traveling, and the recognition also includes recognition of regulatory signs, so high reliability with no false recognition is also required. .

Therefore, for example, Patent Document 3 stores the position of a road sign in map information, estimates the position of the road sign in a captured image when the vehicle approaches that position, and performs image processing. An example of a car navigation device that shortens recognition processing time and improves recognition accuracy by limiting a target range is disclosed.
JP 2006-38558 A JP 2003-123197 A JP 2006-309450 A

  However, when there are a large number of template images such as road signs, the recognition processing time cannot be sufficiently shortened only by limiting the target range of image processing in the captured image (Patent Document 3). In addition, in the method disclosed in Patent Document 3, the position information of the road sign recorded in the map information is used. However, if the position information is old or incorrect, It is not always possible to correctly recognize road signs.

  In view of the problems of the prior art as described above, the present invention is capable of correctly recognizing a road sign and recognizing the road sign even if the map information does not include information related to the road sign. It is an object of the present invention to provide a car navigation device, a road sign recognition method, and a program capable of shortening the time required to arrive.

  The car navigation device of the present invention includes an information processing device, a storage device that stores road map information, a positioning device that measures the current position of the vehicle, and a camera that captures the front of the traveling direction of the vehicle. Prepare. Then, the storage device includes template image information of a road sign installed in the vicinity of the road, a road condition assumed in advance (a relative positional relationship between a feature point of a road including an intersection and a vehicle position, and the road Recognition order information that defines the recognition order of the template image of the road sign for each situation) represented by at least one of the attributes. Further, the information processing device includes a road condition determination unit that determines a road condition at the current position of the vehicle on the road on which the vehicle is traveling, based on the current position of the vehicle obtained by the positioning device and the map information, Recognizing order determination means for determining the recognition order of the template image with reference to the recognition order information according to the road condition determined by the road condition determining means, and the recognition order for the captured image captured by the camera Image recognition means for recognizing the template image in accordance with the recognition order determined by the determination means.

  According to the present invention, the information processing apparatus determines the road condition of the road on which the vehicle is traveling by the road condition determination unit. At this time, in the storage device, a recognition order for preferentially recognizing a road sign (a template image) that is likely to exist for each road condition according to the road condition is stored as recognition order information. . Therefore, when the recognition order determining means refers to the recognition order information and determines the appropriate road sign recognition order according to the road condition determined by the road condition determining means, the image recognition processing means is determined. Recognize road signs according to the recognition order. Accordingly, the image recognition processing means preferentially recognizes road signs that are likely to exist according to the road conditions of the road on which the vehicle is traveling. Therefore, it is possible to shorten the time until the road sign is recognized.

  According to the present invention, even if the map information does not include information related to the road sign, the road sign can be correctly recognized and the time until the road sign is recognized can be shortened.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a diagram showing an example of the configuration of a car navigation device according to an embodiment of the present invention (hereinafter abbreviated as “car navigation device” in the description of the embodiment). As shown in FIG. 1, a car navigation apparatus 1 according to this embodiment is mounted on a vehicle 2 and recognizes a road sign 3 installed on a roadside or the like in addition to a normal car navigation function. Depending on the road condition of the road on which the vehicle 2 is traveling, the traveling condition of the vehicle 2, etc., a function is provided to provide warning information to the driver as appropriate.

  As shown in FIG. 1, the car navigation apparatus 1 includes a storage unit 10 composed of a large-capacity hard disk device, a flash memory, and the like, and a process composed of a CPU (Central Processing Unit), memory (for example, RAM (Random Access Memory)), and the like. Unit 20, GPS (Global Positioning System) receiver 30 for measuring the current position of vehicle 2, and sensor interface 40 connected to various sensors and control devices such as a vehicle speed sensor and direction sensor direction indicator of vehicle 2; The camera 50 includes a CCD (Charge Coupled Device) image sensor, a display device 60 including an LCD (Liquid Crystal Display), and a speaker 70 that outputs audio information.

  The storage unit 10 stores map information 11, recognition order information 12, template image information 13, and the like. Here, the map information 11 includes road information including intersection position data, road shape data (position data of a point sequence representing the shape), road type (a distinction between expressways, national roads, prefectural roads, guidance, and the like). In addition, it includes topographical information on mountains, rivers, coasts, etc., and location information on buildings and facilities such as city halls, schools, and parks.

  The template image information 13 is information in which standard image data is registered for each type of road sign 3, and is used as a template when the road sign 3 is recognized. The recognition order information 12 is information that defines the recognition order of the template image of the road sign 3 according to a predetermined road condition. An example of specific contents of the recognition order information 12 will be described later.

  The processing unit 20 includes a road condition determination unit 21, a recognition order determination unit 22, an image recognition processing unit 23, a traveling condition determination unit 24, a current position acquisition unit 25, a sensor information acquisition unit 26, an image acquisition unit 27, and a warning output unit 28. It is configured to include functional blocks such as In FIG. 1, as functional blocks included in the processing unit 20, functional blocks related to a road sign recognition function are illustrated, and functional blocks related to a so-called car navigation function (for example, a route search unit, a guidance route display unit, etc.) ) Is omitted. The functions of these functional blocks are realized by a CPU (not shown) included in the processing unit 20 executing a predetermined program stored in a memory such as a RAM. Detailed functions of these functional blocks will be sequentially described in the description of FIG.

  FIG. 2 is a diagram illustrating an example of a situation where the car navigation apparatus 1 according to the present embodiment is applied. In FIG. 2, for example, in a portion where the road 110 is curved, a sign of a curve caution 130 is set in front of the road 110, and a stop sign 132, an entry prohibition sign 134, and a sign are not present near the intersections 111 and 112. A traffic light 135 or the like is installed. Further, speed limit signs 131 and the like are often installed where there are no intersections on the road 110.

  That is, since the road sign 3 is installed according to the road conditions such as the shape of the road and the presence or absence of an intersection, the car navigation device 1 needs to grasp the road situation in front of the road on which the vehicle 2 is traveling. Thus, it is possible to predict the road sign 3 that has a high possibility of appearing forward. In that case, until the road sign 3 is recognized by prioritizing the recognition process of the road sign 3 having a high possibility of appearance with respect to the captured image in front of the vehicle 2 acquired by the camera 50. Can be shortened.

  Next, the configuration of the template image information 13 and the recognition order information 12 stored in the storage unit 10 will be described with reference to FIGS. 3 and 4. Here, FIG. 3 is a diagram illustrating an example of the configuration of the template image information 13, and FIG. 4 is a diagram illustrating an example of the configuration of the recognition order information 12.

As shown in FIG. 3, the template image information 13 includes a template ID and a template image. The template ID is information for identifying a template image. In the example of FIG. 3, the template ID is represented by symbols “A”, “B”,. The template image is image data for each type of road sign 3. Incidentally, in the example of FIG. 3, six types of template images are registered as a road sign template image: a curve attention sign 130, a speed limit sign 131, a stop sign 132, an entry prohibition sign 134, a traffic light 135, and a one-way sign 136. Has been. However, the template image need not be limited to these six types.

  Next, as shown in FIG. 4, the recognition order information 12 includes a road situation number, a road situation, and a recognition order. Here, the road condition number is a number for identifying the road condition. In addition, road conditions are represented by road feature point attributes (intersections, curves, inflection points, etc.), relative positional relationships between the feature points and vehicle positions, road attributes (highways, general roads, etc.), etc. Information, for example, 100m before the intersection, 100m before the curve, and on the expressway, there is a traffic light at the next intersection. The recognition order is information associated with the road condition, and the recognition order of the template image of the road sign 3 is determined for a photographed image ahead of the vehicle 2 while traveling on the road with the road condition. Information.

  For example, when the road condition number in FIG. 4 is “1”, the road condition in which the vehicle 2 is traveling is “within 100 m before the intersection with a traffic light”, so the recognition order of the template image is E (traffic light) → C (Temporary stop) → D (prohibition of entry) → F (one-way) → B (speed limit) → A (curve attention). That is, since there is an intersection with a traffic signal within 100 m ahead of the traveling direction of the vehicle 2, the possibility that a traffic signal exists is very high. Therefore, first, a traffic light template is recognized. At this time, the information that the traffic signal exists is based on the map information 11, but the map information 11 is not always accurate. There may be no traffic light. In that case, there is a high possibility that a sign indicating suspension or entry prohibition is installed. Therefore, the recognition order in this case is set so as to recognize a temporary stop or entry prohibition template following the traffic light.

  When the road condition number is “2”, the road condition in which the vehicle 2 is traveling is “within 100 m before the intersection without a traffic light”, so the recognition order of the template images is changed from C (temporary stop) to D (entrance). Prohibition) → F (one-way) → E (traffic light) → B (speed limit) → A (curve attention). In this case, since there is no traffic light at the front intersection, the recognition of the traffic light template is postponed from the recognition of the template such as a pause.

  When the road condition number is “3”, the road condition in which the vehicle 2 is traveling is “within 10 m before the intersection to within the intersection”, so the recognition order of the template images is D (prohibited entry) → F (one (Traffic) → E (traffic light) → C (temporary stop) → B (speed limit) → A (curve attention). In this case, since the vehicle 2 has almost entered the intersection, there is a possibility that there is an entry prohibition or a one-way sign on the road ahead of the intersection.

  When the road condition number is “4”, the road condition in which the vehicle 2 is traveling is “within 100 m before the curve”. Therefore, the recognition order of the template images is changed from A (Curve attention) → B (Speed limit) → F (one way) → E (signal) → C (temporary stop) → D (prohibition of entry). In this case, since there is a curve ahead, there is a high possibility that there will be signs such as curve attention and speed limit.

  Further, when the road condition number is “5”, the road condition in which the vehicle 2 is traveling is “traveling on an expressway”, so the recognition order of the template image is B (speed limit) → A (curve attention) → F (one-way) → D (prohibition of entry) → E (signal) → C (temporary stop). In this case, since the vehicle 2 is traveling on the highway, there is a high possibility that there are signs such as speed limit and curve attention.

  The road situation number “6” is a case where the road situation does not belong to any of the above cases, and the recognition order of the template image is B (speed limit) → E (traffic light) → C (temporary). Stop) → D (Prohibition of entry) → F (One way) → A (Curve attention). The road situation in this case can be considered as a road portion that is a general road and does not have an intersection, so there is a high possibility that there is a speed limit sign. However, since the map information 11 may be different from the current situation, the priority of recognition is increased for traffic lights and temporary stops that may lead to accidents if they are missed.

  Then, with reference to FIG. 1 and FIGS. 5-8, operation | movement of the car navigation apparatus 1 which concerns on this embodiment is demonstrated. Here, FIG. 5 is a diagram illustrating an example of a flowchart of image recognition processing and road condition determination processing in the car navigation device 1, and FIG. 6 is an internal view of the processing unit 20 and the storage unit 10 of the car navigation device 1 according to the present embodiment. FIG. 7 is a diagram showing an example of the configuration, FIG. 7 is a diagram showing an example of a warning output condition table, and FIG. 8 is a diagram showing an example of a flowchart of warning output processing.

  In FIG. 1, the sensor information acquisition unit 26 acquires information on the traveling speed and traveling direction of the vehicle 2 output from a vehicle speed sensor, a direction sensor (gyro), or the like mounted on the vehicle 2 via the sensor interface 40. To do. In addition, the current position acquisition unit 25 corrects the position information of the vehicle 2 obtained from the GPS receiver 30 based on the traveling speed and traveling direction information of the vehicle 2 acquired by the sensor information acquisition unit 26, so that the vehicle 2 Get the current position of.

  The road condition determination unit 21 determines the road condition in which the vehicle 2 is traveling based on the current position acquired by the current position acquisition unit 25 every predetermined time interval (for example, 0.5 seconds). Details of the process for determining the road condition are shown as a flowchart of the road condition determination process in FIG. 5 described later.

  Further, in FIG. 1, the image acquisition unit 27 acquires captured images in front of the traveling direction of the vehicle 2 captured by the camera 50 at predetermined time intervals (for example, 0.5 seconds). And the image recognition process part 23 performs the recognition process of the template image of the road sign 3 registered into the template image information 13 about the picked-up image. The details of the template image recognition process are shown as a flowchart of the image recognition process of FIG. 5 described later.

  The processing unit 20 performs the road condition determination processing and the image recognition processing in parallel. In addition to this, navigation processing is performed as an original function of the car navigation device 1 such as guidance route search and guidance route display. There is a need to do. Therefore, in this embodiment, as shown in FIG. 6, the processing unit 20 is configured by two CPUs, and the function of the image recognition processing unit 23 (including the image acquisition unit 27) is realized by the CPU 1 (20a). The block function is realized by the CPU 2 (20b).

  In such a case, the CPU 1 (20a) and the CPU 2 (20b) are connected to the storage unit 10 through a common bus, and can access the recognition order information 12 and the template image information 13 together. The storage unit 10 is provided with a shared memory 15 for exchanging information between the road condition determination process and the image recognition process. In this case, it is assumed that the shared memory 15 stores a road situation number obtained as a result of the road situation determination process and a template ID obtained as a result of the image recognition process.

  Therefore, the CPU 1 (20a) executes the image recognition process shown in FIG. 5 as the process of the image recognition processing unit 23. As shown in FIG. 5, the CPU 1 (20 a) first acquires a captured image in front of the traveling direction of the vehicle 2 captured by the camera 50 via the image acquisition unit 27 (step S <b> 11). Next, the CPU 1 (20a) appropriately performs filtering processing such as noise removal, edge enhancement, and binarization on the acquired captured image (step S12).

  Next, the CPU 1 (20a) refers to the shared memory 15, acquires the road condition number set by the road condition determination process, and further refers to the recognition order information 12 to correspond to the road condition number. Get the recognition order of images. Then, the CPU 1 (20a) selects one of the template images according to the recognition order (step S13).

  Next, CPU1 (20a) performs the matching process of the template image selected at step S13 with respect to the picked-up image acquired at step S11 (step S14). As a result of the matching processing, if there is an image portion that matches the template image in the photographed image, it is determined that the template image has been successfully recognized, and the template ID of the template image that has been successfully recognized is assigned to the CPU 2 ( The warning output process of 20b) is notified (step S15).

  When notifying the template ID to the warning output process (step S28), the CPU 1 (20a) writes the template ID in the shared memory 15 and then uses the interrupt A signal (see FIG. 6). CPU2 (20b) is notified. Upon receiving the interrupt A signal, the CPU 2 (20b) executes a warning output process (step S28), details of which will be described later.

  Next, the CPU 1 (20a) determines whether or not matching processing has been performed for all template images in accordance with the recognition order set in step S13 (step S16), and has not performed matching processing for all template images. Sometimes (No in step S16), the process returns to step S13, and the processes in and after step S13 are executed again. If all template images have been matched (Yes in step S16), the process returns to step S11, and the processes in and after step S11 are repeatedly executed.

  On the other hand, CPU2 (20b) performs the road condition determination process shown in FIG. As shown in FIG. 5, the CPU 2 (20b) first acquires the current position of the vehicle 2 via the current position acquisition unit 25 (step S21). Next, referring to the map information 11 based on the current position, the road position where the vehicle 2 is traveling is specified (map matching: step S22). Here, the road position includes not only a link number for identifying the road but also a distance traveled by the vehicle 2 from one end (that is, an intersection) of the road. Therefore, the distance to the next intersection and the distance to the curve or bending point on the road can be known from the road position.

  Therefore, the CPU 2 (20b) determines whether or not there is an intersection within 100m ahead of the current position (step S23), and when there is an intersection (Yes in step S23), the intersection is referred to the map information 11 Information such as the presence / absence of traffic lights and the type of intersecting road is acquired (step S24). If there is no intersection (No in step S23), step S24 is skipped.

  Similarly, the CPU 2 (20b) determines whether or not there is a curve within 100m ahead of the current position (step S25). When there is a curve (Yes in step S25), the CPU 2 (20b) refers to the map information 11 to determine the curve. The type of curve (left curve, right curve, S-curve, etc.) is determined (step S26). If there is no curve (No in step S25), step S26 is skipped.

  As described above, the CPU 2 (20b) acquires the road condition on the road on which the vehicle 2 is traveling, refers to the recognition order information 12, and acquires the road condition number corresponding to the road condition. And CPU2 (20b) notifies the acquired road condition number to an image recognition process (step S27). At the time of the notification, the CPU 2 (20b) writes the acquired road condition number in the shared memory 15 and then interrupts the CPU 1 (20a) using the interrupt B signal (see FIG. 6). Insert.

  Thereafter, the CPU 2 (20b) returns to step S21, and repeatedly executes the processing after step 21.

  Next, the warning output process will be described in detail. The feature of the warning output process in the present embodiment is that when a road sign is recognized by the image recognition processing unit 23, the information related to the road sign is not immediately output as it is. The reason for doing so is that it is judged that when a road sign is recognized every time information is output, it is felt noisy for the driver of the vehicle 2 and hinders driving.

  Therefore, in the present embodiment, when the CPU 2 (20b) receives notification from the CPU 1 (20a) that the road sign is recognized, the information related to the traveling state of the vehicle 2 according to the recognized road sign. To get. Then, the CPU 2 (20b) determines that the vehicle 2 is not necessarily driven properly based on the information on the traveling state, or that the driver can be encouraged to drive appropriately by notifying the road sign. In such a case, a warning about the recognized road sign (for example, “the speed limit is exceeded”) is output via the display device 60 or the speaker 70.

  Here, the traveling state includes information on the road state of the road on which the vehicle 2 is traveling and information indicating the traveling state of the vehicle 2. Information representing the traveling state at this time is information input via the sensor interface 40, and includes, for example, traveling speed of the vehicle 2, traveling direction, lighting information of a direction indicator, and the like.

  In the present embodiment, as shown in FIG. 7, information that associates the recognized road sign, travel information, and warning information is created in advance as a warning output condition table (not shown in FIG. 1), It is stored in the storage unit 10. That is, a condition is provided in advance for determining that the vehicle 2 is not necessarily being driven appropriately or that the driver can be encouraged to drive appropriately by notifying the road sign. Then, the processing unit 20 detects the road sign recognized by the image recognition processing unit 23 and the traffic situation determined by the traveling state determination unit 24 (in the example of FIG. 7, the road position of the vehicle 2 (here, the intersection at the front). The warning information to be output can be obtained by referring to the warning output condition table based on the travel speed, lighting information of the direction indicator, etc.).

  By the way, according to FIG. 7, when “traffic light is red” is recognized as a road sign, a warning indicating that “traffic light is red” when the vehicle 2 is located within 100 m before the intersection. Is output. Conversely, when “the traffic light is green or yellow” is recognized, no warning is output.

  When “pause” is recognized as the road sign, the road position (distance to the front intersection) and the traveling speed of the vehicle 2 and the traveling speed are acquired as the traveling state, and the vehicle 2 at the traveling speed is safely When the distance that can be stopped exceeds the distance to the front intersection, a warning prompting “pause” is output. That is, when it is determined that the traveling state is not safe, a warning is output.

In the example of FIG. 7, the traveling speed is determined corresponding to three points where the distance to the front intersection is 50 m, 30 m, and 20 m, but the value is calculated based on the following equation. . That is, it is known that when the traveling vehicle 2 brakes, the stopping distance is expressed by the following equation.
Stop distance = idling distance + braking distance idling distance = driver's reaction time (approximately 0.75 seconds) x running speed braking distance = (running speed) ² / ( ² x 9.8 x friction coefficient)

According to these equations, for example, when the traveling speed is 60 km / h, the stopping distance is 32.7 m. However, this stop distance is a stop distance when a strong brake is applied, and is not necessarily an appropriate stop distance. Also, this distance may become longer depending on the road surface condition. Therefore, in order for the vehicle 2 to stop properly, a longer stop distance is required. Therefore, here, a target stop distance represented by the following equation is defined.
Target stop distance = stop distance + 1 second travel distance And, if the target stop distance obtained from the travel speed of the vehicle 2 is longer than the distance to the front intersection, it is determined that the vehicle 2 cannot be stopped properly, Added a warning.

  Further, when “no entry” is recognized as the road sign, the road position of the vehicle 2 (distance to the intersection ahead) and the lighting state of the direction indicator are acquired as the driving situation. For example, if the straight road is “no entry” and the direction indicator is not lit even within 30 m before the intersection, “the road ahead is prohibited” or “direction” A warning indicating “indicator lighting instruction” is output.

  In addition, when “one-way” is recognized as a road sign, the traveling direction of the vehicle 2 is acquired as the traveling state, and when the vehicle is traveling backward on the traveling road, A warning indicating "medium" is output. When the road sign “Curve attention” is recognized, a warning indicating “Curve attention” is output regardless of the traveling state of the vehicle 2. When “restricted speed” is recognized as a road sign, the traveling speed of the vehicle 2 is acquired as the traveling state, and when the traveling speed is greater than the recognized restricted speed, “restricted speed limit” A warning indicating is output.

  Of the above road signs, when “traffic light is red”, “temporary stop” and “no entry” are recognized, a warning is output while measuring the distance to the intersection, but “one-way” When “Curve attention” and “Over speed limit” are recognized, and even when a predetermined condition is satisfied, a warning is output immediately upon recognition. Among them, “Curve attention” is a sign for predicting danger, so that if it is recognized, a warning is output unconditionally.

  In addition, when the CPU 1 (20a) recognizes a plurality of road signs in the shooting screen in front of the vehicle 2, the CPU 2 (20b) determines the traveling state of the vehicle 2 at that time, and recognizes all the recognized signs. The warning output conditions in FIG. 7 are checked for the road signs, and all warning outputs that satisfy the conditions are output.

  FIG. 8 is a flowchart showing the warning output process described above. As described above, when the CPU 1 (20a) recognizes the road sign, the CPU 2 (20b) writes the recognized road sign template ID in the shared memory 15 (see FIG. 6). When a notification that the sign is recognized is received, first, the template ID is acquired by referring to the shared memory 15 (step S31).

  Next, the CPU 2 (20 b) performs information on the traveling speed and traveling direction of the vehicle 2 obtained via the sensor interface 40 as processing of the traveling state determination unit 24, and also processing of the road state determination unit 21. The obtained road position of the vehicle 2 (distance to the front intersection), that is, information on the traveling state is acquired (step S32).

  Next, the CPU 2 (20b) refers to the warning output condition table stored in the storage unit 10, and outputs a warning output condition (road position, travel speed, and other information in FIG. 7) corresponding to the template ID. And further, warning information is acquired (step S33). When the traveling state information acquired in step S32 satisfies the warning output condition, the warning information is output (step S34).

  As described above, according to the present embodiment, the processing unit 20 determines the road condition of the vehicle 2 and recognizes the road sign that is predicted to exist on the road according to the road condition. The time to reach recognition can be shortened. Also, considering the case where there are more than six kinds of road signs, the image recognition process shown in FIG. 5 is performed for all road signs within a predetermined time (for example, 0.5 seconds). In some cases, the matching process cannot be completed for the template, and the matching process is interrupted. However, in this embodiment, since the road sign predicted to exist is recognized with priority, even if the matching process is terminated in the middle, the matching process after the termination is a matching process for a non-existing road sign. There is a large possibility, and it is hardly affected by censoring during processing. Therefore, in this embodiment, the road sign predicted to exist can be recognized more reliably.

  Hereinafter, a supplementary description will be given of an embodiment obtained by extending this embodiment.

  FIG. 9 is a diagram showing an embodiment in which the recognition of road signs in the school zone is expanded. In general, the school zone is often set within a range of 500 m from the school facility. Therefore, here, for example, when a range of a square 81 having a side of 1 km is provided around the school 80, and the vehicle 2 travels on the general roads 82, 83, and 84 included in this range, the road in the school zone The sign 450 is preferentially recognized.

  For this purpose, first, a template image of a school zone road sign 450 is registered in the template image information 13 of FIG. The processing unit 20 not only acquires road status information as processing of the road status determination unit 21 but also refers to the map information 11 within a predetermined range centered on the current position of the vehicle 2 (for example, It is determined whether or not there is a school in a 1 km square. When there is a school within the range, the recognition order of the road signs 450 in the school zone is set high as processing of the recognition order determination unit.

  As described above, the school zone 81 can quickly and reliably recognize the road sign 450 in the school zone. Note that the processing described above is not applied when the vehicle 2 is traveling on an expressway.

  FIG. 10 is a diagram showing an example of recognizing a template image by limiting the recognition target range in accordance with the type of road sign in the captured images 90 and 90a. For example, as shown in FIG. 10A, the temporary road sign 132 that is a normal road sign is often installed on the left hand in front of the traveling direction of the vehicle 2. Therefore, when such a road sign 132 is recognized, the recognition of the road sign 132 is performed by limiting the recognition target range to, for example, a range 91 of about 1/4 of the upper left side of the captured image 90.

  As shown in FIG. 10 (b), the traffic light 135 is installed on a high pole and is often installed on the lane. The traffic signal 135 is recognized only when the target range is, for example, the range 92 of about ½ to ３ on the left side of the captured image 90a.

  In order to change the recognition target range according to the type of road sign, information for designating the range to be recognized according to each template ID of the template image information 13 (see FIG. 3) (for example, upper left 1 / 4, upper 1/3, etc.) may be set. In that case, the image recognition processing unit 23 can obtain information specifying a recognition range with reference to the template image information 13 based on the template ID. The image recognition process may be performed within the designated range.

  As described above, it is possible to shorten the time until the road sign is recognized by limiting the range of the recognition target.

The figure which showed the example of the structure of the car navigation apparatus which concerns on embodiment. The figure which showed the example of the condition where the car navigation apparatus which concerns on embodiment is applied. The figure which showed the example of the structure of template image information. The figure which showed the example of the structure of recognition order information. The figure which showed the example of the flowchart of the image recognition process in a car navigation apparatus, and a road condition determination process. The figure which showed the example of the internal structure of the process part and memory | storage part of a car navigation apparatus. The figure which showed the example of the warning output condition table. The figure which showed the example of the flowchart of a warning output process. The figure which showed embodiment extended about recognition of the road sign of a school zone. The figure which showed the example which restrict | limits the range of the recognition target of a picked-up image according to the kind of road sign, and recognizes a template image.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Car navigation apparatus 2 Vehicle 3 Road sign 10 Memory | storage part 11 Map information 12 Recognition order information 13 Template image information 15 Shared memory 20 Processing part 21 Road condition determination part 22 Recognition order determination part 23 Image recognition processing part 24 Running condition determination part 25 Present Position acquisition unit 26 Sensor information acquisition unit 27 Image acquisition unit 28 Warning output unit 30 GPS receiver 40 Sensor interface 50 Camera 60 Display device 70 Speaker

Claims (12)

An information processing device mounted on a vehicle, a storage device that stores road map information, a positioning device that measures the current position of the vehicle, and a camera that captures the front in the traveling direction of the vehicle, A car navigation device for guiding the vehicle by displaying a guidance route to a destination set by a driver of the vehicle on a display device,
The storage device
Template image information of road signs installed in the vicinity of the road,
The recognition order of the template image of the road sign for each road situation assumed in advance with respect to the road situation represented by at least one of the relative positional relationship between the feature point of the road including the intersection and the vehicle position and the attribute of the road Recognition order information that defines
Remember
The information processing apparatus includes:
Road condition determination means for determining the road condition at the current position of the road on which the vehicle is traveling, based on the current position obtained by the positioning device and the map information;
Recognizing order determining means for determining the recognizing order of the template images with reference to the recognizing order information according to the road condition determined by the road condition determining means;
Image recognition means for recognizing the template image according to the recognition order determined by the recognition order determination means for the captured image captured by the camera;
A car navigation device characterized by comprising: The car navigation device further includes:
A sensor interface connected to a sensor for acquiring information indicating the running state of the vehicle,
The information processing apparatus further includes:
A traveling state determination unit that determines a traveling state of the vehicle based on a road state at the current position of the vehicle and information indicating the traveling state of the vehicle acquired through the sensor interface with respect to a road on which the vehicle is traveling. When,
The template image is recognized by the image recognizing means, and the traveling state of the vehicle determined by the traveling state determining means satisfies a predetermined condition set in advance according to a road sign corresponding to the recognized template image. Warning output means for outputting a warning related to the road sign when satisfied,
The car navigation device according to claim 1, further comprising: The information processing apparatus further includes:
Facility determination means for referring to the map information based on the current position of the vehicle acquired by the positioning device and determining whether or not there is a predetermined facility within a predetermined distance from the current position;
When it is determined that the predetermined facility is present by the facility determining means, when the recognition order of the template image is determined by the recognition order determining means, the template image of the road sign associated with the predetermined facility is determined. The car navigation device according to claim 1, wherein the recognition order is moved forward. The information processing apparatus includes:
The car navigation device according to claim 1, wherein when the template recognition of the captured image is performed by the image recognition unit, a recognition target area in the captured image is changed according to a type of the template image. An information processing device mounted on a vehicle, a storage device that stores road map information, a positioning device that measures the current position of the vehicle, and a camera that captures the front in the traveling direction of the vehicle, A road sign recognition method in a car navigation device for displaying a guidance route to a destination set by a driver of a vehicle on a display device and guiding the vehicle,
The storage device
Template image information of road signs installed in the vicinity of the road,
The recognition order of the template image of the road sign for each road situation assumed in advance with respect to the road situation represented by at least one of the relative positional relationship between the feature point of the road including the intersection and the vehicle position and the attribute of the road Recognition order information that defines
Remember
The information processing apparatus includes:
A road condition determination step for determining the road condition at the current position of the road on which the vehicle is traveling, based on the current position obtained by the positioning device and the map information;
A recognition order determination step for determining the recognition order of the template images with reference to the recognition order information according to the road condition determined in the road condition determination step;
An image recognition step for recognizing the template image according to the recognition order determined in the recognition order determination step for the captured image captured by the camera;
The road sign recognition method characterized by performing this. The car navigation device further includes:
A sensor interface connected to a sensor for acquiring information indicating the running state of the vehicle,
The information processing apparatus further includes:
A traveling state determination step for determining the traveling state of the vehicle based on the road state at the current position of the vehicle and information indicating the traveling state of the vehicle acquired via the sensor interface for the road on which the vehicle is traveling. When,
The template image is recognized in the image recognition step, and the traveling state of the vehicle determined in the traveling state determination step satisfies a predetermined condition set in advance according to a road sign corresponding to the recognized template image. A warning output step for outputting a warning related to the road sign when satisfied,
The road sign recognition method according to claim 5, wherein: The information processing apparatus further includes:
A facility determination step of referring to the map information based on the current position of the vehicle acquired by the positioning device and determining whether or not there is a predetermined facility within a predetermined distance from the current position;
When it is determined in the facility determination step that the predetermined facility is present, when the recognition order of the template image is determined in the recognition order determination step, a template image of a road sign associated with the predetermined facility is determined. The road sign recognition method according to claim 5, wherein the recognition order is moved forward. The information processing apparatus includes:
The road sign recognition method according to claim 5, wherein when the template image is recognized for the captured image in the image recognition step, a recognition target area in the captured image is changed according to a type of the template image. An information processing device mounted on a vehicle, a storage device that stores road map information, a positioning device that measures the current position of the vehicle, and a camera that captures the front in the traveling direction of the vehicle, A road sign recognition program in a car navigation device that displays a guidance route to a destination set by a driver driver of a vehicle on a display device and guides the vehicle,
In the storage device,
Map information configured to include road position, connection relationship and attribute information, and feature position information;
Template image information of road signs installed in the vicinity of the road,
The recognition order of the template image of the road sign for each road situation assumed in advance with respect to the road situation represented by at least one of the relative positional relationship between the feature point of the road including the intersection and the vehicle position and the attribute of the road Recognition order information that defines
Remember
In the information processing apparatus,
A road condition determination step for determining the road condition at the current position of the road on which the vehicle is traveling, based on the current position obtained by the positioning device and the map information;
A recognition order determination step for determining the recognition order of the template images with reference to the recognition order information according to the road condition determined in the road condition determination step;
An image recognition step for recognizing the template image according to the recognition order determined in the recognition order determination step for the captured image captured by the camera;
Road sign recognition program for running. The car navigation device further includes:
A sensor interface connected to a sensor for acquiring information indicating the running state of the vehicle,
In addition to the information processing apparatus,
A traveling state determination step for determining the traveling state of the vehicle based on the road state at the current position of the vehicle and information indicating the traveling state of the vehicle acquired via the sensor interface for the road on which the vehicle is traveling. When,
The template image is recognized in the image recognition step, and the traveling state of the vehicle determined in the traveling state determination step satisfies a predetermined condition set in advance according to a road sign corresponding to the recognized template image. A warning output step for outputting a warning related to the road sign when satisfied,
The road sign recognition program according to claim 9, wherein: In addition to the information processing apparatus,
Referring to the map information based on the current position of the vehicle acquired by the positioning device, and executing a facility determination step of determining whether there is a predetermined facility within a predetermined distance from the current position;
When it is determined in the facility determination step that the predetermined facility is present, when the recognition order of the template image is determined in the recognition order determination step, a template image of a road sign associated with the predetermined facility is determined. The road sign recognition program according to claim 9, wherein the recognition order is moved forward. In the information processing apparatus,
The road sign recognition program according to claim 9, wherein when the template image is recognized for the photographed image in the image recognition step, a recognition target area in the photographed image is changed according to a type of the template image.

Images