Notice recognition device, notice recognition system, and notice recognition method

Abstract

The purpose of the present disclosure is to shorten the time required to recognize a traffic sign, when a traffic sign in the vehicle periphery is recognized by pattern-matching a captured image with a template image. A sign recognition device (101) of the present disclosure comprises: a reception unit (12) that is mounted in a vehicle (301) and receives an identification signal transmitted from a traffic sign (201) installed on the road; a display position determination unit (13) that estimates position information of the traffic sign (201) on the basis of the identification signal; and a recognition unit (14) that recognizes the traffic sign (201) shown in a captured image of the periphery of the vehicle (301), by executing pattern matching, in accordance with a predetermined priority order, between the captured image and a plurality of template images which were prepared for each type of the traffic sign (201). The recognition unit (14) identifies, on the basis of the position information of the traffic sign, a partial area of the captured image as a target area where the traffic sign is captured, and sets the target area as a target for pattern matching.

Description

[Technical field]

[0001] TECHNICAL FIELD This disclosure relates to traffic sign recognition. [Background technology]

[0002] On the road, there are traffic notices such as traffic lights, road signs, and road markings that display information necessary for traffic. By visually checking these traffic notices, drivers can grasp the information necessary for traffic and ensure smooth traffic. If drivers miss traffic notices, smooth traffic cannot be guaranteed. Therefore, a system that detects traffic notices around the vehicle and notifies the driver is useful.

[0003] Driving support systems that recognize road signs around a vehicle and warn the driver of the road signs as necessary have already been put to practical use. For example, Patent Document 1 discloses a system that recognizes signs ahead of a vehicle by pattern matching an image of the area in front of the vehicle captured by an on-board camera with template images of road signs or the like that are previously installed on the vehicle. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 11-306498 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the road sign recognition method described in Patent Document 1 subjects all areas in the captured image to pattern matching, which results in a problem that it takes a long time to recognize road signs and it is not possible to notify the driver of road signs at an appropriate time.

[0006] The present disclosure has been made to solve the above-mentioned problems, and aims to reduce the time required to recognize traffic signs around a vehicle by pattern matching a captured image with a template image. [Means for solving the problem]

[0007] The notice recognition device disclosed herein includes an image acquisition unit that acquires a captured image of the periphery of a vehicle, a receiver unit that is mounted on the vehicle and receives an identification signal transmitted from a traffic notice installed on a road, a notice position determination unit that estimates position information of the traffic notice indicating the position of the traffic notice relative to the vehicle based on the identification signal, and a recognition unit that recognizes the traffic notice shown in the captured image by pattern matching a plurality of template images prepared for each type of traffic notice according to a predetermined priority order with the captured image. The recognition unit identifies a part of the captured image as the target range in which the traffic notice was captured based on the traffic notice position information estimated by the notice position determination unit, and sets the target range as the target of pattern matching. The identification signal received by the receiver includes three or more identification signals received at different times at three or more different running positions of the vehicle. . Effect of the Invention

[0008] The notice recognition device of the present disclosure estimates the position information of the traffic notice based on the identification signal received from the traffic notice, and specifies the target range of the captured image to be pattern-matched with the template image based on the position information. Therefore, according to the notice recognition device of the present disclosure, the processing time for pattern matching can be shortened and the traffic notice can be recognized quickly compared to the case where the entire range of the captured image is matched with the template image. As a result, the information on the traffic notice can be notified to the vehicle occupants at the appropriate time.

[0009] The objects, features, aspects, and advantages of the present disclosure will become more apparent from the following detailed description and the accompanying drawings. [Brief description of the drawings]

[0010] [Figure 1]It is a block diagram showing the configuration of the bulletin board recognition system according to Embodiment 1. [Diagram 2] It is a flowchart showing the operation of the bulletin board recognition device according to Embodiment 1. [Diagram 3] It is a diagram showing the horizontal angle and distance of a road sign with respect to a camera. [Figure 4] It is a diagram showing the recognition target range in a captured image. [Diagram 5] It is a diagram showing a state where the bulletin board recognition device receives radio waves from a traffic bulletin board at three different locations. [Figure 6] It is a diagram showing an example of displaying traffic bulletin board information on an HUD. [Figure 7] It is a diagram showing a state where the bulletin board recognition device misrecognizes the position of a traffic bulletin board by receiving a reflected wave. [Figure 8] It is a block diagram showing the configuration of the bulletin board recognition system according to Embodiment 2. [Figure 9] It is a flowchart showing the operation of the bulletin board recognition device according to Embodiment 2. [Figure 10] It is a diagram showing a state where the bulletin board recognition device changes the priority order of template images that match a captured image of a camera based on type information received from a road sign. [Figure 11] It is a block diagram showing the configuration of the bulletin board recognition system according to Embodiment 3. [Figure 12] It is a flowchart showing the operation of the bulletin board recognition device according to Embodiment 3. [Figure 13] It is a diagram showing the hardware configuration of the bulletin board recognition device and the road sign.

Embodiments for Carrying Out the Invention

[0011] <A. Embodiment 1> <A-1. Configuration> FIG. 1 is a block diagram showing the configuration of a bulletin board recognition system 1001 according to Embodiment 1. The bulletin board recognition system 1001 includes a traffic bulletin board 201 and a vehicle 301.

[0012] The notice recognition system 1001 is a system that recognizes traffic notices 201 around a vehicle 301 and notifies the driver of the vehicle 301 of information about the recognized traffic notices 201. The traffic notices 201 are notices placed on roads that display traffic information necessary for drivers to ensure smooth traffic flow. The traffic notices 201 are, for example, traffic lights, road signs, or road markings. In the following description, the traffic notices 201 are assumed to be road signs.

[0013] In addition to the posting recognition device 101, the vehicle 301 is provided with a camera 51, a display 53, and a speaker 54. In Fig. 1, the posting recognition device 101 will be described as a device mounted on the vehicle 301. However, components other than the receiver 12 of the posting recognition device 101 do not necessarily have to be provided in the vehicle 301, and may be provided in a server external to the vehicle 301, for example.

[0014] The notice recognition device 101 is connected to a camera 51, a display 53, and a speaker 54, and is configured to be able to use these.

[0015] The camera 51 captures the surroundings of the vehicle 301. The surroundings of the vehicle 301 typically include the road ahead of the vehicle 301, and include a range up to a point 300 m ahead of the vehicle 301, for example.

[0016] The display 53 and the speaker 54 are output devices that notify the driver of the vehicle 301 of information on the traffic notices 201 around the vehicle 301 that are recognized by the notice recognition device 101. Only one of the display 53 and the speaker 54 may be provided in the notice recognition device 101.

[0017] The display 53 displays information about the traffic notices 201 around the vehicle 301 recognized by the notice recognition device 101. The display 53 is, for example, a Center Information Display (CID) or a Head-Up Display (HUD).

[0018] The speaker 54 outputs by voice information about the traffic notices 201 around the vehicle 301 that have been recognized by the notice recognition device 101. The display 53 and the speaker 54 are both mounted on the vehicle 301, and are configured so that the driver of the vehicle 301 can recognize the information about the traffic notices 201 output by the display 53 and the speaker 54.

[0019] Next, a description will be given of the configuration of the notice recognition device 101. The notice recognition device 101 is configured to include an image acquisition unit 11, a reception unit 12, a notice position determination unit 13, a recognition unit 14, a template storage unit 15, and an output control unit 16.

[0020] The image acquisition unit 11 acquires captured images of the surroundings of the vehicle 301 from the camera 51 .

[0021] The receiving unit 12 performs wireless communication with the traffic notice 201. The receiving unit 12 receives radio waves transmitted from the traffic notice 201, demodulates the radio waves, and outputs a demodulated signal to the notice position determination unit 13. The radio waves transmitted from the traffic notice 201 include an identification signal indicating that the source of the radio waves is the traffic notice 201. The receiving unit 12 demodulates this identification signal and outputs it to the notice position determination unit 13. In other words, the receiving unit 12 receives the identification signal from the traffic notice 201.

[0022] The notice position determination unit 13 determines the relative position of the traffic notice 201 with respect to the vehicle 301 based on the identification signal received by the receiving unit 12 from the traffic notice 201. The relative position of the traffic notice 201 determined by the notice position determination unit 13 includes the direction of the traffic notice 201 with respect to the vehicle 301 and the distance between the vehicle 301 and the traffic notice 201.

[0023] The template storage unit 15 stores a plurality of template images for each type of traffic notice 201. The template image of the traffic notice 201 is an image of the exterior of the traffic notice 201. As described above, the traffic notice 201 is classified into traffic lights, road signs, and road markings. For traffic lights, template images may be stored for different types, such as pedestrian traffic lights or vertical traffic lights, for different shapes. For road signs and road markings, template images are stored for each type.

[0024] The recognition unit 14 acquires a captured image of the periphery of the vehicle 301 from the image acquisition unit 11, and identifies an area in the captured image in which the traffic notice 201 is likely to appear as a target area based on the relative position of the traffic notice 201 with respect to the vehicle 301 acquired from the notice position determination unit 13. The recognition unit 14 then performs pattern matching of only the target area of ​​the captured image with a plurality of template images stored in the template storage unit 15, and recognizes that a traffic notice 201 of the type of the template image is present in the periphery of the vehicle 301 when a certain template image matches the target area. The recognition unit 14 then outputs the type of the recognized traffic notice 201 and the relative position of the traffic notice 201 with respect to the vehicle 301 to the output control unit 16.

[0025] The order of template images in which the recognition unit 14 performs pattern matching with the target area of ​​the captured image (hereinafter referred to as the "matching order") is determined in advance. The recognition unit 14 acquires a desired template image from the template storage unit 15 in accordance with the matching order, and performs pattern matching of the template image with the target area of ​​the captured image. If the two do not match, the recognition unit 14 acquires the next template image from the template storage unit 15, and performs pattern matching of the template image with the target area of ​​the captured image. Note that the recognition unit 14 may perform pattern matching one image at a time, or may perform pattern matching in units of multiple images.

[0026] The matching order is determined, for example, in descending order of the risk that the driver misses the traffic sign 201. For example, when the traffic sign 201 is a road sign, the matching order may be determined in the order of the template image of the road sign "Prohibited Vehicle Entry", the template image of the road sign related to speed, and the template image of the road sign of stop. The magnitude of the risk that the driver misses the traffic sign 201 may be determined by the amount of fine when violating the traffic rules posted on the traffic sign 201.

[0027] The output control unit 16 acquires from the recognition unit 14 the recognition result of the traffic sign 201 by the recognition unit 14 and the relative position of the traffic sign 201 with respect to the vehicle 301, and causes these to be output from the display 53 or the speaker 54.

[0028] Next, the configuration of the traffic sign 201 will be described. The traffic sign 201 includes a transmission unit 21 and a control unit 22.

[0029] The transmission unit 21 performs wireless communication with the reception unit 12 of the sign recognition device 101. The transmission unit 21 transmits an identification signal of the traffic sign 201 to the reception unit 12 in accordance with an instruction from the control unit 22. The identification signal includes at least information indicating that the transmission source is a traffic sign. The transmission unit 21 transmits the identification signal to the reception unit 12, for example, at a fixed period.

[0030] The control unit 22 controls the transmission of the identification signal by the transmission unit 21.

[0031] <A-2. Operation> FIG. 2 is a flowchart showing the operation of the sign recognition device 101. Hereinafter, the operation of the sign recognition device 101 will be described along the flowchart of FIG. 2. The flowchart of FIG. 2 starts, for example, together with the start of the travel of the vehicle 301.

[0032] First, in step S101, it is determined whether the reception unit 12 has received an identification signal. If the reception unit 12 has not received the identification signal in step S101, the process of the sign recognition device 101 proceeds to step S105.

[0033] In step S105, the recognition unit 14 performs pattern matching on the entire range of the captured image of the camera 51 acquired by the image acquisition unit 11 with the multiple template images stored in the template storage unit 15 in order according to the default matching order. The default matching order (priority order) is determined, for example, in order of increasing risk if the driver misses the traffic notice 201. Here, the risk is, for example, the amount of fine for violating the traffic information posted by the traffic notice 201. If a pattern matching with the template image is found in the captured image, the recognition unit 14 ends the pattern matching at that point, and the processing of the notice recognition device 101 proceeds to step S106. If the pattern in the captured image matches the template image, the recognition unit 14 determines that the traffic notice corresponding to the template image is present at a position corresponding to the matched pattern in the captured image. That is, the recognition unit 14 detects the traffic notice 201 corresponding to the matched template image. The recognition unit 14 ends the pattern matching at that point, and the processing of the notice recognition device 101 proceeds to step S106. Furthermore, if the recognition unit 14 performs pattern matching on all template images but finds that the target range of the captured image does not match the target range of the captured image, the pattern matching ends and the processing of the notice recognition device 101 proceeds to step S106.

[0034] If the receiving unit 12 receives an identification signal in step S101, the processing of the notice recognition device 101 proceeds to step S102.

[0035] In step S102, the notice position determination unit 13 estimates the relative position of the traffic notice 201 with respect to the vehicle 301 based on the identification signal received by the receiver 12. Specifically, the notice position determination unit 13 calculates the direction and distance of the traffic notice 201 with respect to the vehicle 301 based on the radio wave intensity of the identification signal received by the receiver 12.

[0036] 3 is a diagram showing a method for calculating the position of the traffic notice 201 based on the radio wave intensity of the identification signal. For example, as shown in FIG. 3, the receiver 12 receives radio waves from the traffic notice 201 at three different positions as the vehicle 301 travels. First, based on the intensity of the radio wave received by the receiver 12 the first time, it is found that the traffic notice 201 is located on a circle 62 with a radius of 50 m from the reception point. Next, based on the intensity of the radio wave received by the receiver 12 the second time, it is found that the traffic notice 201 is located on a circle 63 with a radius of 40 m from the reception point. Furthermore, based on the intensity of the radio wave received by the receiver 12 the third time, it is found that the traffic notice 201 is located on a circle 64 with a radius of 30 m from the reception point. Therefore, the position of the traffic notice 201 is specified as the intersection of the circles 62, 63, and 64.

[0037] Returning to the description of the flowchart in Fig. 2, after step S102, in step S103, the recognition unit 14 specifies a part of the image captured by the camera 51 as a target range for pattern matching, based on the relative position of the traffic notice 201 estimated by the notice position determination unit 13.

[0038] 4 shows the horizontal positional relationship between camera 51 mounted on vehicle 301 and traffic notice 201. The horizontal angle of view of camera 51 is defined as θ1. Assume that camera 51 is installed facing the traveling direction of vehicle 301. The horizontal angle that a line connecting traffic notice 201 and camera 51 makes with the traveling direction of vehicle 301 is defined as the horizontal deviation angle θ2 of traffic notice 201.

[0039] 5 shows a target range 61 in an image captured by camera 51. The left-right position of target range 61 in the captured image is determined based on the horizontal angle of view θ1 of camera 51 and the horizontal deviation angle θ2 of traffic notice 201. The up-down position of target range 61 in the captured image is determined based on the vertical angle of view of camera 51 and the distance of traffic notice 201 to vehicle 301.

[0040] Returning to the description of the flowchart in FIG. 2, in step S104 after step S103, the recognition unit 14 performs pattern matching of the target range of the captured image of the camera 51 with the multiple template images stored in the template storage unit 15 in order according to the default matching order. If the pattern in the target range of the captured image matches the template image, the recognition unit 14 determines that the traffic notice corresponding to the template image is present at a position corresponding to the matched pattern in the target range. That is, the recognition unit 14 detects the traffic notice 201 corresponding to the matched template image. The recognition unit 14 ends the pattern matching at that point, and the processing of the notice recognition device 101 proceeds to step S106. Also, if the recognition unit 14 performs pattern matching on all template images but does not match the target range of the captured image, the pattern matching ends, and the processing of the notice recognition device 101 proceeds to step S106.

[0041] After the pattern matching in step S104 or step S105, in step S106, the recognition unit 14 determines whether or not the traffic notice 201 is detected. If the traffic notice 201 is not detected in step S106, the processing of the notice recognition device 101 returns to step S108. If the traffic notice 201 is detected in step S106, the processing of the notice recognition device 101 proceeds to step S107.

[0042] In step S107, the output control unit 16 acquires the type and position information of the detected traffic notice 201 from the recognition unit 14, and causes the information to be output from the display 53 or the speaker .

[0043] FIG. 6 shows an example of the display of the type and position of the traffic sign 201 in the HUD which is an example of the display 53. FIG. 6 is an example of the display when the recognition unit 14 detects road signs of speed limit, stop, and crosswalk on the road ahead of the vehicle 301. A double dashed line frame is displayed on the HUD area so as to surround the road sign. Also, for the road sign of the speed limit, a display object 65 that displays the road sign in conjunction with the current speed of the vehicle 301 is displayed in the HUD area that overlaps the road ahead of the vehicle 301. Thereby, the driver can recognize the road signs ahead of the vehicle 301 and can drive along the road signs.

[0044] Returning to the description of the flowchart of FIG. 2. After step S107, in step S108, the sign recognition device 101 determines whether or not the travel of the vehicle 301 has ended. If the travel of the vehicle 301 has not ended in step S108, the process of the sign recognition device 101 returns to step S101. If the travel of the vehicle 301 has ended in step S108, the process of the sign recognition device 101 ends.

[0045] <A-3. Modification Example> As shown in FIG. 7, the radio wave transmitted from the transmission unit 21 of the traffic sign 201 may reach the vehicle 301 not only as a direct wave but also as a reflected wave reflected by a building. Therefore, there is a possibility that the sign recognition device 101 may misrecognize the arrival direction of the reflected wave as the direction of the traffic sign 201, or may misrecognize the distance to the traffic sign 201 as being short due to the radio wave intensity of the attenuated reflected wave.

[0046] Therefore, it is desirable that the radio wave of the sign signal transmitted from the transmission unit 21 of the traffic sign 201 has directivity to the target road of the traffic information posted by the traffic sign 201. Thereby, misrecognition of the position of the traffic sign 201 by the reflected wave in the sign recognition device 101 is suppressed.

[0047] <A-4. Effect> The notice recognition device 101 according to the first embodiment includes an image acquisition unit 11 that acquires a captured image of the periphery of the vehicle 301, a receiver unit 12 that is mounted on the vehicle 301 and receives an identification signal transmitted from a traffic notice 201 installed on a road, a notice position determination unit 13 that estimates position information of the traffic notice 201 including at least the direction of the traffic notice 201 relative to the vehicle based on the identification signal, and a recognition unit 14 that recognizes the traffic notice 201 shown in the captured image by pattern matching a plurality of template images prepared for each type of traffic notice 201 with the captured image according to a predetermined priority order. The recognition unit 14 identifies a part of the captured image as a target range in which the traffic notice 201 was captured based on the position information of the traffic notice 201 estimated by the notice position determination unit 13, and subjects the target range to pattern matching.

[0048] Therefore, the notice recognition device 101 can reduce the processing time for pattern matching compared to the case where the entire range of the captured image is pattern matched with the template image. As a result, the traffic notice 201 can be quickly recognized, and information about the traffic notice 201 can be notified to the driver at an appropriate time.

[0049] In addition, in the notice recognition device 101 according to the first embodiment, the identification signal received by the receiver 12 may include three or more identification signals received at three or more different traveling positions of the vehicle 301. In this case, the notice recognition device 101 can estimate the position of the traffic notice 201 with high accuracy.

[0050] In addition, the radio waves used when transmitting the identification signal from the surrounding traffic notices 201 may have directionality with respect to the target road of the surrounding traffic notices. In this case, it is possible to prevent the notice recognition device 101 from erroneously recognizing the position of the traffic notice 201 based on the reception strength or arrival direction of the reflected wave from a building or the like.

[0051] The bulletin board recognition system according to Embodiment 1 includes a traffic bulletin board 201 installed on a road for transmitting an identification signal, and a bulletin board recognition device 101 for recognizing traffic bulletin boards around the vehicle 301. Therefore, according to the bulletin board recognition system according to Embodiment 1, the processing time of pattern matching can be shortened compared with the case of performing pattern matching between the entire range of the captured image and the template image. As a result, it is possible to quickly recognize the traffic bulletin board 201 and notify the driver of the information of the traffic bulletin board 201 at an appropriate timing.

[0052] The bulletin board recognition method according to Embodiment 1 includes: an image acquisition unit 11 that acquires a captured image around the vehicle 301; a reception unit 12 mounted on the vehicle 301 that receives an identification signal transmitted from a traffic bulletin board 201 installed on a road; a bulletin board position determination unit 13 that estimates position information of the traffic bulletin board 201 representing the position of the traffic bulletin board 201 with respect to the vehicle 301 based on the identification signal; a recognition unit 14 that recognizes the traffic bulletin board 201 shown in the captured image by performing pattern matching between the captured image and a plurality of template images prepared for each type of the traffic bulletin board 201 in accordance with a predetermined priority order; and the recognition unit 14 specifies a partial range of the captured image as a target range in which the traffic bulletin board 201 is captured based on the position information of the traffic bulletin board 201 estimated by the bulletin board position determination unit 13, and sets the target range as a target for pattern matching. Therefore, according to the bulletin board recognition method according to Embodiment 1, the processing time of pattern matching can be shortened compared with the case of performing pattern matching between the entire range of the captured image and the template image. As a result, it is possible to quickly recognize the traffic bulletin board 201 and notify the driver of the information of the traffic bulletin board 201 at an appropriate timing.

[0053] <B. Embodiment 2> <B-1. Configuration> FIG. 8 is a block diagram showing the configuration of a bulletin board recognition system 1002 according to Embodiment 2. The bulletin board recognition system 1002 includes a traffic bulletin board 202 and a vehicle 301. In the present embodiment, the same reference numerals are given to the same configurations as those in Embodiment 1, and the description of the parts common to Embodiment 1 is omitted as appropriate.

[0054] In addition to the configuration of the traffic sign 201 according to the first embodiment, the traffic sign 202 includes an identification signal storage unit 23. The identification signal storage unit 23 stores type information 231 representing the type of the traffic sign 202.

[0055] The transmission unit 21 transmits the identification signal of the traffic sign 202 in response to an instruction from the control unit 22. This identification signal includes the type information 231. That is, the transmission unit 21 extracts the type information 231 from the identification signal storage unit 23 and transmits it including the type information 231 in the identification signal.

[0056] In the sign recognition system 1002, in addition to the sign recognition device 102, a camera 51, a display 53, and a speaker 54 are provided in the vehicle 301.

[0057] The sign recognition device 102 has the same configuration as the sign recognition device 101 according to the first embodiment, but the processing of the recognition unit 14 is partly different. The recognition unit 14 of the present embodiment changes the priority order of the template images for performing pattern matching according to the type information 231 included in the identification signal received by the reception unit 12 from the default. Specifically, the recognition unit 14 changes the priority order so that the template image of the traffic sign 202 of the type indicated by the type information 231 is preferentially subjected to pattern matching.

[0058] <B-2. Operation> FIG. 9 is a flowchart showing the operation of the sign recognition device 102. The flow in FIG. 9 starts, for example, together with the start of the travel of the vehicle 301. The flow in FIG. 9 is obtained by adding steps S201 and S202 between step S103 and step S104 of the flow in FIG. 2 described in the first embodiment.

[0059] In step S201, the recognition unit 14 determines whether or not the type information 231 is included in the identification signal received by the reception unit 12. If the type information 231 is included in the identification signal in step S201, the processing of the sign recognition device 102 proceeds to step S202.

[0060] In step S202, the recognition unit 14 changes the priority of the template image for pattern matching with the target range of the captured image from the default.

[0061] FIG. 10 is a diagram showing an example of changing the priority. In the example of FIG. 10, it is assumed that the identification signal of the road sign received by the receiving unit 12 from the transmitting unit 21 includes the type information 231 of the road sign. Further, the type information 231 indicates a speed limit road sign as the type of the road sign. The recognition unit 14 changes the priority of the template image so that the template image of the speed limit road sign becomes the top priority. For example, if the default priority order for template image matching is in the order of a no-entry road sign, a speed limit road sign, and a stop sign, the recognition unit 14 changes the priority order to the speed limit road sign, the no-entry road sign, and the stop sign.

[0062] <B-3. Effect> In the second embodiment, the identification signal transmitted from the traffic sign 202 includes type information representing the type of the traffic sign 202. Then, the recognition unit 14 of the sign recognition device 102 changes the priority of the pattern matching of the template image so that the template image of the traffic sign 202 of the type represented by the type information becomes the top priority. Thereby, the recognition unit 14 can shorten the processing time of the pattern matching between the captured image and the template image. As a result, the sign recognition device 102 can quickly recognize the traffic sign 202 and notify the driver of the information of the traffic sign 202 at an appropriate timing.

[0063] <C. Third Embodiment> <C-1. Configuration> FIG. 11 is a block diagram showing the configuration of a sign recognition system 1003 according to the third embodiment. The sign recognition system 1003 includes a traffic sign 203 and a vehicle 301. In the present embodiment, the same reference numerals are given to the same configurations as in the first and second embodiments, and the description of the common parts with the first and second embodiments is omitted as appropriate.

[0064] The traffic notice 203 of the third embodiment differs from the traffic notice 202 of the second embodiment in the stored information stored in the identification signal storage unit 23. The identification signal storage unit 23 of the traffic notice 203 stores target road information 232 in addition to the type information 231. The target road information 232 is information on the target road of the traffic information posted by the traffic notice 203.

[0065] In response to an instruction from the control unit 22, the transmission unit 21 of the third embodiment transmits an identification signal of the traffic notice 203. This identification signal includes the type information 231 and the target road information 232. That is, the transmission unit 21 retrieves the type information 231 and the target road information 232 from the identification signal storage unit 23, and transmits the same together with the type information 231 and the target road information 232 included in the identification signal.

[0066] In the notice recognition system 1003, a vehicle 301 is provided with a notice recognition device 103 as well as a camera 51, a GPS (Global Positioning System) sensor 52, a display 53, and a speaker .

[0067] The GPS sensor 52 receives GPS signals from GPS satellites, measures the position of the vehicle 301 , and outputs the position to the notice recognition device 103 .

[0068] The notice recognition device 103 includes a vehicle position determination unit 17 in addition to the components of the notice recognition device 102 according to the second embodiment. Also, the processing of the recognition unit 14 in the notice recognition device 103 is partially different from that in the second embodiment.

[0069] The vehicle position determination unit 17 determines whether the vehicle 301 is traveling on the target road of the traffic notice 203 based on the position information of the vehicle 301 obtained from the GPS sensor 52 and the target road information 232 of the traffic notice 203 obtained from the receiving unit 12.

[0070] If the vehicle 301 is traveling on the target road of the traffic sign 203, the recognition unit 14 of the present embodiment changes the priority order of matching of the template image according to the type of the traffic sign 203. If the vehicle 301 is not traveling on the target road of the traffic sign 203, the recognition unit 14 matches the template image with the captured image according to the default priority order.

[0071] <C-2. Operation> FIG. 12 is a flowchart showing the operation of the sign recognition device 103. The flow in FIG. 12 starts, for example, together with the start of travel of the vehicle 301. The flow in FIG. 12 is obtained by adding steps S301, S302, and S303 between step S202 and step S104 of the flow in FIG. 9 described in Embodiment 2.

[0072] In step S301, the vehicle position determination unit 17 determines whether the target road information 232 is included in the identification signal received by the reception unit 12. If the target road information 232 is included in the identification signal in step S301, the process of the sign recognition device 103 proceeds to step S302.

[0073] In step 302, the vehicle position determination unit 17 determines whether the vehicle 301 is traveling on the target road based on the GPS signal acquired from the GPS sensor 52 and the target road information 232 acquired from the reception unit 12.

[0074] If the vehicle 301 is traveling on the target road in step S302, the process of the sign recognition device 103 proceeds to step S104.

[0075] If the vehicle 301 is not traveling on the target road in step S302, the process of the sign recognition device 103 proceeds to step S303.

[0076] In step S303, the recognition unit 14 returns the priority order of matching the template image changed in step S202 with the captured image to the default.

[0077] <C-3. Effect> In Embodiment 3, the identification signal received by the bulletin board recognition device 103 from the traffic bulletin board 203 includes target road information 232 representing the target road of the traffic bulletin board 203. When the target road represented by the target road information 232 is not the road on which the vehicle 301 travels, the recognition unit 14 does not change the priority order of the pattern matching of the template image. Therefore, according to the bulletin board recognition device 103, even if the identification information from the traffic bulletin board 203 that has nothing to do with the vehicle 301 is received, it is possible to avoid performing the pattern matching process according to the traffic bulletin board 203.

[0078] <D. Hardware Configuration> Each of the above-described bulletin board recognition devices 101-103 and traffic bulletin boards 201-203 is realized by a processor 81, a memory 82, and a communication device 83 shown in FIG. 13.

[0079] In the bulletin board recognition devices 101-103, the receiving unit 12 is realized by the communication device 83, the template storage unit 15 is realized by the memory 82, and the image acquisition unit 11, the posting position determination unit 13, the recognition unit 14, and the output control unit 16 are realized by the processor 81 executing a program stored in the memory 82.

[0080] In the traffic bulletin boards 201-203, the transmitting unit 21 is realized by the communication device 83, the identification signal storage unit 23 is realized by the memory 82, and the control unit 22 is realized by the processor 81 executing a program stored in the memory 82.

[0081] The processor 81 is, for example, a central processing unit, a processing device, an arithmetic device, a microprocessor, a microcomputer, a DSP (Digital Signal Processor), or the like.

[0082] The programs stored in the memory 82 cause the computer to execute the methods of each part of the notice recognition device 101-103 or the traffic notice 201-203. Here, the memory 82 may be, for example, a non-volatile or volatile semiconductor memory such as a random access memory (RAM), a read only memory (ROM), a flash memory, an erasable programmable read only memory (EPROM), an electrically erasable programmable read only memory (EEPROM), a hard disk drive (HDD), a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, a digital versatile disk (DVD) and its drive device, or any storage medium to be used in the future.

[0083] Furthermore, while the notice recognition devices 101-103 have been described above as in-vehicle devices, they can also be applied to a system constructed by appropriately combining a PND (Portable Navigation Device), a communication terminal (e.g., a mobile terminal such as a mobile phone, a smartphone, or a tablet), the functions of applications installed thereon, a server, etc. In this case, the functions or components of the notice recognition devices 101-103 described above may be distributed among the devices that construct the system, or may be concentrated in one of the devices.

[0084] In addition, each embodiment can be freely combined, and each embodiment can be appropriately modified or omitted. The above description is an example in all respects. It is understood that countless modified examples not illustrated can be assumed. [Explanation of symbols]

[0085] 11 image acquisition unit, 12 receiving unit, 13 notice position determination unit, 14 recognition unit, 15 template storage unit, 16 output control unit, 17 vehicle position determination unit, 21 transmitting unit, 22 control unit, 23 identification signal storage unit, 51 camera, 52 GPS sensor, 53 display, 54 speaker, 61 target range, 81 processor, 82 memory, 83 communication device, 101-103 notice recognition device, 201-203 traffic notice, 231 type information, 232 target road information, 301 vehicle, 1001-1003 notice recognition system.

Claims

1. An image acquisition unit that acquires a captured image of the surroundings of the vehicle; a receiving unit mounted on the vehicle and configured to receive an identification signal transmitted from a traffic notice installed on a road; a notice position determination unit that estimates position information of the traffic notice, which indicates a position of the traffic notice relative to the vehicle, based on the identification signal; a recognition unit that recognizes the traffic notice shown in the captured image by pattern matching a plurality of template images prepared for each type of the traffic notice with the captured image in accordance with a predetermined priority order; The recognition unit identifies a part of the captured image as a target range in which the traffic notice is captured based on the position information of the traffic notice estimated by the notice position determination unit, and performs pattern matching on the target range. The identification signal received by the receiving unit includes three or more identification signals received at different timings at three or more different traveling positions of the vehicle. Poster recognition device.

2. The radio waves used to transmit the identification signal from the traffic notice have directivity with respect to a target road of the traffic notice. The notice recognition device according to claim 1 .

3. The identification signal includes type information representing a type of the traffic notice, the recognition unit changes the priority of pattern matching of the template images so that the template images of the traffic notices of the types represented in the type information have the highest priority. The notice recognition device according to claim 1 .

4. The identification signal includes target road information indicating a target road of the traffic notice, the recognition unit does not change the priority of pattern matching of the template image when the target road indicated in the target road information is not a road on which the vehicle travels. The notice recognition device according to claim 3.

5. An output control unit is further provided to output information about the traffic notice recognized by the recognition unit to an output device mounted on the vehicle. The notice recognition device according to claim 1 .

6. A traffic notice installed on a road and transmitting an identification signal; A notice recognition system including a notice recognition device that recognizes the traffic notices around a vehicle, The poster recognition device includes: An image acquisition unit that acquires a captured image of the surroundings of the vehicle; a receiving unit mounted on the vehicle and receiving the identification signal transmitted from the traffic notice; a notice position determination unit that estimates position information of the traffic notice, which indicates a position of the traffic notice relative to the vehicle, based on the identification signal; a recognition unit that recognizes the traffic notice shown in the captured image by pattern matching a plurality of template images prepared for each type of the traffic notice with the captured image in accordance with a predetermined priority order; The recognition unit identifies a part of the captured image as a target range based on the position information of the traffic notice estimated by the notice position determination unit, and performs pattern matching on the target range. The identification signal received by the receiving unit includes three or more identification signals received at different timings at three or more different traveling positions of the vehicle. Poster recognition system.

7. The image acquisition unit acquires a photographed image of the surroundings of the vehicle, A receiving unit mounted on the vehicle receives an identification signal transmitted from a traffic notice installed on a road, a notice position determination unit that estimates position information of the traffic notice, which indicates a position of the traffic notice relative to the vehicle, based on the identification signal; a recognition unit that recognizes the traffic notice shown in the captured image by pattern matching a plurality of template images prepared for each type of the traffic notice with the captured image in accordance with a predetermined priority order; The recognition unit identifies a part of the captured image as a target range in which the traffic notice is captured based on the position information of the traffic notice estimated by the notice position determination unit, and subjects the target range to the pattern matching; The identification signal received by the receiving unit includes three or more identification signals received at different timings at three or more different traveling positions of the vehicle. How to recognize postings.

Images