JP2009104330A - Hidden vehicle detection system, road side device, onboard device and hidden vehicle detection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect a hidden vehicle with high precision. <P>SOLUTION: A communication information generation part of a road side device generates communication information as information including the location and size of a vehicle with prescribed terminal equipment mounted thereon. An image information generation part of the road side device generates image information as information including the location and size of the vehicle detected by analyzing the image by a camera which images the traffic condition of a road. A vehicle arrangement information generation part of the road side device generates vehicle arrangement information corresponding to each vehicle by collating the communication information with the image information. A reliability addition part of a road side device adds reliability showing the level of information precision to each vehicle arrangement information When the reliability is a prescribed value or less, a hidden vehicle detection part of an onboard device changes the size included in the vehicle arrangement information, and detects the hidden vehicle. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a blind spot vehicle detection system, a roadside device, an in-vehicle device, and a blind spot vehicle detection method for detecting a blind spot vehicle existing at a position that becomes a blind spot from the own vehicle by an obstacle vehicle that is present between itself and the dead vehicle. The present invention relates to a blind spot vehicle detection system, a roadside device, an in-vehicle device, and a blind spot vehicle detection method that can detect a vehicle with high accuracy.

  There has been known a technique for collecting position information of a vehicle or a pedestrian holding a position reporting terminal that transmits the position of the user via wireless communication and warning the driver of the vehicle. There is also known a technique for warning a driver of an obstacle, an oncoming vehicle, or the like detected by analyzing an image from a camera installed in the vehicle or in the vicinity of a road.

  An attempt has been made to detect an object (for example, an oncoming vehicle or a pedestrian) existing in a blind spot as seen from the own vehicle by combining these methods. For example, Patent Document 1 discloses a warning system that detects a pedestrian in a blind spot position based on the position of a portable terminal carried by a pedestrian and the position of an obstacle imaged by an in-vehicle camera, and warns the driver. Is disclosed.

JP 2004-46426 A

  However, the technique disclosed in Patent Document 1 detects a blind spot pedestrian who does not hold a position reporting terminal, although it combines position information based on an image from a camera and position information based on a position reporting terminal. There is a problem that you can not. That is, the thing of patent document 1 is a technique on the assumption that the pedestrian who exists in a blind spot position seeing the own vehicle is carrying the position report terminal.

  However, not all pedestrians on the road or in the vicinity of the road hold such position reporting terminals. For this reason, in order to detect a blind spot pedestrian who does not hold the position reporting terminal, it is necessary to combine it with the analysis result of the image by the camera. The same applies to the case of using the one of Patent Document 1 to detect a blind spot vehicle.

  Here, the information obtained by analyzing the camera image depends on the field of view that depends on the weather or time, the accuracy of the camera, and the accuracy of the recognition program, so it is more accurate than the information obtained via the location reporting terminal. Is low information. Therefore, it is difficult to detect a blind spot vehicle or a blind spot pedestrian with high accuracy by simply combining information obtained by analyzing a camera image and information obtained via a position notification terminal.

  Therefore, how to realize a blind spot vehicle detection system, a roadside device, an in-vehicle device, and a blind spot vehicle detection method capable of detecting a blind spot vehicle with high accuracy has become a major issue.

  The present invention has been made to solve the above-described problems caused by the prior art, and provides a blind spot vehicle detection system, a roadside device, an in-vehicle device, and a blind spot vehicle detection method capable of detecting a blind spot vehicle with high accuracy. The purpose is to do.

  In order to solve the above-described problems and achieve the object, the present invention is a blind spot vehicle detection system that detects a blind spot vehicle existing at a position that becomes a blind spot from the own vehicle by a faulty vehicle existing between the subject vehicle and the vehicle. Detected by image analysis by means of communication via information generation means for generating via-communication information that is information including the position and size of a vehicle equipped with a predetermined terminal device, and a camera that captures traffic conditions on the road Image via information generating means for generating image via information, which is information including the position and size of the vehicle, and vehicle arrangement information corresponding to each vehicle by collating the communication via information with the image via information A reliability adding means for adding reliability indicating the level of accuracy of information to each vehicle arrangement information, and when the reliability is a predetermined value or less, the vehicle arrangement information Change the size included, characterized in that a blind spot vehicle detection means for detecting the blind spot vehicle.

  According to the present invention, it is detected by generating information via communication that is information including the position and size of a vehicle on which a predetermined terminal device is mounted, and performing image analysis on an image captured by a camera that captures traffic conditions on the road. When generating via-image information, which is information including the position and size of the vehicle, and comparing the via-communication information with the via-image information to generate vehicle arrangement information corresponding to each vehicle, for each vehicle arrangement information If the reliability is less than a predetermined value and the reliability is less than a predetermined value, the size included in the vehicle arrangement information is changed to detect the blind spot vehicle. By combining efficiently, it is possible to detect a blind spot vehicle included only in the information via communication or only in the information via image, with high accuracy.

  Exemplary embodiments of a blind spot vehicle detection system, a roadside device, an in-vehicle device, and a blind spot vehicle detection method according to the present invention will be described below in detail with reference to the accompanying drawings. In the first embodiment, the case where the road situation is imaged with a camera connected to the roadside device, and in the second embodiment, the case where the road situation is imaged with a camera mounted on the host vehicle will be described. .

  FIG. 1 is a diagram illustrating an arrangement of a blind spot vehicle detection system according to the first embodiment. As shown in the figure, the blind spot vehicle detection system 1 includes a roadside device 10 installed on the ground and an in-vehicle device 20 mounted on the host vehicle. Here, the roadside device 10 includes a current camera such as a CCD (Charge Coupled Devices) camera that captures road conditions, and other vehicles (see “A” and “D” in the figure) equipped with a position notification terminal. A communication device that collects information such as position, current speed, vehicle size, and vehicle type is connected. In addition, "B" and "C" in the figure represent vehicles that are not equipped with location reporting terminals.

  The roadside device 10 generates information related to vehicle arrangement (hereinafter referred to as “communication via information”) from information such as a vehicle position and a vehicle size acquired via a communication device, and performs image analysis on an image from the camera. Information such as the vehicle position and vehicle size of the vehicle detected by the above (hereinafter referred to as “image via information”) is generated. Then, the vehicle placement information corresponding to each vehicle is generated by comparing the communication via information and the image via information. In addition, when the roadside apparatus 10 produces | generates vehicle arrangement | positioning information, although the reliability showing the level of information accuracy is added, this point is later mentioned using FIG.

  Subsequently, the roadside device 10 wirelessly transmits the vehicle arrangement information to which the reliability is added to the in-vehicle device 20, and the in-vehicle device 20 determines the blind spot vehicle from the vehicle arrangement information to which the reliability is added and the information related to the own vehicle position. Perform detection processing.

  As described above, the blind spot vehicle detection system 1 uses the communication route information generated by the roadside device 10 via the communication device. This communication route information is obtained by using another vehicle (“A” in FIG. And “D”). That is, it does not include information on other vehicles (see “B” and “C” in the figure) that are not equipped with a location reporting terminal. Therefore, it is necessary to supplement the information about the vehicle that is not included in the information via communication with the information via the image that is information derived from the camera image.

  However, the information via the image, which is information derived from the camera image, depends on the field of view that depends on the weather or time, the accuracy of the camera, and the accuracy of the recognition program. In general, the information is less accurate. In addition, the via-image information may be a case where the imaged vehicle is blocked by another vehicle (hidden behind the other vehicle), and the position information of the blocked vehicle becomes more accurate. For this reason, it is not preferable from the viewpoint of information accuracy to simply combine the information via communication and the information via image.

  Therefore, in the blind spot vehicle detection system 1, when the roadside device 10 synthesizes the communication via information and the image via information to generate the final vehicle arrangement information, a reliability indicating the level of information accuracy is added. It is said. And when the vehicle-mounted apparatus 20 detects a blind spot vehicle, it determines whether it is a blind spot vehicle using this reliability. Further, in the roadside device 10, the reliability is given in consideration of the shielding rate in the information via the image (the rate at which the image is shielded by other vehicles).

  Next, features of the blind spot vehicle detection system 1 according to the first embodiment will be described with reference to FIG. FIG. 2 is a diagram illustrating characteristics of the blind spot vehicle detection system 1 according to the first embodiment. In the figure, vehicle arrangement (vehicle position and vehicle size) when the road surface is viewed from above is shown. Further, “A” to “D” in the figure correspond to the vehicles “A” to “D” in FIG. 1.

  As shown in “(1) Vehicle arrangement by road-to-vehicle communication” in the figure, the vehicle arrangement based on the information via communication includes only vehicles (“A” and “D”) equipped with a position reporting terminal. (See 21 in the figure). On the other hand, as shown in “(2) Vehicle arrangement by image analysis”, the vehicle arrangement based on the via-image information includes vehicles (“B”, “C” and “C” which are present in the field of view of the camera shown in FIG. "D") is included (see 22a in the figure). Here, in FIG. 22B, an image captured by the camera is shown, but “C” and “D” are shielded by “B” existing between the camera and the camera. That is, the position information corresponding to “C” and “D” is less accurate than the position information corresponding to “B”.

  Then, as shown in “(3) Combined vehicle arrangement”, the position information on the four vehicles “A” to “D” can be obtained by combining (1) and (2). Each position information is given two types of reliability “normal” or “low” indicating that the reliability is lower than “normal”. Here, since “A” and “D” are included in the information via communication (see (1) in the figure), the reliability is “normal”. In addition, although “B” is not included in the information via communication, the reliability is “normal” because there is no shielding by other vehicles in the information via image. On the other hand, the reliability of “D” is “low” because it is not included in the information via communication and the information via image is blocked by another vehicle.

  As described above, by assigning reliability to the vehicle arrangement information and performing blind spot determination using the vehicle arrangement information to which the reliability is provided, it is possible to perform a blind spot determination process in consideration of information accuracy. Therefore, when the information is low accuracy, it is possible to perform a blind spot determination process (for example, it is not a blind spot vehicle based on the position information, but a blind spot vehicle is determined) based on the position information. The details of the blind spot determination processing will be described later with reference to FIGS. In the first embodiment, the reliability is set to two levels of “normal” and “low”, but may be set to three or more levels.

  Next, the configuration of each device included in the blind spot vehicle detection system 1 according to the first embodiment will be described with reference to FIG. FIG. 3 is a diagram illustrating the configuration of each device included in the blind spot vehicle detection system 1 according to the first embodiment. As shown in FIG. 1, the blind spot vehicle detection system 1 includes a roadside device 10 and an in-vehicle device 20.

  The roadside device 10 is connected to the camera and the communication device shown in FIG. 1 and generates final vehicle arrangement information by combining the information via the image and the information via the communication. 20 is a ground device that performs processing to be transmitted to 20. As shown in the figure, the roadside apparatus 10 includes a receiving unit 11, an input unit 12, a transmission unit 13, a control unit 14, and a storage unit 15. The control unit 14 further includes a communication via information generation unit 14a, an image via information generation unit 14b, a reliability addition processing unit 14c, and a vehicle arrangement information generation unit 14d. The storage unit 15 includes a communication unit. The route information 15a, the image route information 15b, and the vehicle arrangement information 15c are stored.

  The receiving unit 11 is connected to the communication device shown in FIG. 1 and passes information such as a vehicle position and a vehicle size sent from a vehicle equipped with a position reporting terminal to the via-communication information generation unit 14a of the control unit 14. I do. The input unit 12 is connected to the camera shown in FIG. 1, receives image data relating to traffic conditions captured by the camera, and performs a process of passing the image data to the via-image information generation unit 14 b of the control unit 14. The transmission unit 13 performs a process of wirelessly transmitting the vehicle arrangement information 15 c generated by the vehicle arrangement information generation unit 14 d of the control unit 14 and stored in the storage unit 15 to the in-vehicle device 20.

  The control unit 14 generates the communication via information 15a and the image via information 15b based on the data received by the receiving unit 11 and the data received by the input unit 12, and synthesizes the generated communication via information 15a and the image via information 15b. At this time, processing for generating the vehicle arrangement information 15c to which the reliability is given is performed. The via-communication information generation unit 14 a is a processing unit that receives information derived from the location reporting terminal from the receiving unit 11, generates the via-communication information 15 a based on the received information, and stores it in the storage unit 15.

  Here, an example of the communication via information 15a will be described with reference to FIG. FIG. 4 is a diagram illustrating an example of the communication via information 15a. As shown in the figure, the communication via information 15a includes, for example, a “serial number” indicating a serial number of each data, a “representative position” indicating a representative position such as a center point of each vehicle, and a “size” indicating the size of the vehicle. Is configured as a table including The “representative position” and “size” are two-dimensional (plan view of the road as viewed from above) corresponding to the X axis / Y axis of the vehicle arrangement shown in FIG. It is good also as a three-dimensional thing including the direction data. Further, the representative position after a predetermined time including the vehicle speed may be calculated.

  Returning to FIG. 3, the via-image information generation unit 14b will be described. The via-image information generating unit 14b receives the camera image from the input unit 12, analyzes the received image, detects the position and size of each vehicle included in the image, and generates the via-image information 15b. 15 is a processing unit that performs a process of storing the data in the storage unit 15.

  Here, an example of the via-image information 15b will be described with reference to FIG. FIG. 5 is a diagram illustrating an example of the via-image information 15b. As shown in the figure, the via-image information 15b includes, for example, a “serial number” indicating a serial number of each data, a “representative position” indicating a representative position such as a center point of each vehicle, and a “size” indicating the size of the vehicle. ”And a“ shielding rate ”that represents a ratio at which an image of the specific vehicle is shielded by another vehicle.

  The “shielding rate” is “0%” when there is no shielding by other vehicles as shown by serial number 1 in the figure, and the shielding area by other vehicles is 20% as shown by serial number 2 in the figure. “20%”. The “representative position” and “size” are two-dimensional (plan view of the road as viewed from above) corresponding to the X axis / Y axis of the vehicle arrangement shown in FIG. It is good also as a three-dimensional thing including the direction data. Moreover, it is good also as including the vehicle speed calculated | required by comparing the image of a time series.

  Returning to FIG. 3, the reliability addition processing unit 14c will be described. The reliability addition processing unit 14c receives the communication via information 15a and the image via information 15b as input data, adds a reliability indicating the level of information accuracy to each of these data, and then sends the data to the vehicle arrangement information generation unit 14d. It is a processing unit that performs a passing process.

  The reliability addition processing unit 14c sets the reliability to “normal” for each vehicle data included in the communication via information 15a. For each vehicle data included only in the via-image information 14b, when the “shielding rate” shown in FIG. 5 is lower than a predetermined threshold, the reliability is set to “normal” and is equal to or higher than the predetermined threshold. The reliability is “low”. Note that by setting the predetermined threshold value to 0, the reliability of each vehicle data included only in the via-image information 14b can be set to “low”.

  In the first embodiment, the reliability is set to two levels of “normal” and “low” which is lower than “normal”. However, the reliability may be three or more levels. For example, when the reliability of each vehicle data included in the communication via information 15a is “high” higher than “normal” and the shielding rate is lower than a predetermined threshold for each vehicle data included only in the image via information 14b “Normal”, or “low” lower than “normal” may be set when the shielding ratio is equal to or higher than a predetermined threshold.

  The vehicle arrangement information generation unit 14d generates vehicle arrangement information 15c, which is information related to the final vehicle arrangement, based on the communication via information 15a and the image via information 15b added by the reliability addition processing unit 14c. 15 is a processing unit that performs a process of storing the data in the storage unit 15.

  Here, an example of the vehicle arrangement information 15c will be described with reference to FIG. FIG. 6 is a diagram illustrating an example of the vehicle arrangement information 15c. As shown in the figure, the vehicle arrangement information 15c represents, for example, a “vehicle identifier” that uniquely identifies each vehicle, a “representative position” that represents a representative position such as a center point of each vehicle, and a size of the vehicle. It is configured as a table including “size” and “reliability” added by the reliability addition processing unit 14c. The “representative position” and “size” are two-dimensional (plan view of the road as viewed from above) corresponding to the X axis / Y axis of the vehicle arrangement shown in FIG. It is good also as a three-dimensional thing including the direction data. Moreover, it is good also as including vehicle speed.

  The storage unit 15 is a storage unit configured by a storage device such as a RAM (Random Access Memory) or an HDD (Hard Disk Drive). The storage unit 15 stores the communication via information 15a (see FIG. 4), the image via information 15b (see FIG. 5), and the vehicle arrangement information 15c (see FIG. 6) already described. Note that the communication via information 15a, the image via information 15b, and the vehicle arrangement information 15c are updated as needed by the communication via information generation unit 14a, the image via information generation unit 14b, and the vehicle arrangement information generation unit 14d, respectively.

  The in-vehicle device 20 is a device mounted on the host vehicle, receives the vehicle placement information 15c to which the reliability is given from the roadside device 10, and based on the received vehicle placement information 15c and the own vehicle position, the blind spot as seen from the own vehicle. It is an apparatus that performs processing for detecting a vehicle. In the figure, only the components necessary for explaining the feature points of the in-vehicle device 20 are described, but a GPS (Global Positioning System) function and a map display function provided in a general car navigation device. It shall have.

  As shown in the figure, the in-vehicle device 20 includes a receiving unit 21, a display 22, a control unit 23, and a storage unit 24. Moreover, the control part 23 is further provided with the blind spot vehicle detection part 23a, and the memory | storage part 24 memorize | stores the vehicle arrangement | positioning information 24a and the own vehicle position information 24b. In addition, although the figure shows about the case where the detection result of the blind spot vehicle detection unit 23a is displayed on the display 22, it may be notified by voice through a voice output device (not shown).

  The receiving unit 21 performs a process of storing the vehicle arrangement information 15c received from the roadside device via radio in the storage unit 24 as the vehicle arrangement information 24a. The display 22 is a display such as a touch panel display, and is a display device used in a general car navigation system.

  The control unit 23 includes a blind spot vehicle detection unit 23a. The blind spot vehicle detection unit 23a performs a process of detecting a blind spot vehicle existing at a blind spot position viewed from the own vehicle based on the vehicle arrangement information 24a and the own vehicle position information 24b in the storage unit 24, and outputs the detection result to the display 22. Perform the process. The blind spot vehicle detection unit 23a may combine the detection result with map information regarding a road or a building and output the combined result to the display 22.

  Here, an outline of the blind spot vehicle determination process performed by the blind spot vehicle detection unit 23a will be described with reference to FIGS. FIG. 7 is a diagram illustrating an example 1 of the blind spot vehicle determination process, and FIG. 8 is a diagram illustrating an example 2 of the blind spot vehicle determination process. 7 and 8 illustrate a part of the vehicle arrangement information 15c (same as the vehicle arrangement information 24a) shown in FIG.

  As shown in FIG. 7, the blind spot vehicle detection unit 23 a arbitrarily selects one set from each vehicle data included in the vehicle arrangement information 24 a, and sets one of the “obstacle vehicles” and the other “ The vehicle is a blind spot determination target vehicle. In the case shown in the figure, “α” is an “obstacle vehicle” and “β” is a “dead zone determination target vehicle”.

  Here, in the case shown in FIG. 7, the “reliability” of the vehicle arrangement information 24a corresponding to “α” and “β” is both “low”. For this reason, the “size” of the vehicle arrangement information 24a corresponding to the obstacle vehicle (α) is enlarged at a predetermined enlargement rate. That is, by estimating the “size” of the obstacle vehicle based on data with low reliability, the chance that the blind spot determination target vehicle is determined to be a blind spot vehicle is increased.

  On the other hand, the “size” of the vehicle arrangement information 24a corresponding to the blind spot determination target vehicle (β) is reduced at a predetermined reduction rate. That is, by estimating the “size” of the blind spot determination target vehicle based on data with low reliability, the chance that the blind spot determination target vehicle is determined to be a blind spot vehicle is increased.

  In other words, if the reliability is below a predetermined threshold, whether the blind vehicle can be seen from the host vehicle after changing the size of the obstacle vehicle to a larger size and the size of the vehicle for the blind spot determination to a smaller size. To determine whether the vehicle is a blind spot vehicle. Then, the blind spot vehicle is detected by repeating these processes for all the vehicle data included in the vehicle arrangement information 24a.

  FIG. 8 shows a blind spot determination process when “β” shown in FIG. 7 is selected as an obstacle vehicle. Here, the “reliability” of the vehicle arrangement information 24a corresponding to “γ” selected as the blind spot determination target vehicle is “normal”. For this reason, the “size” of the vehicle arrangement information 24a corresponding to “γ” is not changed. On the other hand, since “β” is selected as an obstacle vehicle in FIG. 8, “size” is expanded at a predetermined expansion rate.

  Returning to the description of FIG. 3, the storage unit 24 will be described. The storage unit 24 is a storage unit configured by a storage device such as a RAM (Random Access Memory) or an HDD (Hard Disk Drive). The vehicle arrangement information 24 a is vehicle arrangement information 15 c received from the roadside device 10 via the receiving unit 21. The own vehicle position information 24b is information (for example, latitude and longitude) indicating the current position of the in-vehicle device 20 acquired by a GPS (Global Positioning System) function (not shown) or a sensor such as a vehicle speed sensor. Updated as needed with movement.

  Next, a processing procedure performed in the roadside apparatus 10 will be described with reference to FIG. FIG. 9 is a flowchart showing a processing procedure in the roadside apparatus 10. As shown in the figure, when the receiving unit 11 receives vehicle information by road-to-vehicle communication (step S101), the communication via information generation unit 14a generates the communication via information 15a (step S102). Further, when the input unit 12 receives the camera image (step S103), the image via information generation unit 14b generates the image via information 15b in consideration of the shielding rate (step S104). Note that step S101 to step S102 and step S103 to step S104 may be processed in parallel.

  Subsequently, the reliability addition processing unit 14c adds reliability to the communication via information 15a and the image via information 15b (step S105), and the vehicle arrangement information generating unit 14d generates the vehicle arrangement information 15c (step S106). . And the transmission part 13 transmits the vehicle arrangement | positioning information 15c toward the vehicle-mounted apparatus 20 (step S107), and complete | finishes a process.

  Next, a processing procedure performed in the in-vehicle device 20 will be described with reference to FIG. FIG. 10 is a flowchart showing a processing procedure in the in-vehicle device 20. As shown in the figure, when the receiving unit 21 receives the vehicle arrangement information 15c (step S201), it is stored in the storage unit 24 as the vehicle arrangement information 24a. And the blind spot vehicle detection part 23a starts the blind spot vehicle detection process based on the vehicle arrangement | positioning information 24a and the own vehicle position information 24b.

  The blind spot vehicle detection unit 23a selects the obstacle vehicle and the blind spot candidate vehicle (step S202), and determines whether or not the reliability of the vehicle arrangement information 24a corresponding to the obstacle vehicle is “low” (step S203). When the reliability is “low” (step S203, Yes), the vehicle size of the obstacle vehicle is enlarged based on a predetermined enlargement ratio (step S204). If the reliability is not “low” (No in step S203), the process proceeds to step S205 without performing the process in step S204.

  Subsequently, it is determined whether or not the reliability of the vehicle arrangement information 24a corresponding to the blind spot candidate vehicle is “low” (step S205). If the reliability is “low” (step S205, Yes). Then, the vehicle size of the blind spot candidate vehicle is reduced based on a predetermined reduction rate (step S206). On the other hand, if the reliability is not “low” (No in step S205), the process proceeds to step S207 without performing the process in step S206.

  Then, it is determined whether or not the shielding rate of the blind spot candidate vehicle by the obstacle vehicle is equal to or greater than a predetermined threshold (step S207), and if it is equal to or greater than the predetermined threshold (step S207, Yes), the blind spot candidate vehicle is determined to be a blind spot. It determines with a vehicle (step S208). If it is less than the predetermined threshold (No at Step S207), the process proceeds to Step S209 without determining that the blind spot candidate vehicle is a blind spot vehicle.

  Then, it is determined whether or not the determination has been completed for all the vehicles (step S209). When the determination has been completed for all the vehicles (step S209, Yes), the process is terminated. On the other hand, when the determination for all the vehicles has not been completed yet (No at Step S209), the processes after Step S202 are repeated.

  As described above, in the first embodiment, the communication via information generation unit of the roadside device generates communication via information that is information including the position and size of the vehicle on which the predetermined terminal device is mounted. A route information generation unit generates image route information, which is information including the position and size of a vehicle detected by image analysis of an image captured by a camera that captures road traffic conditions, and generates vehicle placement information of the roadside device. When the unit collates the information via communication and the information via image to generate vehicle arrangement information corresponding to each vehicle, the reliability addition processing unit of the roadside device has high or low information accuracy with respect to each vehicle arrangement information. The blind spot vehicle detection unit of the in-vehicle device detects the blind spot vehicle by changing the size included in the vehicle arrangement information when the reliability is equal to or less than a predetermined value.

  Therefore, by efficiently combining information with different accuracy, it is possible to detect a blind spot vehicle included only in the information via communication or only in the information via image, with high accuracy. Further, the in-vehicle device only needs to perform the blind spot vehicle detection process by performing the imaging of road conditions by the camera, image processing, matching processing between communication via information and image via information, and reliability addition processing by the roadside device. For this reason, an in-vehicle device can be realized at low cost.

  Incidentally, in the above-described first embodiment, the case where the road situation is imaged with the camera connected to the roadside device has been described, but the road situation may be imaged with a camera mounted on the host vehicle. Therefore, in a second embodiment described below, a case where a road situation is imaged with a camera mounted on the host vehicle will be described. Note that, in the following description, the description of the items already described in the first embodiment will be omitted or only a brief description.

  FIG. 11 is a diagram illustrating a configuration of each device included in the blind spot vehicle detection system according to the second embodiment. Note that the roadside device 110 shown in FIG. 11 is not connected to a camera that captures road conditions, and instead, the in-vehicle device 20 is connected to a camera that captures road conditions. Then, in the roadside device 110, only the generation processing of the communication via information 115a corresponding to the communication via information 15a in FIG. 3 and the transmission processing for transmitting the communication via information 155a to the in-vehicle device 120 are performed.

  On the other hand, in the vehicle-mounted device 120, vehicle arrangement information based on the communication via information 115a received by wireless communication from the roadside device 110 and the image via information 125b generated based on the image data from the input unit 123 connected to the vehicle-mounted camera. 125c is generated. That is, in the second embodiment, a processing unit other than the reception unit 11, the communication via information generation unit 14 a and the transmission unit 13 included in the roadside device 10 in the first embodiment, that is, the input unit 12, the image via information generation unit 14 b, The case where the reliability addition process part 14c and the vehicle arrangement | positioning information generation part 14d are moved to the vehicle-mounted apparatus 20 is shown.

  As illustrated in FIG. 11, the roadside apparatus 110 includes a reception unit 111, a transmission unit 113, a control unit 114, and a storage unit 115. The control unit 114 further includes a communication via information generation unit 114a, and the storage unit 115 stores the communication via information 115a. Here, the reception unit 111 is the same as the reception unit 11 in FIG. 3, the transmission unit 113 is the same as the transmission unit 13, the communication via information generation unit 114a is the same as the communication via information generation unit 14a, and the communication via information 115a is the same as the communication via information. Each corresponds to 15a.

  The in-vehicle device 120 includes a receiving unit 121, a display 122, an input unit 123 that receives image data from the in-vehicle camera, a control unit 124, and a storage unit 125. The control unit 124 further includes an image via information generation unit 124a, a reliability addition processing unit 124b, a vehicle arrangement information generation unit 124c, and a blind spot vehicle detection unit 124d. Information 125a, via-image information 125b, vehicle arrangement information 125c, and host vehicle position information 125d are stored.

  Here, the receiving unit 121 is the same as the receiving unit 21 in FIG. 3, the display 122 is the same as the display 22, the image passing information generation unit 124a is the same as the image passing information generation unit 14b, and the reliability addition processing unit 124b is also the reliability addition. Similarly to the processing unit 14c, the vehicle arrangement information generation unit 124c corresponds to the vehicle arrangement information generation unit 14d, and the blind spot vehicle detection unit 124d corresponds to the blind spot vehicle detection unit 23a. Further, the image via information 125b corresponds to the image via information 15b in FIG. 3, the vehicle arrangement information 125c corresponds to the vehicle arrangement information 15c or the vehicle arrangement information 24a, and the own vehicle position information 125d corresponds to the own vehicle position information 24b. ing.

  Thus, even when the road situation is imaged from the in-vehicle camera on the vehicle on which the in-vehicle device 120 is mounted, the blind spot vehicle can be detected efficiently as in the case of the first embodiment. . Moreover, the information regarding the arrangement | positioning of the vehicle based on a vehicle-mounted camera image can be supplemented with the information regarding the arrangement | positioning of the vehicle originating in a position report terminal.

  As described above, the blind spot vehicle detection system, roadside device, in-vehicle device, and blind spot vehicle detection method according to the present invention are useful for detection of blind spot vehicles and blind spot pedestrians.

It is a figure which shows arrangement | positioning of the blind spot vehicle detection system which concerns on Example 1. FIG. It is a figure which shows the characteristic of the blind spot vehicle detection system which concerns on Example 1. FIG. It is a figure which shows the structure of each apparatus contained in the blind spot vehicle detection system which concerns on Example 1. FIG. It is a figure which shows an example of information via communication. It is a figure which shows an example of information via an image. It is a figure which shows an example of vehicle arrangement | positioning information. It is a figure which shows the example 1 of a blind spot vehicle determination process. It is a figure which shows the example 2 of a blind spot vehicle determination process. It is a flowchart which shows the process sequence in a roadside apparatus. It is a flowchart which shows the process sequence in a vehicle-mounted apparatus. It is a figure which shows the structure of each apparatus contained in the blind spot vehicle detection system which concerns on Example 2. FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Blind spot vehicle detection system 10 Roadside device 11 Receiving part 12 Input part 13 Transmitting part 14 Control part 14a Communication via information generation part 14b Image via information generation part 14c Reliability addition processing part 14d Vehicle arrangement information generation part 15 Storage part 15a Via communication Information 15b Information via image 15c Vehicle arrangement information 20 On-vehicle device 21 Reception unit 22 Display 23 Control unit 23a Blind spot vehicle detection unit 24 Storage unit 24a Vehicle arrangement information 24b Own vehicle position information 110 Roadside device 111 Reception unit 113 Transmission unit 114 Control unit 114a Communication via information generation unit 115 Storage unit 115a Communication via information 120 In-vehicle device 121 Reception unit 122 Display 123 Input unit 124 Control unit 124a Image via information generation unit 124b Reliability addition processing unit 124c Vehicle arrangement information generation unit 124d Blind spot vehicle detection Through parts 125 storage unit 125a via communication information 125b image information 125c vehicle location information 125d vehicle position information

Claims (7)

A blind spot vehicle detection system for detecting a blind spot vehicle existing at a position that becomes a blind spot from the own vehicle by an obstacle vehicle existing between the host vehicle,
Via-communication information generation means for generating via-communication information that is information including the position and size of a vehicle equipped with a predetermined terminal device;
Via-image information generating means for generating via-image information that is information including the position and size of the vehicle detected by image analysis of an image captured by a camera that captures the traffic situation of the road;
A reliability adding means for adding a reliability indicating the level of information accuracy to each vehicle arrangement information when generating the vehicle arrangement information corresponding to each vehicle by collating the communication via information and the image via information. When,
A blind spot vehicle detection system comprising: blind spot vehicle detection means for detecting the blind spot vehicle by changing the size included in the vehicle arrangement information when the reliability is equal to or less than a predetermined value. The image via information generation means includes
Including the shielding rate by other vehicles for the information via the image,
The reliability adding means includes
When the vehicle arrangement information corresponding to the vehicle included only in the information via the image and the shielding rate is equal to or higher than a predetermined threshold, a low reliability lower than the normal reliability, which is a normal reliability, The blind spot vehicle detection system according to claim 1, wherein the normal reliability is added to the vehicle arrangement information. The reliability adding means includes
The vehicle corresponding to the vehicle included only in the information via the image, the normal reliability that is the normal reliability for the vehicle arrangement information corresponding to the vehicle included in both the information via the communication and the information via the image. The blind spot vehicle detection system according to claim 1, wherein low reliability, which is lower than the normal reliability, is added to the arrangement information. The blind spot vehicle detection means includes:
The vehicle size included in the vehicle arrangement information to which the low reliability is added by the reliability adding unit is enlarged at a predetermined enlargement ratio when the vehicle arrangement information corresponds to the obstacle vehicle, and the vehicle 4. The blind spot vehicle detection system according to claim 1, wherein when the arrangement information corresponds to a blind spot candidate vehicle that is a candidate for the blind spot vehicle, the vehicle is reduced at a predetermined reduction rate. A roadside device used in a blind spot vehicle detection system for detecting a blind spot vehicle existing at a position that becomes a blind spot from the own vehicle by an obstacle vehicle existing between the host vehicle,
Via-communication information generation means for generating via-communication information that is information including the position and size of a vehicle equipped with a predetermined terminal device;
Via-image information generating means for generating via-image information that is information including the position and size of the vehicle detected by image analysis of an image captured by a camera that captures the traffic situation of the road;
A reliability adding means for adding a reliability indicating the level of information accuracy to each vehicle arrangement information when generating the vehicle arrangement information corresponding to each vehicle by collating the communication via information and the image via information. A roadside device comprising: An in-vehicle device used in a blind spot vehicle detection system for detecting a blind spot vehicle existing at a position that becomes a blind spot from the own vehicle by an obstacle vehicle existing between the own vehicle,
Information via communication, which is information including the position and size of the vehicle on which the predetermined terminal device is mounted, and the position and size of the vehicle detected by image analysis of the image captured by the camera that captures the traffic situation on the road When the vehicle placement information corresponding to each vehicle is generated by collating the information via the image, which is information, the vehicle placement information to which the reliability indicating the level of information accuracy is added to each vehicle placement information is received. Vehicle arrangement information receiving means,
An in-vehicle apparatus comprising: a blind spot vehicle detecting unit that detects the blind spot vehicle by changing the size included in the vehicle arrangement information when the reliability is equal to or less than a predetermined value. A blind spot vehicle detection method for detecting a blind spot vehicle existing at a position that becomes a blind spot from the own vehicle by an obstacle vehicle existing between the host vehicle,
A via-communication information generation step for generating via-communication information, which is information including the position and size of a vehicle equipped with a predetermined terminal device;
A via-image information generating step for generating via-image information that is information including the position and size of the vehicle detected by image analysis of an image taken by a camera that captures the traffic situation of the road;
A reliability addition step of adding a reliability indicating the level of information accuracy to each vehicle arrangement information when generating the vehicle arrangement information corresponding to each vehicle by collating the information via communication and the information via the image When,
A blind spot vehicle detection method comprising: a blind spot vehicle detecting step of detecting the blind spot vehicle by changing the size included in the vehicle arrangement information when the reliability is equal to or less than a predetermined value.

Images