Object detection device, roadside device, and data processing device

Abstract

To provide an object detection device, a road side device equipped with the object detection device, and a data processor capable of easily and surely discriminating a two wheeler from a bicycle.SOLUTION: The object detection apparatus 10 includes the data acquisition unit DA that acquires data of a moving body MB in the target area DD, the velocity calculation unit VC that calculates the moving velocity on the basis of the data of the moving body MB, and the identification unit TI that identifies whether the moving body MB is a two wheeler or a bicycle according to the moving velocity of the moving body MB, whereby it is possible to easily and reliably identify a two wheeler and a bicycle.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to an object detection device that is installed, for example, near a road to perform object detection to obtain information about objects present on the road and its surroundings, a roadside device equipped with an object detection device, and a data processing device that processes information about objects. [Background technology]

[0002] As an example of an object detection device, a mobile object information notification device (see Patent Document 1) is known that detects a mobile object by irradiating it with laser light and notifies the detected information. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2020-35397 Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the above Patent Document 1, among the moving objects to be detected, two-wheeled vehicles (motorcycles) and bicycles, which are difficult to distinguish, are handled by combining RFID information with irradiation of laser light.

[0005] The present invention has been made in consideration of the above points, and aims to provide an object detection device that enables simple and reliable discrimination between two-wheeled vehicles and bicycles, a roadside device equipped with an object detection device, and a data processing device. [Means for solving the problem]

[0006] To achieve the above objective, the object detection device includes a data acquisition unit that acquires data on moving objects in a target area, a speed calculation unit that calculates the moving speed of the moving object based on the data on the moving object, and an identification unit that identifies whether the moving object is a two-wheeled vehicle or a bicycle based on the moving speed of the moving object.

[0007] The object detection device can easily and reliably distinguish between two-wheeled vehicles and bicycles by identifying whether the vehicle is a two-wheeled vehicle or a bicycle based on the moving speed of the moving object calculated from data on the moving object in the target area.

[0008] In a specific aspect of the present invention, the identification unit identifies whether the moving object is a motorcycle or a bicycle depending on whether the moving object's speed is equal to or greater than a set speed threshold. In this case, it is possible to quickly and accurately identify whether the moving object is a motorcycle or a bicycle depending on the measurement result of the moving speed.

[0009] In another aspect of the present invention, when the identification unit determines that the moving object is a two-wheeled vehicle or a bicycle based on the moving speed of the moving object, the identification unit maintains the determination in a held state, thereby enabling early determination of whether the moving object is a two-wheeled vehicle or a bicycle.

[0010] In yet another aspect of the present invention, the data acquisition unit is a distance measurement unit that acquires point cloud data by distance measurement. In this case, identification can be performed based on the point cloud data acquired by distance measurement.

[0011] In yet another aspect of the present invention, an ID assigning unit assigns an ID based on shape characteristics of a moving object, a data acquiring unit continuously acquires data, and a speed calculating unit determines the identity of the moving object to which the ID has been assigned based on the shape characteristics and calculates the moving speed of the moving object based on the displacement between successively acquired data for the same moving object. In this case, a single moving object to be detected can be accurately tracked.

[0012] To achieve the above object, a roadside device includes any one of the object detection devices described above.

[0013] By being equipped with any of the above object detection devices, the roadside device can easily and reliably distinguish between two-wheeled vehicles and bicycles by identifying whether the vehicle is a two-wheeled vehicle or a bicycle based on the moving speed of the moving object calculated from data on the moving object in the target area.

[0014] In a specific aspect of the present invention, the object detection result for the moving body, including the identification result of whether the moving body is a motorcycle or a bicycle, performed by the identification unit, is notified to another moving body. In this case, the identification result of whether the moving body is a motorcycle or a bicycle can be notified to the other moving body.

[0015] To achieve the above object, a data processing device identifies whether a moving object is a two-wheeled vehicle or a bicycle according to a moving speed calculated based on data of the moving object in a target area.

[0016] The data processing device can easily and reliably distinguish between two-wheeled vehicles and bicycles by identifying whether the vehicle is a two-wheeled vehicle or a bicycle based on the moving speed of the moving body calculated from data on the moving body in the target area. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram for explaining an outline of an object detection device according to an embodiment and a roadside device including the object detection device; [Figure 2] FIG. 2 is a block diagram showing an example of the configuration of a roadside device. [Figure 3] FIG. 2 is a block diagram for explaining the object detection device and a data processing device that constitutes the object detection device from a functional aspect. [Figure 4] 10 is a data table for explaining an example of various types of data to be handled. [Figure 5] 4 is a flowchart for explaining a series of processes in a roadside device including an object detection device. [Figure 6] FIG. 1 is a conceptual diagram for explaining an overview of an object detection device. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an object detection device according to an embodiment and a roadside device equipped with the object detection device will be described below with reference to Fig. 1 etc. Fig. 1 is a conceptual diagram for providing an overview of an object detection device 10 according to this embodiment and a roadside device 100 equipped with the object detection device 10, and shows an intersection CS on a road where the roadside device 100 is installed.

[0019] The roadside device 100 sets a predetermined range including the intersection CS and its surrounding area as a monitoring range where it should detect traffic conditions. An example configuration will be described in detail later, but the roadside device 100 is an information providing device that uses a LiDAR (Light Detection And Ranging) or the like, which is a ranging unit (ranging device), to capture traffic conditions in the monitoring range and provide the acquired various information to the outside, and is equipped with an object detection device 10 and a control device 50.

[0020] The object detection device 10 performs object detection to obtain information about objects present on roads and their surroundings, and includes a LiDAR 11 and other components constituting a ranging unit. The ranging unit (LiDAR 11) acquires point cloud data by continuously measuring distances. A single or multiple ranging units may be provided depending on the monitoring range, etc. In the illustrated example, one LiDAR 11 is configured to monitor a specific area of ​​a road upstream of an intersection CS, i.e., a target area DD to be monitored. The LiDAR 11 functions as a data acquisition unit DA that acquires data about moving objects MB present in the target area DD. As will be described in detail with reference to FIGS. 2 and 3, the object detection device 10 performs various processes, such as calculations to determine the shape and moving speed of the moving object MB, based on the point cloud data about the moving object MB acquired by continuous ranging by the LiDAR 11, and is able to determine the type of the moving object MB based on the processing results. The object detection device 10 outputs to the control device 50 various pieces of information (target information) extracted from the point cloud data as described above.

[0021] The control device 50 is composed of, for example, various circuit boards, and aggregates information about the moving object MB from the object detection device 10 described above and information from the signal controller SC that controls the four traffic lights TL installed at the intersection CS. Based on this information, the control device 50 integrates and organizes information about traffic conditions at the intersection CS and its surroundings and transmits it to various locations. In the illustrated example, the roadside device 100 functions as an information providing device (alarm device) by wirelessly outputting and transmitting various necessary data to external devices such as a vehicle VE (autonomous vehicle AD) that is passing through the intersection CS, a display device DS such as a signage SN installed at the intersection CS, or a traffic volume recording device TR. In other words, the control device 50 functions as an I2V (Infrastructure to Vehicle) control device that provides information from the infrastructure side to the vehicle side.

[0022] Among the information to be provided by the roadside device 100 as described above, the target information handled by the object detection device 10 is assumed to be information regarding the type of moving object MB, as described above. The type of moving object MB can be identified, for example, as a large four-wheeled vehicle, a standard-sized vehicle, a two-wheeled vehicle (motorcycle), a bicycle, or a pedestrian. However, when identifying the type of moving object MB, it is difficult to distinguish between two-wheeled objects and bicycles, particularly when they are located far from the installation position of the object detection device 10, because the difference in shape is small. In contrast, the object detection device 10 of this embodiment identifies whether the moving object MB is a two-wheeled object or a bicycle according to the moving speed of the moving object MB (due to the difference in moving speed), thereby enabling simple and reliable identification of two-wheeled objects and bicycles. In other words, two-wheeled objects and bicycles are identified by focusing on the difference in speed while traveling. Furthermore, the roadside device 100 of this embodiment is equipped with the object detection device 10, and is therefore able to provide information on the moving object MB that has been classified as to whether it is a two-wheeled vehicle or a bicycle.

[0023] In the example shown in the figure, the roadside device 100 communicates with a vehicle VE (automated vehicle AD) as it passes through an intersection CS, and provides it with various acquired information that serves as driving assistance (such as information about blind spots that cannot be seen from the automatically driven vehicle AD). More specifically, in the figure, the vehicle VE (automated vehicle AD) is attempting to enter the intersection CS from a direction opposite to the target area DD and make a right turn, as indicated by the dashed arrow AR1. This direction intersects with the direction of travel of a moving object MB traveling straight from the target area DD. The roadside device 100 is aware of this situation through two-way wireless communication. In this case, it is extremely important for traffic safety that the roadside device 100 accurately communicate the traffic conditions in the target area DD to the vehicle VE (automated vehicle AD) as early as possible. In this embodiment, as described above, the object detection device 10 accurately identifies whether the moving body MB traveling in the target area DD is a two-wheeled vehicle or a bicycle as vehicle type information included in the target information, thereby providing information that enables the traffic situation to be grasped more accurately.

[0024] Hereinafter, the functional aspects of each component of the roadside device 100 will be described with reference to the block diagram shown in Fig. 2. As shown in the figure, in this example, the object detection device 10 of the roadside device 100 includes a LiDAR processing unit 12 that performs various data processing in addition to a LiDAR 11. The control device 50 also includes a main control unit 51 that performs various data processing, and a communication unit 52 that functions as a notification unit AN by communicating with the outside.

[0025] First, of the object detection device 10, the LiDAR 11, which is the ranging unit, is composed of, for example, a laser, a galvanometer mirror, a polygon mirror, etc., and as mentioned above, performs sensing on the target area DD, and as a data acquisition unit DA, enables the acquisition of point cloud data as ranging data necessary to understand the situation of a moving object MB present in the target area DD.

[0026] Furthermore, the LiDAR processing unit 12 of the object detection device 10 is composed of various electronic circuit elements and functions as a data processing unit DU that analyzes the point cloud data acquired by the LiDAR 11 and extracts target information about the moving object MB, such as the position, traveling direction, and moving speed of the moving object MB. Note that an example of the configuration for data analysis and extraction in the LiDAR processing unit 12 will be described later with reference to FIG. 3.

[0027] Next, the main control unit 51 of the control device 50 is composed of various electronic circuit elements, etc., and performs processes such as collecting and analyzing information to be reported to the outside as the roadside device 100. In the example shown in the figure, the main control unit 51 is connected to the object detection device 10, which is an internal device, and the signal controller SC, which is an external device, and acquires target information, which is information about a moving object MB, from the object detection device 10, and acquires light color information, which is information about a signal lamp TL (see FIG. 1), from the signal controller SC. The main control unit 51 processes and edits the collected information to content appropriate for the destination (the autonomous vehicle AD, signage SN, or traffic volume recording device TR in FIG. 1), and outputs the information via the communication unit 52 (notification unit AN), which is an interface with the external device. For example, it is conceivable that information such as the entry status of moving bodies MB into an intersection CS from the oncoming lane, etc. and the timing of switching on and off of traffic lights TL could be provided to an autonomous vehicle AD, information on the possibility of danger occurring due to traffic conditions could be provided to a signage SN, and information such as the number (number of people) of each type of moving body MB could be provided to a traffic volume recording device TR.

[0028] Hereinafter, with reference to the block diagram shown in FIG. 3, data processing in the roadside device 100, particularly in the object detection device 10, will be described in detail.

[0029] As shown in Figure 3, in the object detection device 10, the LiDAR processing unit 12 is a data processing device DU that includes or functions as a data reception unit DR, a distance image processing unit DP (object detection processing unit OD), a speed calculation unit VC, an ID assignment unit IG, a vehicle type identification unit VI (identification unit TI), and a mobile data management unit MM (data management unit DM), and performs a series of processes on the point cloud data acquired by the LiDAR 11 as the data acquisition unit DA to generate various data and manage the various data.

[0030] First, when the data receiving unit DR of the LiDAR processing unit 12 receives point cloud data of the moving object MB from the LiDAR 11, the distance image processing unit DP performs processing to generate a distance image. That is, imaging (distance imaging) processing is performed to enable identification (individuation) of each moving object MB from the point cloud data. Here, the distance image processing unit DP also functions as an object detection processing unit OD that performs object detection to identify each moving object MB in the distance image, i.e., processing to extract a portion corresponding to each moving object MB from the generated distance image. Note that, as described above, ranging in the LiDAR 11 is performed continuously. Accordingly, the distance image generation and object detection processing in the distance image processing unit DP are performed sequentially on a frame-by-frame basis for the ranging data (point cloud data) acquired through continuous operation.

[0031] Furthermore, an ID is assigned to each moving object MB in the above-mentioned distance image by the ID assigning unit IG. The identity of each moving object MB is determined based on its shape characteristics and the relative positional relationship between the distance images generated successively. That is, once the ID assigning unit IG assigns an ID corresponding to the shape characteristics to an object corresponding to a moving object MB extracted by processing in the distance image processing unit DP (object detection processing unit OD), the identity can be confirmed by checking for the presence of an object in subsequent distance images that has the shape characteristics corresponding to the assigned ID, making it possible to track the moving object MB between successive images.

[0032] For each object detected in the range image generated by the range image processor DP, i.e., for each identified moving object MB, the speed calculation unit VC calculates the moving speed of each moving object MB. That is, for moving object MBs that have been assigned IDs based on shape features and whose identity has been confirmed between successive images, the speed calculation unit VC calculates vector values ​​such as the moving speed and direction of travel of the moving object MB by comparing the displacement (difference in position change) between the images. In other words, the speed calculation unit VC determines the identity of the moving object MB to which an ID has been assigned based on shape features, and calculates the moving speed of the moving object MB based on the displacement between successively acquired data for the same moving object MB.

[0033] The vehicle type identification unit VI identifies the vehicle type of the moving object MB based on its shape characteristics and moving speed. In this example, the vehicle type identification unit VI (identification unit TI) classifies the moving object MB into a large vehicle, a standard vehicle, a bicycle, or a pedestrian based on the individual shape characteristics of the individual moving object MB. That is, various parameters related to shape and dimensions, as well as thresholds and other criteria for determining the vehicle type, are predefined, and vehicle type identification is performed according to the predefined criteria. However, among the vehicle type identification based on the shape characteristics described above, the term "bicycle" may include both two-wheeled vehicles (motorcycles) and bicycles, which are identified based on different criteria. Here, the vehicle type identification unit VI, which serves as an identification unit TI for identifying two-wheeled vehicles and bicycles, does not distinguish between them based on shape characteristics alone, but identifies whether the moving object MB is a two-wheeled vehicle or a bicycle based on the moving speed of the moving object MB. More specifically, the vehicle type identification unit VI (identification unit TI) identifies whether the moving object MB is a two-wheeled vehicle or a bicycle based on whether the moving speed of the moving object MB is equal to or greater than a predefined speed threshold (e.g., 30 km / h or greater). Here, in terms of classification based on shape characteristics, both two-wheeled vehicles and regular bicycles are classified as "bicycles," and then a separate flag (two-wheeled vehicle flag) is set based on the calculated travel speed to distinguish whether the "bicycle" is a regular bicycle or a two-wheeled vehicle.

[0034] In this embodiment, the vehicle type identification unit VI determines whether a moving object MB is identical in continuously acquired distance images. For example, as shown in FIG. 1, when a moving object MB moving from upstream to downstream is captured by a fixed-point LiDAR 11, the distance measurement data becomes coarser the further upstream (farther away), and the closer to the downstream side, i.e., the closer to the distance measurement side, the higher the distance measurement accuracy. Therefore, by performing vehicle type identification for each frame, more accurate vehicle type identification is possible.

[0035] Furthermore, in this embodiment, when it is determined that the moving body MB is a two-wheeled vehicle or a bicycle based on the moving speed of the moving body MB, the determination is maintained in a held state. That is, as in the example described above, once it is detected that the moving speed of the moving body MB is equal to or greater than a preset speed threshold (30 km / h or greater) (the two-wheeled vehicle flag is set), the moving body MB will maintain the two-wheeled vehicle flag set even if it subsequently travels at a speed below the speed threshold, and will not be treated as a normal bicycle, but will maintain the state of being identified as a two-wheeled vehicle as long as it is classified as a "bicycle" in the vehicle type identification based on shape characteristics.

[0036] The various data acquired and generated as described above are sequentially stored in a mobile data management unit MM (data management unit DM) that is configured with various storage devices, for example.

[0037] Here, as described above and as shown in the figure, the roadside device 100 can use the target information processed by the object detection device 10 to provide information to various external devices such as a vehicle VE (autonomous driving vehicle AD), signage SN, or traffic volume recording device TR.

[0038] From another perspective, the LiDAR processing unit 12 can also be said to function as a data processing device that performs data processing to acquire target information from the point cloud data acquired by the LiDAR 11.

[0039] Hereinafter, various data handled in the above embodiment will be described with reference to a data table, an example of which is shown in Fig. 4. Each of the data tables α and β shows target information related to moving objects MB present in the target area DD, extracted by analyzing point cloud data for the target area DD at a certain point in time, and is stored in the moving object data management unit MM (data management unit DM) shown in Fig. 3.

[0040] Here, the data handled includes, for example, a "target ID," a "type," and a "motorcycle flag." The "target ID" is assigned by the ID assignment unit IG, and the "type" is selected by the vehicle type identification unit VI (identification unit TI) based on shape characteristics. The "motorcycle flag," like the "type," is selected by the vehicle type identification unit VI (identification unit TI), but indicates whether the vehicle is a motorcycle or a bicycle based on its traveling speed. Here, even if the "type," determined from a shape perspective, is determined to be a "bicycle," as described above, if the "motorcycle flag" is set, i.e., "present," it is treated as a "motorcycle" rather than a "motorcycle." Note that, in the above, if the "motorcycle flag" is "absent," it is treated as a "bicycle" rather than a "motorcycle."

[0041] In addition to the above, each data table α, β also includes "position data," "velocity data," and "direction data" that indicate the location, velocity, and direction of the moving object MB within the target area DD, as data identified based on the calculation results of the distance image processing unit DP (object detection processing unit OD) and the velocity calculation unit VC.

[0042] The contents of each data table α and β will be explained in more detail below. In the figure, data table α shows target information related to moving objects MBs present in the target area DD at a certain point in time, while data table β shows target information related to moving objects MBs present in the target area DD at a later point in time than data table α. Note that in the figure, some of the differences between data table β and data table α are indicated by hatching. In the illustrated example, data table α shows that a total of N moving objects MBs have been extracted. In contrast, data table β shows, as indicated by hatching, that one new moving object MB has been added compared to data table α, bringing the total to N+1. Furthermore, for moving objects MBs already being monitored at the time shown in data table α, for example, the moving object MB with target ID "8" has its "Type" listed as "Bicycle," but its "Motorcycle Flag" has changed from "None" to "Yes." In other words, it can be seen that what was previously treated as a regular bicycle has been changed to a motorcycle. As described above, the data content changes from frame to frame.

[0043] A series of processes in the roadside device 100 equipped with the object detection device 10 as described above will be described below with reference to the flowchart shown in FIG.

[0044] First, when the main body of the roadside device 100 is started up, the LiDAR 11 serving as the data acquisition unit DA of the object detection device 10 performs distance measurement to acquire point cloud data (step S101).

[0045] Next, the point cloud data acquired in step S101 is subjected to data processing such as distance measurement image generation by the distance image processing unit DP (object detection processing unit OD) of the LiDAR processing unit 12, and object detection is performed (step S102). Furthermore, the LiDAR processing unit 12 calculates various data (vector calculation for moving speed and traveling direction, etc.) for the target moving object MB as exemplified in Fig. 4, thereby tracking the moving object MB (step S103), and thereby the moving speed of the detected object is calculated by the speed calculation unit VC (step S104). Note that the tracking in step S103 is performed for all detected moving object MBs, but if the moving speed cannot be calculated in step S104, this process is skipped.

[0046] Here, it is confirmed whether or not a target ID has already been assigned to the object (moving body MB) being processed by the ID assignment unit IG (step S105). If not (step S105: No), a target ID is assigned to the object (step S106). If already assigned (step S105: Yes), the process proceeds to the next identification process (step S107) without any special processing.

[0047] In step S107, the vehicle type identification unit VI performs vehicle type identification based on shape characteristics. Here, as described above, each moving object MB assigned a target ID is classified into "large vehicle," "standard vehicle," "bicycle," and "pedestrian" according to a preset threshold value for shape characteristics, and those that cannot be classified are classified as "other." However, among these, "bicycle" includes both two-wheeled vehicles and regular bicycles that are not two-wheeled vehicles.

[0048] Next, as a result of the identification process in step S107, it is confirmed whether the type is "bicycle" (step S108). If it is "bicycle" (step S108: Yes), it is further confirmed whether the moving speed of the moving object MB to which the target ID is assigned is equal to or higher than a predetermined threshold (for example, 30 km / h or higher) (step S109). If it is equal to or higher than the threshold (step S109: Yes), the "bicycle" is treated as a "two-wheeled vehicle" (step S110), and a "two-wheeled vehicle flag" for treating it as such is set (step S111). That is, in the table data of FIG. 4, the "two-wheeled vehicle flag" column is set to "Yes".

[0049] On the other hand, in step S109, if the moving speed of the moving object MB to which the target ID is assigned is not equal to or greater than the threshold (step S109: No), it is confirmed whether the "two-wheeled vehicle flag" is set (step S112). If the "two-wheeled vehicle flag" is not set (step S112: No), no special processing is performed and the process proceeds to the next step. In other words, in this case, the type remains "bicycle" and the "two-wheeled vehicle flag" field is set to "none," and the object is treated as a normal "bicycle" rather than a motorcycle. Note that in step S112, if the "two-wheeled vehicle flag" is already set (step S112: Yes), the processes of steps S110 and S111 are performed to maintain the state of being treated as a "two-wheeled vehicle." In other words, once the "two-wheeled vehicle flag" is set, the object is maintained as being treated as a "two-wheeled vehicle" even if the moving speed subsequently falls below the threshold.

[0050] After determining in step S108 that the type is other than "bicycle" (step S108: No), or after going through the processing of step S111, or after determining in step S112 that the "two-wheeled vehicle flag" is not set (step S112: No), data on each moving body MB is selected to be output to the outside as target information, and the target information is output by the control device 50 (step S113). After outputting in step S113, the roadside unit 100 returns to the initial step S101 and repeats the series of operations described above.

[0051] An overview of the object detection device 10 of this embodiment will be described below with reference to the conceptual diagram shown in FIG.

[0052] As shown in the figure and as already described, the object detection device 10 of this embodiment includes a data acquisition unit DA that acquires data of a moving body MB in a target area DD, a speed calculation unit VC that calculates the moving speed of the moving body MB based on the data of the moving body MB, and an identification unit TI that identifies whether the moving body MB is a two-wheeled vehicle or a bicycle based on the moving speed of the moving body MB.

[0053] In the above embodiment, the data acquisition unit DA is configured, for example, by a LiDAR 11 as a distance measurement unit that continuously measures distances to a target area DD. Furthermore, a case has been described in which the LiDAR processing unit 12, configured by electronic circuits, etc., performs object detection of a moving object MB from a distance image generated for each frame based on point cloud data (distance measurement data) acquired by the LiDAR 11, thereby calculating the moving speed and identifying whether the moving object is a motorcycle or a bicycle. In this case, the LiDAR processing unit 12 functions as a speed calculation unit VC and an identification unit TI.

[0054] As described above, the object detection device 10 of this embodiment identifies whether a moving object is a two-wheeled vehicle or a bicycle based on the moving speed of the moving object MB calculated from the data of the moving object MB in the target area DD, which makes it possible to easily yet reliably identify two-wheeled vehicles and bicycles compared to methods that combine RFID information with other types of information to determine the movement, methods that determine the movement based on height or lane position, and methods that optimize the learning model.

[0055] Furthermore, in the roadside device 100 of this embodiment, by being equipped with the object detection device 10 as described above, it becomes possible to easily and reliably distinguish between two-wheeled vehicles and bicycles, and for example, to provide information to the outside based on such distinction.

[0056] On the other hand, when viewed from a different perspective, the LiDAR processing unit 12 of the object detection device 10 can also be said to function as a data processing device DU that performs various data processing, such as generating target information for providing information from distance measurement, i.e., surveying results, using the LiDAR 11. In other words, the data processing device DU configured by the LiDAR processing unit 12 can also be considered to identify whether the moving object MB is a two-wheeled vehicle or a bicycle according to the moving speed calculated based on data of the moving object MB in the target area DD.

[0057] 〔others〕 The present invention is not limited to the above-described embodiment, and can be embodied in various forms without departing from the spirit and scope of the present invention.

[0058] First, among the above, the moving body MB as an object to be monitored that exists in the target area DD is captured based on point cloud data (ranging data) from a ranging unit such as LiDAR 11, but the method of capturing the moving body MB is not limited to this, and it is also possible to use a two-dimensional image from a normal camera or to combine ranging by LiDAR and imaging by a camera.

[0059] Furthermore, in the above embodiment, the object detection device 10 including the LiDAR processing unit 12 is described as being provided in the roadside device 100, but the present invention is not limited to this. For example, it is also conceivable to store analyzed data in the LiDAR processing unit 12 as a data processing device DU, and use the stored data when analyzing, for example, an accident or the like. Regarding the accident analysis described above, it is also conceivable to install only a device equivalent to the LiDAR 11 capable of acquiring distance measurement data at the site where the roadside device 100 is installed, transmit the distance measurement data from the site to an external device such as a traffic volume recording device TR, and provide a data processing device DU in the traffic volume recording device TR or the like to perform the necessary data processing.

[0060] Furthermore, it is also conceivable that the distance measurement data (point cloud data) and various information obtained as a result of analyzing this data may be managed and processed in the cloud.

[0061] In the above description, the roadside device 100 is installed at an intersection CS, but the location is not limited to this and the device can be installed at various locations. For example, the present invention can be applied to a target area DD that includes a lane junction or a lane change section.

[0062] Furthermore, the shape of the intersection CS is merely an example, and is not limited to this and can be applied to cases with various shapes and structures. [Explanation of symbols]

[0063] 10...object detection device, 11...LiDAR, 12...LiDAR processing unit, 50...control device, 51...main control unit, 52...communication unit, 100...roadside device, AD...autonomous driving vehicle, AN...alarm unit, AR1...arrow, CS...intersection, DA...data acquisition unit, DD...target area, DM...data management unit, DP...distance image processing unit, DR...data acceptance unit, DS...display device, DU...data processing device, IG...ID assignment unit, MB...mobile body, MM...mobile body data management unit, OD...object detection processing unit, SC...signal controller, SN...signage, TI...identification unit, TL...signal lamp, TR...traffic volume recording device, VC...speed calculation unit, VE...vehicle, VI...vehicle type identification unit, α, β...data table

Claims

1. a data acquisition unit that acquires data of moving objects in a target area; a speed calculation unit that calculates a moving speed based on data of the moving object; an identification unit that identifies whether the moving object is a two-wheeled vehicle or a bicycle according to the moving speed of the moving object; An object detection device comprising:

2. The object detection device according to claim 1 , wherein the identification unit identifies whether the moving object is a two-wheeled vehicle or a bicycle depending on whether the moving speed of the moving object is equal to or greater than a set speed threshold.

3. The object detection device according to claim 2 , wherein when the identification unit determines that the moving object is a two-wheeled vehicle or a bicycle based on the moving speed of the moving object, the identification unit maintains the determination in a held state.

4. The object detection device according to claim 1 , wherein the data acquisition unit is a distance measurement unit that acquires point cloud data by distance measurement.

5. an ID assigning unit that assigns an ID based on a shape feature of the moving object; the data acquisition unit continuously acquires data; 2. The object detection device according to claim 1, wherein the speed calculation unit determines the identity of the moving object to which an ID has been assigned based on shape characteristics, and calculates the moving speed of the moving object based on the displacement between consecutively acquired data about the same moving object.

6. A roadside device comprising the object detection device according to any one of claims 1 to 5.

7. The roadside device according to claim 6, wherein the result of object detection regarding the moving body, including the result of the identification by the identification unit as to whether the moving body is a two-wheeled vehicle or a bicycle, is notified to other moving bodies.

8. A data processing device that identifies whether a moving object is a two-wheeled vehicle or a bicycle based on a moving speed calculated based on data of the moving object in a target area.

Images