WO2015104924A1 - Moving-body detection device, computer program, and moving-body detection method - Google Patents

Moving-body detection device, computer program, and moving-body detection method Download PDF

Info

Publication number
WO2015104924A1
WO2015104924A1 PCT/JP2014/082071 JP2014082071W WO2015104924A1 WO 2015104924 A1 WO2015104924 A1 WO 2015104924A1 JP 2014082071 W JP2014082071 W JP 2014082071W WO 2015104924 A1 WO2015104924 A1 WO 2015104924A1
Authority
WO
WIPO (PCT)
Prior art keywords
sensing
moving object
moving
pulse
unit
Prior art date
Application number
PCT/JP2014/082071
Other languages
French (fr)
Japanese (ja)
Inventor
下浦　弘
Original Assignee
住友電気工業株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 住友電気工業株式会社 filed Critical 住友電気工業株式会社
Publication of WO2015104924A1 publication Critical patent/WO2015104924A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G06V20/54—Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G08G1/0116—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G08G1/0133—Traffic data processing for classifying traffic situation
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • G08G1/0141—Measuring and analyzing of parameters relative to traffic conditions for specific applications for traffic information dissemination
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/04—Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors

Definitions

  • the present invention relates to a moving body sensing device for sensing a moving body, a computer program for realizing the moving body sensing device, and a moving body sensing method.
  • an image-type vehicle detector using an image processing technique has been widely used.
  • an apparatus for detecting a vehicle by detecting a vehicle head portion such as a front surface of the vehicle from image information captured by a video camera and aligning the vehicle head portion of a mobile model prepared in advance with the detected vehicle head portion is disclosed. (See Patent Document 1).
  • the present invention has been made in view of such circumstances, and provides a moving body sensing apparatus capable of sensing a moving body with high accuracy, a computer program for realizing the moving body sensing apparatus, and a moving body sensing method.
  • the purpose is to do.
  • the moving body sensing device acquires data for detecting a moving body by an acquisition unit, and based on the acquired data, each of a plurality of predetermined areas on the road surface is a moving body.
  • a determination unit that determines whether or not the mobile object candidate region indicates presence, and a mobile object or a mobile object group that is defined by a plurality of regions that are determined to be mobile object candidate regions by the determination unit is the road.
  • a generator for generating the pulses at a required interval.
  • a computer program is a computer program for causing a computer to detect a moving object, and is based on data for detecting the moving object by the computer.
  • a step of determining whether each of the plurality of areas is a moving object candidate area indicating the presence of a moving object, and a moving object or a moving object group defined by the plurality of areas determined to be moving object candidate areas A step of identifying a time of arrival at which a predetermined sensing area for sensing a moving object on the road surface is reached and a time of departure from the sensing area; and a pulse having a required width between the time of arrival and the time of departure. Generating at a required interval.
  • data for sensing a moving body is acquired by an acquisition unit, and each of a plurality of predetermined areas on a road surface is based on the acquired data.
  • the determination unit determines whether or not the moving object candidate area indicates presence, and the moving object or moving object group defined by the plurality of areas determined to be the moving object candidate area includes the road surface.
  • the moving body sensing device acquires data for detecting a moving body by the acquisition unit, and based on the acquired data, each of a plurality of predetermined areas on the road surface
  • a determination unit that determines whether or not a mobile object candidate area indicates the presence of a mobile object, and a mobile object or a mobile object group that is defined by a plurality of areas that are determined to be mobile object candidate areas by the determination unit.
  • a specifying unit for specifying a time of arrival at a predetermined detection area for detecting a moving object on the road surface and a time of departure from the detection area; and a point of time from the time of arrival specified by the specific unit
  • a generation unit for generating a pulse having a required width at a required interval.
  • a computer program is a computer program for causing a computer to sense a moving object, and is based on data for sensing the moving object by the computer.
  • a step of determining whether each of the plurality of predetermined areas is a moving object candidate area indicating the presence of a moving object; and a moving object or moving object defined by the plurality of areas determined to be moving object candidate areas A step of identifying a time point when the group reaches a predetermined sensing area for sensing a moving object on the road surface and a time point when the group leaves the sensing area; Generating a plurality of pulses at a required interval.
  • the acquisition unit acquires data for detecting a moving body, and each of a plurality of predetermined areas on the road surface moves based on the acquired data.
  • a step in which the determination unit determines whether or not the moving object candidate area indicates the presence of a body, and the moving object or moving object group defined by the plurality of areas determined to be moving object candidate areas is the road
  • a specific unit specifies a time of arrival at which a predetermined sensing area for sensing a moving body on the surface is reached and a time of separation from the sensing area, and a pulse having a required width between the time of arrival and the time of departure.
  • a step of generating a generator at a required interval is
  • the determination unit determines whether each of a plurality of predetermined areas on the road surface is a moving object candidate area indicating the presence of the moving object, based on the data for sensing the moving object acquired by the acquiring unit.
  • the acquisition unit is an image sensor (for example, an imaging device such as a video camera), but is not limited thereto, and for example, a millimeter wave sensor, a laser sensor, or the like can be used.
  • the moving body is a vehicle including a two-wheeled vehicle.
  • the predetermined plurality of areas (also referred to as small areas) on the road surface are areas corresponding to road surface areas obtained by dividing the measurement area of the moving body on the road surface into a plurality of predetermined sizes, and the size on the road surface is For example, a rectangular region of about 50 cm ⁇ 50 cm can be used, but the size of the region is not limited to this.
  • the determination unit determines whether each region is a moving object candidate region based on the data acquired by the acquisition unit. That is, the determination unit determines whether or not a moving object exists in each region in the measurement area.
  • the identifying unit leaves the sensing region when the moving body or the moving body group defined by the plurality of regions determined as the moving object candidate region by the determining unit reaches a predetermined sensing region on the road surface. Identify when to leave. For example, it is possible to demarcate each large area as a moving body group in which a moving body or a plurality of moving bodies form a group by grouping each area determined to be a moving body candidate area. it can.
  • the sensing area is an area having a required sensing length along the traveling direction of the moving body for sensing the presence of the moving body.
  • the specifying unit specifies an arrival time at which the top of the moving body or the moving body group reaches the sensing area, and specifies a leaving time at which the end of the moving body or the moving body group leaves the sensing area.
  • the generation unit generates a pulse having a required width at a required interval from the arrival time specified by the specification unit to the departure time.
  • the required width is the time width during which the pulse is on.
  • the required interval is the period of the pulse, and is the time from when the pulse is turned on to when the next pulse is turned on after the required width.
  • a large number of vehicles including two-wheeled vehicles are generated by repeatedly generating pulses (also referred to as sensing pulses) at a required interval until the moving body or group of moving bodies arrives at the sensing area and then leaves, and outputs the generated pulses. When the vehicle travels through the sensing area, it is possible to sense (measure) individual vehicles separately.
  • the moving body sensing device includes a speed calculation unit that calculates a speed of the moving body or the moving body group, and the required interval based on the speed calculated by the speed calculation unit.
  • the speed calculation unit calculates the speed of the moving object or moving object group.
  • the speed of the moving object or the moving object group can be obtained, for example, by the moving distance between different time points of the area determined to be the moving object candidate area, or between different time points of the defined moving object or moving object group. It can obtain
  • the interval calculation unit calculates a required pulse interval based on the speed calculated by the speed calculation unit. Thereby, regardless of the speed of the moving body or the moving body group, pulses can be generated at an appropriate required interval according to the speed.
  • the mobile body sensing device calculates traffic based on the correspondence between the speed of a mobile body or a group of mobile bodies and the traffic volume and the speed calculated by the speed calculator.
  • a volume calculation unit is provided, and the interval calculation unit calculates the required interval based on the traffic volume calculated by the traffic volume calculation unit.
  • the traffic volume calculation unit calculates the traffic volume based on the correspondence between the speed of the moving body or the moving body group and the traffic volume and the speed calculated by the speed calculation unit.
  • the correspondence between the speed and the traffic volume can be determined and stored in advance, or a relational expression between the speed and the traffic volume may be obtained by calculation.
  • the relationship between the traffic volume and the speed can be expressed by, for example, a quadratic function including the critical speed at the critical point where the traffic volume is maximum and the maximum traffic volume at the critical speed. That is, when the speed increases from 0 to the critical speed, the traffic volume also increases. When the speed further increases beyond the critical speed, the traffic volume decreases.
  • the generation unit stops generating a pulse and the peak value is constant. This signal is generated.
  • the generation unit stops generating a pulse and generates a signal having a constant peak value.
  • the speed threshold may be a value that can determine whether or not the moving object or the moving object group is stopped. Stopping the generation of a pulse generates a signal having a constant peak value instead of a pulse that repeats an on state and an off state.
  • the constant peak value includes the case where the peak value is zero. That is, to generate a signal having a constant peak value means to generate a continuous signal having a predetermined peak value (on-state signal) and to generate a continuous signal having a peak value of 0 (off-state signal). means.
  • whether an on-state signal or an off-state signal is generated may be set in advance. It is also possible to determine whether to generate an on-state signal or an off-state signal according to the relationship between the time point when the calculated speed is determined to be smaller than the predetermined speed threshold and the pulse. For example, when it is determined that the moving body or the moving body group stops in the middle of the pulse width of the pulse (in the middle of the pulse), an on-state signal (a continuous signal having a predetermined peak value) is generated. In addition, when it is determined that the moving body or the moving body group has stopped between one pulse and the next pulse (pulse is in an off state), a signal in an off state (continuous signal with a peak value of 0) is generated. can do. Thereby, even when a mobile body or a mobile body group stops, suitable sensing information can be provided to an external traffic signal control device or the like.
  • the moving body sensing device determines the required width of the pulse based on the sensing length of the sensing area, the predetermined moving body length, and the speed calculated by the speed calculation unit. A part.
  • the specific unit is configured to detect when the dimension of the mobile body or the mobile body group in the sensing area in the road width direction is smaller than a predetermined width threshold. Is specified as the time of withdrawal.
  • the identifying unit identifies the time when the dimension of the moving body or group of moving bodies in the sensing area in the road width direction becomes smaller than a predetermined width threshold as the departure time.
  • the predetermined width threshold can be, for example, about the width of the vehicle (four-wheeled vehicle).
  • the width of the vehicle group in the road width direction is considered to be larger (wider) than the width of one vehicle. Therefore, the time when the dimension of the moving body or the moving body group in the road width direction becomes smaller than the predetermined width threshold is considered as the time when the end of the vehicle group leaves the sensing area. be able to.
  • the generating unit generates a pulse after the pulse when the specifying unit specifies a departure time in the middle of the pulse having the required width. It is supposed to stop.
  • the generation unit stops generating the pulse after the pulse when the specific unit specifies the departure time in the middle of the pulse having the required width. That is, when an arbitrary pulse is in the on state and the departure time is specified, the pulse is generated and output as it is, but the next pulse of the pulse is not generated. Thereby, after the end of the vehicle group leaves the sensing area, the generation of pulses can be stopped.
  • the specific unit may be configured such that when the dimension of the mobile body or the mobile body group in the sensing area in the road width direction is larger than a predetermined width threshold value. Is specified as the arrival time.
  • the identifying unit identifies the time when the dimension of the moving body or the group of moving bodies in the sensing area in the road width direction is larger than a predetermined width threshold as the arrival time.
  • the predetermined width threshold can be, for example, about the width of the vehicle (four-wheeled vehicle).
  • the width of the vehicle group in the road width direction is considered to be larger (wider) than the width of one vehicle. Therefore, when the dimension of the moving body or the moving body group in the road width direction becomes larger than the predetermined width threshold, it is considered that the head of the vehicle group reaches the sensing area, so that time is specified as the reaching time point. be able to.
  • the moving body sensing device includes an extraction unit that extracts a plurality of feature points in the region based on data acquired by the acquisition unit, and a feature point extracted by the extraction unit.
  • a distance calculation unit that calculates a separation distance; and a vector calculation unit that calculates a movement vector of each feature point extracted by the extraction unit, wherein the determination unit is configured such that the distance calculated by the distance calculation unit is a predetermined distance threshold value. If the number of feature points that are shorter and the movement vector calculated by the vector calculation unit is within a predetermined range is greater than a predetermined threshold, it is determined that the area is a moving object candidate area.
  • the extraction unit extracts a plurality of feature points in the region based on the data acquired by the acquisition unit. That is, for example, a plurality of feature points are extracted in an area corresponding to a rectangular road surface area of about 50 cm ⁇ 50 cm.
  • the region can be further divided into a plurality of pixel blocks, and feature points can be extracted for each pixel block.
  • the feature points can be extracted by techniques such as edge extraction (spatial differentiation), peripheral incremental extraction, and background difference extraction.
  • the distance calculation unit calculates a separation distance between the feature points extracted by the extraction unit. Further, the vector calculation unit calculates a movement vector of each feature point extracted by the extraction unit.
  • the movement vector can be obtained by tracking each extracted feature point with time and by the movement distance and movement direction of the feature point moved between different time points.
  • the determination unit determines that the region is a moving object candidate region when the calculated distance is shorter than a predetermined distance threshold and the number of feature points having the calculated movement vector within a predetermined range is larger than the predetermined threshold.
  • the distance between the feature points is within a certain range (shorter than the distance threshold), and the movement vector of the feature points is approximate (the movement direction and the movement distance are within a predetermined range).
  • a predetermined threshold for example, if the ratio of pixel blocks from which feature points satisfying the condition to the total number of pixel blocks in the area are greater than the threshold, the area is moved. It determines with it being a body candidate area
  • FIG. 1 is a block diagram illustrating an example of the configuration of the mobile body sensing device 100 according to the present embodiment
  • FIG. 2 is an explanatory diagram illustrating an example of an installation state of the mobile body sensing device 100 according to the present embodiment. As shown in FIG.
  • the moving body sensing device 100 includes a control unit 11 that controls the entire device, an interface unit 12, a feature point extraction unit 13, a determination unit 14, a specification unit 15, a storage unit 16, a speed calculation unit 17, A pulse width determination unit 18, a pulse interval calculation unit 19, a pulse signal generation unit 20, an output unit 21, and the like are provided.
  • a video camera 200 as an acquisition unit is connected to the moving body sensing device 100.
  • the video camera 200 has a predetermined area near the road in a state where imaging conditions such as a predetermined height and a lens optical axis direction (for example, depression angle and rotation angle) are set with the road as a field of view. It is installed at the point.
  • the video camera 200 can image a vehicle existing in a measurement area on a road including a sensing line (also referred to as a sensing area).
  • the sensing line is an area having a required sensing length along the traveling direction of the vehicle for sensing the presence of the vehicle (also referred to as a moving body).
  • the video camera 200 sends image data (also referred to as data) obtained by imaging to the moving body sensing device 100.
  • the moving body sensing device 100 senses a vehicle in a measurement area or a group of vehicles (also referred to as a moving body group) in a measurement area using a captured image captured by the video camera 200 and detects each vehicle. It generates and outputs a sensing pulse equivalent to separating the vehicle.
  • the vehicle includes not only a four-wheeled vehicle but also a two-wheeled vehicle.
  • the moving body sensing device 100 and the video camera 200 are configured as separate devices. However, the present invention is not limited to this, and the moving body sensing device 100 is integrated with the video camera 200. The structure which makes
  • the interface unit 12 acquires the image data sent from the video camera 200 and stores it in the storage unit 16.
  • the image data is stored in the storage unit 16 on a frame-by-frame basis in synchronization with the frame rate of the video camera 200 (interval at the time of imaging, for example, 30 frames per second).
  • FIG. 3 is an explanatory diagram showing an example of a captured image.
  • the video camera 200 is provided at a required height from the road surface, and the video camera 200 captures an image from the front of a vehicle or a group of vehicles traveling on a two-lane road. In the captured image, the entire measurement area including the sensing line is captured.
  • the traveling direction of the vehicle in each lane can be set as will be described later.
  • FIG. 4 is a schematic diagram illustrating an example of a vehicle group traveling on a road
  • FIG. 5 is an explanatory diagram illustrating an example of a method of calculating the speed of the vehicle group.
  • FIGS. 4 and 5 a case will be described in which a plurality of two-wheeled vehicles are traveling in a group as a vehicle group.
  • each motorcycle travels on the road surface and each motorcycle travels a certain distance between time t-1 and time t.
  • the distance is the amount of movement of each motorcycle.
  • a plurality of feature points (P1 to P10 in the example of FIG. 5 are extracted for convenience) based on a captured image obtained by capturing with the video camera 200 at time t ⁇ 1.
  • a feature point extraction method will be described later. If each feature point extracted at time t-1 is tracked and the feature points P1 to P10 are specified at time t, the amount of movement of the feature points P1 to P10 from time t-1 to time t is obtained. be able to. As shown in FIG.
  • the movement vectors of the feature points P1 to P10 can be obtained.
  • the speed for example, the average speed of the speeds of the two-wheeled vehicles
  • the movement vector can be obtained by using a technique such as an optical flow, for example.
  • FIG. 6 is a schematic diagram showing an example of a road surface including a measurement area.
  • a two-lane road is presented, but the number of lanes is not limited to the example of FIG. 6.
  • the measurement area including the sensing lines is divided into a grid (mesh) and divided into a plurality of road surface areas.
  • length ⁇ width is, for example, 50 cm ⁇ 50 cm on the road surface, but the size of the road surface area is not limited to this.
  • the length of the measurement line is, for example, about 30 m to 60 m
  • the road width per lane is, for example, about 3 m to 3.5 m.
  • the control unit 11 can set the traveling direction of the vehicle (upward or downward direction of the road, also referred to as the traveling direction) for each lane.
  • the traveling direction of the vehicle upward or downward direction of the road, also referred to as the traveling direction
  • only one lane is illustrated, but the same applies to the opposite lane.
  • FIG. 7 is a schematic diagram showing an example of a captured image of the small area ⁇ S corresponding to one road surface area on the road surface.
  • the road surface area ( ⁇ S) shown in FIG. 6 is divided into 100 pixel blocks of 10 ⁇ 10 in length and width.
  • one pixel block in FIG. 7 is a collection of a plurality of pixels obtained by imaging a 5 cm ⁇ 5 cm area on the road surface. it can.
  • the number of pixel blocks is not limited to the example of FIG.
  • FIG. 8 is a schematic diagram illustrating an example of a movement vector of feature points extracted in a pixel block.
  • the feature point extracting unit 13 Based on image data (sensed data) obtained by imaging (sensing) the vehicle with the video camera 200, the feature point extracting unit 13 configures each pixel constituting the small area ⁇ S corresponding to each road surface area in the measurement area. Extract feature points in the block. That is, as described above, a plurality of feature points are extracted in the small area ⁇ S corresponding to the rectangular road surface area of about 50 cm ⁇ 50 cm.
  • the small region ⁇ S can be further divided into a plurality of pixel blocks, and feature points can be extracted for each pixel block.
  • the feature points can be extracted by techniques such as edge extraction (spatial differentiation), peripheral incremental extraction, and background difference extraction.
  • the feature point extraction unit 13 has a function as a vector calculation unit, and calculates a movement vector of each extracted feature point.
  • the movement vector can be obtained by tracking each extracted feature point with time and by the movement distance and movement direction of the feature point moved between different time points.
  • points in each pixel block represent feature points
  • arrows represent movement vectors. Note that a blank pixel block represents a pixel block for which a feature point could not be extracted.
  • the feature point extraction unit 13 has a function as a distance calculation unit, and calculates a separation distance between the extracted feature points.
  • the determination unit 14 is a vehicle candidate area (also referred to as a moving object candidate area) in which each of a plurality of predetermined road surface areas on the road surface indicates the presence of the vehicle based on image data obtained by imaging the vehicle with the video camera 200. ). That is, the determination unit 14 determines whether a vehicle is present in the small area ⁇ S corresponding to each road surface area in the measurement area.
  • the determination unit 14 determines that the distance between the feature points calculated by the feature point extraction unit 13 is shorter than a predetermined distance threshold, and the movement vector calculated by the feature point extraction unit 13 is within a predetermined range.
  • the number of feature points is larger than a predetermined threshold, it is determined that the small area ⁇ S (corresponding road surface area) is a vehicle candidate area.
  • the distance between the feature points is within a certain range (shorter than the distance threshold), and the movement vector of the feature points is approximate (the movement direction and the movement distance are within a predetermined range).
  • the feature points satisfying the above are grouped (for example, labeled) and the number of grouped feature points is larger than a predetermined threshold, for example, the feature points satisfying the condition for the total number of pixel blocks in the small region ⁇ S are
  • the determination unit 14 determines that the small area ⁇ S is a vehicle candidate area. Since it is determined whether or not a vehicle exists for each small area ⁇ S corresponding to a relatively small road surface area, for example, even when many vehicles including two-wheeled vehicles are traveling in groups, It can be detected with high accuracy.
  • FIG. 9 is a schematic diagram showing an example of a method for determining whether or not the small region ⁇ S is a vehicle candidate region.
  • the total number of pixel blocks in the small region ⁇ S is 100.
  • the number of pixel blocks from which feature points are extracted is 81.
  • the ratio of pixel blocks from which feature points that satisfy the condition to the total number of pixel blocks in the small area ⁇ S are extracted is larger than 70%.
  • the traveling direction of each lane in the measurement area (for example, the upward and downward directions of the road) is set in advance, and the feature point extraction unit 13 sets the direction of the movement vector among the calculated movement vectors. Only a movement vector that is the same as or approximate to the traveling direction (for example, an angle between the moving vector and the traveling direction is about 10 °) may be extracted.
  • FIG. 10 is a schematic diagram showing an example of a vehicle group defined by vehicle candidate areas.
  • FIG. 10 shows a case where the vehicle group is defined in FIG.
  • the vehicle group is defined (configured and specified) by combining a plurality of road surface areas determined as vehicle candidate areas as one large area among all road surface areas in the measurement area.
  • a set of road surface areas with a pattern represents a vehicle group.
  • the specifying unit 15 arrives when a vehicle or a vehicle group defined by a plurality of regions (road surface region, small region ⁇ S) determined to be vehicle candidate regions by the determining unit 14 reaches a predetermined sensing line on the road surface. Identify the time of departure and the time of departure from the sensing line. That is, the specifying unit 15 specifies the arrival time when the head of the vehicle or the vehicle group reaches the sensing line, and specifies the departure time when the end of the vehicle or the vehicle group leaves the sensing line.
  • regions road surface region, small region ⁇ S
  • FIG. 11 is a schematic diagram illustrating an example of a time point when a vehicle group existing in the second lane reaches the sensing line
  • FIG. 12 is a schematic diagram illustrating an example of a time point when the vehicle group existing in the first lane reaches the sensing line.
  • the specifying unit 15 specifies the time when the dimension of the vehicle or the vehicle group in the road width direction on the sensing line becomes larger than a predetermined width threshold for each lane as the arrival time.
  • the predetermined width threshold can be, for example, about the width of the vehicle (four-wheeled vehicle).
  • the width of the vehicle group in the road width direction is considered to be larger (wider) than the width of one vehicle. Therefore, the time when the dimension of the defined vehicle or vehicle group in the road width direction becomes larger than the predetermined width threshold is considered as the time when the head of the vehicle group reaches the sensing line, so that time is specified as the arrival time. can do.
  • the arrival time point TS2 when the head of the vehicle group traveling in the second lane reaches the detection line (the output start point of the second lane detection pulse described later) TS2 is specified.
  • the arrival time point TS1 when the head of the vehicle group traveling in the first lane reaches the detection line (the output start point of a detection pulse of the first lane described later) TS1 is specified.
  • the length in the road width direction of the vehicle or the vehicle group in the sensing line may be a continuous length, or if the dimension in the road width direction is not continuous (discrete), it may be an accumulated length. Good.
  • FIG. 13 is a schematic diagram showing an example of when the vehicle group leaves the sensing line.
  • the specifying unit 15 specifies the time when the dimension of the vehicle or the vehicle group in the road width direction on the sensing line is smaller than a predetermined width threshold as the departure time.
  • the predetermined width threshold can be, for example, about the width of the vehicle (four-wheeled vehicle).
  • the width of the vehicle group in the road width direction is considered to be larger (wider) than the width of one vehicle. Therefore, the time when the dimension of the defined vehicle or vehicle group in the road width direction becomes smaller than the predetermined width threshold is considered as the time when the end of the vehicle group leaves the sensing line. can do.
  • the arrival time point TE at which the tail of the vehicle group traveling in the first lane and the second lane leaves the detection line (the detection pulse output stop point described later) TE is specified.
  • the departure time when the vehicle group leaves the sensing line is the same timing in the first lane and the second lane, but may be different timing.
  • the width threshold value for specifying the arrival time point and the threshold value for specifying the departure time point may be the same value or different values.
  • the pulse signal generation unit 20 has a function as a generation unit.
  • the pulse signal generation unit 20 generates a sensing pulse (also referred to as a pulse) having a required width at a required interval from the arrival time specified by the specifying unit 15 to the departure time.
  • the required width is the pulse width of the sensing pulse, and is the time width during which the sensing pulse is in the on state.
  • the required interval is the period of the sensing pulse, and is the time from when the sensing pulse is turned on until the next sensing pulse is turned on after the required width is turned off.
  • FIG. 14 is a time chart showing an example of a sensing pulse signal generated by the moving body sensing device 100 of the present embodiment.
  • the pulse signal generation unit 20 starts generating the second lane sensing pulse at the arrival time TS2 when the head of the vehicle group reaches the sensing line.
  • the pulse signal generation unit 20 starts generating the first lane sensing pulse at the arrival time TS1 when the head of the vehicle group reaches the sensing line.
  • Each sensing pulse has a required width of Tw and a required interval of Td.
  • the pulse signal generation unit 20 stops generating the detection pulse at the departure time TE when the end of the vehicle group leaves the detection line. Note that the required width Tw and the required interval Td of the pulse signal shown in FIG.
  • the required width Tw and the required interval Td can vary from pulse to pulse.
  • a time difference (Td ⁇ Tw) may be calculated as the required interval.
  • the speed calculation unit 17 calculates the speed of the defined vehicle or vehicle group.
  • the speed of the vehicle or the vehicle group can be obtained by, for example, the method illustrated in FIG.
  • the speed of the small area may be obtained from the movement distance between different time points of any small area determined as the vehicle candidate area, and the speed of each of the plurality of small areas determined as the vehicle candidate area You may obtain
  • the vehicle candidate area far from the sensing line may be excluded by using the average speed of the vehicle group within a required distance (for example, 10 m to 15 m) from the sensing line. Thereby, even when the length of the vehicle group is relatively long, the speed of the vehicle group can be obtained with high accuracy.
  • the speed of the vehicle group can be obtained as follows. That is, a relationship between the number of feature points extracted by the feature point extraction unit 13 and the actual vehicle group density (the number of vehicles per unit length) is obtained in advance, and a relational expression (speed / density) The speed of the vehicle group can also be obtained using a density curve or KV curve).
  • FIG. 15 is an explanatory diagram showing an example of the correspondence between speed and density.
  • the horizontal axis indicates the density
  • the vertical axis indicates the speed.
  • the curve shown in FIG. 15 is referred to as a KV curve.
  • Density is the number of vehicles present per unit length. It is empirically known that there is a negative correlation between speed and density. For example, when the speed increases (fast), the driver tends to increase the inter-vehicle distance in order to maintain safety. Therefore, as the speed increases, the density decreases, resulting in a curve as illustrated in FIG.
  • speed Vc and density Kc are speed (critical speed) and density (critical density) at a critical point where the traffic volume is maximum. A region where the density is smaller than the critical density Kc is a natural flow region, and a region where the density is larger than the critical density Kc is a traffic flow region.
  • the pulse width determining unit 18 has a function as a determining unit, and determines the required width of the sensing pulse based on the sensing length of the sensing line, the predetermined vehicle length, and the speed calculated by the speed calculating unit 17.
  • the pulse interval calculation unit 19 has a function as an interval calculation unit, and calculates a required interval of the sensing pulse based on the speed calculated by the speed calculation unit 17. Thereby, regardless of the speed of the vehicle or the vehicle group, the sensing pulse can be generated at an appropriate required interval according to the speed.
  • the pulse interval calculation unit 19 also has a function as a traffic volume calculation unit, and is based on the correspondence between the speed of the vehicle or the vehicle group and the traffic volume and the speed calculated by the speed calculation unit 17. Calculate the amount.
  • FIG. 17 is an explanatory diagram showing an example of the correspondence between the traffic volume and the average speed.
  • the relationship between the traffic volume and the average speed is expressed by, for example, a quadratic function including the critical speed Vc at the critical point where the traffic volume is maximum and the maximum traffic volume Qmax at the critical speed Vc. Can do. That is, when the average speed increases from 0 to the critical speed Vc, the traffic volume also increases, and when the average speed further increases above the critical speed Vc, the traffic volume decreases.
  • the correspondence relationship between the average speed and the traffic volume can be determined and stored in advance, or a relational expression between the average speed and the traffic volume may be obtained by calculation.
  • the required interval of the sensing pulses is based on, for example, a fixed period (saturated traffic flow rate: 2 seconds / unit, for example) instead of the configuration calculated by obtaining the traffic volume from the vehicle group speed (average speed). May generate 1 sense pulse every 2 seconds).
  • the pulse signal generation unit 20, the pulse width determination unit 18, and the pulse interval calculation unit 19 use the speed calculated by the speed calculation unit 17.
  • the speed calculated by the speed calculation unit 17 is, for example, (1) an average speed after generating (outputting) the previous pulse, (2) the latest speed, and (3) a past fixed time (for example, 5 seconds). Like the average speed, it can be calculated by selecting one of the methods (1) to (3). This will be specifically described below.
  • FIG. 18 is a time chart showing an example of the timing for determining the required width and required interval of the sensing pulse.
  • the required width Tw1 and required interval Td1 of the first sensing pulse are determined immediately before time t1 or time t1, which is the timing of generation (output) of the sensing pulse.
  • the required width Tw2 and the required interval Td2 of the second sensing pulse are determined immediately before time t2 or time t2, which is the timing of generating (outputting) the sensing pulse.
  • the speed calculation unit 17 since the frequency or period at which the speed calculation unit 17 calculates the speed is shorter than the required interval (period) of the sensing pulse (for example, every 100 ms), the speed calculation unit 17 performs the period from time t1 to time t2. In this case, the speed is calculated a plurality of times.
  • the speed calculated by the speed calculation unit 17 (the speed of each normal vehicle when converted into a normal vehicle) and the number of vehicles that are the number of pulses generated by the pulse signal generation unit 20 are set to a required cycle (for example, 1 minute). Etc.) and may be transmitted to a central device that manages traffic signal control.
  • FIG. 19 is a time chart showing an example of the end timing of the sensing pulse.
  • the pulse signal generating unit 20 stops generating the sensing pulse after the sensing pulse. That is, when an arbitrary sensing pulse is in the on state and the departure time is specified, the sensing pulse is generated and output as it is as a sensing pulse of the required width Tw without being aborted. No sense pulse is generated. Thereby, after the end of the vehicle group leaves the sensing line, the generation of the sensing pulse can be stopped.
  • the pulse signal generation unit 20 stops generating pulses and generates a signal having a constant peak value.
  • the speed threshold may be a value that can determine whether or not the vehicle or the vehicle group is stopped.
  • FIG. 20 is a time chart showing an example of stopping the generation of sensing pulses when the vehicle group stops.
  • the generation of the pulse is stopped, and a signal having a constant peak value is generated.
  • the constant peak value includes the case where the peak value is zero. That is, to generate a signal with a constant peak value, instead of a pulse that repeats the ON state and the OFF state, a continuous signal with a peak value of 0 (off-state signal) is generated as shown in the upper chart of FIG. This means that a continuous signal (on-state signal) having a predetermined peak value is generated as shown in the lower chart of FIG. Note that whether an on-state signal or an off-state signal is generated may be set in advance.
  • an on-state signal or an off-state signal it is also possible to determine whether to generate an on-state signal or an off-state signal according to the relationship between the time point when the calculated speed is determined to be smaller than the predetermined speed threshold and the sensing pulse. For example, when it is determined that the vehicle or the vehicle group has stopped between one sensing pulse and the next sensing pulse (the sensing pulse is in an off state) as shown in the upper chart of FIG. Thereafter, an off-state signal (a continuous signal having a peak value of 0) is generated.
  • an on-state signal predetermined value
  • an on-state signal A continuous signal of peak values
  • FIG. 21 is a flowchart illustrating an example of a processing procedure of the moving body sensing device 100 according to the present embodiment.
  • the control unit 11 acquires data (for example, image data) for sensing the vehicle or the vehicle group (S11), and extracts a plurality of feature points for each small area of the road surface (S12).
  • the small area of the road surface is a small area ⁇ S on the captured image corresponding to each of a plurality of road surface areas obtained by dividing the road surface in the measurement area in a grid pattern.
  • the control unit 11 calculates a movement vector of each extracted feature point (S13), and determines whether each of the small areas ⁇ S is a vehicle candidate area (S14).
  • the control unit 11 identifies a vehicle group (including vehicles) defined by a plurality of small regions ⁇ S determined to be a vehicle candidate region (S15), and calculates the speed of the vehicle group (S16).
  • the control unit 11 specifies the arrival time point when the defined vehicle group reaches the sensing line (S17), and generates and outputs a sensing pulse having a required width at a required interval (S18).
  • the control unit 11 specifies the departure time point when the defined vehicle group leaves the sensing line (S19), stops generating and outputting the sensing pulse (S20), and ends the process.
  • the moving body sensing device 100 of the present embodiment can also be realized using a general-purpose computer including a CPU, a RAM, and the like. That is, as shown in FIG. 21, a computer program defining each processing procedure is recorded on a computer program recording medium such as a CD, DVD, USB memory, etc., and the computer program is loaded into a RAM provided in the computer. By executing the computer program on the CPU, the moving body sensing device 100 can be realized on the computer.
  • the moving body sensing device 100 of the present embodiment even when a large number of vehicles including two-wheeled vehicles travel while forming a vehicle group, the individual pulses are separated to generate sensing pulses for sensing the number of vehicles. can do. And by using the detection pulse which the mobile body detection apparatus 100 of this Embodiment produces
  • the video camera is used as the acquisition unit.
  • the acquisition unit is not limited to the video camera, and the vehicle is used for a plurality of road surface areas obtained by dividing the measurement area of the road surface in a grid pattern. As long as it is possible to determine whether or not there is a millimeter wave sensor, a laser sensor or the like can be used.
  • the sensing pulse can be generated and output.
  • a sensing pulse can be generated and output as if two ordinary vehicles passed.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Traffic Control Systems (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

This invention provides a moving-body detection device, a computer program, and a moving-body detection method whereby moving bodies can be detected with a high degree of precision. This moving-body detection device has the following: a determination unit that uses an acquisition unit to acquire data for detecting moving bodies and uses the acquired data to determine, for each of a plurality of prescribed regions of a road surface, whether or not that region is a moving-body candidate region, indicating the presence of a moving body; an identification unit that identifies an entry time at which a moving body or moving-body group demarcated by a plurality of regions that were determined to be moving-body candidate regions will enter a prescribed detection region of the road surface for detecting moving bodies and an exit time at which said moving body or moving-body group will exit said detection region; and a generation unit that generates pulses of a required width at required intervals from the identified entry time to the identified exit time.

Description

移動体感知装置、コンピュータプログラム及び移動体感知方法Moving object sensing device, computer program, and moving object sensing method
　本発明は、移動体を感知する移動体感知装置、該移動体感知装置を実現するためのコンピュータプログラム及び移動体感知方法に関する。 The present invention relates to a moving body sensing device for sensing a moving body, a computer program for realizing the moving body sensing device, and a moving body sensing method.
　円滑な交通を実現し、あるいは交通事故を未然に防止するために詳細な交通情報を提供することが求められている。例えば、道路の所定の地点を通過する車両の通過台数又は平均速度等のマクロな交通パラメータを取得することができる装置が求められている。 It is required to provide detailed traffic information in order to realize smooth traffic or prevent traffic accidents. For example, there is a need for a device that can acquire macro traffic parameters such as the number of vehicles passing through a predetermined point on a road or the average speed.
　このような装置の一例として、画像処理技術を利用した画像式車両感知器の普及が進んでいる。例えば、ビデオカメラで撮像された画像情報から、車両の前面などの車頭部分を検出し、検出した車頭部分に予め用意した移動体モデルの車頭部分を整合させることにより、車両を検出する装置が開示されている（特許文献１参照）。 As an example of such an apparatus, an image-type vehicle detector using an image processing technique has been widely used. For example, an apparatus for detecting a vehicle by detecting a vehicle head portion such as a front surface of the vehicle from image information captured by a video camera and aligning the vehicle head portion of a mobile model prepared in advance with the detected vehicle head portion is disclosed. (See Patent Document 1).
特開２０１３－８９１２９号公報JP 2013-89129 A
　しかしながら、特許文献１の装置にあっては、車両の車頭部分を検出する必要があるが、二輪車を含む多数の車両が車群をなして走行しているような場合では、二輪車を含む多数の車両を個別に分離することが難しく、分離した車両毎に車頭部分を検出することが困難な場合もある。そして、正確な台数を検出することができない場合には、交通量も正確に把握することができず、適切な交通信号制御に資することができない可能性がある。 However, in the device of Patent Document 1, it is necessary to detect the head portion of the vehicle. However, in the case where a large number of vehicles including a two-wheeled vehicle are traveling in a vehicle group, a large number of two-wheeled vehicles are included. It may be difficult to separate the vehicles individually, and it may be difficult to detect the vehicle head portion for each separated vehicle. And when an exact number cannot be detected, traffic volume cannot be grasped correctly and it may be unable to contribute to appropriate traffic signal control.
　本発明は、斯かる事情に鑑みてなされたものであり、精度良く移動体を感知することができる移動体感知装置、該移動体感知装置を実現するためのコンピュータプログラム及び移動体感知方法を提供することを目的とする。 The present invention has been made in view of such circumstances, and provides a moving body sensing apparatus capable of sensing a moving body with high accuracy, a computer program for realizing the moving body sensing apparatus, and a moving body sensing method. The purpose is to do.
　本発明の実施の形態に係る移動体感知装置は、移動体を感知するためのデータを取得部で取得し、取得したデータに基づいて、道路面上の所定の複数の領域それぞれが移動体の存在を示す移動体候補領域であるか否かを判定する判定部と、該判定部で移動体候補領域であると判定された複数の領域で画定される移動体又は移動体群が、前記道路面の移動体を感知するための所定の感知領域に到達する到達時点及び前記感知領域から離脱する離脱時点を特定する特定部と、該特定部で特定した到達時点から離脱時点の間、所要幅のパルスを所要間隔で生成する生成部とを備える。 The moving body sensing device according to the embodiment of the present invention acquires data for detecting a moving body by an acquisition unit, and based on the acquired data, each of a plurality of predetermined areas on the road surface is a moving body. A determination unit that determines whether or not the mobile object candidate region indicates presence, and a mobile object or a mobile object group that is defined by a plurality of regions that are determined to be mobile object candidate regions by the determination unit is the road A specific unit for identifying a time of arrival at which a predetermined sensing area for detecting a moving body of a surface is reached and a time of separation from the sensing area, and a required width between the time of arrival identified by the specific unit and the time of departure And a generator for generating the pulses at a required interval.
　本発明の実施の形態に係るコンピュータプログラムは、コンピュータに、移動体の感知をさせるためのコンピュータプログラムであって、コンピュータに、移動体を感知するためのデータに基づいて、道路面上の所定の複数の領域それぞれが移動体の存在を示す移動体候補領域であるか否かを判定するステップと、移動体候補領域であると判定された複数の領域で画定される移動体又は移動体群が、前記道路面の移動体を感知するための所定の感知領域に到達する到達時点及び前記感知領域から離脱する離脱時点を特定するステップと、特定した到達時点から離脱時点の間、所要幅のパルスを所要間隔で生成するステップとを実行させる。 A computer program according to an embodiment of the present invention is a computer program for causing a computer to detect a moving object, and is based on data for detecting the moving object by the computer. A step of determining whether each of the plurality of areas is a moving object candidate area indicating the presence of a moving object, and a moving object or a moving object group defined by the plurality of areas determined to be moving object candidate areas A step of identifying a time of arrival at which a predetermined sensing area for sensing a moving object on the road surface is reached and a time of departure from the sensing area; and a pulse having a required width between the time of arrival and the time of departure. Generating at a required interval.
　本発明の実施の形態に係る移動体感知方法は、移動体を感知するためのデータを取得部で取得し、取得したデータに基づいて、道路面上の所定の複数の領域それぞれが移動体の存在を示す移動体候補領域であるか否かを判定部が判定するステップと、移動体候補領域であると判定された複数の領域で画定される移動体又は移動体群が、前記道路面の移動体を感知するための所定の感知領域に到達する到達時点及び前記感知領域から離脱する離脱時点を特定部が特定するステップと、特定した到達時点から離脱時点の間、所要幅のパルスを所要間隔で生成部が生成するステップとを含む。 In the moving body sensing method according to the embodiment of the present invention, data for sensing a moving body is acquired by an acquisition unit, and each of a plurality of predetermined areas on a road surface is based on the acquired data. The determination unit determines whether or not the moving object candidate area indicates presence, and the moving object or moving object group defined by the plurality of areas determined to be the moving object candidate area includes the road surface. A step in which a specific unit specifies an arrival time when reaching a predetermined detection area for detecting a moving object and a departure time when leaving the detection area, and a pulse having a required width is required between the specified arrival time and the departure time. Generating by the generation unit at intervals.
　本発明によれば、精度良く移動体を感知することができる。 According to the present invention, it is possible to detect a moving object with high accuracy.
本実施の形態の移動体感知装置の構成の一例を示すブロック図である。It is a block diagram which shows an example of a structure of the mobile body sensing apparatus of this Embodiment. 本実施の形態の移動体感知装置の設置状態の一例を示す説明図である。It is explanatory drawing which shows an example of the installation state of the mobile body sensing apparatus of this Embodiment. 撮像画像の一例を示す説明図である。It is explanatory drawing which shows an example of a captured image. 道路を走行する車群の一例を示す模式図である。It is a schematic diagram which shows an example of the vehicle group which drive | works a road. 車群の速度の算出方法の一例を示す説明図である。It is explanatory drawing which shows an example of the calculation method of the speed of a vehicle group. 計測エリアを含む道路面の一例を示す模式図である。It is a schematic diagram which shows an example of the road surface containing a measurement area. 道路面上の一の路面領域に対応する小領域の撮像画像の一例を示す模式図である。It is a schematic diagram which shows an example of the captured image of the small area | region corresponding to one road surface area | region on a road surface. 画素ブロックで抽出された特徴点の移動ベクトルの一例を示す模式図である。It is a schematic diagram which shows an example of the movement vector of the feature point extracted by the pixel block. 小領域が車両候補領域であるか否かを判定する方法の一例を示す模式図である。It is a schematic diagram which shows an example of the method of determining whether a small area | region is a vehicle candidate area | region. 車両候補領域で画定される車群の一例を示す模式図である。It is a schematic diagram which shows an example of the vehicle group demarcated by a vehicle candidate area | region. 第２車線に存在する車群が感知ラインに到達した時点の一例を示す模式図である。It is a schematic diagram which shows an example when the vehicle group which exists in a 2nd lane arrived at the sensing line. 第１車線に存在する車群が感知ラインに到達した時点の一例を示す模式図である。It is a schematic diagram which shows an example when the vehicle group which exists in a 1st lane arrived at the sensing line. 車群が感知ラインから離脱した時点の一例を示す模式図である。It is a schematic diagram which shows an example at the time of a vehicle group having left | separated from the sensing line. 本実施の形態の移動体感知装置が生成する感知パルス信号の一例を示すタイムチャートである。It is a time chart which shows an example of the sensing pulse signal which the mobile body sensing apparatus of this Embodiment produces | generates. 速度と密度との対応関係の一例を示す説明図である。It is explanatory drawing which shows an example of the correspondence of speed and density. 感知パルスの所要幅の決定方法の一例を示す模式図である。It is a schematic diagram which shows an example of the determination method of the required width | variety of a sensing pulse. 交通量と平均速度との対応関係の一例を示す説明図である。It is explanatory drawing which shows an example of the correspondence of traffic volume and average speed. 感知パルスの所要幅及び所要間隔を決定するタイミングの一例を示すタイムチャートである。It is a time chart which shows an example of the timing which determines the required width and required space | interval of a sensing pulse. 感知パルスの終了タイミングの一例を示すタイムチャートである。It is a time chart which shows an example of the end timing of a sensing pulse. 車群が停止した場合の感知パルスの生成を停止する例を示すタイムチャートである。It is a time chart which shows the example which stops the production | generation of the sensing pulse when a vehicle group stops. 本実施の形態の移動体感知装置の処理手順の一例を示すフローチャートである。It is a flowchart which shows an example of the process sequence of the mobile body sensing apparatus of this Embodiment.
［本願発明の実施形態の説明］
　（１）本発明の実施の形態に係る移動体感知装置は、移動体を感知するためのデータを取得部で取得し、取得したデータに基づいて、道路面上の所定の複数の領域それぞれが移動体の存在を示す移動体候補領域であるか否かを判定する判定部と、該判定部で移動体候補領域であると判定された複数の領域で画定される移動体又は移動体群が、前記道路面の移動体を感知するための所定の感知領域に到達する到達時点及び前記感知領域から離脱する離脱時点を特定する特定部と、該特定部で特定した到達時点から離脱時点の間、所要幅のパルスを所要間隔で生成する生成部とを備える。 [Description of Embodiment of Present Invention]
(1) The moving body sensing device according to the embodiment of the present invention acquires data for detecting a moving body by the acquisition unit, and based on the acquired data, each of a plurality of predetermined areas on the road surface A determination unit that determines whether or not a mobile object candidate area indicates the presence of a mobile object, and a mobile object or a mobile object group that is defined by a plurality of areas that are determined to be mobile object candidate areas by the determination unit. A specifying unit for specifying a time of arrival at a predetermined detection area for detecting a moving object on the road surface and a time of departure from the detection area; and a point of time from the time of arrival specified by the specific unit And a generation unit for generating a pulse having a required width at a required interval.
　また、本発明の実施の形態に係るコンピュータプログラムは、コンピュータに、移動体の感知をさせるためのコンピュータプログラムであって、コンピュータに、移動体を感知するためのデータに基づいて、道路面上の所定の複数の領域それぞれが移動体の存在を示す移動体候補領域であるか否かを判定するステップと、移動体候補領域であると判定された複数の領域で画定される移動体又は移動体群が、前記道路面の移動体を感知するための所定の感知領域に到達する到達時点及び前記感知領域から離脱する離脱時点を特定するステップと、特定した到達時点から離脱時点の間、所要幅のパルスを所要間隔で生成するステップとを実行させる。 A computer program according to an embodiment of the present invention is a computer program for causing a computer to sense a moving object, and is based on data for sensing the moving object by the computer. A step of determining whether each of the plurality of predetermined areas is a moving object candidate area indicating the presence of a moving object; and a moving object or moving object defined by the plurality of areas determined to be moving object candidate areas A step of identifying a time point when the group reaches a predetermined sensing area for sensing a moving object on the road surface and a time point when the group leaves the sensing area; Generating a plurality of pulses at a required interval.
　また、本発明の実施の形態に係る移動体感知方法は、移動体を感知するためのデータを取得部で取得し、取得したデータに基づいて、道路面上の所定の複数の領域それぞれが移動体の存在を示す移動体候補領域であるか否かを判定部が判定するステップと、移動体候補領域であると判定された複数の領域で画定される移動体又は移動体群が、前記道路面の移動体を感知するための所定の感知領域に到達する到達時点及び前記感知領域から離脱する離脱時点を特定部が特定するステップと、特定した到達時点から離脱時点の間、所要幅のパルスを所要間隔で生成部が生成するステップとを含む。 In the moving body sensing method according to the embodiment of the present invention, the acquisition unit acquires data for detecting a moving body, and each of a plurality of predetermined areas on the road surface moves based on the acquired data. A step in which the determination unit determines whether or not the moving object candidate area indicates the presence of a body, and the moving object or moving object group defined by the plurality of areas determined to be moving object candidate areas is the road A step in which a specific unit specifies a time of arrival at which a predetermined sensing area for sensing a moving body on the surface is reached and a time of separation from the sensing area, and a pulse having a required width between the time of arrival and the time of departure. And a step of generating a generator at a required interval.
　判定部は、取得部で取得した移動体を感知するためのデータに基づいて、道路面上の所定の複数の領域それぞれが移動体の存在を示す移動体候補領域であるか否かを判定する。取得部は、画像センサ（例えば、ビデオカメラ等の撮像装置）であるが、これに限定されるものではなく、例えば、ミリ波センサ、レーザセンサ等を用いることもできる。移動体は、二輪車を含む車両である。道路面の所定の複数の領域（小領域とも称する）は、道路面での移動体の計測エリアを所定の大きさで複数分割した路面領域に対応する領域であり、道路面での大きさが、例えば、５０ｃｍ×５０ｃｍ程度の矩形状の領域とすることができるが、領域の大きさはこれに限定されるものではない。判定部は、取得部で取得したデータに基づいて、各領域が移動体候補領域であるか否かを判定する。すなわち、判定部は、計測エリア内の各領域に移動体が存在しているか否かを判定する。 The determination unit determines whether each of a plurality of predetermined areas on the road surface is a moving object candidate area indicating the presence of the moving object, based on the data for sensing the moving object acquired by the acquiring unit. . The acquisition unit is an image sensor (for example, an imaging device such as a video camera), but is not limited thereto, and for example, a millimeter wave sensor, a laser sensor, or the like can be used. The moving body is a vehicle including a two-wheeled vehicle. The predetermined plurality of areas (also referred to as small areas) on the road surface are areas corresponding to road surface areas obtained by dividing the measurement area of the moving body on the road surface into a plurality of predetermined sizes, and the size on the road surface is For example, a rectangular region of about 50 cm × 50 cm can be used, but the size of the region is not limited to this. The determination unit determines whether each region is a moving object candidate region based on the data acquired by the acquisition unit. That is, the determination unit determines whether or not a moving object exists in each region in the measurement area.
　特定部は、判定部で移動体候補領域であると判定された複数の領域で画定される移動体又は移動体群が、道路面の所定の感知領域に到達する到達時点及び前記感知領域から離脱する離脱時点を特定する。例えば、移動体候補領域であると判定された領域それぞれを纏めて１つの大領域とすることにより、当該大領域を移動体又は複数の移動体が群をなした移動体群として画定することができる。感知領域は、移動体の存在感知のため移動体の走行方向に沿って所要の感知長を有する領域である。特定部は、移動体又は移動体群の先頭が感知領域に到達する到達時点を特定するとともに、移動体又は移動体群の末尾が感知領域から離脱する離脱時点を特定する。 The identifying unit leaves the sensing region when the moving body or the moving body group defined by the plurality of regions determined as the moving object candidate region by the determining unit reaches a predetermined sensing region on the road surface. Identify when to leave. For example, it is possible to demarcate each large area as a moving body group in which a moving body or a plurality of moving bodies form a group by grouping each area determined to be a moving body candidate area. it can. The sensing area is an area having a required sensing length along the traveling direction of the moving body for sensing the presence of the moving body. The specifying unit specifies an arrival time at which the top of the moving body or the moving body group reaches the sensing area, and specifies a leaving time at which the end of the moving body or the moving body group leaves the sensing area.
　生成部は、特定部で特定した到達時点から離脱時点の間、所要幅のパルスを所要間隔で生成する。所要幅は、パルスがオン状態である時間幅である。また、所要間隔は、パルスの周期であり、パルスがオンとなった時点から、所要幅の後オフとなり、次のパルスがオンとなる時点までの時間である。移動体又は移動体群が感知領域に到達してから離脱するまでの間、パルス（感知パルスとも称する）を所要間隔で繰り返し生成し、生成したパルスを出力することにより、二輪車を含む多数の車両が感知領域を通過して走行する場合、個々の車両を分離して感知（計測）することができる。 The generation unit generates a pulse having a required width at a required interval from the arrival time specified by the specification unit to the departure time. The required width is the time width during which the pulse is on. The required interval is the period of the pulse, and is the time from when the pulse is turned on to when the next pulse is turned on after the required width. A large number of vehicles including two-wheeled vehicles are generated by repeatedly generating pulses (also referred to as sensing pulses) at a required interval until the moving body or group of moving bodies arrives at the sensing area and then leaves, and outputs the generated pulses. When the vehicle travels through the sensing area, it is possible to sense (measure) individual vehicles separately.
　（２）本発明の実施の形態に係る移動体感知装置は、前記移動体又は移動体群の速度を算出する速度算出部と、該速度算出部で算出した速度に基づいて、前記所要間隔を算出する間隔算出部とを備える。 (2) The moving body sensing device according to an embodiment of the present invention includes a speed calculation unit that calculates a speed of the moving body or the moving body group, and the required interval based on the speed calculated by the speed calculation unit. An interval calculation unit for calculating.
　速度算出部は、移動体又は移動体群の速度を算出する。移動体又は移動体群の速度は、例えば、移動体候補領域であると判定された領域の異なる時点間における移動距離により求めることができ、あるいは、画定した移動体又は移動体群の異なる時点間における移動距離により求めることができる。間隔算出部は、速度算出部で算出した速度に基づいて、パルスの所要間隔を算出する。これにより、移動体又は移動体群の速度にかかわらず、速度に応じた適切な所要間隔でパルスを生成することができる。 The speed calculation unit calculates the speed of the moving object or moving object group. The speed of the moving object or the moving object group can be obtained, for example, by the moving distance between different time points of the area determined to be the moving object candidate area, or between different time points of the defined moving object or moving object group. It can obtain | require by the movement distance in. The interval calculation unit calculates a required pulse interval based on the speed calculated by the speed calculation unit. Thereby, regardless of the speed of the moving body or the moving body group, pulses can be generated at an appropriate required interval according to the speed.
　（３）本発明の実施の形態に係る移動体感知装置は、移動体又は移動体群の速度と交通量との対応関係及び前記速度算出部で算出した速度に基づいて交通量を算出する交通量算出部を備え、前記間隔算出部は、前記交通量算出部で算出した交通量に基づいて前記所要間隔を算出するようにしてある。 (3) The mobile body sensing device according to the embodiment of the present invention calculates traffic based on the correspondence between the speed of a mobile body or a group of mobile bodies and the traffic volume and the speed calculated by the speed calculator. A volume calculation unit is provided, and the interval calculation unit calculates the required interval based on the traffic volume calculated by the traffic volume calculation unit.
　交通量算出部は、移動体又は移動体群の速度と交通量との対応関係及び速度算出部で算出した速度に基づいて交通量を算出する。速度と交通量との対応関係は、予め定めておき記憶しておくこともでき、あるいは速度と交通量との関係式を演算で求めるようにしてもよい。交通量と速度との関係は、例えば、交通量が最大となる臨界点での臨界速度及び臨界速度での最大交通量を含む２次関数で表すことができる。すなわち、速度が０から臨界速度に増加すると交通量も増加し、速度が臨界速度よりさらに増加すると交通量は減少する。間隔算出部は、交通量算出部で算出した交通量に基づいて所要間隔を算出する。例えば、算出した交通量が１分当たり３０台、すなわち６０秒当たり３０台とすると、所要間隔は、２秒（＝６０／３０）となる。これにより、交通量に応じて適切な所要間隔でパルスを生成することができる。 The traffic volume calculation unit calculates the traffic volume based on the correspondence between the speed of the moving body or the moving body group and the traffic volume and the speed calculated by the speed calculation unit. The correspondence between the speed and the traffic volume can be determined and stored in advance, or a relational expression between the speed and the traffic volume may be obtained by calculation. The relationship between the traffic volume and the speed can be expressed by, for example, a quadratic function including the critical speed at the critical point where the traffic volume is maximum and the maximum traffic volume at the critical speed. That is, when the speed increases from 0 to the critical speed, the traffic volume also increases. When the speed further increases beyond the critical speed, the traffic volume decreases. The interval calculation unit calculates the required interval based on the traffic volume calculated by the traffic volume calculation unit. For example, if the calculated traffic volume is 30 units per minute, that is, 30 units per 60 seconds, the required interval is 2 seconds (= 60/30). Thereby, pulses can be generated at appropriate intervals according to the traffic volume.
　（４）本発明の実施の形態に係る移動体感知装置は、前記生成部は、前記速度算出部で算出した速度が所定の速度閾値より小さい場合、パルスの生成を停止し、波高値が一定の信号を生成するようにしてある。 (4) In the mobile body sensing device according to the embodiment of the present invention, when the speed calculated by the speed calculation unit is smaller than a predetermined speed threshold, the generation unit stops generating a pulse and the peak value is constant. This signal is generated.
　生成部は、速度算出部で算出した速度が所定の速度閾値より小さい場合、パルスの生成を停止し、波高値が一定の信号を生成する。速度閾値は、移動体又は移動体群が停止しているか否かを判定することができる程度の値であればよい。パルスの生成を停止するとは、オン状態及びオフ状態を繰り返すパルスに代えて、波高値が一定の信号を生成する。波高値が一定とは、波高値が０の場合も含む。すなわち、波高値が一定の信号を生成するとは、所定の波高値の連続信号（オン状態の信号）を生成すること、及び波高値が０の連続信号（オフ状態の信号）を生成することを意味する。なお、オン状態の信号を生成するか、オフ状態の信号を生成するかは、予め設定することができるようにしてもよい。また、算出した速度が所定の速度閾値より小さいと判定した時点とパルスとの関係に応じて、オン状態の信号を生成するかオフ状態の信号を生成するかを決定することもできる。例えば、パルスのパルス幅の途中（パルスがオン状態の途中）で移動体又は移動体群が停止したと判定した場合には、オン状態の信号（所定の波高値の連続信号）を生成する。また、一のパルスと次のパルスとの間（パルスがオフ状態）で移動体又は移動体群が停止したと判定した場合には、オフ状態の信号（波高値が０の連続信号）を生成することができる。これにより、移動体又は移動体群が停止した場合でも、適切な感知情報を外部の交通信号制御装置などへ提供することができる。 When the speed calculated by the speed calculation unit is smaller than a predetermined speed threshold, the generation unit stops generating a pulse and generates a signal having a constant peak value. The speed threshold may be a value that can determine whether or not the moving object or the moving object group is stopped. Stopping the generation of a pulse generates a signal having a constant peak value instead of a pulse that repeats an on state and an off state. The constant peak value includes the case where the peak value is zero. That is, to generate a signal having a constant peak value means to generate a continuous signal having a predetermined peak value (on-state signal) and to generate a continuous signal having a peak value of 0 (off-state signal). means. Note that whether an on-state signal or an off-state signal is generated may be set in advance. It is also possible to determine whether to generate an on-state signal or an off-state signal according to the relationship between the time point when the calculated speed is determined to be smaller than the predetermined speed threshold and the pulse. For example, when it is determined that the moving body or the moving body group stops in the middle of the pulse width of the pulse (in the middle of the pulse), an on-state signal (a continuous signal having a predetermined peak value) is generated. In addition, when it is determined that the moving body or the moving body group has stopped between one pulse and the next pulse (pulse is in an off state), a signal in an off state (continuous signal with a peak value of 0) is generated. can do. Thereby, even when a mobile body or a mobile body group stops, suitable sensing information can be provided to an external traffic signal control device or the like.
　（５）本発明の実施の形態に係る移動体感知装置は、前記感知領域の感知長、所定の移動体長及び前記速度算出部で算出した速度に基づいて、前記パルスの所要幅を決定する決定部を備える。 (5) The moving body sensing device according to the embodiment of the present invention determines the required width of the pulse based on the sensing length of the sensing area, the predetermined moving body length, and the speed calculated by the speed calculation unit. A part.
　決定部は、感知領域の感知長、所定の移動体長及び速度算出部で算出した速度に基づいて、パルスの所要幅を決定する。感知領域の感知長をＷ、移動体長をＬ及び移動体又は移動体群の速度をＶとすると、パルスの所要幅Ｔｗは、Ｔｗ＝（Ｗ＋Ｌ）／Ｖの式で求めることができる。すなわち、パルスの所要幅Ｔｗは、移動体の先頭が感知領域に到達する時点から、移動体の末尾が感知領域から離脱する時点までの移動体の移動距離（Ｗ＋Ｌ）を速度Ｖで除算することにより求めることができる。これにより、車両などの移動体が纏まって車群を構成して走行している場合でも、１台毎に車両（移動体）を分離したのと同等のパルスを生成して出力することができる。 The determining unit determines the required width of the pulse based on the sensing length of the sensing region, the predetermined moving body length, and the speed calculated by the speed calculating unit. If the sensing length of the sensing region is W, the moving body length is L, and the speed of the moving body or group of moving bodies is V, the required width Tw of the pulse can be obtained by the equation Tw = (W + L) / V. That is, the required pulse width Tw is obtained by dividing the moving distance (W + L) of the moving object from the time when the head of the moving object reaches the sensing area to the time when the end of the moving object leaves the sensing area by the velocity V. It can ask for. As a result, even when a moving body such as a vehicle is traveling while forming a vehicle group, it is possible to generate and output a pulse equivalent to the separation of the vehicle (moving body) for each vehicle. .
　（６）本発明の実施の形態に係る移動体感知装置は、前記特定部は、前記感知領域での前記移動体又は移動体群の道路幅方向の寸法が所定の幅閾値より小さくなった時点を前記離脱時点として特定するようにしてある。 (6) In the mobile body sensing device according to the embodiment of the present invention, the specific unit is configured to detect when the dimension of the mobile body or the mobile body group in the sensing area in the road width direction is smaller than a predetermined width threshold. Is specified as the time of withdrawal.
　特定部は、感知領域での移動体又は移動体群の道路幅方向の寸法が所定の幅閾値より小さくなった時点を離脱時点として特定する。所定の幅閾値は、例えば、車両（四輪車）の幅程度とすることができる。二輪車を含む車両が群を構成して走行している場合、道路幅方向の車群の幅は、１台の車両の幅よりも大きい（広い）と考えられる。そこで、移動体又は移動体群の道路幅方向の寸法が所定の幅閾値より小さくなった時点は、車群の末尾が感知領域から離脱する時点と考えられるので、当該時点を離脱時点として特定することができる。 The identifying unit identifies the time when the dimension of the moving body or group of moving bodies in the sensing area in the road width direction becomes smaller than a predetermined width threshold as the departure time. The predetermined width threshold can be, for example, about the width of the vehicle (four-wheeled vehicle). When vehicles including two-wheeled vehicles are traveling in a group, the width of the vehicle group in the road width direction is considered to be larger (wider) than the width of one vehicle. Therefore, the time when the dimension of the moving body or the moving body group in the road width direction becomes smaller than the predetermined width threshold is considered as the time when the end of the vehicle group leaves the sensing area. be able to.
　（７）本発明の実施の形態に係る移動体感知装置は、前記生成部は、前記所要幅のパルスの途中で前記特定部が離脱時点を特定した場合、前記パルスの後のパルスの生成を停止するようにしてある。 (7) In the mobile body sensing device according to an embodiment of the present invention, the generating unit generates a pulse after the pulse when the specifying unit specifies a departure time in the middle of the pulse having the required width. It is supposed to stop.
　生成部は、所要幅のパルスの途中で特定部が離脱時点を特定した場合、当該パルスの後のパルスの生成を停止する。すなわち、任意のパルスがオン状態であるときに、離脱時点が特定された場合、当該パルスはそのまま生成して出力するが、当該パルスの次のパルスは生成しないようにする。これにより、車群の末尾が感知領域から離脱した後は、パルスの生成を停止させることができる。 The generation unit stops generating the pulse after the pulse when the specific unit specifies the departure time in the middle of the pulse having the required width. That is, when an arbitrary pulse is in the on state and the departure time is specified, the pulse is generated and output as it is, but the next pulse of the pulse is not generated. Thereby, after the end of the vehicle group leaves the sensing area, the generation of pulses can be stopped.
　（８）本発明の実施の形態に係る移動体感知装置は、前記特定部は、前記感知領域での前記移動体又は移動体群の道路幅方向の寸法が所定の幅閾値より大きくなった時点を前記到達時点として特定するようにしてある。 (8) In the mobile body sensing device according to the embodiment of the present invention, the specific unit may be configured such that when the dimension of the mobile body or the mobile body group in the sensing area in the road width direction is larger than a predetermined width threshold value. Is specified as the arrival time.
　特定部は、感知領域での移動体又は移動体群の道路幅方向の寸法が所定の幅閾値より大きくなった時点を到達時点として特定する。所定の幅閾値は、例えば、車両（四輪車）の幅程度とすることができる。二輪車を含む車両が群を構成して走行している場合、道路幅方向の車群の幅は、１台の車両の幅よりも大きい（広い）と考えられる。そこで、移動体又は移動体群の道路幅方向の寸法が所定の幅閾値より大きくなった時点は、車群の先頭が感知領域に到達する時点と考えられるので、当該時点を到達時点として特定することができる。 The identifying unit identifies the time when the dimension of the moving body or the group of moving bodies in the sensing area in the road width direction is larger than a predetermined width threshold as the arrival time. The predetermined width threshold can be, for example, about the width of the vehicle (four-wheeled vehicle). When vehicles including two-wheeled vehicles are traveling in a group, the width of the vehicle group in the road width direction is considered to be larger (wider) than the width of one vehicle. Therefore, when the dimension of the moving body or the moving body group in the road width direction becomes larger than the predetermined width threshold, it is considered that the head of the vehicle group reaches the sensing area, so that time is specified as the reaching time point. be able to.
　（９）本発明の実施の形態に係る移動体感知装置は、前記取得部で取得したデータに基づいて前記領域で特徴点を複数抽出する抽出部と、該抽出部で抽出した特徴点同士の離隔距離を算出する距離算出部と、前記抽出部で抽出した各特徴点の移動ベクトルを算出するベクトル算出部とを備え、前記判定部は、前記距離算出部で算出した距離が所定の距離閾値より短く、前記ベクトル算出部で算出した移動ベクトルが所定範囲内にある特徴点の数が所定の閾値より大きい場合、前記領域が移動体候補領域であると判定するようにしてある。 (9) The moving body sensing device according to the embodiment of the present invention includes an extraction unit that extracts a plurality of feature points in the region based on data acquired by the acquisition unit, and a feature point extracted by the extraction unit. A distance calculation unit that calculates a separation distance; and a vector calculation unit that calculates a movement vector of each feature point extracted by the extraction unit, wherein the determination unit is configured such that the distance calculated by the distance calculation unit is a predetermined distance threshold value. If the number of feature points that are shorter and the movement vector calculated by the vector calculation unit is within a predetermined range is greater than a predetermined threshold, it is determined that the area is a moving object candidate area.
　抽出部は、取得部で取得したデータに基づいて領域で特徴点を複数抽出する。すなわち、例えば、５０ｃｍ×５０ｃｍ程度の矩形状の路面領域に対応する領域において、特徴点を複数抽出する。例えば、当該領域をさらに複数の画素ブロックに分割し、画素ブロック毎に特徴点を抽出することができる。特徴点は、例えば、エッジ抽出（空間微分）、周辺増分抽出、背景差分抽出などの手法により抽出することができる。距離算出部は、抽出部で抽出した特徴点同士の離隔距離を算出する。また、ベクトル算出部は、抽出部で抽出した各特徴点の移動ベクトルを算出する。移動ベクトルは、抽出した各特徴点を時間経過とともに追跡し、異なる時点間で移動した特徴点の移動距離及び移動方向により求めることができる。判定部は、算出された距離が所定の距離閾値より短く、算出した移動ベクトルが所定範囲内にある特徴点の数が所定の閾値より大きい場合、領域が移動体候補領域であると判定する。 The extraction unit extracts a plurality of feature points in the region based on the data acquired by the acquisition unit. That is, for example, a plurality of feature points are extracted in an area corresponding to a rectangular road surface area of about 50 cm × 50 cm. For example, the region can be further divided into a plurality of pixel blocks, and feature points can be extracted for each pixel block. The feature points can be extracted by techniques such as edge extraction (spatial differentiation), peripheral incremental extraction, and background difference extraction. The distance calculation unit calculates a separation distance between the feature points extracted by the extraction unit. Further, the vector calculation unit calculates a movement vector of each feature point extracted by the extraction unit. The movement vector can be obtained by tracking each extracted feature point with time and by the movement distance and movement direction of the feature point moved between different time points. The determination unit determines that the region is a moving object candidate region when the calculated distance is shorter than a predetermined distance threshold and the number of feature points having the calculated movement vector within a predetermined range is larger than the predetermined threshold.
　すなわち、抽出した特徴点のうち、特徴点同士の距離が一定以内（距離閾値より短い）であって、特徴点の移動ベクトルが近似している（移動方向と移動距離が所定範囲内）という条件を充足する特徴点の数が所定の閾値より大きい場合、例えば、領域内の画素ブロックの総数に対する当該条件を充足する特徴点が抽出された画素ブロックの割合が閾値より大きい場合、当該領域を移動体候補領域であると判定する。比較的小さい路面領域に対応する領域毎に移動体が存在しているか否かを判定するので、例えば、二輪車を含む多くの車両が群をなして走行している場合でも、車群全体を精度良く検出することができる。 That is, among the extracted feature points, the distance between the feature points is within a certain range (shorter than the distance threshold), and the movement vector of the feature points is approximate (the movement direction and the movement distance are within a predetermined range). If the number of feature points satisfying the condition is greater than a predetermined threshold, for example, if the ratio of pixel blocks from which feature points satisfying the condition to the total number of pixel blocks in the area are greater than the threshold, the area is moved. It determines with it being a body candidate area | region. Since it is determined whether or not there is a moving body for each area corresponding to a relatively small road surface area, for example, even when many vehicles including two-wheeled vehicles are traveling in groups, the entire vehicle group is accurately detected. It can be detected well.
［本願発明の実施形態の詳細］
　以下、本発明に係る移動体感知装置、移動体感知装置を実現するためのコンピュータプログラム及び移動体感知方法の実施の形態を示す図面に基づいて説明する。なお、以下に記載する実施の形態の少なくとも一部を任意に組み合わせることができる。図１は本実施の形態の移動体感知装置１００の構成の一例を示すブロック図であり、図２は本実施の形態の移動体感知装置１００の設置状態の一例を示す説明図である。図１に示すように、移動体感知装置１００は、装置全体を制御する制御部１１、インタフェース部１２、特徴点抽出部１３、判定部１４、特定部１５、記憶部１６、速度算出部１７、パルス幅決定部１８、パルス間隔算出部１９、パルス信号生成部２０、出力部２１などを備える。移動体感知装置１００には、取得部としてのビデオカメラ２００を接続してある。 [Details of the embodiment of the present invention]
DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a moving object sensing device, a computer program for realizing the moving object sensing device, and a moving object sensing method according to embodiments of the present invention will be described with reference to the drawings. Note that at least some of the embodiments described below can be arbitrarily combined. FIG. 1 is a block diagram illustrating an example of the configuration of the mobile body sensing device 100 according to the present embodiment, and FIG. 2 is an explanatory diagram illustrating an example of an installation state of the mobile body sensing device 100 according to the present embodiment. As shown in FIG. 1, the moving body sensing device 100 includes a control unit 11 that controls the entire device, an interface unit 12, a feature point extraction unit 13, a determination unit 14, a specification unit 15, a storage unit 16, a speed calculation unit 17, A pulse width determination unit 18, a pulse interval calculation unit 19, a pulse signal generation unit 20, an output unit 21, and the like are provided. A video camera 200 as an acquisition unit is connected to the moving body sensing device 100.
　図２に示すように、ビデオカメラ２００は、道路を視野として、所定の高さ、レンズの光軸方向（例えば、俯角及び回転角）などの撮像条件が設定された状態で道路付近の所要の地点に設置してある。ビデオカメラ２００は、感知ライン（感知領域とも称する）を含む道路上の計測エリア内に存在する車両を撮像することができる。感知ラインは、車両（移動体とも称する）の存在感知のため車両の走行方向に沿って所要の感知長を有する領域である。ビデオカメラ２００は、撮像して得られた画像データ（データとも称する）を移動体感知装置１００へ送出する。 As shown in FIG. 2, the video camera 200 has a predetermined area near the road in a state where imaging conditions such as a predetermined height and a lens optical axis direction (for example, depression angle and rotation angle) are set with the road as a field of view. It is installed at the point. The video camera 200 can image a vehicle existing in a measurement area on a road including a sensing line (also referred to as a sensing area). The sensing line is an area having a required sensing length along the traveling direction of the vehicle for sensing the presence of the vehicle (also referred to as a moving body). The video camera 200 sends image data (also referred to as data) obtained by imaging to the moving body sensing device 100.
　移動体感知装置１００は、ビデオカメラ２００で撮像した撮像画像を用いて計測エリア内の車両又は複数の車両が纏まって群をなした車群（移動体群とも称する）を感知し、１台毎に車両を分離したのと同等の感知パルスを生成して出力する。なお、本実施の形態では、車両には、四輪車だけでなく二輪車も含むものとする。また、本実施の形態では、移動体感知装置１００とビデオカメラ２００とは別個の装置として構成されているが、これに限定されるものではなく、移動体感知装置１００は、ビデオカメラ２００と一体をなす構成であってもよい。 The moving body sensing device 100 senses a vehicle in a measurement area or a group of vehicles (also referred to as a moving body group) in a measurement area using a captured image captured by the video camera 200 and detects each vehicle. It generates and outputs a sensing pulse equivalent to separating the vehicle. In the present embodiment, the vehicle includes not only a four-wheeled vehicle but also a two-wheeled vehicle. In the present embodiment, the moving body sensing device 100 and the video camera 200 are configured as separate devices. However, the present invention is not limited to this, and the moving body sensing device 100 is integrated with the video camera 200. The structure which makes | forms may be sufficient.
　インタフェース部１２は、ビデオカメラ２００が送出した画像データを取得し、記憶部１６に記憶する。画像データは、ビデオカメラ２００のフレームレート（撮像時点の間隔、例えば、１秒間に３０フレーム）と同期して、１フレーム単位毎に記憶部１６に記憶される。 The interface unit 12 acquires the image data sent from the video camera 200 and stores it in the storage unit 16. The image data is stored in the storage unit 16 on a frame-by-frame basis in synchronization with the frame rate of the video camera 200 (interval at the time of imaging, for example, 30 frames per second).
　図３は撮像画像の一例を示す説明図である。図３の例では、ビデオカメラ２００を道路面から所要の高さに設けてあり、ビデオカメラ２００は、２車線の道路を走行する車両又は車群の前方から撮像する。撮像画像には、感知ラインを含む計測エリア全体が撮像されている。なお、図３において、後述のように、各車線での車両の進行方向を設定することもできる。 FIG. 3 is an explanatory diagram showing an example of a captured image. In the example of FIG. 3, the video camera 200 is provided at a required height from the road surface, and the video camera 200 captures an image from the front of a vehicle or a group of vehicles traveling on a two-lane road. In the captured image, the entire measurement area including the sensing line is captured. In FIG. 3, the traveling direction of the vehicle in each lane can be set as will be described later.
　まず、車群の速度の算出方法について説明する。図４は道路を走行する車群の一例を示す模式図であり、図５は車群の速度の算出方法の一例を示す説明図である。図４、図５の例では、車群として、複数の二輪車が纏まって群をなして走行している場合について説明する。 First, a method for calculating the speed of the vehicle group will be described. FIG. 4 is a schematic diagram illustrating an example of a vehicle group traveling on a road, and FIG. 5 is an explanatory diagram illustrating an example of a method of calculating the speed of the vehicle group. In the example of FIGS. 4 and 5, a case will be described in which a plurality of two-wheeled vehicles are traveling in a group as a vehicle group.
　図４に示すように、道路面を二輪車それぞれが走行し、時刻ｔ－１から時刻ｔまでの間に各二輪車がある距離だけ走行したとする。当該距離は各二輪車の移動量である。そして、時刻ｔ－１において、ビデオカメラ２００で撮像して得られた撮像画像に基づいて複数の特徴点（図５の例では、便宜上Ｐ１～Ｐ１０とする）を抽出したとする。特徴点の抽出方法は後述する。時刻ｔ－１において抽出した各特徴点を追跡し、時刻ｔにおいて当該特徴点Ｐ１～Ｐ１０を特定したとすると、時刻ｔ－１から時刻ｔまでの間の特徴点Ｐ１～Ｐ１０の移動量を求めることができる。そして、図５に示すように、時刻ｔ－１から時刻ｔまでの間の特徴点Ｐ１～Ｐ１０の移動距離及び移動方向がわかると、特徴点Ｐ１～Ｐ１０の移動ベクトルを求めることができ、特徴点Ｐ１～Ｐ１０それぞれの移動ベクトルから車群（図５の例では二輪車群）の速度（例えば、各二輪車の速度の平均速度）を求めることができる。移動ベクトルは、例えば、オプティカルフローなどの手法を用いることにより求めることができる。 As shown in FIG. 4, it is assumed that each motorcycle travels on the road surface and each motorcycle travels a certain distance between time t-1 and time t. The distance is the amount of movement of each motorcycle. Then, it is assumed that a plurality of feature points (P1 to P10 in the example of FIG. 5 are extracted for convenience) based on a captured image obtained by capturing with the video camera 200 at time t−1. A feature point extraction method will be described later. If each feature point extracted at time t-1 is tracked and the feature points P1 to P10 are specified at time t, the amount of movement of the feature points P1 to P10 from time t-1 to time t is obtained. be able to. As shown in FIG. 5, when the movement distances and movement directions of the feature points P1 to P10 from time t-1 to time t are known, the movement vectors of the feature points P1 to P10 can be obtained. The speed (for example, the average speed of the speeds of the two-wheeled vehicles) of the vehicle group (two-wheeled vehicle group in the example of FIG. 5) can be obtained from the movement vectors at the points P1 to P10. The movement vector can be obtained by using a technique such as an optical flow, for example.
　次に特徴点の抽出方法について説明する。図６は計測エリアを含む道路面の一例を示す模式図である。図６の例では、片側２車線の道路を提示しているが、車線の数は図６の例に限定されるものではない。図６に示すように、感知ラインを含む計測エリアを格子状（メッシュ状）に区切り、複数の路面領域に分割する。路面領域の大きさは、道路面上で、長さ×幅が、例えば、５０ｃｍ×５０ｃｍであるが、路面領域の大きさはこれに限定されるものではない。また、計測ラインの長さは、例えば、３０ｍ～６０ｍ程度であり、１車線当たりの道路幅は、例えば、３ｍ～３．５ｍ程度であるが、図６の例では、道路の走行方向及び道路幅方向の路面領域の数は便宜上少なく例示している。また、図６において、例えば、制御部１１は、車線毎に車両の進行方向（道路の上り方向又は下り方向であり、走行方向とも称する）を設定することができる。また、図６の例では、片側の車線のみ例示しているが、反対車線についても同様である。 Next, the feature point extraction method will be described. FIG. 6 is a schematic diagram showing an example of a road surface including a measurement area. In the example of FIG. 6, a two-lane road is presented, but the number of lanes is not limited to the example of FIG. 6. As shown in FIG. 6, the measurement area including the sensing lines is divided into a grid (mesh) and divided into a plurality of road surface areas. As for the size of the road surface area, length × width is, for example, 50 cm × 50 cm on the road surface, but the size of the road surface area is not limited to this. In addition, the length of the measurement line is, for example, about 30 m to 60 m, and the road width per lane is, for example, about 3 m to 3.5 m. In the example of FIG. The number of road surface regions in the width direction is small for convenience. In FIG. 6, for example, the control unit 11 can set the traveling direction of the vehicle (upward or downward direction of the road, also referred to as the traveling direction) for each lane. In the example of FIG. 6, only one lane is illustrated, but the same applies to the opposite lane.
　図７は道路面上の一の路面領域に対応する小領域ΔＳの撮像画像の一例を示す模式図である。図７の例では、図６に示す路面領域（ΔＳ）を、縦横１０×１０の１００個の画素ブロックに分けている。この場合、路面領域の大きさが、５０ｃｍ×５０ｃｍであるとすると、図７の１つの画素ブロックは、路面上の５ｃｍ×５ｃｍの領域を撮像して得られた複数の画素の集まりということができる。なお、画素ブロックの数は図７の例に限定されるものではない。 FIG. 7 is a schematic diagram showing an example of a captured image of the small area ΔS corresponding to one road surface area on the road surface. In the example of FIG. 7, the road surface area (ΔS) shown in FIG. 6 is divided into 100 pixel blocks of 10 × 10 in length and width. In this case, if the size of the road surface area is 50 cm × 50 cm, one pixel block in FIG. 7 is a collection of a plurality of pixels obtained by imaging a 5 cm × 5 cm area on the road surface. it can. The number of pixel blocks is not limited to the example of FIG.
　図８は画素ブロックで抽出された特徴点の移動ベクトルの一例を示す模式図である。特徴点抽出部１３は、ビデオカメラ２００で車両を撮像（感知）して得られた画像データ（感知データ）に基づいて、計測エリア内の各路面領域に対応する小領域ΔＳを構成する各画素ブロックにおいて特徴点を抽出する。すなわち、前述のように、５０ｃｍ×５０ｃｍ程度の矩形状の路面領域に対応する小領域ΔＳにおいて、特徴点を複数抽出する。この場合、図８に示すように、小領域ΔＳをさらに複数の画素ブロックに分割し、画素ブロック毎に特徴点を抽出することができる。特徴点は、例えば、エッジ抽出（空間微分）、周辺増分抽出、背景差分抽出などの手法により抽出することができる。 FIG. 8 is a schematic diagram illustrating an example of a movement vector of feature points extracted in a pixel block. Based on image data (sensed data) obtained by imaging (sensing) the vehicle with the video camera 200, the feature point extracting unit 13 configures each pixel constituting the small area ΔS corresponding to each road surface area in the measurement area. Extract feature points in the block. That is, as described above, a plurality of feature points are extracted in the small area ΔS corresponding to the rectangular road surface area of about 50 cm × 50 cm. In this case, as shown in FIG. 8, the small region ΔS can be further divided into a plurality of pixel blocks, and feature points can be extracted for each pixel block. The feature points can be extracted by techniques such as edge extraction (spatial differentiation), peripheral incremental extraction, and background difference extraction.
　また、特徴点抽出部１３は、ベクトル算出部としての機能を有し、抽出した各特徴点の移動ベクトルを算出する。移動ベクトルは、図５の例で説明したように、抽出した各特徴点を時間経過とともに追跡し、異なる時点間で移動した特徴点の移動距離及び移動方向により求めることができる。図８の例では、各画素ブロック内の点が特徴点を表し、矢印が移動ベクトルを表す。なお、画素ブロックの中で空白のものは特徴点を抽出することができなかった画素ブロックを表す。 Further, the feature point extraction unit 13 has a function as a vector calculation unit, and calculates a movement vector of each extracted feature point. As described in the example of FIG. 5, the movement vector can be obtained by tracking each extracted feature point with time and by the movement distance and movement direction of the feature point moved between different time points. In the example of FIG. 8, points in each pixel block represent feature points, and arrows represent movement vectors. Note that a blank pixel block represents a pixel block for which a feature point could not be extracted.
　また、特徴点抽出部１３は、距離算出部としての機能を有し、抽出した特徴点同士の離隔距離を算出する。 Further, the feature point extraction unit 13 has a function as a distance calculation unit, and calculates a separation distance between the extracted feature points.
　判定部１４は、ビデオカメラ２００で車両を撮像して得られた画像データに基づいて、道路面上の所定の複数の路面領域それぞれが車両の存在を示す車両候補領域（移動体候補領域とも称する）であるか否かを判定する。すなわち、判定部１４は、計測エリア内の各路面領域に対応する小領域ΔＳに車両が存在しているか否かを判定する。 The determination unit 14 is a vehicle candidate area (also referred to as a moving object candidate area) in which each of a plurality of predetermined road surface areas on the road surface indicates the presence of the vehicle based on image data obtained by imaging the vehicle with the video camera 200. ). That is, the determination unit 14 determines whether a vehicle is present in the small area ΔS corresponding to each road surface area in the measurement area.
　より具体的には、判定部１４は、特徴点抽出部１３で算出された特徴点間の距離が所定の距離閾値より短く、特徴点抽出部１３で算出された移動ベクトルが所定範囲内にある特徴点の数が所定の閾値より大きい場合、小領域ΔＳ（対応する路面領域）が車両候補領域であると判定する。 More specifically, the determination unit 14 determines that the distance between the feature points calculated by the feature point extraction unit 13 is shorter than a predetermined distance threshold, and the movement vector calculated by the feature point extraction unit 13 is within a predetermined range. When the number of feature points is larger than a predetermined threshold, it is determined that the small area ΔS (corresponding road surface area) is a vehicle candidate area.
　すなわち、抽出した特徴点のうち、特徴点同士の距離が一定以内（距離閾値より短い）であって、特徴点の移動ベクトルが近似している（移動方向と移動距離が所定範囲内）という条件を充足する特徴点をグループ化（例えば、ラベリング）し、グループ化した特徴点の数が所定の閾値より大きい場合、例えば、小領域ΔＳ内の画素ブロックの総数に対する当該条件を充足する特徴点が抽出された画素ブロックの割合が閾値より大きい場合、判定部１４は、当該小領域ΔＳを車両候補領域であると判定する。比較的小さい路面領域に対応する小領域ΔＳ毎に車両が存在しているか否かを判定するので、例えば、二輪車を含む多くの車両が群をなして走行している場合でも、車群全体を精度良く検出することができる。 That is, among the extracted feature points, the distance between the feature points is within a certain range (shorter than the distance threshold), and the movement vector of the feature points is approximate (the movement direction and the movement distance are within a predetermined range). When the feature points satisfying the above are grouped (for example, labeled) and the number of grouped feature points is larger than a predetermined threshold, for example, the feature points satisfying the condition for the total number of pixel blocks in the small region ΔS are When the ratio of the extracted pixel block is larger than the threshold value, the determination unit 14 determines that the small area ΔS is a vehicle candidate area. Since it is determined whether or not a vehicle exists for each small area ΔS corresponding to a relatively small road surface area, for example, even when many vehicles including two-wheeled vehicles are traveling in groups, It can be detected with high accuracy.
　図９は小領域ΔＳが車両候補領域であるか否かを判定する方法の一例を示す模式図である。図９の例では、小領域ΔＳ内の画素ブロックの総数は１００個である。そして、特徴点が抽出された画素ブロックの数は８１個である。この８１個の画素ブロックそれぞれの特徴点のうち、図９に例示するように、特徴点Ｐ１１、Ｐ１２、Ｐ１３、Ｐ１４、Ｐ１５、Ｐ１６の６個の特徴点の移動ベクトルは近似していないので除外する。また、特徴点Ｐ１７は他の特徴点から離れているので、特徴点Ｐ１７も除外する。そうすると、前述の条件を充足する特徴点の数は、７４（＝８１－６－１）個となる。ここで、前述の閾値を、例えば、７０％とすると、小領域ΔＳ内の画素ブロックの総数に対する当該条件を充足する特徴点が抽出された画素ブロックの割合が７０％より大きいので、小領域ΔＳは車両候補領域であると判定することができる。同様の判定処理を計測エリア内のすべての路面領域（小領域ΔＳ）について行うことにより、計測エリアを格子状に分けた比較的小さな路面領域毎に車両の存在を判定することができる。 FIG. 9 is a schematic diagram showing an example of a method for determining whether or not the small region ΔS is a vehicle candidate region. In the example of FIG. 9, the total number of pixel blocks in the small region ΔS is 100. The number of pixel blocks from which feature points are extracted is 81. Among the feature points of each of the 81 pixel blocks, as illustrated in FIG. 9, the movement vectors of the six feature points of feature points P11, P12, P13, P14, P15, and P16 are not approximated and thus excluded. To do. Since the feature point P17 is far from other feature points, the feature point P17 is also excluded. Then, the number of feature points that satisfy the above-described condition is 74 (= 81-6-1). Here, if the above-mentioned threshold is 70%, for example, the ratio of pixel blocks from which feature points that satisfy the condition to the total number of pixel blocks in the small area ΔS are extracted is larger than 70%. Can be determined to be a vehicle candidate area. By performing the same determination process for all road surface areas (small areas ΔS) in the measurement area, it is possible to determine the presence of the vehicle for each relatively small road surface area in which the measurement area is divided into a grid.
　なお、図９に例示した方法に加えて、あるいは当該方法に代えて、以下の方法を用いることもできる。すなわち、予め計測エリアの各車線の進行方向（例えば、道路の上り方向、下り方向など）を設定しておき、特徴点抽出部１３は、算出した移動ベクトルのうち、移動ベクトルの方向が設定した進行方向と同一又は近似（例えば、移動ベクトルと進行方向のとのなす角度が、１０°程度など）する移動ベクトルだけを抽出するようにしてもよい。 In addition to the method illustrated in FIG. 9 or in place of the method, the following method can also be used. That is, the traveling direction of each lane in the measurement area (for example, the upward and downward directions of the road) is set in advance, and the feature point extraction unit 13 sets the direction of the movement vector among the calculated movement vectors. Only a movement vector that is the same as or approximate to the traveling direction (for example, an angle between the moving vector and the traveling direction is about 10 °) may be extracted.
　図１０は車両候補領域で画定される車群の一例を示す模式図である。図１０は、図６において車群を画定した場合を示す。図１０に示すように、車群は、計測エリア内のすべての路面領域のうち、車両候補領域と判定された複数の路面領域を１つの大領域として纏めた領域で画定（構成、特定）される。図１０の例では、模様を付した路面領域の集合が車群を表している。 FIG. 10 is a schematic diagram showing an example of a vehicle group defined by vehicle candidate areas. FIG. 10 shows a case where the vehicle group is defined in FIG. As shown in FIG. 10, the vehicle group is defined (configured and specified) by combining a plurality of road surface areas determined as vehicle candidate areas as one large area among all road surface areas in the measurement area. The In the example of FIG. 10, a set of road surface areas with a pattern represents a vehicle group.
　次に、図１０に示したような車群が感知ラインに到達する到達時点、及び車群が感知ラインから離脱する離脱時点の特定方法について説明する。特定部１５は、判定部１４で車両候補領域であると判定された複数の領域（路面領域、小領域ΔＳ）で画定される車両又は車群が、道路面の所定の感知ラインに到達する到達時点及び感知ラインから離脱する離脱時点を特定する。すなわち、特定部１５は、車両又は車群の先頭が感知ラインに到達する到達時点を特定するとともに、車両又は車群の末尾が感知ラインから離脱する離脱時点を特定する。 Next, a description will be given of a method of identifying the arrival time point when the vehicle group reaches the sensing line as shown in FIG. 10 and the departure time point when the vehicle group leaves the sensing line. The specifying unit 15 arrives when a vehicle or a vehicle group defined by a plurality of regions (road surface region, small region ΔS) determined to be vehicle candidate regions by the determining unit 14 reaches a predetermined sensing line on the road surface. Identify the time of departure and the time of departure from the sensing line. That is, the specifying unit 15 specifies the arrival time when the head of the vehicle or the vehicle group reaches the sensing line, and specifies the departure time when the end of the vehicle or the vehicle group leaves the sensing line.
　図１１は第２車線に存在する車群が感知ラインに到達した時点の一例を示す模式図であり、図１２は第１車線に存在する車群が感知ラインに到達した時点の一例を示す模式図である。特定部１５は、車線毎に感知ラインでの車両又は車群の道路幅方向の寸法が所定の幅閾値より大きくなった時点を到達時点として特定する。所定の幅閾値は、例えば、車両（四輪車）の幅程度とすることができる。二輪車を含む車両が群を構成して走行している場合、道路幅方向の車群の幅は、１台の車両の幅よりも大きい（広い）と考えられる。そこで、画定された車両又は車群の道路幅方向の寸法が所定の幅閾値より大きくなった時点は、車群の先頭が感知ラインに到達する時点と考えられるので、当該時点を到達時点として特定することができる。 FIG. 11 is a schematic diagram illustrating an example of a time point when a vehicle group existing in the second lane reaches the sensing line, and FIG. 12 is a schematic diagram illustrating an example of a time point when the vehicle group existing in the first lane reaches the sensing line. FIG. The specifying unit 15 specifies the time when the dimension of the vehicle or the vehicle group in the road width direction on the sensing line becomes larger than a predetermined width threshold for each lane as the arrival time. The predetermined width threshold can be, for example, about the width of the vehicle (four-wheeled vehicle). When vehicles including two-wheeled vehicles are traveling in a group, the width of the vehicle group in the road width direction is considered to be larger (wider) than the width of one vehicle. Therefore, the time when the dimension of the defined vehicle or vehicle group in the road width direction becomes larger than the predetermined width threshold is considered as the time when the head of the vehicle group reaches the sensing line, so that time is specified as the arrival time. can do.
　図１１の例では、第２車線を走行する車群の先頭が感知ラインに到達した到達時点（後述の第２車線の感知パルスの出力開始時点）ＴＳ２を特定している。また、図１２の例では、第１車線を走行する車群の先頭が感知ラインに到達した到達時点（後述の第１車線の感知パルスの出力開始時点）ＴＳ１を特定している。なお、感知ラインでの車両又は車群の道路幅方向の寸法は、連続する長さでもよく、道路幅方向の寸法が連続していない（離散している）場合には、累積した長さでもよい。 In the example of FIG. 11, the arrival time point TS2 when the head of the vehicle group traveling in the second lane reaches the detection line (the output start point of the second lane detection pulse described later) TS2 is specified. In the example of FIG. 12, the arrival time point TS1 when the head of the vehicle group traveling in the first lane reaches the detection line (the output start point of a detection pulse of the first lane described later) TS1 is specified. In addition, the length in the road width direction of the vehicle or the vehicle group in the sensing line may be a continuous length, or if the dimension in the road width direction is not continuous (discrete), it may be an accumulated length. Good.
　図１３は車群が感知ラインから離脱した時点の一例を示す模式図である。特定部１５は、感知ラインでの車両又は車群の道路幅方向の寸法が所定の幅閾値より小さくなった時点を離脱時点として特定する。所定の幅閾値は、例えば、車両（四輪車）の幅程度とすることができる。二輪車を含む車両が群を構成して走行している場合、道路幅方向の車群の幅は、１台の車両の幅よりも大きい（広い）と考えられる。そこで、画定された車両又は車群の道路幅方向の寸法が所定の幅閾値より小さくなった時点は、車群の末尾が感知ラインから離脱する時点と考えられるので、当該時点を離脱時点として特定することができる。 FIG. 13 is a schematic diagram showing an example of when the vehicle group leaves the sensing line. The specifying unit 15 specifies the time when the dimension of the vehicle or the vehicle group in the road width direction on the sensing line is smaller than a predetermined width threshold as the departure time. The predetermined width threshold can be, for example, about the width of the vehicle (four-wheeled vehicle). When vehicles including two-wheeled vehicles are traveling in a group, the width of the vehicle group in the road width direction is considered to be larger (wider) than the width of one vehicle. Therefore, the time when the dimension of the defined vehicle or vehicle group in the road width direction becomes smaller than the predetermined width threshold is considered as the time when the end of the vehicle group leaves the sensing line. can do.
　図１３の例では、第１車線及び第２車線を走行する車群の末尾が感知ラインから離脱した到達時点（後述の感知パルスの出力停止時点）ＴＥを特定している。図１３の例では、車群が感知ラインから離脱する離脱時点が、第１車線及び第２車線で同じタイミングであるが、異なるタイミングとなる場合もある。また、到達時点を特定する幅閾値と離脱時点を特定する閾値とは同じ値でもよく、異なる値でもよい。 In the example of FIG. 13, the arrival time point TE at which the tail of the vehicle group traveling in the first lane and the second lane leaves the detection line (the detection pulse output stop point described later) TE is specified. In the example of FIG. 13, the departure time when the vehicle group leaves the sensing line is the same timing in the first lane and the second lane, but may be different timing. Also, the width threshold value for specifying the arrival time point and the threshold value for specifying the departure time point may be the same value or different values.
　パルス信号生成部２０は、生成部としての機能を有する。パルス信号生成部２０は、特定部１５で特定した到達時点から離脱時点の間、所要幅の感知パルス（パルスとも称する）を所要間隔で生成する。所要幅は、感知パルスのパルス幅であり、感知パルスがオン状態である時間幅である。また、所要間隔は、感知パルスの周期であり、感知パルスがオンとなった時点から、所要幅の後オフとなり、次の感知パルスがオンとなる時点までの時間である。 The pulse signal generation unit 20 has a function as a generation unit. The pulse signal generation unit 20 generates a sensing pulse (also referred to as a pulse) having a required width at a required interval from the arrival time specified by the specifying unit 15 to the departure time. The required width is the pulse width of the sensing pulse, and is the time width during which the sensing pulse is in the on state. The required interval is the period of the sensing pulse, and is the time from when the sensing pulse is turned on until the next sensing pulse is turned on after the required width is turned off.
　図１４は本実施の形態の移動体感知装置１００が生成する感知パルス信号の一例を示すタイムチャートである。パルス信号生成部２０は、車群の先頭が感知ラインに到達した到達時点ＴＳ２において、第２車線の感知パルスの生成を開始する。パルス信号生成部２０は、車群の先頭が感知ラインに到達した到達時点ＴＳ１において、第１車線の感知パルスの生成を開始する。感知パルスは、いずれも所要幅がＴｗ、所要間隔がＴｄである。また、パルス信号生成部２０は、車群の末尾が感知ラインから離脱した離脱時点ＴＥにおいて、感知パルスの生成を停止する。なお、図１４に示すパルス信号の所要幅Ｔｗ、所要間隔Ｔｄは、簡便のため同一にしてあるが、実際には、パルス毎に所要幅Ｔｗ及び所要間隔Ｔｄは変わり得る。なお、パルス間隔算出部１９で所要間隔Ｔｄを算出する代わりに、所要間隔として時間差（Ｔｄ－Ｔｗ）を算出するようにしてもよい。 FIG. 14 is a time chart showing an example of a sensing pulse signal generated by the moving body sensing device 100 of the present embodiment. The pulse signal generation unit 20 starts generating the second lane sensing pulse at the arrival time TS2 when the head of the vehicle group reaches the sensing line. The pulse signal generation unit 20 starts generating the first lane sensing pulse at the arrival time TS1 when the head of the vehicle group reaches the sensing line. Each sensing pulse has a required width of Tw and a required interval of Td. Further, the pulse signal generation unit 20 stops generating the detection pulse at the departure time TE when the end of the vehicle group leaves the detection line. Note that the required width Tw and the required interval Td of the pulse signal shown in FIG. 14 are the same for the sake of simplicity, but in practice, the required width Tw and the required interval Td can vary from pulse to pulse. Instead of calculating the required interval Td by the pulse interval calculation unit 19, a time difference (Td−Tw) may be calculated as the required interval.
　出力部２１は、パルス信号生成部２０で生成した感知パルスを、交通信号制御装置、情報提供装置などの外部の装置へ出力する。車両又は車群が感知ラインに到達してから離脱するまでの間、感知パルスを所要間隔で繰り返し生成し、生成した感知パルスを出力することにより、二輪車を含む多数の車両が感知ラインを通過して走行する場合、個々の車両を分離して感知（計測）することができる。 The output unit 21 outputs the sensing pulse generated by the pulse signal generation unit 20 to an external device such as a traffic signal control device or an information providing device. A number of vehicles including two-wheeled vehicles pass through the sensing line by repeatedly generating the sensing pulse at a required interval from when the vehicle or a group of vehicles arrives at the sensing line until it leaves the vehicle, and outputs the generated sensing pulse. When traveling, it is possible to sense (measure) individual vehicles separately.
　速度算出部１７は、画定された車両又は車群の速度を算出する。車両又は車群の速度は、例えば、図５で例示したような方法で求めることができる。また、車両候補領域であると判定された任意の小領域の異なる時点間における移動距離により小領域の速度を求めてもよく、車両候補領域であると判定された複数の小領域それぞれの速度の平均により求めてもよく、あるいは、画定した車両又は車群の異なる時点間における移動距離により速度を求めてもよい。また、感知ラインから所要の距離（例えば、１０ｍ～１５ｍなど）以内の車群の平均速度を用い、感知ラインから遠方にある車両候補領域は除外するようにしてもよい。これにより、車群の長さが比較的長い場合でも、車群の速度を精度良く求めることができる。 The speed calculation unit 17 calculates the speed of the defined vehicle or vehicle group. The speed of the vehicle or the vehicle group can be obtained by, for example, the method illustrated in FIG. In addition, the speed of the small area may be obtained from the movement distance between different time points of any small area determined as the vehicle candidate area, and the speed of each of the plurality of small areas determined as the vehicle candidate area You may obtain | require by an average, or you may obtain | require a speed by the movement distance between the different time points of the defined vehicle or vehicle group. Further, the vehicle candidate area far from the sensing line may be excluded by using the average speed of the vehicle group within a required distance (for example, 10 m to 15 m) from the sensing line. Thereby, even when the length of the vehicle group is relatively long, the speed of the vehicle group can be obtained with high accuracy.
　また、車群の速度は、以下のようにして求めることができる。すなわち、特徴点抽出部１３で抽出した特徴点の数と、実際の車群密度（単位長さあたりの車両の存在台数）との関係を予め求めておき、速度と密度の関係式（速度・密度曲線、Ｋ－Ｖ曲線とも称する）を用いて車群の速度を求めることもできる。 Also, the speed of the vehicle group can be obtained as follows. That is, a relationship between the number of feature points extracted by the feature point extraction unit 13 and the actual vehicle group density (the number of vehicles per unit length) is obtained in advance, and a relational expression (speed / density) The speed of the vehicle group can also be obtained using a density curve or KV curve).
　図１５は速度と密度との対応関係の一例を示す説明図である。図１５において、横軸は密度を示し、縦軸は速度を示す。図１５に示す曲線をＫ－Ｖ曲線と称する。密度は、単位長さ当たりに存在する車両の台数である。速度と密度との間には負の相関関係があることが経験的に知られている。例えば、速度が大きく（速く）なると運転者は安全性を保つため車間距離を大きくとる性向がある。したがって、速度が大きくなるほど密度は小さくなり、図１５に例示するような曲線となる。また、図１５において、速度Ｖｃ、密度Ｋｃは、交通量が最大となる臨界点での速度（臨界速度）及び密度（臨界密度）である。密度が臨界密度Ｋｃより小さい領域は、自然流領域であり、密度が臨界密度Ｋｃより大きい領域は、渋滞流領域である。 FIG. 15 is an explanatory diagram showing an example of the correspondence between speed and density. In FIG. 15, the horizontal axis indicates the density, and the vertical axis indicates the speed. The curve shown in FIG. 15 is referred to as a KV curve. Density is the number of vehicles present per unit length. It is empirically known that there is a negative correlation between speed and density. For example, when the speed increases (fast), the driver tends to increase the inter-vehicle distance in order to maintain safety. Therefore, as the speed increases, the density decreases, resulting in a curve as illustrated in FIG. In FIG. 15, speed Vc and density Kc are speed (critical speed) and density (critical density) at a critical point where the traffic volume is maximum. A region where the density is smaller than the critical density Kc is a natural flow region, and a region where the density is larger than the critical density Kc is a traffic flow region.
　パルス幅決定部１８は、決定部としての機能を有し、感知ラインの感知長、所定の車長及び速度算出部１７で算出した速度に基づいて、感知パルスの所要幅を決定する。 The pulse width determining unit 18 has a function as a determining unit, and determines the required width of the sensing pulse based on the sensing length of the sensing line, the predetermined vehicle length, and the speed calculated by the speed calculating unit 17.
　図１６は感知パルスの所要幅の決定方法の一例を示す模式図である。感知ラインの感知長をＷ、車長をＬ及び車両又は車群の速度をＶとすると、感知パルスの所要幅Ｔｗは、Ｔｗ＝（Ｗ＋Ｌ）／Ｖの式で求めることができる。すなわち、感知パルスの所要幅Ｔｗは、車両の先頭が感知ラインに到達する時点から、車両の末尾が感知ラインから離脱する時点までの車両の移動距離（Ｗ＋Ｌ）を速度Ｖで除算することにより求めることができる。これにより、車両が纏まって車群を構成して走行している場合でも、１台毎に車両を分離したのと同等の感知パルスを生成して出力することができる。 FIG. 16 is a schematic diagram showing an example of a method for determining the required width of the sensing pulse. If the sensing length of the sensing line is W, the vehicle length is L, and the speed of the vehicle or vehicle group is V, the required width Tw of the sensing pulse can be obtained by the equation Tw = (W + L) / V. That is, the required width Tw of the sensing pulse is obtained by dividing the moving distance (W + L) of the vehicle from the time when the head of the vehicle reaches the sensing line to the time when the end of the vehicle leaves the sensing line by the speed V. be able to. As a result, even when the vehicles are traveling together in a vehicle group, it is possible to generate and output a sensing pulse equivalent to separating the vehicles for each vehicle.
　パルス間隔算出部１９は、間隔算出部としての機能を有し、速度算出部１７で算出した速度に基づいて、感知パルスの所要間隔を算出する。これにより、車両又は車群の速度にかかわらず、速度に応じた適切な所要間隔で感知パルスを生成することができる。 The pulse interval calculation unit 19 has a function as an interval calculation unit, and calculates a required interval of the sensing pulse based on the speed calculated by the speed calculation unit 17. Thereby, regardless of the speed of the vehicle or the vehicle group, the sensing pulse can be generated at an appropriate required interval according to the speed.
　より具体的には、パルス間隔算出部１９は、交通量算出部としての機能も有し、車両又は車群の速度と交通量との対応関係及び速度算出部１７で算出した速度に基づいて交通量を算出する。 More specifically, the pulse interval calculation unit 19 also has a function as a traffic volume calculation unit, and is based on the correspondence between the speed of the vehicle or the vehicle group and the traffic volume and the speed calculated by the speed calculation unit 17. Calculate the amount.
　図１７は交通量と平均速度との対応関係の一例を示す説明図である。図１７に示すように、交通量と平均速度との関係は、例えば、交通量が最大となる臨界点での臨界速度Ｖｃ及び臨界速度Ｖｃでの最大交通量Ｑｍａｘを含む２次関数で表すことができる。すなわち、平均速度が０から臨界速度Ｖｃに増加すると交通量も増加し、平均速度が臨界速度Ｖｃよりさらに増加すると交通量は減少する。平均速度と交通量との対応関係は、予め定めておき記憶しておくこともでき、あるいは平均速度と交通量との関係式を演算で求めるようにしてもよい。 FIG. 17 is an explanatory diagram showing an example of the correspondence between the traffic volume and the average speed. As shown in FIG. 17, the relationship between the traffic volume and the average speed is expressed by, for example, a quadratic function including the critical speed Vc at the critical point where the traffic volume is maximum and the maximum traffic volume Qmax at the critical speed Vc. Can do. That is, when the average speed increases from 0 to the critical speed Vc, the traffic volume also increases, and when the average speed further increases above the critical speed Vc, the traffic volume decreases. The correspondence relationship between the average speed and the traffic volume can be determined and stored in advance, or a relational expression between the average speed and the traffic volume may be obtained by calculation.
　パルス間隔算出部１９は、算出した交通量に基づいて所要間隔を算出する。例えば、算出した交通量が１分当たり３０台、すなわち６０秒当たり３０台とすると、所要間隔は、２秒（＝６０／３０）となる。これにより、交通量に応じて適切な所要間隔で感知パルスを生成することができる。なお、感知パルスの所要間隔は、車群の速度（平均速度）から交通量を求めることにより算出する構成に代えて、例えば、固定周期（飽和交通流率：例えば、２秒／台に基づいて、２秒毎に１感知パルスを生成する）を用いてもよい。 The pulse interval calculation unit 19 calculates a required interval based on the calculated traffic volume. For example, if the calculated traffic volume is 30 units per minute, that is, 30 units per 60 seconds, the required interval is 2 seconds (= 60/30). Thereby, a sensing pulse can be generated at an appropriate required interval according to the traffic volume. The required interval of the sensing pulses is based on, for example, a fixed period (saturated traffic flow rate: 2 seconds / unit, for example) instead of the configuration calculated by obtaining the traffic volume from the vehicle group speed (average speed). May generate 1 sense pulse every 2 seconds).
　前述の例において、パルス信号生成部２０、パルス幅決定部１８、パルス間隔算出部１９は、速度算出部１７で算出した速度を用いる。速度算出部１７で算出する速度は、例えば、（１）前回パルスを生成（出力）してからの平均速度、（２）最新の速度、（３）過去一定時間（例えば、５秒間など）の平均速度の如く、（１）～（３）のいずれかの方法を選択して算出することができる。以下、具体的に説明する。 In the above example, the pulse signal generation unit 20, the pulse width determination unit 18, and the pulse interval calculation unit 19 use the speed calculated by the speed calculation unit 17. The speed calculated by the speed calculation unit 17 is, for example, (1) an average speed after generating (outputting) the previous pulse, (2) the latest speed, and (3) a past fixed time (for example, 5 seconds). Like the average speed, it can be calculated by selecting one of the methods (1) to (3). This will be specifically described below.
　図１８は感知パルスの所要幅及び所要間隔を決定するタイミングの一例を示すタイムチャートである。図１８の例において、１つ目の感知パルスの所要幅Ｔｗ１及び所要間隔Ｔｄ１は、感知パルスの生成（出力）のタイミングである時刻ｔ１又は時刻ｔ１の直前に決定される。同様に、２つ目の感知パルスの所要幅Ｔｗ２及び所要間隔Ｔｄ２は、感知パルスの生成（出力）のタイミングである時刻ｔ２又は時刻ｔ２の直前に決定される。一方、速度算出部１７で速度を算出する頻度又は周期は、感知パルスの所要間隔（周期）よりも短いので（例えば、１００ｍｓの都度）、速度算出部１７は、時刻ｔ１から時刻ｔ２までの間において、速度を複数回算出することになる。 FIG. 18 is a time chart showing an example of the timing for determining the required width and required interval of the sensing pulse. In the example of FIG. 18, the required width Tw1 and required interval Td1 of the first sensing pulse are determined immediately before time t1 or time t1, which is the timing of generation (output) of the sensing pulse. Similarly, the required width Tw2 and the required interval Td2 of the second sensing pulse are determined immediately before time t2 or time t2, which is the timing of generating (outputting) the sensing pulse. On the other hand, since the frequency or period at which the speed calculation unit 17 calculates the speed is shorter than the required interval (period) of the sensing pulse (for example, every 100 ms), the speed calculation unit 17 performs the period from time t1 to time t2. In this case, the speed is calculated a plurality of times.
　前述の（１）前回パルスを生成（出力）してからの平均速度を採用する場合には、安定した値が得られるが、パルスの所要間隔が長くなると、速度算出部１７での算出処理が増大し、遅れが問題となり得る。また、（２）最新の速度を採用する場合には、遅れの問題はないが、瞬間速度であるため値が不安定になるおそれがある。現実的には、（３）過去一定時間（例えば、５秒間など）の平均速度を採用するのが望ましいが、移動体感知装置の使用条件等により適宜選択すればよい。 When the average speed from the previous (1) generation (output) of the previous pulse is adopted, a stable value can be obtained. However, if the required interval between pulses becomes longer, the calculation process in the speed calculation unit 17 is performed. Increases and delays can be a problem. (2) When the latest speed is adopted, there is no problem of delay, but the value may become unstable because of the instantaneous speed. Actually, (3) it is desirable to adopt an average speed for a certain period of time in the past (for example, 5 seconds).
　また、速度算出部１７で算出した速度（普通車に換算したときの各普通車の速度）、パルス信号生成部２０で生成したパルスの数である車両の台数を所要の周期（例えば、１分など）で集計して、交通信号制御を管理する中央装置などへ送信するようにしてもよい。 Further, the speed calculated by the speed calculation unit 17 (the speed of each normal vehicle when converted into a normal vehicle) and the number of vehicles that are the number of pulses generated by the pulse signal generation unit 20 are set to a required cycle (for example, 1 minute). Etc.) and may be transmitted to a central device that manages traffic signal control.
　図１９は感知パルスの終了タイミングの一例を示すタイムチャートである。図１９に示すように、パルス信号生成部２０は、所要幅Ｔｗの感知パルスの途中で特定部１５が離脱時点を特定した場合、当該感知パルスの後の感知パルスの生成を停止する。すなわち、任意の感知パルスがオン状態であるときに、離脱時点が特定された場合、当該感知パルスは打ち切ることなく、所要幅Ｔｗの感知パルスとしてそのまま生成して出力するが、当該感知パルスの次の感知パルスは生成しないようにする。これにより、車群の末尾が感知ラインから離脱した後は、感知パルスの生成を停止させることができる。 FIG. 19 is a time chart showing an example of the end timing of the sensing pulse. As shown in FIG. 19, when the identifying unit 15 identifies the departure time in the middle of the sensing pulse having the required width Tw, the pulse signal generating unit 20 stops generating the sensing pulse after the sensing pulse. That is, when an arbitrary sensing pulse is in the on state and the departure time is specified, the sensing pulse is generated and output as it is as a sensing pulse of the required width Tw without being aborted. No sense pulse is generated. Thereby, after the end of the vehicle group leaves the sensing line, the generation of the sensing pulse can be stopped.
　パルス信号生成部２０は、速度算出部１７で算出した速度が所定の速度閾値より小さい場合、パルスの生成を停止し、波高値が一定の信号を生成する。速度閾値は、車両又は車群が停止しているか否かを判定することができる程度の値であればよい。 When the speed calculated by the speed calculation unit 17 is smaller than a predetermined speed threshold, the pulse signal generation unit 20 stops generating pulses and generates a signal having a constant peak value. The speed threshold may be a value that can determine whether or not the vehicle or the vehicle group is stopped.
　図２０は車群が停止した場合の感知パルスの生成を停止する例を示すタイムチャートである。パルスの生成を停止し、波高値が一定の信号を生成する。波高値が一定とは、波高値が０の場合も含む。すなわち、波高値が一定の信号を生成するとは、オン状態及びオフ状態を繰り返すパルスに代えて、図２０の上段のチャートのように、波高値が０の連続信号（オフ状態の信号）を生成すること、及び図２０の下段のチャートのように、所定の波高値の連続信号（オン状態の信号）を生成することを意味する。なお、オン状態の信号を生成するか、オフ状態の信号を生成するかは、予め設定することができるようにしてもよい。 FIG. 20 is a time chart showing an example of stopping the generation of sensing pulses when the vehicle group stops. The generation of the pulse is stopped, and a signal having a constant peak value is generated. The constant peak value includes the case where the peak value is zero. That is, to generate a signal with a constant peak value, instead of a pulse that repeats the ON state and the OFF state, a continuous signal with a peak value of 0 (off-state signal) is generated as shown in the upper chart of FIG. This means that a continuous signal (on-state signal) having a predetermined peak value is generated as shown in the lower chart of FIG. Note that whether an on-state signal or an off-state signal is generated may be set in advance.
　また、算出した速度が所定の速度閾値より小さいと判定した時点と感知パルスとの関係に応じて、オン状態の信号を生成するかオフ状態の信号を生成するかを決定することもできる。例えば、図２０の上段のチャートのように、一の感知パルスと次の感知パルスとの間（感知パルスがオフ状態）で車両又は車群が停止したと判定した場合には、当該判定の時点以降は、オフ状態の信号（波高値が０の連続信号）を生成する。 It is also possible to determine whether to generate an on-state signal or an off-state signal according to the relationship between the time point when the calculated speed is determined to be smaller than the predetermined speed threshold and the sensing pulse. For example, when it is determined that the vehicle or the vehicle group has stopped between one sensing pulse and the next sensing pulse (the sensing pulse is in an off state) as shown in the upper chart of FIG. Thereafter, an off-state signal (a continuous signal having a peak value of 0) is generated.
　また、図２０の下段のチャートのように、感知パルスのパルス幅の途中（感知パルスがオン状態の途中）で車両又は車群が停止したと判定した場合には、オン状態の信号（所定の波高値の連続信号）を生成する。これにより、車両又は車群が停止した場合でも、適切な感知情報を外部の交通信号制御装置又は情報提供装置などへ提供することができる。 Also, as shown in the lower chart of FIG. 20, when it is determined that the vehicle or the vehicle group has stopped in the middle of the pulse width of the sensing pulse (in the middle of the sensing pulse), an on-state signal (predetermined value) A continuous signal of peak values). Thereby, even when the vehicle or the vehicle group stops, appropriate sensing information can be provided to an external traffic signal control device or an information providing device.
　次に、本実施の形態の移動体感知装置１００の動作について説明する。図２１は本実施の形態の移動体感知装置１００の処理手順の一例を示すフローチャートである。なお、便宜上、処理の主体を制御部１１として説明する。制御部１１は、車両又は車群を感知するためのデータ（例えば、画像データ）を取得し（Ｓ１１）、道路面の小領域毎に特徴点を複数抽出する（Ｓ１２）。道路面の小領域とは、計測エリア内の道路面を格子状に分割した複数の路面領域それぞれに対応する撮像画像上の小領域ΔＳである。 Next, the operation of the moving body sensing device 100 of the present embodiment will be described. FIG. 21 is a flowchart illustrating an example of a processing procedure of the moving body sensing device 100 according to the present embodiment. For convenience, the subject of processing will be described as the control unit 11. The control unit 11 acquires data (for example, image data) for sensing the vehicle or the vehicle group (S11), and extracts a plurality of feature points for each small area of the road surface (S12). The small area of the road surface is a small area ΔS on the captured image corresponding to each of a plurality of road surface areas obtained by dividing the road surface in the measurement area in a grid pattern.
　制御部１１は、抽出した各特徴点の移動ベクトルを算出し（Ｓ１３）、小領域ΔＳそれぞれが車両候補領域であるか否かを判定する（Ｓ１４）。制御部１１は、車両候補領域であると判定された複数の小領域ΔＳで画定される車群（車両も含む）を特定し（Ｓ１５）、車群の速度を算出する（Ｓ１６）。 The control unit 11 calculates a movement vector of each extracted feature point (S13), and determines whether each of the small areas ΔS is a vehicle candidate area (S14). The control unit 11 identifies a vehicle group (including vehicles) defined by a plurality of small regions ΔS determined to be a vehicle candidate region (S15), and calculates the speed of the vehicle group (S16).
　制御部１１は、画定した車群が感知ラインに到達した到達時点を特定し（Ｓ１７）、所要幅の感知パルスを所要間隔で生成して出力する（Ｓ１８）。制御部１１は、画定した車群が感知ラインから離脱した離脱時点を特定し（Ｓ１９）、感知パルスの生成、出力を停止し（Ｓ２０）、処理を終了する。 The control unit 11 specifies the arrival time point when the defined vehicle group reaches the sensing line (S17), and generates and outputs a sensing pulse having a required width at a required interval (S18). The control unit 11 specifies the departure time point when the defined vehicle group leaves the sensing line (S19), stops generating and outputting the sensing pulse (S20), and ends the process.
　本実施の形態の移動体感知装置１００は、ＣＰＵ、ＲＡＭなどを備えた汎用コンピュータを用いて実現することもできる。すなわち、図２１に示すような、各処理手順を定めたコンピュータプログラムをＣＤ、ＤＶＤ、ＵＳＢメモリ等のコンピュータプログラム記録媒体に記録しておき、当該コンピュータプログラムをコンピュータに備えられたＲＡＭにロードし、コンピュータプログラムをＣＰＵで実行することにより、コンピュータ上で移動体感知装置１００を実現することができる。 The moving body sensing device 100 of the present embodiment can also be realized using a general-purpose computer including a CPU, a RAM, and the like. That is, as shown in FIG. 21, a computer program defining each processing procedure is recorded on a computer program recording medium such as a CD, DVD, USB memory, etc., and the computer program is loaded into a RAM provided in the computer. By executing the computer program on the CPU, the moving body sensing device 100 can be realized on the computer.
　本実施の形態の移動体感知装置１００によれば、二輪車を含む多数の車両が車群を構成して走行する場合でも、個々の車両を分離して、車両の台数を感知する感知パルスを生成することができる。そして、本実施の形態の移動体感知装置１００が生成する感知パルスを用いることにより、渋滞を精度良く判定することができ、ギャップ感応を含む交通信号制御を適切に行うことが可能となる。 According to the moving body sensing device 100 of the present embodiment, even when a large number of vehicles including two-wheeled vehicles travel while forming a vehicle group, the individual pulses are separated to generate sensing pulses for sensing the number of vehicles. can do. And by using the detection pulse which the mobile body detection apparatus 100 of this Embodiment produces | generates, a traffic jam can be determined with a sufficient precision and it becomes possible to perform traffic signal control including a gap sensitivity appropriately.
　上述の実施の形態では、取得部としてビデオカメラを用いる構成であったが、取得部はビデオカメラに限定されるものではなく、道路面の計測エリアを格子状に分割した複数の路面領域について車両が存在するか否かを判定することができるものであれば、ミリ波センサ又はレーザセンサなどを用いることもできる。 In the above-described embodiment, the video camera is used as the acquisition unit. However, the acquisition unit is not limited to the video camera, and the vehicle is used for a plurality of road surface areas obtained by dividing the measurement area of the road surface in a grid pattern. As long as it is possible to determine whether or not there is a millimeter wave sensor, a laser sensor or the like can be used.
　本実施の形態では、車群だけでなく１台の車両（特に車長が長い大型の車両など）であっても感知し、例えば、普通車に相当する台数が感知ラインを通過したのと同等の感知パルスを生成して出力することができる。例えば、１台の大きな車両が感知ラインを通過した場合には、２台の普通車が通過した如く感知パルスを生成して出力することができる。 In the present embodiment, not only a group of vehicles but also one vehicle (especially a large vehicle with a long vehicle length) is sensed. For example, the number of vehicles equivalent to ordinary vehicles passes the sensing line. The sensing pulse can be generated and output. For example, when one large vehicle passes the sensing line, a sensing pulse can be generated and output as if two ordinary vehicles passed.
　開示された実施の形態は、すべての点で例示であって制限的なものではないと考えられるべきである。本発明の範囲は上記した説明ではなくて特許請求の範囲によって示され、特許請求の範囲と均等の意味及び範囲内でのすべての変更が含まれることが意図される。 The disclosed embodiments are to be considered in all respects as illustrative and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.
　１００　移動体感知装置
　１１　制御部
　１２　インタフェース部
　１３　特徴点抽出部
　１４　判定部
　１５　特定部
　１６　記憶部
　１７　速度算出部
　１８　パルス幅決定部
　１９　パルス間隔算出部
　２０　パルス信号生成部
　２１　出力部
　２００　ビデオカメラ DESCRIPTION OF SYMBOLS 100 Moving body detection apparatus 11 Control part 12 Interface part 13 Feature point extraction part 14 Judgment part 15 Identification part 16 Storage part 17 Speed calculation part 18 Pulse width determination part 19 Pulse interval calculation part 20 Pulse signal generation part 21 Output part 200 Video camera

Claims (11)

  1. 　移動体を感知するためのデータを取得部で取得し、取得したデータに基づいて、道路面上の所定の複数の領域それぞれが移動体の存在を示す移動体候補領域であるか否かを判定する判定部と、
    　該判定部で移動体候補領域であると判定された複数の領域で画定される移動体又は移動体群が、前記道路面の移動体を感知するための所定の感知領域に到達する到達時点及び前記感知領域から離脱する離脱時点を特定する特定部と、
    　該特定部で特定した到達時点から離脱時点の間、所要幅のパルスを所要間隔で生成する生成部と
    　を備える移動体感知装置。     Data for sensing a moving object is acquired by an acquisition unit, and whether or not each of a plurality of predetermined areas on the road surface is a moving object candidate area indicating the presence of a moving object is determined based on the acquired data A determination unit to perform,
    The arrival time point when the moving object or the moving object group defined by the plurality of areas determined as the moving object candidate area by the determination unit reaches a predetermined sensing area for sensing the moving object on the road surface, and A specifying unit for specifying a point of time of leaving from the sensing area;
    A moving body sensing device comprising: a generating unit that generates a pulse having a required width at a required interval from an arrival time specified by the specifying unit to a departure time.
  2. 　前記移動体又は移動体群の速度を算出する速度算出部と、
    　該速度算出部で算出した速度に基づいて、前記所要間隔を算出する間隔算出部と
    　を備える請求項１に記載の移動体感知装置。     A speed calculation unit for calculating the speed of the mobile body or the mobile body group;
    The mobile body sensing device according to claim 1, further comprising: an interval calculation unit that calculates the required interval based on the speed calculated by the speed calculation unit.
  3. 　移動体又は移動体群の速度と交通量との対応関係及び前記速度算出部で算出した速度に基づいて交通量を算出する交通量算出部を備え、
    　前記間隔算出部は、
    　前記交通量算出部で算出した交通量に基づいて前記所要間隔を算出するようにしてある請求項２に記載の移動体感知装置。     A traffic volume calculation unit that calculates the traffic volume based on the correspondence between the speed of the moving body or the moving body group and the traffic volume and the speed calculated by the speed calculation unit,
    The interval calculation unit
    The mobile body sensing device according to claim 2, wherein the required interval is calculated based on the traffic volume calculated by the traffic volume calculation unit.
  4. 　前記生成部は、
    　前記速度算出部で算出した速度が所定の速度閾値より小さい場合、パルスの生成を停止し、波高値が一定の信号を生成するようにしてある請求項２又は請求項３に記載の移動体感知装置。     The generator is
    The moving body detection according to claim 2 or 3, wherein when the speed calculated by the speed calculation unit is smaller than a predetermined speed threshold, generation of a pulse is stopped and a signal having a constant peak value is generated. apparatus.
  5. 　前記感知領域の感知長、所定の移動体長及び前記速度算出部で算出した速度に基づいて、前記パルスの所要幅を決定する決定部を備える請求項２から請求項４までのいずれか１項に記載の移動体感知装置。 5. The apparatus according to claim 2, further comprising a determination unit that determines a required width of the pulse based on a detection length of the detection region, a predetermined moving body length, and a speed calculated by the speed calculation unit. The moving body sensing device described.
  6. 　前記特定部は、
    　前記感知領域での前記移動体又は移動体群の道路幅方向の寸法が所定の幅閾値より小さくなった時点を前記離脱時点として特定するようにしてある請求項１から請求項５までのいずれか１項に記載の移動体感知装置。     The specific part is:
    The time point when the dimension in the road width direction of the moving body or the moving body group in the sensing area becomes smaller than a predetermined width threshold value is specified as the leaving time point. 2. A moving body sensing device according to item 1.
  7. 　前記生成部は、
    　前記所要幅のパルスの途中で前記特定部が離脱時点を特定した場合、前記パルスの後のパルスの生成を停止するようにしてある請求項１から請求項６までのいずれか１項に記載の移動体感知装置。     The generator is
    The generation of a pulse after the pulse is stopped when the specifying unit specifies a departure time in the middle of the pulse having the required width. Moving body sensing device.
  8. 　前記特定部は、
    　前記感知領域での前記移動体又は移動体群の道路幅方向の寸法が所定の幅閾値より大きくなった時点を前記到達時点として特定するようにしてある請求項１から請求項７までのいずれか１項に記載の移動体感知装置。     The specific part is:
    The time point when the dimension in the road width direction of the movable body or the movable body group in the sensing area becomes larger than a predetermined width threshold value is specified as the arrival time point. 2. A moving body sensing device according to item 1.
  9. 　前記取得部で取得したデータに基づいて前記領域で特徴点を複数抽出する抽出部と、
    　該抽出部で抽出した特徴点同士の離隔距離を算出する距離算出部と、
    　前記抽出部で抽出した各特徴点の移動ベクトルを算出するベクトル算出部と
    　を備え、
    　前記判定部は、
    　前記距離算出部で算出した距離が所定の距離閾値より短く、前記ベクトル算出部で算出した移動ベクトルが所定範囲内にある特徴点の数が所定の閾値より大きい場合、前記領域が移動体候補領域であると判定するようにしてある請求項１から請求項８までのいずれか１項に記載の移動体感知装置。     An extraction unit that extracts a plurality of feature points in the region based on the data acquired by the acquisition unit;
    A distance calculation unit for calculating a separation distance between the feature points extracted by the extraction unit;
    A vector calculation unit that calculates a movement vector of each feature point extracted by the extraction unit,
    The determination unit
    When the distance calculated by the distance calculation unit is shorter than a predetermined distance threshold and the number of feature points within the predetermined range of the movement vector calculated by the vector calculation unit is larger than the predetermined threshold, the region is a moving object candidate region The mobile body sensing device according to any one of claims 1 to 8, wherein the mobile body sensing device is determined to be.
  10. 　コンピュータに、移動体の感知をさせるためのコンピュータプログラムであって、
    　コンピュータに、
    　移動体を感知するためのデータに基づいて、道路面上の所定の複数の領域それぞれが移動体の存在を示す移動体候補領域であるか否かを判定するステップと、
    　移動体候補領域であると判定された複数の領域で画定される移動体又は移動体群が、前記道路面の移動体を感知するための所定の感知領域に到達する到達時点及び前記感知領域から離脱する離脱時点を特定するステップと、
    　特定した到達時点から離脱時点の間、所要幅のパルスを所要間隔で生成するステップと
    　を実行させるコンピュータプログラム。     A computer program for causing a computer to detect a moving object,
    On the computer,
    Determining whether each of a plurality of predetermined areas on the road surface is a moving object candidate area indicating the presence of the moving object, based on data for sensing the moving object;
    From the arrival time when the moving object or the moving object group defined by the plurality of areas determined as the moving object candidate area reaches the predetermined sensing area for sensing the moving object on the road surface, and from the sensing area Identifying when to leave,
    A computer program that executes a step of generating a pulse having a required width at a required interval from a specified arrival time to a departure time.
  11. 　移動体を感知するためのデータを取得部で取得し、取得したデータに基づいて、道路面上の所定の複数の領域それぞれが移動体の存在を示す移動体候補領域であるか否かを判定部が判定するステップと、
    　移動体候補領域であると判定された複数の領域で画定される移動体又は移動体群が、前記道路面の移動体を感知するための所定の感知領域に到達する到達時点及び前記感知領域から離脱する離脱時点を特定部が特定するステップと、
    　特定した到達時点から離脱時点の間、所要幅のパルスを所要間隔で生成部が生成するステップと
    　を含む移動体感知方法。     Data for sensing a moving object is acquired by an acquisition unit, and whether or not each of a plurality of predetermined areas on the road surface is a moving object candidate area indicating the presence of a moving object is determined based on the acquired data A step of determining by the unit;
    From the arrival time when the moving object or the moving object group defined by the plurality of areas determined as the moving object candidate area reaches the predetermined sensing area for sensing the moving object on the road surface, and from the sensing area A step for the specific part to identify the point of departure to leave,
    A generating unit that generates a pulse having a required width at a required interval between a specified arrival time and a departure time.
PCT/JP2014/082071 2014-01-08 2014-12-04 Moving-body detection device, computer program, and moving-body detection method WO2015104924A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2014-001937 2014-01-08
JP2014001937A JP2015130118A (en) 2014-01-08 2014-01-08 Moving body sensing device, computer pogram and moving body sensing method

Publications (1)

Publication Number Publication Date
WO2015104924A1 true WO2015104924A1 (en) 2015-07-16

Family

ID=53523762

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2014/082071 WO2015104924A1 (en) 2014-01-08 2014-12-04 Moving-body detection device, computer program, and moving-body detection method

Country Status (2)

Country Link
JP (1) JP2015130118A (en)
WO (1) WO2015104924A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2023079628A (en) * 2021-11-29 2023-06-08 株式会社情報技術 Traffic condition acquisition system

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005258515A (en) * 2004-03-09 2005-09-22 Koito Ind Ltd Device for sensing traveling object
JP2007115191A (en) * 2005-10-24 2007-05-10 Sumitomo Electric Ind Ltd Vehicle-sensing system, inspection unit, and inspection method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005258515A (en) * 2004-03-09 2005-09-22 Koito Ind Ltd Device for sensing traveling object
JP2007115191A (en) * 2005-10-24 2007-05-10 Sumitomo Electric Ind Ltd Vehicle-sensing system, inspection unit, and inspection method

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2023079628A (en) * 2021-11-29 2023-06-08 株式会社情報技術 Traffic condition acquisition system

Also Published As

Publication number Publication date
JP2015130118A (en) 2015-07-16

Similar Documents

Publication Publication Date Title
EP3258214B1 (en) Object detection device
WO2016149938A1 (en) Video monitoring method, video monitoring system and computer program product
CA2837314C (en) Method for measuring a height profile of a vehicle passing on a road
US10043064B2 (en) Method and apparatus of detecting object using event-based sensor
GB2585800A (en) People counting and tracking systems and methods
KR102595604B1 (en) Method and apparatus of detecting object using event-based sensor
JP6014440B2 (en) Moving object recognition device
CN104378539B (en) The video structural extraction of semantics video camera and its method of scene adaptive
CN104392224A (en) Crack detection method for road surface
JP6315308B2 (en) Control object identification device, mobile device control system, and control object recognition program
US20140002658A1 (en) Overtaking vehicle warning system and overtaking vehicle warning method
CN103226891A (en) Video-based vehicle collision accident detection method and system
JPWO2010109831A1 (en) Drive recorder
JP2017142735A (en) Pedestrian action identification device for vehicle
CN102768802B (en) Method for judging road vehicle jam based on finite-state machine (FSM)
KR101866381B1 (en) Apparatus and Method for Pedestrian Detection using Deformable Part Model
JP2016218758A5 (en)
CN115147587A (en) Obstacle detection method and device and electronic equipment
Satzoda et al. Overtaking & receding vehicle detection for driver assistance and naturalistic driving studies
WO2015104924A1 (en) Moving-body detection device, computer program, and moving-body detection method
JP2017208007A5 (en)
JP6599644B2 (en) Motion determination device, motion determination method, and motion determination program
TWI293746B (en)
JP2002367077A (en) Device and method for deciding traffic congestion
JP4074571B2 (en) Traffic flow measuring device

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14877957

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14877957

Country of ref document: EP

Kind code of ref document: A1