JP4052415B2 - Vehicle detection device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a vehicle detection apparatus at a tollgate on a highway, an entrance / exit of a toll parking lot, or the like.
[0002]
[Prior art]
At toll gates on expressways and entrances and exits of toll parking lots, the existence of passing vehicles is detected for toll billing, vehicle entry / exit management, etc., and the direction of travel, speed, dimensions, etc. are detected to select the vehicle type. A vehicle detection device for discrimination is installed.
[0003]
An example of a conventional photoelectric tube type vehicle detection device will be described with reference to FIG. FIG. 5 is a perspective view of a photoelectric tube type vehicle detection device. In the figure, vehicle separators 121 using photoelectric tubes are installed on both sides of the lane 120 so as to face each other, and a tread board is provided on the road surface of the lane 120. 122 is embedded.
[0004]
The vehicle 100 enters the lane 120 and the light emitted from one of the vehicle separators 121 and received by the other is blocked to detect the entry of the vehicle, and the tread 122 determines whether the vehicle 100 is moving forward or backward. The
[0005]
Such a vehicle detection device is further equipped with various auxiliary sensors, so that not only the vehicle 100 enters but also the vehicle type is determined.
[0006]
However, such a method using a phototube is applicable only when the vehicles 100 enter in a line in order like a toll gate, and a plurality of vehicles 100 like the main line of a highway are parallel. The application to the place where it passes through has a problem that the light beam for detection is not applicable because there is a state in which a plurality of vehicles are blocked simultaneously or continuously.
[0007]
An example of a conventional overhead vehicle detection device will be described with reference to FIG. FIG. 6 is a perspective view of the overhead type vehicle detection device. In the figure, a gantry 104 is installed above the lane 120 so as to cross the lane 120, and the gantry 104 is positioned directly above the lane 120. An overhead sensor 103 is installed.
[0008]
If the gantry 104 is installed across multiple lanes, the overhead sensor 103 is provided for each lane 120. Therefore, there are a plurality of lanes 120 and the highway main line or multiple gates where the vehicle 100 passes in parallel. It can also be applied to tollgates having a parking lot and entrances and exits of pay parking lots.
[0009]
The overhead sensor 103 irradiates a light beam (hereinafter referred to as “beam”) such as an LED (Light Emitting Diode) or a laser beam on the lane 120 from above the lane 120 and receives the reflected light. Thus, the presence of the vehicle 100 traveling in the lane 120 is detected. In this case, the two beams A102a and B102b before and after the lane 120 are directed from the upper lane toward the lane surface and across the lane. Irradiation is performed by scanning (scanning) in an extremely short cycle (for example, 2 or 3 milliseconds), and the traveling direction of the vehicle 100 is measured by the order in which the two beams A102a and B102b are blocked.
[0010]
In addition, the distance from the reflected light of the beam to the passing vehicle 100 is measured, the vehicle width, shape, and height of the vehicle are calculated, and the time difference between the two beams A102a and B102b is determined. The passing speed is calculated, and the vehicle length of the vehicle 100 is calculated from the result. The vehicle type can also be determined from the above-described various pieces of specification information of the vehicle 100.
[0011]
However, in the case of the overhead type vehicle detection device described above, the vehicle length measurement accuracy greatly depends on the vehicle passing speed, and the vehicle 100 travels at an extremely low speed, such as when there is a traffic jam, or when the vehicle 100 is stationary just below the overhead sensor 103. Etc., there is a principle drawback that the vehicle length measurement accuracy is greatly reduced.
[0012]
That is, in this vehicle length calculation method, the vehicle length is calculated by dividing the time of passing two beams by the vehicle speed on the assumption that the vehicle during the beam detection time moves at a constant speed. While traveling, there is generally little change in speed per hour and there is little time to pass through a certain section, so the amount of change in vehicle speed within that time is also small, so beam detection during high-speed driving, or at least during normal cruise driving Although it can be approximated that the vehicle 100 during the time moves at a substantially constant speed, the error increases when the vehicle speed change of the vehicle 100 during the beam detection time is large (it does not move at a constant speed).
[0013]
The case where the vehicle speed change of the vehicle 100 is large is a case where the vehicle 100 travels at an ultra-low speed, starts acceleration, decelerates, stops, or temporarily stops during a traffic jam or the like, as described above. Occurs at a certain speed or below). Therefore, it is necessary to take some measures against a significant drop in vehicle length measurement accuracy that is always possible at a certain speed or less.
[0014]
[Problems to be solved by the invention]
The present invention eliminates the drawbacks of the conventional vehicle detection device as described above, and can be applied to a case where there are a plurality of lanes and the vehicle passes in parallel. It is another object of the present invention to provide a vehicle detection device that can accurately measure the vehicle length.
[0015]
[Means for Solving the Problems]
(1) The present invention has been made to solve the above-mentioned problems. As a first means, the vehicle scans in the direction across the lane from above the lane to the lane for each lane through which the vehicle passes. A light projecting / receiving unit that irradiates a beam of light and detects the vehicle from the reflected light to measure the distance to the vehicle, and a camera that is provided on the lane and captures an image on the lane, and the image is captured by the camera An image processing unit having an artificial retina algorithm for processing a video and measuring a vehicle traveling direction and a vehicle speed in real time, and controlling the projecting / receiving unit and the image processing unit and measuring the same unit of time measured by the projecting / receiving unit configure the overhead sensor cross-sectional shape of the vehicle and a said controller unit a real-time vehicle speed measured by the image processing unit by using the vehicle length calculated by integrating with discriminates the vehicle It is intended to provide a vehicle detection apparatus according to claim.
[0016]
According to the above means, the present invention can be applied even when there are a plurality of lanes and the vehicle passes in parallel, and the vehicle speed of the vehicle can be measured in real time by an artificial retina algorithm regardless of the light projecting / receiving unit. The vehicle length can be calculated accurately regardless of the change.
[0017]
(2) As a second means, for each lane through which the vehicle passes, two beams that scan in the direction across the lane from above the lane to above the lane are irradiated before and after the lane, and the vehicle is reflected from the reflected light. Equipped with a light emitting and receiving unit that detects and measures the distance to the vehicle and the vehicle speed and calculates the vehicle length, and a wide-angle lens that is installed on the lane and can accommodate the entire shape of the vehicle passing through the lane An image processing unit having an image processing circuit for measuring the length of the vehicle by processing an image captured by the camera with the wide-angle lens instead of the light projecting / receiving unit when the vehicle speed of the vehicle is equal to or less than a certain speed; When the vehicle speed of the vehicle is higher than the constant speed, the vehicle length calculated by the light projector / receiver is used when the vehicle speed of the vehicle is lower than the constant speed. by using the vehicle length measured in the image processing unit There is provided a vehicle detection apparatus characterized by configuring the overhead sensor and a controller for performing determination of both.
[0018]
According to the second means, it can be applied even when there are a plurality of lanes and the vehicle passes in parallel, and the vehicle length of the vehicle is usually calculated from the result measured by the light projecting and receiving unit with two beams. However, the vehicle length can be measured by performing image processing on the video of the camera with a wide-angle lens at a certain speed or less of the same vehicle speed at which errors are likely to occur.
[0019]
DETAILED DESCRIPTION OF THE INVENTION
A vehicle detection apparatus according to a first embodiment of the present invention will be described based on FIG. 1 and FIG. FIG. 1 is a perspective view of an overhead-type vehicle detection device of the present embodiment, and FIG. 2 is a conceptual diagram of the configuration of the overhead sensor.
[0020]
As shown in FIG. 1, in the vehicle detection device 1 according to the present embodiment, an overhead sensor 3 is installed on a gantry 4 that extends over a lane 20 and is installed above the lane 20. Unlike the example, a single beam is projected onto the lane 20 below, the reflected light is received, and image processing using camera images is used in combination.
[0021]
As shown in FIG. 2, the overhead sensor 3 irradiates the beam 2 and detects the vehicle 100 from the reflected light, and measures the distance to the vehicle 100, and an image having an artificial retina algorithm. It comprises a processing unit 7 and a controller unit 5 that controls the light projecting / receiving unit 6 and the image processing unit 7 to comprehensively discriminate the vehicle.
[0022]
The light projecting / receiving unit 6 irradiates the vehicle 100 traveling on the lane 20 with the beam 2 and receives the reflected light thereof, and measures the distance from the reflected light of the beam to the passing vehicle 100, An object detection / distance measurement circuit 9 for calculating the shape recognition, vehicle width, and vehicle height of the vehicle 100 is provided.
[0023]
The beam 2 is irradiated onto the lane 20 from above the lane 20, and in this case, an extremely short cycle (for example, a few milliseconds) such that the beam 2 is directed from above the lane 20 toward the road surface of the lane and crosses the lane. The vehicle 100 that has been scanned and irradiated and passes through the lane 20 is always detected.
[0024]
The image processing unit 7 includes a camera 11 and includes an artificial retina algorithm 10 that processes the captured video. This artificial retina algorithm 10 does not recognize a video as a collection of pixels, but treats a certain cluster of pixels as an object and specializes in extracting movement of the cluster and a change amount of color, etc. Is to do.
[0025]
Note that such an artificial retina algorithm 10 itself has already been developed and sold as an image processing chip (IC) or module (substrate). Without the capacity memory, the image processing unit 7 can easily determine the moving direction of the vehicle 100 and the moving distance per unit time, that is, the vehicle speed.
[0026]
The artificial retina algorithm is generally not good at accurately recognizing the shape because of the small number of pixels, but is good at capturing the movement of the object. The vehicle speed can be measured and determined in real time.
[0027]
In the vehicle detection device of the present embodiment, the vehicle length of the vehicle 100 can be calculated with high accuracy by combining speed information measured in real time by the image processing unit 7 and the vehicle shape measured by the light projecting / receiving unit 6. Specifically, by integrating the cross-sectional shape of the vehicle 100 in the unit of equal time measured by the light projecting / receiving unit 6 using the real-time vehicle speed measured by the image processing unit 7, the vehicle speed is not required to be constant. Therefore, accurate vehicle shape recognition and vehicle length calculation are possible regardless of the vehicle speed.
[0028]
It should be noted that the light projecting / receiving unit 6, the image processing unit 7, and the controller unit 5 in the overhead sensor 3 do not have to be housed in a single unit as shown, and may be configured in separate units.
[0029]
Next, a vehicle detection apparatus according to a second embodiment of the present invention will be described based on FIGS. 3 and 4. FIG. 3 is a perspective view of the overhead-type vehicle detection device of the present embodiment, and FIG. 4 is a conceptual diagram of the configuration of the overhead sensor.
[0030]
As shown in FIG. 3, in the vehicle detection device 1 ′ of the present embodiment, as in the first embodiment, the overhead sensor 3 ′ is placed on the gantry 4 ′ that extends over the lane 20 and extends over the lane 20. However, unlike the first embodiment, the overhead sensor 3 ′ projects two beams to the lower lane 20 before and after the direction of the lane, and reflects the reflected light. It is a composite system that receives light and further uses image processing with a wide-angle camera image.
[0031]
As shown in FIG. 4, the overhead sensor 3 ′ of the present embodiment emits two beams A2a and B2b, detects the vehicle 100 from the reflected light, and determines the distance to the vehicle 100 and the vehicle speed. The vehicle 100 is comprehensively controlled by controlling the light projecting / receiving unit 6 ′ to measure, the image processing unit 7 ′ having the camera 11 ′ with the wide-angle lens and the image processing circuit 10 ′, the light projecting / receiving unit 6 ′, and the image processing unit 7 ′. And a controller unit 5 ′ for determining the above.
[0032]
The light projecting / receiving unit 6 ′ passes from the reflected light of the beam and the light projecting / receiving device 8 ′ that irradiates the vehicle 100 traveling in the lane 20 with the front and rear beams A2a and B2b and receives the reflected light. An object detection / distance measurement circuit 9 ′ that detects the vehicle 100, measures the distance to the vehicle 100, and calculates the shape recognition, vehicle width, and vehicle height of the vehicle 100 is provided.
[0033]
The two beams A2a and B2b are irradiated from above the lane 20 onto the lane 20, but each beam 2a, 2b is directed to the road surface of the lane from above the lane 20 and crosses the lane (for example, an extremely short cycle). 2 and 3 milliseconds), and the vehicle 100 passing through the lane 20 is surely detected.
[0034]
Further, the object detection / distance measurement circuit 9 ′, when the vehicle 100 passes at a high speed, passes the vehicle 100 from the time difference in which the passing vehicle 100 blocks the two beams A2a and B2b as in the above-described conventional example. The speed is calculated, and the vehicle length of the vehicle 100 is calculated from the result.
[0035]
However, in the present embodiment, an image processing unit 7 ′ is provided in order to compensate for a significant decrease in vehicle length measurement accuracy that may occur when the vehicle 100 is below a certain speed.
[0036]
The image processing unit 7 ′ includes a camera 11 ′ with a wide-angle lens and includes an image processing circuit 10 ′ that processes the captured video.
[0037]
The wide-angle lens-equipped camera 11 ′ can monitor a wide range of lanes 20, but the monitoring range of the wide-angle lens-equipped camera 11 ′ in the vehicle detection device 1 ′ of the present embodiment can be any range that passes through the lane 20. This is a range in which the entire shape of the vehicle type can be accommodated within the field of view.
[0038]
The image processing circuit 10 'is a general circuit configured by a memory that stores an image captured by the camera as a pixel, and a processor (such as a CPU) that extracts a vehicle from the pixel and measures the vehicle length. . Originally, in image processing, a high-speed CPU is required to process an image of a passing vehicle in real time. However, as will be described later, image processing can be performed more inexpensively by limiting the application of image processing to a certain speed or less. The circuit 10 ′ can be configured.
[0039]
The image processing unit 7 ′ only performs image processing of the vehicle 100 captured by the wide-angle lens-equipped camera 11 ′ only when the light projecting / receiving unit 6 ′ determines that the vehicle 100 is traveling at a constant speed or less. In this case, the image processing circuit 10 ′ measures the length of the vehicle.
[0040]
The image processing unit 7 ′ generally takes a processing time, but the function is applied without causing a problem by limiting the application to the case where the vehicle 100 is at a constant speed or less.
[0041]
Here, “below a certain speed” means that measurement is performed by a conventional light emitting / receiving unit 9 ′ by running at a very low speed, starting acceleration, decelerating / stopping, or temporarily stopping in a traffic jam as described above. In the speed range in which the vehicle length measurement accuracy may greatly decrease due to changes in vehicle speed, depending on the conditions of the expressway, parking lot entrance, etc., for example, “10 km / h or less”, etc. It is set in the controller unit 5 ′.
[0042]
From the above, when the vehicle 100 is below a certain speed in the specification information such as the vehicle shape, vehicle height, vehicle width and vehicle length measured by the light projecting / receiving unit 6 ′, the vehicle length measured by the image processing unit 7 ′. By supplementing information, the accuracy of vehicle length measurement is significantly improved. As a result, vehicle speed recognition and vehicle length calculation can be performed accurately regardless of the vehicle speed without requiring constant speed. It becomes.
[0043]
It should be noted that the light projecting / receiving unit 6 ′, the image processing unit 7 ′, and the controller unit 5 ′ in the overhead sensor 3 ′ may not be housed in one unit as shown in the figure, and may be configured as separate units. Also good.
[0044]
The embodiment of the present invention has been described above, but the present invention is not limited to the above embodiment, and it goes without saying that various modifications may be made within the scope of the present invention.
[0045]
【The invention's effect】
(1) As described above, according to the first aspect of the present invention, the vehicle detector is irradiated with one beam that scans in the direction crossing the lane from above the lane to the lane for each lane through which the vehicle passes, and its reflected light. A light projecting / receiving unit that detects the vehicle from the vehicle and measures the distance to the vehicle, a camera that is provided on the lane and that captures an image on the lane, and performs image processing on the image captured by the camera and the traveling direction of the vehicle And an image processing unit having an artificial retina algorithm for measuring the vehicle speed in real time, the light projecting / receiving unit and the image processing unit are controlled, and the cross-sectional shape of the vehicle in equal time units measured by the light projecting / receiving unit is since so as to constitute the overhead sensor and a controller for performing determination of the vehicle using the vehicle length calculated by integrating with the real-time vehicle speed measured by the image processing unit, lane are several It can be applied even when both of them pass in parallel, and the vehicle speed of the vehicle can be measured in real time with an artificial retina algorithm regardless of the light projecting / receiving unit, so the measured vehicle speed regardless of the vehicle speed and its change. Based on this, an accurate vehicle length can be obtained.
[0046]
(2) Further, according to the invention of claim 2, the vehicle detection device irradiates two beams for scanning in the direction crossing the lane from above the lane to the lane for each lane through which the vehicle passes. A light projecting / receiving unit that detects the vehicle from the reflected light and measures the distance to the vehicle and the vehicle speed of the vehicle to calculate the vehicle length; and the entire shape of the vehicle that is provided on the lane and passes through the lane. When the vehicle speed of the vehicle with the wide-angle lens stored in the field of view is below a certain speed, the image taken by the camera with the wide-angle lens instead of the light projecting / receiving unit is processed to measure the vehicle length of the vehicle An image processing unit having a processing circuit, the light projecting / receiving unit, and the image processing unit are controlled . When the vehicle speed of the vehicle is higher than the constant speed, the vehicle length calculated by the light projecting / receiving unit is used. When the vehicle speed is less than the predetermined speed, the image processing unit Since so as to constitute the overhead sensor and a controller for performing determination of the vehicle using the vehicle length measured, can also be applied when lane where multiple there vehicle passes in parallel, and the vehicle The vehicle length is usually calculated from the result measured by the light projecting and receiving unit with two beams. When the vehicle speed is below a certain speed, the image of the camera with a wide-angle lens is processed to measure the vehicle length. Regardless of the exact car length is obtained.
[Brief description of the drawings]
FIG. 1 is an explanatory view of a vehicle detection device according to a first embodiment of the present invention, and is an overall perspective view thereof.
FIG. 2 is a conceptual diagram of a configuration of an overhead sensor according to the first embodiment.
FIG. 3 is an explanatory view of a vehicle detection device according to a second embodiment of the present invention, and is an overall perspective view thereof.
FIG. 4 is a conceptual diagram of a configuration of an overhead sensor according to a second embodiment.
FIG. 5 is an explanatory view of an example of a conventional photoelectric tube type vehicle detection device, and is an overall perspective view thereof.
FIG. 6 is an explanatory diagram of an example of a conventional overhead vehicle detection device, and is an overall perspective view thereof.
[Explanation of symbols]
1 Vehicle detection device 2 Beam 2a Beam A
2b Beam B
3, 3 'overhead sensor 4, 4' gantry 5, 5 'controller unit 6, 6' light emitting / receiving unit 7, 7 'image processing unit 8, 8' light emitting / receiving unit 9, 9 'object detection / distance measuring circuit 10 artificial Retina algorithm 10 'Image processing circuit 11 Camera 11' Camera with wide-angle lens 20 Lane 100 Vehicle

Claims (2)

A light projecting / receiving unit that irradiates one beam that scans in the direction across the lane from above the lane to the lane for each lane that the vehicle passes, and detects the vehicle from the reflected light to measure the distance to the vehicle; An image processing unit having an artificial retina algorithm that is provided on the lane and images a video captured on the lane and processes the image of the image captured by the camera and measures the vehicle traveling direction and vehicle speed in real time ; A vehicle calculated by controlling the light projecting / receiving unit and the image processing unit and integrating the cross-sectional shape of the vehicle in equal time units measured by the light projecting / receiving unit using the real-time vehicle speed measured by the image processing unit. A vehicle detection apparatus comprising: a controller unit for determining the vehicle using a length; and an overhead sensor. For each lane that the vehicle passes, irradiate two beams that scan in the direction across the lane from above the lane to the lane, and detect the vehicle from the reflected light to detect the vehicle and the distance to the vehicle. A light emitting / receiving unit that measures the vehicle speed of the vehicle and calculates the vehicle length; a camera with a wide angle lens that is provided on the lane and that can be accommodated within the field of view of the entire shape of the vehicle passing through the lane; In this case, instead of the light projecting / receiving unit, an image processing unit having an image processing circuit for measuring the length of the vehicle by processing an image captured by the camera with the wide-angle lens, the light projecting / receiving unit, and the image processing unit When the vehicle speed of the vehicle is higher than the constant speed, the vehicle length calculated by the light projecting / receiving unit is used, and when the vehicle speed of the vehicle is equal to or lower than the constant speed, the vehicle length measured by the image processing unit is used. controller for discriminating the vehicle using Vehicle detection apparatus characterized by configuring the overhead sensor with and.

Images