JPH05325092A - Vehicle sensor - Google Patents

Abstract

PURPOSE:To surely recognize the traffic condition by simultaneously monitoring plural lanes to detect the number of passing vehicles per unit time. CONSTITUTION:A strut 13 is provided in the side part of a road 11, and a picture input device 14 is set to its upper end part to photograph white lines 12, which are drawn across the road 11, and running vehicles 15. Picture information photographed by the picture input device 14 is inputted to a picture processor 16 provided in the base part of the strut 13. This picture processor 16 processes picture information photographed by the picture input device 14 and senses passage of vehicles 15 in accordance with the change of picture information of white lines 12 due to vehicle passage and transmits the result to a centralized processing and monitor device 18 provided in a monitor center through a cable 17. The centralized processing and monitor device 18 obtains the number of passing vehicles per unit time based on the vehicle sense signal to recognize the traffic condition.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle detection device applied to traffic volume measurement for traffic control.

[0002]

2. Description of the Related Art As a vehicle detection device (traffic counter) for traffic control, an ultrasonic system and a microwave system are generally used in addition to the optical system.

As shown in FIG. 10 (a), an optical vehicle detection device is provided with a light projector 1 and a light receiver 2 on opposite sides of a road 1 so as to face each other. Is cut off, the state is detected and the number of vehicles passing through is counted. In addition, an ultrasonic vehicle detection device is shown in FIG.
As shown in 0 (b), ultrasonic transmitter / receivers 4 are provided on both sides of the road 1.
The ultrasonic transmitter / receiver 4 emits ultrasonic pulses at regular time intervals, and the reflected wave from the vehicle 3 is received to detect passage of the vehicle. The microwave type vehicle detection device uses microwaves instead of ultrasonic waves, and has substantially the same configuration as the ultrasonic type.

[0004]

However, in the above-mentioned conventional optical vehicle detection device and ultrasonic or microwave type vehicle detection device, one device cannot simultaneously monitor a plurality of lanes. So on a road with multiple lanes,
It was difficult to introduce a vehicle detection device.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a vehicle detection device capable of simultaneously monitoring a plurality of lanes and detecting the number of passing vehicles per unit time to grasp the traffic condition. ..

[0006]

A vehicle sensing device according to the present invention is an image input device for obtaining image information by picking up an image of a certain area including a white line drawn in a transverse direction on a road surface, and the image input device. Generate a binary image from input image information 2
Value image generation circuit, a projection circuit for horizontally projecting the binary image output from the binary image generation circuit, a differentiation circuit for differentiating the projection signal output from the projection circuit, and an output from the differentiation circuit A pulse signal generation circuit that binarizes the differentiated signal generated to generate a pulse signal, a pulse count circuit that detects the pulse signal output from the pulse signal generation circuit, and a change in the count value of the pulse count circuit And a pulse count detection circuit for outputting a vehicle detection signal.

[0007]

The vehicle detection and the number of passing vehicles are performed by utilizing the fact that the white line cannot be detected by the image processing as the vehicle passes. In addition, when multiple white lines are drawn in the crossing direction of the road, the length of the vehicle
Speed detection can be performed.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is an overall configuration diagram showing an embodiment of the present invention. In FIG. 1, 11 is a road having a plurality of, for example, three traveling lanes 11a, 1b, 11c.
A plurality of, for example, three white lines 12 are drawn on the 1 in the transverse direction at regular intervals. The intervals of the white lines 12 are set so that, for example, three vehicles can fit in the entire length of a large vehicle and two vehicles can fit in the medium vehicle.

A pillar 13 is provided on the side of the road 11 corresponding to the white line 12 in the center. This prop 13
Bends the upper part to be positioned on the road 11, and installs the image input device (camera) 14 at the tip of the road 11.
The white line 12 and the traveling vehicle 15 over a plurality of upper lanes can be photographed. Further, the image processing device 16 is installed at the base of the support column 13 and the image information obtained by photographing with the image input device 14 is inputted. The image processing device 16 processes the image information sent from the image input device 14 and transmits it to the centralized processing / monitoring device 18 provided in the monitoring center via the cable 17.

The image processing device 16 is, as shown in FIG. 2, generated by the binary image generating circuit 21 for generating a binary image from the image input from the image input device 14 and the binary image generating circuit 21. A projection circuit 22 that horizontally projects a binary image, a differentiation circuit 23 that differentiates the projection signal output from the projection circuit 22, and a signal that is differentiated by the differentiation circuit 23 is binarized to generate a pulse signal. A pulse signal generation circuit 24 for generating, a pulse count circuit 25 for counting the positive pulse signals generated by the pulse signal generation circuit 24, and a pulse count detection circuit 26 for detecting a change in the count value of the pulse count circuit 25. Composed. Next, the operation of the above embodiment will be described.

The image input device 14 shown in FIG.
The white line 12 portion drawn by 1 is photographed, and the image information is input to the image processing device 16. The image processing device 16 processes the image information sent from the image input device 14 as follows to obtain a vehicle detection signal.

FIG. 3 shows an example of image processing when there is no vehicle 15 on the road 11, that is, when there is no vehicle in the image. The image processing device 16 shown in detail in FIG. 2 binarizes the image information sent from the image input device 14 by the binary image generation circuit 21, but if there is no vehicle in the image, the image processing device 16 shown in FIG. As shown in a), only the white line 12 drawn in the road crossing direction is binarized.

2 generated by this binary image generation circuit 21
The value image is sent to the projection circuit 22 to be projected in the horizontal direction, and a projection signal having a constant value is obtained as shown in FIG. The projection signal generated by the projection circuit 22 is differentiated by the differentiation circuit 23, but the differentiation signal becomes "0" as shown in FIG. 3 (c). It should be noted that this differentiating process is performed in a portion excluding both ends of the screen. Pulse signal generation circuit 2
Reference numeral 4 denotes a circuit which generates a pulse signal only when the differentiated signal is a constant value h (> 0) or more. Therefore, when the differentiated signal "0" is input, its output signal is As shown in 3 (d), it is always a constant value "0".

When the output signal of the pulse signal generating circuit 24 is held at a constant value as described above, the counting operation of the pulse counting circuit 25 is not performed and the counter value thereof is as shown in FIG. 3 (e). Remains "0". Therefore, the pulse count detection circuit 26 cannot detect the change in the count value of the pulse count circuit 25. The output signal of the pulse count detection circuit 26 is subjected to centralized processing /
It is sent to the monitoring device 18. When the number of positive pulses counted by the pulse counting circuit 25 is “0” as described above,
There is no vehicle in the captured image.

FIG. 4 shows an example of image processing when a dark vehicle is present in the image. When the image information input from the image input device 14 is binarized by the binary image generation circuit 21, if a blackish vehicle is present in the image, the portion of the white line 12 where the vehicle does not exist as shown in FIG. 4A. Is binarized.
When the binary image is input to the projection circuit 22 and projected in the horizontal direction, the portion where the vehicle exists becomes lower than the portion where the vehicle does not exist, as shown in FIG. 4B.

When the projection signal generated by the projection circuit 22 is differentiated by the differentiation circuit 23, positive and negative differential signals having sharp peaks at the edge portion of the vehicle are generated as shown in FIG. 4 (c). This differential signal is used as the pulse signal generation circuit 2
4 is input, one positive pulse signal is generated as shown in FIG. When this pulse signal is input to the pulse counting circuit 25 to count the number of positive pulses,
It becomes "1" as shown in (e). When the number of positive pulses is counted by the pulse counting circuit 25 as described above,
A vehicle exists in the captured image.

FIG. 5 shows an example of image processing when a whitish vehicle is present in the image. When the image information input from the image input device 14 is binarized by the binary image generation circuit 21, the region where the vehicle is present and the white line 12 are binarized as shown in FIG. When the binary image is input to the projection circuit 22 and projected in the horizontal direction, the portion where the vehicle exists has a larger value than the portion where the vehicle does not exist as shown in FIG. 5B. When the projection signal generated by the projection circuit 22 is differentiated by the differentiation circuit 23, a differentiation signal having a sharp peak at the edge portion of the vehicle is generated as shown in FIG. 5 (c). When this differential signal is input to the pulse signal generation circuit 24, one positive pulse signal is generated as shown in FIG. When this positive pulse signal is input to the pulse counting circuit 25 and counted, it becomes "1" as shown in FIG. 5 (e). When the number of positive pulses is counted by the pulse counting circuit 25 as described above, the vehicle is present in the captured image.

FIG. 6 shows an example of image processing when there are two vehicles, that is, a blackish vehicle and a whitish vehicle.
When the image information input from the image input device 14 is binarized by the binary image generation circuit 21, an area where a whitish vehicle exists and a white line 12 where no vehicle exists as shown in FIG. 6A.
Is binarized. When the binary image is input to the projection circuit 22 and projected in the horizontal direction, as shown in FIG. 6B, the maximum value is obtained in a portion where a whitish vehicle exists and the minimum value is obtained in a portion where a dark vehicle exists. To take. When the projection signal generated by the projection circuit 22 is differentiated by the differentiation circuit 23, positive and negative differential signals having sharp peaks at the edge portion of the vehicle are generated as shown in FIG. 6C. When this differentiated signal is input to the pulse signal generation circuit 24, FIG.
Two positive pulse signals are generated as shown in (d). When this positive pulse signal is input to the pulse counting circuit 25 and counted, the count value becomes "2" as shown in FIG. 6 (e). When the number of positive pulses is counted by the pulse counting circuit 25 as described above, the vehicle is present in the captured image.

FIG. 7 shows a method of detecting the number of passing vehicles when two vehicles pass in the front and rear direction. Pulse counting circuit 2
If the number of positive pulses output from 5 changes as shown in the figure, the pulse count detection circuit 26 has increased the number of positive pulses from "0" to "1" at time t2, so that the vehicle has entered. Therefore, the passing vehicle is incremented by "1". Similarly, since the number of positive pulses has increased from "1" to "2" at time t4, the passing vehicle is incremented by "1". At time t6, the number of positive pulses has decreased from "2" to "1", so it is determined that one vehicle has advanced outside the detection area, and the passing vehicle is not counted up. Similarly, since the number of positive pulses has decreased from "2" to "1" at time t8, the passing vehicle is not counted up. When the number of positive pulses increases from “0” to “2”, the number of passing vehicles is incremented by “2”. In this way, the number of passing vehicles is measured by observing the time-series change in the number of positive pulses with the pulse count detection circuit 26.

FIG. 8 shows an example of speed detection in the case of a dark vehicle. The image processing area is divided into white lines as shown in FIG. 8A (areas A and B in the figure), and the rise time of the pulse signal in each area is measured as shown in FIG. 8B. However, the vehicle speed can be measured by utilizing the difference Δt between the rising times. That is, when the interval between the white lines 12 is L, the vehicle speed v can be obtained by the following equation: v = L / Δt. Furthermore, when the number of white lines 12 is three or more, the average speed can be obtained, and the reliability is increased.

When three white lines 12 are provided, by detecting the generation of a pulse signal for each white line,
The size of the vehicle can be determined. That is, FIG.
As shown in (a), A as a white line 12 on the road 11,
When 3 vehicles of B and C are drawn, in the state where the vehicle is not approaching, each of the white lines of A, B and C is detected and pulse signals are generated respectively, and the vehicle is approaching the white line and on the white line. When it is located at, the pulse signal for the white line is turned off.

FIG. 9B shows a change over time in the white line detection pattern when a large vehicle and a medium-sized vehicle enter. In FIG. 9B, at time t1, a pulse signal for each white line A, B, C is output in a state where the vehicle is not approaching. When the vehicle enters the white line C portion at time t2, the pulse signals for the white lines A and B are output and the pulse signal for the white line C is not output. When the vehicle is positioned on the white lines B and C at time t3, only the pulse for the white line A is output. In the case of a large vehicle, the vehicle is positioned on the three white lines A, B and C at time t4, and the pulse signals are all turned off. In the case of a medium-sized car, since the vehicle length is short, the rear end has passed the white line C and the front end has not reached the white line A at time t4, and only the pulse signal for the white line B is output. Similarly, the output state of the pulse signal with respect to the white line changes as the vehicle passes, and the large vehicle and the medium-sized vehicle can be discriminated from the changed state. That is, in the case of a large vehicle, there is a state in which the pulse signals for the white lines A, B and C are turned off at the same time, so it is possible to distinguish between a large vehicle and a medium vehicle. Further, the average speed of the vehicle can be detected from the elapsed time of the white line detection pattern.

Since the passing vehicles can be detected as described above, the number of passing vehicles per unit time can be obtained. The traffic situation can be grasped from the number of passing vehicles per unit time, and it becomes possible to take appropriate measures such as signal intervention when a traffic jam occurs. Further, by suppressing illegally parked vehicles, it is possible to alleviate traffic congestion caused by parked vehicles. Furthermore,
Since the speed of a passing vehicle can be detected, a warning can be given to a high-speed vehicle to promote safe driving, or a traffic violation can be cracked down. Further, the present invention can detect the size of a passing vehicle, and thus can be applied to a toll road charging system.

[0025]

As described in detail above, according to the present invention, a certain area including a white line drawn in a transverse direction on a road surface is imaged, and the passage of a vehicle is detected from the change of image information with respect to the white line due to the passage of the vehicle. Therefore, it is possible to monitor a plurality of lanes at the same time, detect the number of passing vehicles per unit time to grasp the traffic condition, and do not spoil the beauty of the city.

[Brief description of drawings]

FIG. 1 is a diagram showing an overall configuration diagram of a vehicle detection device according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of an image processing apparatus in the embodiment.

FIG. 3 is a diagram showing an example of image processing when there is no vehicle.

FIG. 4 is a diagram showing an example of image processing when a blackish vehicle exists.

FIG. 5 is a diagram showing an example of image processing when a whitish vehicle exists.

FIG. 6 is a diagram showing an example of image processing when two vehicles are present.

FIG. 7 is a diagram showing a method of detecting the number of passing vehicles when two vehicles pass in front and rear.

FIG. 8 is a diagram showing an example of vehicle speed detection.

FIG. 9 is a diagram showing a processing example in the case of distinguishing between a large vehicle and a medium vehicle.

FIG. 10 is a diagram showing a configuration of a conventional vehicle detection device.

[Explanation of symbols]

11 ... Road, 12 ... White line drawn across the road, 13
... Posts, 14 ... Image input device, 15 ... Vehicle, 16 ... Image processing device, 17 ... Cable, 18 ... Centralized processing / monitoring device, 21 ... Binary image generation circuit, 22 ... Projection circuit, 23 ...
Differentiating circuit, 24 ... Pulse signal generating circuit, 25 ... Pulse counting circuit, 26 ... Pulse counting detecting circuit.

Claims (1)

[Claims] 1. An image input device that obtains image information by imaging a certain area including a white line drawn in a transverse direction on a road surface, and a binary image is generated from the image information input from the image input device. Value image generation circuit, a projection circuit for horizontally projecting the binary image output from the binary image generation circuit, a differentiation circuit for differentiating the projection signal output from the projection circuit, and an output from the differentiation circuit A pulse signal generation circuit for binarizing the differentiated signal generated to generate a pulse signal, a pulse count circuit for detecting the pulse signal output from the pulse signal generation circuit, and a change in the count value of the pulse count circuit And a pulse count detection circuit that outputs a vehicle detection signal.