JP2909241B2 - Vehicle type identification device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle type discriminating apparatus applied to a toll road toll collection system or the like.

[0002]

2. Description of the Related Art Conventionally, a vehicle type discriminating apparatus for a toll road determines a vehicle type by measuring the number of axles, treads, license plate sizes, vehicle type classification numbers, and the like. There are the following two examples of applying the vehicle type discriminating apparatus to a toll road.

(1) On a distance-distance road, a vehicle type discriminating device is installed at the entrance, a toll ticket recording the discriminated vehicle type is automatically issued, and a passer-by carries this to the exit, and the pass ticket is machined. The toll is calculated by the processing described above, and the toll is collected by the collection staff. In this case, when it is determined that there is a high possibility that the vehicle type is erroneously determined, a record indicating that the vehicle type is unknown is recorded in the toll ticket, and the vehicle type is set by the collection staff at the exit.

(2) On a uniform toll road, a vehicle type discriminating device is installed at one place such as an entrance or an exit, and a toll corresponding to the discriminated vehicle type is collected. In this case, a toll is collected when the vehicle type is determined, so that the determination result that the vehicle type is unknown is not allowed.

FIG. 5 is a diagram showing the system configuration in each case.
As shown in FIG. 5 and 6, reference numeral 11 denotes a passing vehicle, 1
3 is a vehicle separator, 15 is an axis / wheel number detector, 16 is an imaging device, 17 is an image processing device, 19 is an automatic ticket issuing machine,
Reference numeral 20 denotes an automatic toll collection machine.

[0006]

When the image pickup device is abnormal and cannot take an image of the front of the vehicle, or when an image is obtained but the license plate is dirty and cannot be read, the license plate is read. Since the information is missing and the vehicle type is determined only by the other information such as the number of axes and the number of wheels, there is a high possibility that the vehicle type is unknown (distance system) or the vehicle type is erroneous.

The present invention has been made in view of the above circumstances, and can reduce the erroneous determination of a vehicle type due to a trivial cause such as dirt of an imaging device and the determination result of unknown vehicle type.
It is an object of the present invention to provide a vehicle type discrimination device that can improve the maintainability of the system and the reliability of the entire system.

[0008]

A vehicle type discriminating apparatus according to the present invention comprises: a vehicle detector for detecting passage of a vehicle; and an imaging device for imaging a front portion of the vehicle based on a vehicle detection signal from the vehicle detector. an image processing device for detecting the license plate size and vehicle type classification number from the image data output from the imaging unit, provided in the image processing apparatus, wherein
Detects abnormal signals of the imaging device,
The reading rate of the
The brightness of the member plate is lower than the set value
Detection means for detecting an error and outputting an abnormal status signal
And a tread buried in the lane to detect the number of axles and wheel spans by passing the vehicle, the number plate size and the vehicle type classification number detected by the image processing device, and the number of axles and wheel spans detected by the tread boards as data. As a vehicle type determination control unit that determines a vehicle type, and an abnormality detection unit of the image processing device.
An alarm output unit for notifying the abnormality by display or alarm when an abnormality is detected is further provided.

[0009]

(1) When the imaging apparatus is turned off from the imaging apparatus to the image processing apparatus, if there is a strobe, if there is a strobe, or if an abnormality related to the imaging apparatus occurs, the image processing apparatus can take an image by outputting an abnormal status signal. An abnormality of the device can be detected.

(2) Even if the imaging device itself is normal, there is an obstacle (for example, accumulated snow) in front of the imaging device.
In some cases, even if the captured image is processed, the license plate is hidden and cannot be read.

Although a license plate cannot be read by a motorcycle or the like which does not have a license plate attached to the front portion of the vehicle even in a normal state, if there is an obstacle in front of the image pickup apparatus as described above, the license plate of most vehicles is limited. Cannot be read, and if the image processing apparatus calculates the license plate reading rate, it can be distinguished from the case where the license plate cannot be read in a normal state. (3) Further, if the image pickup device becomes dirty, the captured image becomes dark and the license plate becomes difficult to understand, leading to a decrease in the reading rate.

In such a case, the brightness of the white portion of the license plate (a plate background portion for a white number, a character portion for a green plate) is reduced even before the reading rate is reduced. come.

As described above, in the image processor, (1) when the abnormal status signal of the image pickup device is detected, (2) when the reading rate of the license plate decreases, and (3) the brightness of the license plate decreases ( In the case of darkening, for example, an alarm can be notified to a monitoring person or the like by a display or a buzzer from the alarm output unit.

By notifying an observer of an alarm as described above, it is possible to remove an obstacle in front of the imaging device, clean dirt on the imaging device, and repair a failure of the imaging device at an early stage. , The effect of the system can be minimized.

[0015]

FIG. 1 is a diagram schematically showing a system configuration of a vehicle type discriminating apparatus according to an embodiment of the present invention.

In FIG. 1, reference numeral 1 denotes a lane in which the vehicle type discriminating apparatus of the present invention is installed, and the vehicle 2 travels in the direction of a thick arrow in the figure. A tread 3 is buried in the lane 1 in the lane width direction, and vehicle detectors 4 including a pair of photoelectric sensors are installed on both sides of the tread 3 so as to face each other. Also,
An imaging device 5 is installed in the vicinity of the lane behind the tread plate 3 and captures an image including a license plate attached to the front of the vehicle 2. At the time of this imaging, a strobe light is emitted to obtain a stable image regardless of the weather or day and night. Information on the image captured by the imaging device 5 is provided to an image processing device 6, subjected to predetermined image processing, and then output to a vehicle type determination control unit 7. The alarm output unit 8
When the abnormality of the imaging device 5 or the license plate reading rate is reduced for some reason or the brightness of the license plate portion is reduced, an abnormal status is received from the image processing device 6 and displayed to a monitor or the like by a display or a buzzer. It informs of the abnormality.

FIG. 2 is a block diagram showing a functional configuration of the present apparatus. The vehicle detector 4 vertically arranges a plurality of phototubes and forms a light film on the lane 1 by projecting and receiving light.
Detects the approach of the vehicle 2 by blocking the light film, and the vehicle detection signal S1 is output to the imaging device 5 and the vehicle type discrimination control unit 7. The imaging device 5 receives the vehicle detection signal S
The front part including the license plate part of the vehicle 1 is imaged in synchronization with the rising of
2 is output to the image processing device 6. When the power of the imaging device 5 is turned off or the imaging device 5 is abnormal such as a strobe abnormality, the image processing device 6 is notified by the imaging device abnormality signal S3.
Notify.

The image processing device 6 includes a camera I / F 61, an image memory 62, an image processing operation circuit 63, a CPU 64, a program memory 65, a data memory 66, an alarm I / F 6
And extracts information such as a license plate size and a vehicle type classification number from an image of the front surface of the vehicle captured by the imaging device 5 and outputs the information. Further, when an abnormality of the image pickup device 5 or a decrease in the license plate reading rate or a decrease in the brightness of the white portion of the license plate (the background portion in the case of the white plate and the character portion in the case of the green plate in the case of the white plate) is detected. That is, the video analog signal S2 output from the imaging device 5 is input to the camera I / F through a coaxial cable or the like, where it is converted into a digital value (A / D conversion).
It is stored in the image memory 62.

The image memory 62 stores the digital value 1
This is a memory circuit that stores an image (one frame). From the image data in the image memory 62, the license plate portion is extracted by the image processing operation circuit 63 according to the vehicle number recognition processing program in the program memory, and the license plate size, the vehicle type classification number, and the like are determined from the license plate portion. Then, the average luminance of the white part of the license plate (the background part in the case of the white plate, the character part in the case of the green plate) is calculated.

The CPU 64 controls the above-described vehicle number recognition processing. The CPU 64 outputs a vehicle number recognition result D3 to the vehicle type determination control unit 7. The data memory 66 is a RAM area for storing various parameters required for the vehicle number recognition processing, vehicle number recognition results for the past several hours, and various data such as the average luminance of the white part of the license plate for the past several cases. It is. From the alarm I / F 67, when the CPU 64 detects an abnormality in the imaging device 5, or when the license plate reading rate sharply decreases or the brightness of the license plate white portion decreases even though there is no abnormality in the strobe. When an abnormal condition is detected, an abnormal status signal S5 is output to the alarm output unit 8. Then, upon receiving the abnormal status signal S5, the alarm output unit 8 informs a monitor or the like of an abnormality by a buzzer, a display, or the like.

On the other hand, the tread 3 is composed of a plurality of flat contacts 31 and resistance contacts 32 as shown in FIG.
The vehicle 2 is driven by the voltage signal S4 generated by the treading pressure of the wheels.
The number of axes and the wheel span are detected. That is, the flat contact 31
As shown in FIG. 3 (b), the upper contact 33 and the lower contact 34 are formed of a flat upper contact 33 and a lower contact 34. 34 is short-circuited. The resistance contact 32 is composed of a flat upper contact 35 and a plurality of lower contacts 36, as shown in FIG. 3C, and a resistor 37 is connected between adjacent lower contacts. The two contacts 35 and 36 are made of a material such as stainless steel.
6 is short-circuited at the point of contact, the potential difference between the two contacts 35 and 36 is used to measure the tread position of the wheel. As shown in FIG.
Is provided in the vicinity of both ends of the tread 3, the wheel distance is measured based on the potential difference between the two resistance-type contacts 32.

The vehicle type discrimination control unit 7 obtains axis number data D1 and wheel distance data D2 based on the voltage signal S4 supplied from the tread plate 3, and recognizes a license plate size and a vehicle type classification number supplied from the image processing device 6. The processing result D3 is obtained. The information is integrated to determine the vehicle type, and the result of the determination is output to devices such as the automatic ticket issuing machine 19 in FIG. 5 and the automatic toll collection device 20 in FIG.

Next, the operation of this embodiment will be described. When the vehicle 2 travels in the lane 1 and reaches the position of the vehicle detector 4, the in-vehicle detection signal S1 is output to the imaging device 5 and the vehicle type discrimination control unit 7 when the front end of the vehicle body cuts off the optical film. Starts. That is, the imaging device 5 outputs the vehicle detection signal S1
An image of the front surface of the vehicle including the license plate is taken in accordance with the rise of the vehicle, and this image signal is sent to the camera I / F 61 as a video analog signal S2. At this time, the image processing device 6
CPU64 may result detects imaging instrumentation 置異 normal signal S3,
Alarm status signal S5 is output from alarm I / F67
Output to the unit 8.

The video analog signal S2 is supplied to the camera I / F 61
Is converted into a digital value (A / D conversion), and one screen of the digital image is stored in the image memory 62. When the digital image is stored in the image memory 62, the CP
Under the control of U64, the vehicle number recognition processing program stored in the program memory 65 is started, the license plate portion is extracted by the image processing operation circuit 63, and the license plate size, the vehicle classification number and the like are determined. You. At this time, the calculated average luminance of the white part (the background part in the case of a white plate, the character part in the case of a line plate) of the license plate and the vehicle number recognition processing result are stored in a data memory for abnormality detection processing to be described later. 66.

The CPU 64 checks the license plate reading rate calculated from the vehicle number recognition result stored in the data memory 66 and the average luminance of the white portion of the license plate at regular intervals. That is, even when the imaging device and the like are normal, it is impossible to read a license plate of a motorcycle or a vehicle without a license plate attached to the front of the vehicle. If an obstacle such as is located in front of the imaging device, it is impossible to read the license plate of most vehicles. License plate reading rate decreases. Therefore, when the license plate reading rate suddenly decreases in this manner, the CPU sends an abnormal status signal S5 from the alarm I / F 67.

In photographing, a strobe light is emitted in order to obtain a constant image regardless of the weather or day and night. However, when the image pickup device 5 becomes dirty, the image becomes almost constant unless the license plate is dirty. Since the brightness of the white portion of the license plate that is being used decreases, the brightness of the white portion of the license plate for several consecutive past cases decreases. The brightness of the part decreases. Even in such a case, the CPU 64 provides the alarm I / F 67
The abnormal status signal S5 is sent. When receiving the abnormal status signal S5 from the alarm I / F 67, the alarm output unit 8 informs a monitor or the like of the abnormality with a display or a buzzer.

On the other hand, the vehicle type discrimination control section 7 obtains the number-of-axes data D1 and the wheel-distance data D2 based on the voltage signal S4 supplied from the tread plate 3 during the period when the vehicle detection signal S1 rises. Then, these axis number data D1 and wheel distance data D2 are obtained from the vehicle number recognition result D3,
The vehicle type is determined according to the flowchart shown in FIG. First, as step St1, data on the number of axes is obtained, and it is determined whether or not the vehicle is an oversized vehicle. Here, if the number is four or more, it is determined that the vehicle is an oversized vehicle. On the other hand, if the number of axes is three or less, the wheel distance data D2 is obtained as step St2, and the following five values are determined according to the wheel distance.
Divided into two steps.

First, at step St21, if the wheelbase is extremely small, it is determined that the vehicle is a mini vehicle (motorcycle). Step S
If the wheelbase is small at t22, the vehicle number recognition result D3
When the vehicle type classification number is 40, 41, 43, 50, 80,
The vehicle type is determined to be a mini vehicle or the like, and at other times it is determined to be a normal vehicle. If the wheel span is medium as step St23, it is determined that the vehicle is a medium-sized vehicle when the number of axes is three, and is a normal car when the number of axes is other than three. In step 24, when the wheelbase is large, when the license plate size of the vehicle number recognition result D3 is medium,
If the first digit (the number on the left) of the vehicle type classification number is 3 to 7, it is an ordinary car, 1 or 2, and if it is 2 axes or less, a medium-sized car, 1 or 2
If the number of vehicles is 3 and the vehicle type classification number is unknown, it is determined that the vehicle type is unknown. When the license plate size is large, it is determined that the vehicle is an oversized vehicle. In step St25, when the wheelbase is extra-large, it is determined that the vehicle type is unknown when the license plate size is medium, that the vehicle is extra-large when the license plate size is large, and that the vehicle is large when it is unknown.

As described above, according to the present embodiment, when an abnormality in the image pickup device is detected, the number plate reading rate and the brightness of the white portion of the license plate are periodically checked to determine that they are abnormal. By issuing an alarm at the time, etc. and notifying the monitoring personnel etc. of the abnormality at an early stage,
These abnormalities can be quickly dealt with, and erroneous vehicle type determination and inability to determine vehicle type can be minimized. The present invention can be applied to a system in which it is necessary to determine a vehicle type, such as a toll road shown in the above embodiment, a toll parking lot having a different fee depending on the vehicle type, and the like.

[0030]

As described above in detail, according to the present invention, an abnormality in an image pickup apparatus (power off of an image pickup apparatus, abnormal strobe light, etc.)
Also, even though the imaging device is normal, the number plate reading rate is significantly reduced due to an obstacle (for example, snow) in front of the imaging device, or the white portion of an image captured due to dirt on the imaging device. When an abnormality such as a decrease in brightness is detected, an alarm is issued to a supervisor, and the countermeasures such as cleaning and repair of the imaging device and removal of obstacles in front of the imaging device can be taken for these abnormalities. Can be performed at an early stage, so that the maintainability of the system is improved, and erroneous determination of the vehicle type due to trivial causes such as dirt of the imaging device and the determination result of the unknown vehicle type can be reduced. The reliability of the whole system is also improved.

[Brief description of the drawings]

FIG. 1 is a schematic diagram showing a system configuration of a vehicle type discriminating apparatus according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a functional configuration of the vehicle type discriminating apparatus in the embodiment.

FIG. 3 is a configuration diagram of a tread plate in the embodiment.

FIG. 4 is a flowchart showing a vehicle type determination operation in the embodiment.

FIG. 5 is a system configuration diagram at a conventional road entrance toll-based road system.

FIG. 6 is a system configuration diagram at a conventional uniform toll road entrance.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Lane, 2 ... Vehicle, 3 ... Footboard, 4 ... Vehicle detector, 5 ...
Imaging device, 6 image processing device, 7 vehicle type discrimination control unit, 8
... alarm output unit, 11 ... vehicle, 13 ... vehicle separator, 15 ...
Number of shafts, number of wheels detector, 16: imaging device, 17: image processing device, 19: automatic ticket issuing machine, 20: automatic toll collection device,
31 flat contact, 32 resistance contact, 33 flat upper contact, 34 flat lower contact, 35 upper contact, 36
... lower contact, 37 ... resistor, 61 ... camera I / F, 62 ...
Image memory, 63: image processing operation circuit, 64: CPU,
65 ... program memory, 66 ... data memory, 67 ...
Alarm I / F.

Claims (1)

(57) [Claims] 1. A vehicle detector for detecting passage of a vehicle, an image pickup device for picking up an image of a front portion of the vehicle based on a vehicle detection signal from the vehicle detector, and a license plate based on image data output from the image pickup unit. an image processing apparatus for detecting a size and vehicle type classification number, et al provided in the image processing apparatus
Detecting an abnormal signal of the imaging device,
When the plate reading rate is lower than the set
The brightness of the imaged license plate is lower than the set value
Error that detects an error condition and outputs an error status signal
Detecting means, a tread buried in the lane to detect the number of axles and wheel spans by passing the vehicle, a license plate size and a vehicle type classification number detected by the image processing device, and the number of axles and the wheel spans detected by the treads a vehicle type determination control section for discriminating the vehicle type as data, if more abnormality is detected in the abnormality detecting means of the image processing apparatus, vehicle type, characterized by comprising an alarm output unit for notifying the abnormality by a display or alarm Discriminator.