JPH0474760B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a vehicle type identification device used in toll road toll collection machines, parking lots, and the like.

First, a conventional vehicle type discrimination device will be explained with reference to FIG.

FIG. 1 shows an embodiment of a conventional vehicle type discrimination device. 1a and 1b are islands that separate lanes of toll road toll gates. A is a vehicle detection device, and the vehicle detection device A is composed of optical vehicle detectors 2a, 2b and a footboard 3, and classifies the types of vehicles ₄₁ to _4o entering the vehicle, for example, based on the number of axles, number of wheels, etc. In addition, an ultrasonic vehicle width measuring device may be used to classify vehicle types based on vehicle width. Reference numeral 5 denotes an automatic ticket issuing device which issues different tickets 6 for each vehicle type. 7a, 7b
is a vehicle start detection device. An optical type, an ultrasonic type, a treadle type buried in the road surface, a loop coil type, etc. are used alone or in combination. 8 is a signal light that guides the vehicle starting. Reference numeral 9 denotes a vehicle type memory that stores the vehicle types of vehicles ₄₁ to _4o that have passed through the vehicle detection device A and are waiting for the issuance of a ticket 6 in accordance with the order of entry. Reference numeral 10 denotes a ticket issuance control device which is connected to the automatic ticket issuance device 5 and the signal light 8, receives a starting completion signal from the preceding vehicle, and uses the signal light 8 to guide the following vehicle from stopping to starting, and also controls traffic according to the vehicle type memory 9. Controls issuance of tickets. In addition, when used at the exit side, the ticket issuing device 5 can be replaced with an automatic toll collection machine. Further, in the case of manned collection, the departure detector 7 may be omitted.

The conventional vehicle type discrimination device used in these unmanned and labor-saving systems will be described in further detail.

The current toll road toll system is based on the classification of two types of vehicles: regular vehicles and large vehicles (sometimes divided into ,), but these classifications are based on vehicles with a gross weight of 8 tons or more or a maximum loading weight of 5 tons. It has become more than a ton. However, even with the weighing scale method, there is a problem with loading, so in reality, the tread of the second and subsequent axles and single/double tires (4 wheels - 6 wheels or more), which have the highest correlation to these, are measured using the treadle method. In the past, attempts have been made to put into practical use a method of making an individual or comprehensive judgment, and a method of determining the vehicle type by measuring the maximum vehicle width using an ultrasonic pulse reflection method.

However, these specifications are not legally regulated, just because they currently have a high correlation, and there is no guarantee that they will remain the same as before.

The same applies to other measurable specifications such as vehicle length and vehicle height.

Furthermore, since these specifications cannot be determined unless measurements are taken up to at least the second axis in the vehicle length direction, the distance from the vehicle type identification device to the issuing machine (or receiving machine) is long (practically about 10 m or more). This requires a long island, making the construction cost relatively high, and since multiple small vehicles can enter during the island, vehicle type memory is required.

This complicates the hardware system, leaving problems in terms of cost and stability.In addition, when a vehicle overtakes in an island, the vehicle and type of vehicle concerned become confused, leading to the problem of confusing the system.

The present invention has been made in order to improve the problems described above.The present invention detects the leading edge of an incoming vehicle based on light shielding signals on both sides of the vehicle and outputs a leading edge signal, and also detects the leading edge of the vehicle by detecting the leading edge of the vehicle. A vehicle detection device that makes a logical judgment based on both signals and light shielding signals on both sides of the vehicle and outputs a signal that a vehicle is passing, and a television camera that optically captures the license plate on the front of the vehicle. An image processing device that outputs a vehicle type discrimination signal based on the image signal of the license plate, a vehicle type discrimination signal from the image processing device, a vehicle passing signal from the vehicle detection device, and a signal indicating that the vehicle in front has started. The image processing device is configured to receive a start completion signal from a start detector shown in FIG. A digitization circuit that captures the image signal of the license plate part from the television camera and converts it into binary data, and stores the data from the digitization circuit and retains the memory contents until the next leading edge signal is input. A frame memory, a standard pattern file in which the characters indicating the car type on the license plate are binarized and stored as a standard pattern, and data of an area corresponding to the size of the character pattern representing the car type from the entire screen in the frame memory. A register is sequentially extracted, and the data from the same register is input sequentially,
Two types of non-inversion and inversion from the standard pattern file above.
type of data is input through the inverter circuit,
A verification circuit that compares the input data from the above register with input data from the inverter circuit and outputs a vehicle type discrimination signal based on the matching characters when they match, and the above frame memory, register, inverter circuit, standard pattern file, and verification circuit. The present invention provides a vehicle type discrimination device characterized in that it is constituted by a control device that is connected to each other and controls a data verification process. The vehicle type identification device of the present invention conventionally discriminated the vehicle type by detecting the number of axles, number of wheels, width, etc. This method uses a camera to optically read the numbers and compare them with the numbers on a standard pattern to determine the vehicle model, which eliminates the need for a long-distance detection section of the vehicle model identification device, simplifying the device and improving reliability.

Next, the vehicle type discrimination device of the present invention will be explained with reference to FIGS. 2 to 5.

FIG. 2 is a diagram showing the equipment configuration and arrangement of the present invention, and parts similar to the conventional ones are described using the same reference numerals. 1a and 1b are islands, and A is a vehicle detection device composed of a pair of optical vehicle detectors 2a and 2b and a footboard 3. The detection device A receives a leading edge (rise of light shielding) signal of an incoming vehicle 4. A vehicle passing signal S ₂ logically determined based on S ₁ and the signal from the footboard 3 is output. This is to distinguish, for example, a false signal that is only shaded by a person passing by, and a vehicle signal.

5 is an automatic ticket issuing device, and 6 is a ticket to be issued. 7a and 7b are start detectors, and 8 is a signal light. 10 is a pass ticket issuing control device;
A signal from the image processing device 12, a vehicle passing signal _S2 , and a start detector 7a, 7b without going through the vehicle type memory 9 shown in the figure.
In response to the start detection signal _S3 from the vehicle, the issuance of a pass ticket 6 is controlled for each vehicle.

Next, the central part of the present invention will be explained.

FIG. 3 is a functional block diagram of the image processing device 12, in which reference numeral 11 is a television camera (TV camera), and 12 is an image processing device which receives a command from the above-mentioned vehicle detection device A with the rising signal _S1 of vehicle approach and receives the command from the TV camera. The shutter is momentarily opened to capture one frame of image, read the numbers on the plate, compare this information with the standard pattern, determine the vehicle type, and send a determination signal to the ticket issuing control device 10.

13 is a digitization circuit that includes a preprocessing circuit for the image signal Vd, divides it into arbitrary pixels, and binarizes it into white and black. This circuit performs preprocessing to enhance the contrast of the letters on the plate, sets a threshold at an appropriate gray level, and converts the image signal Vd into a binary image of white and black.

Reference numeral 14 denotes a frame memory, which stores one screen _of the binarized image of the entire screen captured by the TV camera. The memory of the binarized image signal is retained and not updated.

Reference numeral 16 denotes a register having m×n digits, and has a font size (constant) indicating the vehicle type number shown in FIG. 4, and m×n digits as many pixels as can be extracted. m×n by the scanning signal _S4 from the control device 20, which will be described later.
The entire screen of the frame memory 14 is scanned up to the digit, and image data is sequentially cut out.

17 decomposes each number from 0 to 9 indicating the car model on the plate into an m x n digit matrix,
This is a standard pattern file that stores binarized standard patterns.

18 is an inverter circuit that inverts the white and black of the standard pattern. If the base of the plate is white or yellow, the characters will be at a black level when binarized, but if the base is green, the characters will be white, so this circuit inverts this.

Reference numeral 19 denotes a matching circuit, which compares the m×n digit image pattern cut out from the frame memory 14 and shown in FIG.
×n-digit standard pattern is compared via the inverter 18. If the two match within a certain precision among the m×n digits, this number represents the corresponding vehicle type.

Reference numeral 20 denotes a control device, which is a central control section that sequentially controls the circuits within the image processing device 12 to perform the functions described above. The above is the configuration of the present invention. In addition, when used on the exit side, the ticket issuing device 5 is replaced with an automatic toll collection machine, and in the case of manned collection, a departure detector 7a,
Needless to say, it can be applied to systems where 7b is omitted.

Next, detailed functions of the vehicle type discrimination device having the above configuration will be explained.

In FIG. 2, as soon as the front head of the vehicle 4 enters and blocks the optical vehicle detectors 2a and 2b, a leading edge signal _S1 is generated from the vehicle detection device A, and the image processing device 12 is commanded to control the TV camera. Image signal from 11
Start importing Vd and update old data.
Since the TV camera 11 is set to photograph almost the front of each vehicle entering the main lane, the image of the plate is always included in the main screen.

When the vehicle 4 enters further and the front wheels press the footboard 3 to detect forward movement, the vehicle detection device A can determine that it is a vehicle based on the logical product with the light blocking signal, so the vehicle passing signal _S2 is used to control ticket issuance. Output to the device 10. The ticket issuance control device 10 immediately determines the vehicle type based on the _S2 signal, the outgoing detection signal _S3 from the start detection devices 7a and 7b, and the vehicle type determination signal from the image processing device 12, and transmits it to the automatic ticket issuance device 5. A toll ticket 6 is issued according to the type of vehicle. Here, the leading edge signal _S1 , which is the light blocking signal of the optical vehicle detectors 2a and 2b, rises, but if the treadle signal does not rise, for example when a person passes by, the vehicle is detected by the falling light of the light blocking signal. Since it can be determined that this is not the case, the image processing operation of the image processing device 12 is immediately stopped.

When the ticket 6 is issued, the signal light 8 is controlled by the ticket issuance control device 10 and changes from a red light indicating that the vehicle is stopped to a green light indicating that the vehicle can start. The entering vehicle is guided by the signal light 8 and stops in front of the ticket issuing device 5, takes out the ticket 6, and the vehicle 4 starts. When the start is completed, a start detection signal _S3 is fed back from the start detection devices 7a and 7b to the issuance control device 10 to prompt the issuance of a passing ticket 6 for the following vehicle. If the next vehicle has not entered, the issuance control device 10 is returned to the initial state, and if the type of the next vehicle has been determined, the standby state is canceled and the issuance is performed.

Next, we will explain the outline of how to read the numbers representing the vehicle type on the license plate.

FIG. 5 is an example of license plate specifications.
330mm x 165mm for regular cars, 440mm x 220 for large cars
mm, depending on the purpose, white number (black font/white background), yellow number (black font/yellow background), green number (white font/green background)
There are three types. If this is binarized into white and black, the first two will be black/white, but the green number will be white/white.
It becomes a black background and is the opposite of the previous two.

As shown in Figure 5, the plate size is 330mm x
There are two types: 165mm and 440mm x 220mm, but unlike the plate size, the number at the top right (chain line frame 21) that represents the car model is constant at 30mm x 40mm, and the letter pattern is unified. Although there may be two digits, it is the first digit that indicates the car model. If you can read two digits, you need to select the first digit. Although not shown, if a two-digit memory is provided, selection can be made easily.

When the vehicle 4 enters and the image of the front of the vehicle is binarized and stored in the frame memory 14 via the digitization circuit 13, the image from the upper left corner of the entire screen in the frame memory is transmitted by the signal _S4 commanded by the control device 20. The images are sequentially cut out into the register 16 by m×n pixels.

Figure 4 shows the status of the data in the register when one line is completed in the direction of the arrow from left to right, one pixel at a time, and one line is scanned down, and all the numbers representing the car type have been cut out. It is shown as a binary value. These data are shifted pixel by pixel from the register 16 to the matching circuit 19 and compared with the standard pattern. On the other hand, while the data is shifted by one pixel, the standard pattern file storing standard patterns from 0 to 9 is scanned twice. be done. This is because, as described above, when binarized into black and white, there are two types of inverted black and white. During that time, the inverter 18 is switched once to invert white and black.

Therefore, while characters are cut out in the register and the standard pattern is scanned twice and compared in the collation circuit 19, the corresponding numbers are matched and extracted, and the car type is determined.

According to the vehicle type discrimination device according to the present invention as described above, unlike the conventional vehicle type discrimination device in which the vehicle type cannot be determined unless the vehicle enters at least the second axle, the vehicle type can be determined based on the vehicle type in front of the vehicle (at the latest on the first axle). Since this is possible, the distance between the vehicle type identification device (optical vehicle detector and treadle) and the ticket issuing device can be extremely short. This eliminates the need for other vehicles to intervene, unlike in conventional systems, eliminating the need for vehicle type memory, which was previously required, and eliminating confusion in the system due to the order of vehicle types being thrown out of order due to overtaking, etc., increasing system reliability. performance is significantly improved.

Furthermore, the island can be shortened, which reduces construction costs, and since vehicle model memory is no longer required, the system has extremely high practical effects, such as cost reduction and improved stability.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of a conventional vehicle type discrimination device, FIG. 2 is a block diagram showing the configuration of the vehicle type discrimination device of the present invention, and FIG. 3 is a block diagram showing the internal configuration of an image processing device. Figure 4 shows m in the register.
FIG. 5 is an explanatory diagram showing the black and white state of each of xn pixels, and FIG. 5 is a plan view showing an example of a license plate of a vehicle. A...Vehicle detection device, 2a, 2b...Optical vehicle detector, 3...Treadboard, 4...Vehicle, 5...Automatic ticket issuing device, 6...Ticket, 7a, 7b...Start detector, 1
0...passport issuing control device, 11...television camera,
12... Image processing device, 13... Digitization circuit, 1
4...Frame memory, 16...Register, 17...
Standard pattern file, 18... Inverter, 19
... Verification circuit, 20 ... Control device, S ₁ ... Leading edge signal, S ₂
... Vehicle passing signal, S ₃ ... Start detection signal, S ₄ ... Scanning signal, V _d ... Image signal.

Claims (1)

[Claims] 1. Detects the leading edge of an incoming vehicle based on the shading signals on both sides of the vehicle and outputs the leading edge signal, and also makes a logical judgment based on both the signal from the tread pressed by the front wheels of the vehicle and the shading signal on both sides of the vehicle. A vehicle detection device that outputs a signal when a vehicle is passing; a television camera that optically captures the license plate on the front of the vehicle; and an image processing device that outputs a vehicle type identification signal based on the license plate image signal captured by the camera. and the vehicle type discrimination signal from the same image processing device,
A pass ticket that receives a passing vehicle signal from the vehicle detection device and a start completion signal from a start detector indicating that the start of the vehicle in front has been completed, and issues a ticket issuance command to the pass ticket issuing device based on the signals. The image processing device includes a digitization circuit that captures the image signal of the license plate portion from the television camera using the leading edge signal of the vehicle detection device and converts it into binary data; a frame memory that stores the data and holds the same memory content until the next leading edge signal is input; a standard pattern file that binarizes the characters indicating the vehicle type on the license plate and stores them as a standard pattern; and the frame memory. A register that sequentially cuts out data in an area corresponding to the character pattern size representing the car model from the entire screen of the screen, and a register that sequentially inputs data from the same register, as well as two types of non-inverted and inverted data from the standard pattern file. a collation circuit that inputs type data through an inverter circuit, collates the input data from the register with the input data from the inverter circuit, and outputs a vehicle type discrimination signal based on the matching characters when they match; and the frame memory; 1. A vehicle type discriminating device comprising: a register, an inverter circuit, a standard pattern file, and a control device connected to a matching circuit to control data matching processing.