JPH03188599A - Vehicle type deciding device - Google Patents

Abstract

PURPOSE:To shorten the time when the passage of vehicles are stopped for installation of equipments by photographing the front face of a vehicle with the detection signal of a vehicle separator to process the photographed image and deciding integratedly the type of the vehicle after decision of the size of a number plate, recognition of two digits of a vehicle type separation number, and measurement of shape features of the vehicle respectively. CONSTITUTION:When a vehicle is sensed by a vehicle separator 1, an image pickup device 2 photographs the front face of the vehicle with use of the vehicle detection signal. An image processor 3 inputs the still video signals photographed by the device 2 and applies the A/D conversion to these video signals to decide the size of the number plate, to recognize two digits of a vehicle separation number, and to measure the shape features of the vehicle. Then these result of recognition and measurement are integratedly decided for decision of the type of the vehicle. Thus it is possible to shorten the time when the passage of vehicles are stopped for installation of equipments and therefore to apply a vehicle type deciding device to an existing road having a large traffic volume.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a vehicle type discrimination device applied to vehicle type discrimination of toll machine products.

[Prior Art] FIG. 6 is a diagram showing the configuration of a conventional vehicle type discrimination device, in which a vehicle separator 11 detects the entry of a vehicle, an image pickup device 12 and an image processor 13 recognize the vehicle license plate, and a tread plate recognizes the vehicle. The vehicle height detector 14 measures the vehicle height, and the overhang detector 15 measures the distance from the tip of the vehicle to the first axle tire.These recognition results and measurement results are comprehensively judged to determine the vehicle type. is determined.

[Issue to be solved by the inventionffi] As shown in Figure 6, conventional vehicle type identification requires a large number of sensors, and treads etc. need to be embedded in the road, so it takes a long time to install the equipment. It was considered difficult to introduce a vehicle type identification device to an existing road with heavy traffic, as it would require road closures for hours.

The problem of the present invention is to configure the equipment configuration for vehicle type discrimination by a vehicle separator, an image pickup device, and an image processing device, and to measure vehicle width and vehicle height, which were conventionally measured using tread plates and vehicle height detectors, etc. It is an object of the present invention to provide a vehicle type discriminating device that can reduce the time required for vehicle traffic closure for equipment installation and reduce equipment costs by performing image processing.

[Means for Solving the Problems] A vehicle type discrimination device according to the present invention includes a vehicle separator that detects a vehicle passing through a toll gate, an imager that photographs the front of the vehicle using the detection signal, and a vehicle separator that detects a vehicle passing through a tollgate. The vehicle is characterized by being equipped with an image processing machine that processes the image to determine the license plate size, recognize the two-digit vehicle model identification number, and measure the characteristics of the vehicle shape to comprehensively determine the vehicle model. .

[Operation] According to the present invention, since license plate recognition and vehicle shape feature measurement are performed by image processing, the license plate presence range (approximately 1.5 m horizontally x 1 m vertically) required for conventional license plate recognition is included. The imaging field of view is as wide as the full width of the lane (approximately 3 meters horizontally), and the camera of the imaging device has a lower case letter thickness of 6 on the license plate with the above imaging field of view.
If a high-resolution electronic shutter camera capable of resolving images is used and the horizontal field of view is Mm, the required horizontal resolution Ah will be Ah>1000.1176 lines. ), furthermore, the feature measurement of the vehicle shape and the recognition of the above-mentioned image are performed by cutting out the plate, determining the plate size, cutting out the vehicle type classification number, and car type by targeting the range where the license plate exists in the image captured in the above-mentioned imaging field of view. By recognizing the classification number, the vehicle type can be determined by comprehensively determining the recognition result of the license plate and the characteristic measurement result of the vehicle shape.

[Embodiment] FIG. 1 is a diagram showing the configuration of a vehicle type discrimination device according to an embodiment of the present invention. The vehicle separator 1 is an optical sensor for detecting a vehicle passing through a lane, and Multiple light beams are passed through the vehicle, and the vehicle is detected when one of the light beams is blocked by the vehicle in front of it. The image pickup device 2 uses this vehicle detection signal to photograph the front of the vehicle obliquely with a horizontal field of view of approximately the width of the lane (approximately 3 m) at a shutter speed of approximately 1/1000. Since the camera needs to resolve the 6 am width of lowercase letters on the license plate, a high-resolution electronic shutter camera with a horizontal resolution of 3 m ÷ 6 mm - 500 lines or more is used. The image processing device 3 inputs the still image video signal photographed by the imaging device 2, performs AD conversion, performs license plate recognition processing and vehicle shape feature measurement processing, and comprehensively judges the recognition results and measurement results. Determine the car model.

FIG. 2 is a diagram showing a conceptual diagram of vehicle type discrimination according to an embodiment of the present invention, where al indicates the imaging field of view and is an image taken from an angle of the entire vehicle, and bl indicates the range where the license plate exists. ing. When performing license plate recognition processing, the processing is performed on the range b2 obtained by cutting out this bl. This is to speed up the processing, and the processing results include determination of the B3 plate size (large plate/regular plate) and recognition of the two-digit vehicle classification number.

a2 compresses al to the measurement accuracy required for vehicle type identification,
A cutout of the vehicle part is shown, and in this figure, the width of the vehicle is defined as W between the lower left corner of the vehicle and the right tire, and the height of the vehicle is defined as the height from the bottom of the tire to the highest point. Here al is 82
The purpose of compressing the data is to speed up processing. As a result of the processing, a3 vehicle width and W1 vehicle height are obtained.

For overall vehicle type determination, if it can be clearly determined based on the plate size and vehicle classification number, the vehicle type will be determined regardless of the vehicle shape measurement results, and if it cannot be clearly determined by the plate size and vehicle classification number alone, the vehicle shape measurement results will be taken into consideration. and make a judgment.

In the example of b3 in Fig. 2, the regular plate 52 is recognized and the vehicle is determined to be a regular car regardless of the vehicle shape measurement results, but if the license plate cannot be recognized, then the vehicle with the a3 vehicle shape measurement results It is determined that the vehicle is a regular vehicle based on the width W and vehicle height.

Figure 3 is a diagram showing an example of measurement of a large vehicle, etc. that exceeds the imaging field of view. When determining the vehicle type based on the measurement results of vehicle width and vehicle height, the size is classified according to the intended vehicle type classification. , indicates that it is acceptable to clip the imaging field of view to the cup for determination.

Since actual vehicle type discrimination differs depending on how the vehicle type is classified, the concept will be explained here using an example of the vehicle type discrimination table shown in FIG. In the Table of Figure 4, the size M of the license plate recognition result indicates a regular plate,
L indicates a large plate. The number is the recognized vehicle classification number,
X indicates Don't care.

The numerical value given to a light vehicle to an extra-large vehicle indicates the possibility of that vehicle type, and 100 can be considered to be 100% and 0 to be 0%.
k1 to 9 are the probabilities determined from the corresponding vehicle type traffic ratio for the license plate recognition result. Taking the plate recognition result LIX as an example, k] is the probability that it is a large vehicle, and k2 is the probability that it is an extra large vehicle, kl + It becomes 2-100.

In this table example, it is assumed that the check is performed in order starting from M2O.

In Table 2, the vehicle type can be clearly determined from the M40 to M9X license plate recognition results.
The vehicle type is determined regardless of the shape measurement result. LIX~
For others, the probability obtained with Tablel and Tab
The probabilities obtained from the shape measurement results of le2 are added for each vehicle type, and the vehicle type with the highest probability is determined as the determination result.

Table 2 shows an example in which the vehicle width and vehicle height are each divided into three categories according to their size, where vehicle width mouth WS<WM<WL.

The vehicle height has a relationship of H3<HM<HL, and the actual measured values w, h
gives the probability of the corresponding partition.

k10 ~ ni 25 is the same as Table's ni 1 ~ ni 9,
This is the probability determined from the traffic ratio of the corresponding vehicle type to the vehicle width and vehicle height. Taking vehicle width WM as an example, the probability of a regular vehicle is k
12. The probability of a medium-sized car is k13, and the probability of a large-sized car is k14, which is k12 + klB + k14 - 100.

Fig. 5 is a diagram showing the processing flow for vehicle type discrimination described above, and in the vehicle shape feature measurement processing flow, the left tire can be recognized in the feature measurement. Depending on the installation height of the imager, the left tire may be hidden behind the bonnet and cannot be seen, indicating that this can be used as a characteristic of the bus.

[Effects of invention According to the present invention, the following effects can be achieved.

(1) Since conventional footboards and vehicle height detectors are not required, the time required for vehicles to be shut down during equipment installation can be shortened, making it possible to introduce vehicle type discrimination devices to existing roads with heavy traffic. Moreover, equipment costs can also be reduced.

(2) As a proposal for a vehicle type identification device for existing roads with heavy traffic volume, a method that uses only license plate recognition has been proposed, but there is a limit to the recognition rate of license plates, and Even if the plate can be recognized, depending on the type of vehicle, it may not be possible to identify the vehicle type based on license plate information alone.

The present invention measures vehicle shape characteristics to make it possible to identify vehicle types based on vehicle width and height information, even for vehicles that cannot be identified by license plate recognition.
This is intended to improve the vehicle type discrimination rate.

[Brief explanation of drawings]

FIG. 1 is a configuration diagram of a vehicle type discrimination device according to an embodiment of the present invention, FIG. 2 is a conceptual diagram of vehicle type discrimination according to an embodiment of the present invention, and FIG.
Figure 4 is a diagram showing an example of measuring a shape that exceeds the imaging field of view in an embodiment of the present invention, Figure 4 is a diagram showing an example of a vehicle type discrimination table in an embodiment of the present invention, and Figure 5 is a diagram showing an example of a vehicle type discrimination table in an embodiment of the present invention. FIG. 6, a diagram showing a vehicle type discrimination processing flow, is a diagram showing a conventional example. 1... Vehicle separator, 2... Imaging device, 3... Image processing device.

Claims (1)

[Claims] A vehicle separator that detects vehicles passing through a toll gate;
An imager takes a picture of the front of the vehicle using the detection signal, and the image taken by the imager is processed to determine the license plate size, recognize the two-digit vehicle classification number, and measure the characteristics of the vehicle shape. A vehicle type discrimination device characterized by comprising an image processing device for discriminating the vehicle type.