JPH03276070A - Automatic vehicle number reader with speed measuring function - Google Patents

Abstract

PURPOSE:To accurately calculate the speed of a vehicle by detecting the head of the vehicle plural times by image processing and calculating the movement distance, and using data on the head of the vehicle estimated from the sizes of characters constituting a plate. CONSTITUTION:This reader is equipped with a photographing means 3 which photographs a road, vehicle head detecting means 11 - 15, a character segmenting means which segments the characters constituting the license plate from an image, and speed detecting means 16 and 17 which finds the speed of the vehicle by using the estimated height of the vehicle head and the move ment distance of the vehicle. The head of the vehicle is detected plural times by the image processing to calculate the movement distance of the head of the vehicle and the height of the head of the vehicle is estimated from the sizes of the characters which are found by the character segmenting means. Therefore, when the speed of the vehicle is calculated from the movement distance of the head of the vehicle and the time based upon the number of frames of picture planes during the period, the height data on the head of the vehicle is usable.

Description

DETAILED DESCRIPTION OF THE INVENTION <Industrial Application Field> The present invention relates to an automatic vehicle number reading device with a speed measurement function that automatically identifies the plate number of a running vehicle.

<Prior Art> The importance of identifying vehicles traveling on the road has recently become more and more important.

Therefore, an automatic vehicle number reader is being developed that uses television cameras installed on the road to optically capture vehicle license plates and use image processing to identify the number.

For example, a position sensor that detects the presence of a vehicle is installed at a predetermined position on the road, and a lighting device and automatic vehicle number reader are installed downstream from the installation location, and the position sensor detects the passing of the vehicle. A method has been proposed in which, when detected, the vehicle is illuminated with a lighting device and photographed with a camera installed in an automatic vehicle number reader. A plate portion of the photographed image is cut out by plate cutting processing, a character portion therein is cut out by character cutting processing, and character recognition processing is performed by character recognition processing.

However, the above method has a drawback in that a position sensor must be installed separately from the automatic vehicle number reader.

Therefore, in order to omit the position sensor, a method has been developed in which images are constantly taken with a camera and license plates are extracted regardless of the presence or absence of a vehicle from the video information.

However, for example, if the vertical field of view is 3.2 m and the number of screen scans is 60 frames per second, the number of frames capturing a vehicle traveling at 120 km/h is approximately 5.8 frames, so 5.
The screen for these five frames must be examined at high speed within the frame time, that is, 1712 seconds. Therefore, it is necessary to cut out the license plate using very high-speed image processing. Therefore, it is necessary to simplify the plate cutting process. In other words, there is a problem in that it is necessary to increase the speed even at the cost of reducing the accuracy of plate cutting.

Therefore, the applicant has filed a patent application for a vehicle number mill and dynamic reading device that can read vehicle numbers with sufficient accuracy without installing a position sensor or simplifying the plate cutting process. (March 15, 1990 "Automatic vehicle number reading device").

The automatic vehicle number reading device according to this application has a photographing means for photographing the road, and a vehicle head detecting means for detecting the head of the vehicle by processing the photographed image, and cuts out the plate only when the head of the vehicle is detected. , which performs character extraction and recognizes the extracted characters.

<Problems to be Solved by the Invention> However, according to the above-mentioned automatic vehicle number reading device, it is possible to detect the entrance of a vehicle without using a position sensor by detecting the head of the vehicle through image processing. It is not possible to detect the speed of

Although vehicle speed information is often required together with license plate information, conventionally it could only be measured using a separate speed measuring device.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide an automatic vehicle number reading device with a speed measurement function that can detect not only the entry of a vehicle but also the speed of the vehicle.

Means for Solving the Problems> In order to achieve the above object, the automatic vehicle number reading device with speed measurement function of the present invention includes a photographing means for photographing a road and a series of photographed images. A vehicle head detection means for detecting the vehicle head multiple times, a character cutting means for cutting out characters constituting a license plate from a photographed image, and a height of the vehicle head estimated from the size of the characters determined by the character cutting means; The vehicle has a speed detecting means for determining the speed of the vehicle using the moving distance of the vehicle head detected by the vehicle head detecting means.

According to the automatic vehicle number reading device with speed measurement function configured as described above, by detecting the vehicle head multiple times through image processing,
The moving distance of the vehicle head is calculated, and the height of the vehicle head is estimated from the size of the character obtained by the character cutting means. This estimate is based on the following: Vehicles with large letters that make up the plate are generally large vehicles such as trucks and lotus cars, and the height of the head of the vehicle is relatively high. Vehicles with small letters are generally medium-sized or small-sized cars that have a relatively low height of the head of the vehicle. It is based on the fact that

Therefore, when calculating the speed of the vehicle from the distance traveled by the vehicle head and the time based on the number of frames on the screen during that time, it is possible to use the height data of the vehicle head obtained as described above. , it is possible to make more accurate velocity calculations.

<Example> Embodiments will be described in detail below with reference to the accompanying drawings showing examples.

FIG. 2A and FIG. 2B are a plan view and a front view showing an installation example of an automatic vehicle number reading device with a speed measurement function. The automatic vehicle number reading device with a speed measurement function consists of an automatic vehicle number reading device body 1, lighting devices 2a and 2b, a camera device 3, etc. The camera device 3 is installed on the side of the road together with the lighting devices 2a and 2b. It is attached to the upper part of the arm of Pole 4. The automatic vehicle number reading device main body 1 is attached to the side of the installation pole 4. In addition, traffic lights, pedestrian bridges, etc. may be used instead of installed poles.

The camera device 3 is attached at a predetermined angle of inclination so that the field of view of the camera device 3 is set to a predetermined area R on the road.

FIG. 3 is a block diagram showing the internal circuit configuration of the automatic vehicle number reading device main body 1. As shown in FIG. The car number automatic reading device main body 1 is A
/D conversion circuit 11, edge detection circuit 12, binarization circuit 1
3, a first memory 14, an image processing processor 15, a second memory 16, and a general-purpose processor 17.

An image signal captured by the camera is A/D converted by an A/D conversion circuit 11. The converted digital data is supplied to edge detection circuit 12. The edge detection circuit 12 detects portions (edges) where the density of the image changes. The output signal of the edge detection circuit 12 is
3, and is binarized using a predetermined threshold value as a reference.

The binarized signal is stored in the first memory 14 as a preprocessed image. The data stored in the first memory 14 is supplied to an image processor 15, where a profile (vertical projection) is created. This profile is stored in the second memory 16. The data stored in the second memory 16 is input to the general-purpose processor 17,
Here, vehicle head detection processing, plate cutting out processing, character cutting out processing, and character recognition processing are performed.

More specifically, the edge detection circuit 12 extracts only the outline of the vehicle by performing two-dimensional differentiation on an image in which the vehicle or the like is shown. The binarization circuit 13 makes the outline of the vehicle even clearer by binarizing it. The binarized image is shown in FIG. 4A. A profile of the binarized image is created by the image processing processor 15.

This profile is shown in Figure 4B. Vertical axis y in Figure 4B
is in the direction of vehicle travel.

A flowchart (first
(see figure). First, the original image is captured (step (1)), and the head of the vehicle is detected based on the histogram shown in FIG. 4B (step (2)). This vehicle head detection process is
This is a process in which the position y1 where the number of pixels added by scanning the profile from below exceeds a threshold value A is determined to be the head of the vehicle. When the head of the vehicle is detected (step (3)), it is already -
The original image is imported (step (4)). This step (4
) and the time at which the original image is captured in step (1) above corresponds to the scanning time for one frame of the image.

This is because, in this embodiment, the vehicle head detection process is completed within the scanning time of one frame, but if the vehicle head detection process is completed within the scanning time of n (n≧2) frames. If completed, the difference between the time to capture the original image in step (4) and the time to capture the original image in step (1) above will correspond to the scanning time for n frames of the image.

After capturing the original image in step (4), the vehicle head detection process is performed again (step (5)), and the vehicle head y2 is detected. If the head of the vehicle cannot be detected (step (6)), the process returns to the beginning, and if the head of the vehicle is detected, the distance traveled on the screen is calculated (step (7)).

In other words, the already determined vehicle head Y1. ．． Calculate the moving distance on the screen from Y2. This moving distance is determined by L-y2-yl. If L<0, the vehicle is traveling in the opposite direction, so the calculation is terminated.

If L≧0, L is compared with a threshold value Lo (step (8)). If L<Lo, the moving distance is too small and the influence of the vehicle head detection error is greatly reflected in the speed calculation, so the process returns to step (4) to calculate the speed and capture the next original image.

If L≧Lo, the moving distance is assumed to be sufficiently large and the speed is calculated (step (9)). Further, in parallel with this, car number reading processing (steps (10) to (13)) is performed.

In step (10), this vehicle number reading process is performed by limiting the range within the front α and rear β of the vehicle head y to the plate cutting range.
), then perform plate cutting step (11)) in this limited range, then character cutting step (1, 2)) to extract the character area, and finally character recognition ( This is a process of performing step (13)) to identify the number.

In the plate cutting process of step (11) above, for example, on the premise that the edge-detected and binarized image includes horizontal line segments or vertical line segments as described above, the image is Scan the image horizontally to extract line segments, search for bare line segments (horizontal line pairs) whose horizontal lengths are within a certain range, and whose vertical spacing is within a certain range. Line segments are extracted and a bare line segment (vertical line pair) whose length in the vertical direction is within a certain range and whose interval in the horizontal direction is within a certain range is searched. In this case, the method disclosed in Japanese Patent Laid-Open No. 62-123583, for example, may be adopted as the method for extracting the line segment.

Then, find the one most likely to be a plate frame with the smallest vertical and horizontal deviations of the vertical line pair and the horizontal line pair, and if there is one, use it as the plate frame. If there is only one of a bare vertical line and a pair of horizontal lines, the plate frame may be determined based on that one.

In the character cutting process of step (12), for example, as shown in FIG. 6, the inside of the plate determined in step (11) is binarized, a horizontal projection is obtained, and the height of the character is determined.
Furthermore, the vertical projection is determined to determine the width of the characters, and based on the results of these determinations, the character sizes are classified into large characters and medium characters.

In step (9), the moving distance on the screen is converted into the actual moving distance of the vehicle on the road. In this case, one feature of the present invention is that the height ΔH of the vehicle head is varied and calculated based on the character size obtained in the character cutting process of step (12).

The conversion method will be explained below based on FIG.

The height of the camera device 3 is H5, the angle of inclination is θ, the road surface is CO5, the average height of the vehicle head is ΔH1 or ΔH2, and the vertical field of view of the screen photographed by the camera device 3 is Y.

The positions on the road surface corresponding to both ends B and A of the vertical field of view of the screen are C and A, respectively, and the approach position of the vehicle head is EL, DI, F2, respectively, depending on the average height of the vehicle head. D2, the approach position of the vehicle head when -yl is G1 or G2, and the approach position of the vehicle head when -y2 is Fl or F2.

The subscript "1" or "2" corresponds to the vehicle head height ΔH1 or ΔH2.

The vehicle head heights ΔH1 and ΔH2 correspond to whether the characters are large or medium-sized, as described above.

This is based on the fact that vehicles with large plates are generally large cars and have a relatively high head height, while vehicles with medium plates are small or medium-sized cars and have a relatively low head height. be. Therefore, speed calculation is performed by setting the height of the vehicle head in the case of a large plate as ΔH2, and the height of the vehicle head in the case of a medium size plate as ΔH1.

This method of estimating the height of the vehicle head from the size of the letters that make up the plate is superior to the method of measuring the vehicle width and length from a screen to determine whether it is a large vehicle or a small or medium-sized vehicle. . This is because automatic vehicle number reading devices have a narrow field of view of 1.5 to 2 m in the horizontal direction and 3 to 4 m in the vertical direction.
This is because it is difficult to accurately determine the size of a vehicle using vehicle width and vehicle length.

From the arrangement in Figure 5, Ac = HY (H-Y cosθ) sinθ DI El = (H-ΔH1)Y(H-Yc
os θ) sin θ D2 F2 e= (H-ΔB2)Y(H-Yc
os θ) sin θ holds true.

Similarly, DJ Fl −(H−ΔH1) J(H−J,c
os θ) sin θDI Gl − (H−ΔH1)(L+)) (H−(L+J) cos θ) sin θD2 F
2- (H-ΔH2) J(H-J cos
θ) sin θD2 G2- (H-ΔH2) (L+J) (H-(L1) cos θ) sin θ holds, and vehicle approach distance GI Fl = DI Gl −Di FI0
2 F2-D2 G2-D2 Find F2. Here, is the distance between y2 and A on the screen. Therefore, the speed of the vehicle is GI Fl, or 02 F2.
The time to import the original image in step (4) and the step (
It can be calculated by dividing by the difference from the time when the original image is captured in 1).

The obtained speed data is output to the outside of the automatic vehicle number reading device and is used to determine the speed of the vehicle.

Although the present invention has been described above based on the embodiments, the present invention is not limited to the above embodiments. For example, in a vehicle head detection process, a portion of the road surface whose brightness is lower than the standard brightness is recognized as a shadow of the vehicle. It may be detected and used for vehicle entry (see Japanese Patent Laid-Open No. 174199/1983). Various other design changes may be made without departing from the gist of the present invention.

<Effects of the Invention> As described above, according to the automatic vehicle number reading device with speed measurement function of the present invention, by detecting the head of the vehicle multiple times through image processing, the distance traveled can be calculated, and the speed of the vehicle can be calculated. I can do it. Therefore, it is possible to provide vehicle speed information in addition to vehicle number information without using a special device for detecting speed.

In this case, the vehicle head height data estimated from the size of the characters forming the plate can be used, so it is possible to calculate the speed more precisely.

[Brief explanation of the drawing]

Figure 1 is a flowchart showing vehicle head detection processing and speed calculation processing, Figure 2A is a plan view showing an installation example of an automatic vehicle number reader with speed measurement function, Figure 2B is a front view of the same, and Figure 3 is speed measurement. A block diagram of the circuit configuration of an automatic vehicle number reading device with a function; Fig. 4A is an image of a vehicle that has been binarized by edge detection; Fig. 4B is a profile obtained by adding the number of pixels in the vehicle running direction; Fig. 5 The figure is a diagram explaining a method of calculating speed according to the height of the vehicle head, and Figure 6 is a diagram explaining a method of cutting out characters. 1... Vehicle number automatic reading device body, 3... Camera device, 15... Image processing processor, 17... General purpose processor

Claims (1)

[Scope of Claims] 1. An automatic vehicle number reading device that automatically identifies the plate number of a vehicle traveling on the road, comprising a photographing means for photographing the road, and a plurality of vehicle fronts by processing a series of photographed images. a vehicle head detection means for detecting the number of times, a character extraction means for cutting out characters constituting a license plate from a photographed image, a height of the vehicle head estimated from the size of the characters obtained by the character extraction means, and a vehicle head detection means for detecting the vehicle head. 1. An automatic vehicle number reading device with a speed measurement function, comprising speed detection means for determining the speed of the vehicle using the distance traveled by the head of the vehicle detected by the means. 2. The vehicle head detection means extracts pixels representing edges with a high density change rate based on the photographed image, counts the number of pixels in the traveling direction of the vehicle, and calculates a predetermined number of pixels with this counted number of pixels. 2. The automatic vehicle number reading device with a speed measurement function according to claim 1, wherein the vehicle head is detected by comparing the vehicle head with a threshold value.