JPH06111177A - Automatic recognizing device for number plate - Google Patents

Abstract

PURPOSE:To provide an automatic number plate recognizing device capable of easily adjusting a visual field at the time of setting a camera, shortening time required for setting work and accurately recognizing also the number plate of a vehicle traveling over a traffic lane due to the existence of an obstacle or the like. CONSTITUTION:The automatic number plate recognizing device characteristically consists of a camera 1 for picking up an image of a vehicle traveling on a traffic lane, a recognizing means 2 for recognizing the characters of the plate from the image picked up by the camera 1, a character position calculating means 3 for finding out the horizontal direction address of the plate center on the image from the positions of the characters recognized by the means 2, and a visual field control means 4 for finding out the average of addresses of plate centers corresponding to past serveral vehicles and automatically controlling the visual field of the camera 1 so that the plate center on the screen is adjusted to the horizontal center of the screen.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle license plate automatic recognition apparatus.

[0002]

2. Description of the Related Art A vehicle automatic license plate recognition apparatus has a camera installed on a horizontal column that crosses a lane such as an expressway. When the vehicle runs, the camera takes an image of the vehicle. The recognition unit recognizes the characters (vehicle number) on the vehicle license plate from the captured image. Then, the recognized characters of the license plate are recorded. This automatic license plate recognition device is used, for example, to measure travel time (passing required time) between interchanges on a highway and as required traveling time information for a following vehicle.

[0003]

In a conventional license plate automatic recognition apparatus, a camera is fixedly installed on a horizontal support and a license plate of a vehicle passing through a fixed field of view is recognized. Therefore, when fixing the camera,
Strict positioning adjustment is required so that the field of view of the camera corresponds appropriately to the lane width. However, adjusting the field of view when installing the camera is troublesome, and there is a disadvantage that it takes a long time to install and fix the camera. Further, if there is an obstacle (for example, a broken vehicle) in or near the field of view of the camera, the number of vehicles traveling far beyond the lane increases. In this case, the camera may not be able to properly recognize the license plate, which is disadvantageous in that the number of vehicles that can be recognized is reduced.

The present invention solves the above problems, makes it easy to adjust the field of view when installing a camera, requires only a short time for installation work, and is the number of vehicles that run out of the lane due to the influence of obstacles and the like. It is an object of the present invention to provide an automatic license plate recognition device that can accurately recognize a plate.

[0005]

An automatic license plate recognition apparatus according to claim 1 of the present application claims a camera for picking up an image of a vehicle traveling on a road and a number from an image picked up by the camera. It comprises a recognition means for recognizing the plate, a means for obtaining the distribution of the positions of the license plates of the plurality of vehicles recognized by the recognition means, and a visual field control means for automatically controlling the visual field of the camera according to the obtained distribution. There is.

In this license plate automatic recognition device,
Each traveling vehicle is imaged by a camera, and the license plate is recognized by the recognition means from the image. Then, the distribution of the position of each license plate is obtained, and the visual field control means controls the visual field center so that the center of the visual field is located in the region where the number plate is most, according to the distribution of the position.

An automatic license plate recognition apparatus according to claim 4 of the present application includes a first camera for recognizing a license plate which images a vehicle traveling on a road, and a second camera which has a wider field of view than the first camera. A camera, a vehicle tracking unit that detects the vehicle imaged by the second camera and tracks the traveling path of the vehicle, and a position prediction unit that predicts the position of the vehicle at a predetermined time based on the track of the tracked vehicle. The visual field control means for automatically controlling the visual field of the first camera toward the predicted position and the recognition means for recognizing the license plate from the image captured by the first camera.

In this license plate automatic recognition device,
When the vehicle is detected in the image captured by the second camera, the trail of the vehicle thereafter is traced by the tracking means. Then, the position of the vehicle at a predetermined time is predicted from the trace. The visual field control means controls the predicted position of the vehicle so that the first camera visual field enters, the first camera captures an image of the vehicle, and the recognition means recognizes the license plate from the obtained image.

[0009]

1 is a block diagram showing an example of the circuit configuration of an automatic license plate recognizing apparatus according to an embodiment of the present invention.
The automatic license plate recognition device includes a camera 1 installed on a horizontal support 5 (see FIG. 2), a recognition unit (recognition means) 2 for recognizing a vehicle number of a vehicle from an image captured by the camera 1,
A character position calculation unit (character position calculation means) 3 for obtaining an address in the horizontal direction of the screen of the center of the plate Y from the position of the character (vehicle number) on the license plate Y recognized by the recognition unit 2.
Then, the average of the past several addresses of the center of the plate Y calculated by the character position calculation unit 3 is calculated, and the visual field of the camera 1 is controlled to move so that the average of the plate Y centers is in the horizontal center of the screen. And a control section (visual field control means) 4.

As shown in FIG. 2, for example, the camera 1 is installed on a horizontal support column 5 that crosses a lane such as an expressway, and images the traveling vehicle X. The camera 1 may be installed, for example, with the camera 1 having an appropriate length (for example, 50 c
(m length) It is slidably arranged via a slide rail or the like (not shown), and is installed so that the camera 1 horizontally reciprocates on the rail by driving a forward / reverse rotation motor. Alternatively, instead of the horizontal movement system of the camera 1, the camera 1
The view angle of the camera 1 can be displaced with respect to the horizontal support 5.
It may be arranged via a rotary table or the like (not shown), and the camera 1 may be installed so as to rotate by driving the rotary table.

The recognizing section 2 is a known recognizing means,
From the image captured by the camera 1, binarization, character cutting, etc. are performed to recognize the characters (vehicle number / serial number) on the license plate Y. Further, the character position calculation unit 3 obtains the center address of the recognized serial number as the center address of the plate Y, for example. In addition, the visual field control unit 4 obtains the average of the past several addresses of the center of the plate Y calculated by the character position calculation unit 3 so that the average of the center of the plate Y is at the center in the horizontal direction of the screen. The field of view of the camera 1 is moved. As shown in FIGS. 2 and 3, when the field of view of the camera 1 is horizontal d (m) and the number of pixels in the horizontal direction is n, the horizontal distance per pixel is d / n (m).
When the field of view of the camera 1 is the field of view A in FIGS. 2 and 3,
The average of the recognized centers of the plate Y is shifted to the left by K pixels from the center of the screen. At this time, the field of view of the camera 1 is d / n
The plate Y is located at the center of the screen when the field of view B is set by moving it to the left by × K (m) (see FIG. 4).

As a method of moving the field of view of the camera 1, as described above, there are a method of moving the camera 1 itself in the horizontal direction with respect to the horizontal support 5 and a method of displacing the angle of the camera 1. When moving the camera 1 in the horizontal direction, d / n × K on the column 5 shown in FIG.
(M) Move it. Further, when displacing the angle of the camera 1, when the focal length of the lens is m (m),
Rotation angle α

[0013]

[Equation 1]

[0014] In the license plate automatic recognition device having such a configuration, as shown in FIG. 2, when the vehicle Z in trouble is stopped in the lane and the traveling vehicle X passes over the lane significantly, the number of the vehicle X is reduced. As shown in FIG. 3, the plate Y is largely displaced to the left from the center of the screen (K pixels are displaced from the center of the screen), which may cause a situation in which the vehicle number cannot be properly recognized. When the centers of the plurality of license plates Y imaged by the camera 1 are all displaced from the center of the screen, the character position calculation unit 3 obtains the center address of the recognized serial number as the center address of the plate Y, for example. Then, the visual field control unit 4 obtains the average of the centers of the plurality of plates Y,
The visual field of the camera 1 is moved so that the center average is located at the center in the horizontal direction of the screen. That is, the field of view of the camera 1 is d
/ N × K (m) is moved to the left and the field of view B is set so that the plate Y becomes the center of the screen (see FIG. 4).

In the above embodiment, the reference position of the plate is set to the center of the vehicle number, but the reference position is used to calculate how much the plate on the screen deviates from the horizontal center of the screen. Therefore, if it is predetermined, it may be at any position on the plate. Figure 5
It is a block diagram which shows the structure of the license plate automatic recognition apparatus of other Example of this invention. The automatic license plate recognition apparatus of this embodiment includes a camera 1, a control motor 11 for changing the orientation of the camera 1, and an image processing unit 12 for performing processing for license plate recognition from an image captured by the camera 1. It consists of and.

In this license plate automatic recognition device,
If the lanes of the lanes A, B, ..., I on the road shown in FIG. 7 are entered as the field of view of the camera 1, the lanes (areas) of vehicles passing through the field of view within a certain time are accumulated for each lane. Remember, so that the vehicle always passes in the center of the field of view,
The orientation of the camera 1 is controlled by the control motor 11.

Here, when the camera 1 is at the standard position (the camera is parallel to the lane) and the lane E is in the central optimum area as shown in FIG. 7, the lane D moves the camera 5 to the left. Lane C is 10 ° to the left, lane B is 2 to the left
0 °, lane A is 30 ° to the left, lane F is 5 ° to the right, lane G is 10 ° to the right, and lane H is 20 to the right.
By moving the lane I by 30 ° to the right, each lane can be moved to the optimum area of the screen.

Next, the operation of the automatic license plate recognition apparatus of this embodiment will be described with reference to the flow chart shown in FIG. First, the road surface is imaged by the camera 1, an image is captured, and if a vehicle is present in the image, the license plate is read and recognition processing is performed (step ST1). Next, it is determined whether or not the recognition is successful (step ST2), and if the determination is YES, +1 is stored in the passing area (passing lane) (step ST3). For example, if the recognized vehicle passes through the lane F, the lane F is +1.
Is counted. Then, in step ST4, it is determined whether the timing is a fixed time, that is, whether the fixed time has elapsed. Until the fixed time elapses, the process returns to step ST1 and the processes of steps ST1 to ST4 are repeated. After a certain period of time, it is checked which lane the vehicle passes the most, and the control motor 11 controls the angle of the camera 1 so that the lane comes to the optimum area of the vehicle (step ST5). Then, the cumulative counter is cleared (step ST6) to prepare for the next process.

Now, it is assumed that processing is performed for a certain period of time in the screen state shown in FIG. 8A, and as a result, the number of vehicles passing the lane I is the largest, and the number of vehicles passing the lane H is the second largest. Then, first, the camera 1 is moved so that the lane H becomes the optimum area, and the field of view becomes as shown in FIG. 8B. In this case, since the lane I comes into the field of view, the camera 1 moves so that the lane I becomes the optimum area next. Even if there is a license plate outside the field of view like this,
The camera can be moved so that the area is the optimum area within the field of view. This utilizes the fact that when the number of vehicles passing through the lane I is the largest, the number of vehicles passing through the area near the same tends to increase.

Although the lane = area is described in the above embodiment, the area counted for each area does not have to correspond to the lane, and may be an area divided into smaller areas than the lane. FIG. 9 is a block diagram showing the configuration of a license plate automatic recognition apparatus according to still another embodiment of the present invention. In this embodiment, the automatic license plate recognition apparatus includes a recognition camera 1
And a control motor 1 for changing the orientation of the camera 1.
1, a camera 13 having a wider field of view than that of the camera 1, and a track of the vehicle detected by the camera 13 is tracked, the position of the vehicle at a predetermined time is predicted from the track, and the field of view of the camera 1 is controlled by the control motor 11. In addition, the image processing unit 12 performs the recognition process of the license plate based on the image from the camera 1.

Next, the operation of the automatic license plate recognition apparatus of this embodiment will be described with reference to the flow chart shown in FIG. First, an image is input from the vehicle detection camera 13 (step ST11), and it is determined whether or not a vehicle is present in the input image by using the background difference method (step S).
T12). When there is a vehicle, the tracking method is used to recognize the same vehicle and obtain the center of gravity of the position of the vehicle. Then, the obtained position is subjected to curve approximation by a spline function (step ST13). Then, the tangent line of the last obtained center-of-gravity position is obtained (step ST14). Next, the intersection point of the obtained tangent line and the image capture trigger line of the recognition camera 1 in the camera 13 is obtained, the orientation of the camera 1 is determined based on this intersection point, and the control motor 11
The orientation of the camera 1 is controlled so as to be the determined orientation (step ST15). Then, it is determined whether the current vehicle center of gravity position is below the trigger line (step ST
16) If it is below, an image is captured from the camera 1 (step ST17), and character recognition processing is performed on the captured image (step ST18).

FIG. 10 shows the field of view of the vehicle detection camera 13 in this case. In the figure, circles represent the center of gravity of the moving vehicle, and broken lines represent the image captured by the camera 1 for recognition. Trigger lines, triangles indicate the intersection of the trigger line and the locus of the center of gravity. According to the automatic license plate recognition apparatus of this embodiment, the vehicle is tracked and always directed to the vehicle traveling in the field of view of the camera. You can

[0023]

According to the invention described in claim 1 of this application, the distribution of the positions of a plurality of recognized license plates is obtained, and the visual field of the camera is automatically controlled according to the obtained distribution. Adjusting the field of view when installing the camera is easy and easy, and the installation work can be completed in a short time. In addition, the range can be covered from the field of view of the camera, and the license plate of the vehicle traveling out of the lane due to the influence of obstacles can be accurately recognized.

According to the invention of claim 4, the vehicle is detected by the second camera having a wide field of view, the track is tracked, the position of the vehicle at a predetermined time is predicted from the track, and the predicted position is set to the predicted position. Since the field of view of the first camera for recognition is also controlled, even if the camera of the camera for recognition is very narrow or the width of the road is very large, always keep the vehicle in the field of view. It is possible to always recognize without fail.

[Brief description of drawings]

FIG. 1 is a block diagram showing a circuit configuration of an automatic license plate recognition apparatus according to an embodiment of the present invention.

FIG. 2 is an explanatory view showing a state in which the automatic license plate recognition apparatus of the embodiment is installed.

FIG. 3 is an explanatory diagram showing a license plate imaged in a visual field A of a camera of the automatic license plate recognition apparatus according to the embodiment.

FIG. 4 is an explanatory view showing a license plate imaged in a visual field B of a camera of the automatic license plate recognition apparatus of the embodiment.

FIG. 5 is a block diagram showing the configuration of an automatic license plate recognition apparatus according to another embodiment of the present invention.

FIG. 6 is a flowchart for explaining the operation of the automatic license plate recognition apparatus according to the embodiment.

FIG. 7 is a diagram showing an example of a screen corresponding to a visual field of a camera of the automatic license plate recognition apparatus according to the embodiment.

FIG. 8 is a diagram illustrating a relationship between a field of view and a region in the automatic license plate recognition apparatus according to the embodiment.

FIG. 9 is a block diagram showing the configuration of a license plate automatic recognition apparatus according to still another embodiment of the present invention.

FIG. 10 is a diagram for explaining the movement of the visual field and the position of the center of gravity in the automatic license plate recognition apparatus according to the embodiment.

FIG. 11 is a flowchart for explaining the operation of the automatic license plate recognition apparatus according to the embodiment.

[Explanation of symbols]

 1 camera 2 recognition means 3 character position calculation means 4 visual field control means

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Tsuyoshi Kurata No. 678, Teraimizucho, Karakumaru Dori, Nakagyo Ward, Kyoto City Omron Software Co., Ltd.

Claims (4)

[Claims] 1. A camera for capturing an image of a vehicle traveling on a road,
Recognition means for recognizing the license plate from the image taken by the camera, means for obtaining the distribution of the positions of the license plates of the plurality of vehicles recognized by the recognition means, and automatic control of the visual field of the camera according to the obtained distribution Automatic license plate recognition device consisting of visual field control means. 2. The recognizing means is for recognizing the characters on the license plate, and the means for obtaining the distribution of the position of the license plate is a screen horizontal position of the reference position of the plate from the position of the characters recognized by the recognizing means. It comprises a plate position calculating means for obtaining an address in a direction and an average value calculating means for obtaining an average of the past several addresses of the plate calculated by the plate position calculating means. The automatic license plate recognition device according to claim 1, wherein the field of view of the camera is automatically controlled so that the center of the plate on the screen is located at the center in the horizontal direction of the screen. 3. The means for obtaining the distribution of the position of the license plate is a counting means for dividing the field of view into a plurality of regions and counting the number of license plates recognized for each region. The automatic license plate recognition device according to claim 1, wherein the visual field of the camera is automatically controlled according to the count value of the counting means. 4. A first camera for recognizing a license plate for imaging a vehicle traveling on a road, a second camera having a wider field of view than the first camera, and a vehicle imaged by the second camera. A vehicle tracking unit that detects and tracks the traveling path of the vehicle, a position prediction unit that predicts the position of the vehicle at a predetermined time from the track of the tracked vehicle, and a first camera of the first camera toward the predicted position. An automatic license plate recognition device comprising a visual field control means for automatically controlling a visual field and a recognition means for recognizing a license plate from an image captured by a first camera.