JPH08235493A - Vehicle number reader - Google Patents

Abstract

PURPOSE: To decide the position and frame of a number plate with high accuracy and to improve vehicle number recognizing accuracy by obtaining running speed information by photographing plural images of a running vehicle. CONSTITUTION: The plural images of the running vehicle are photographed by a photographing part 11, and photographing signals are converted to multivalue digital data in accordance with density by a signal conversion part 12, and the data are stored transiently in first memory 13. Also, a template consisting of vehicle number data required for the recognition of the vehicle number and the position data of the vehicle number is stored in second memory 14 in advance. An arithmetic recognition part 15 extracts an in-plate character candidate with fixed moving distance and performing linear movement for the plural images by tracking the in-plate character candidate when it is extracted from the multivalue digital data stored in the first memory 13. Thence, the position and the frame of the number plate are decided, and the in-plate character candidate is segmented, and the vehicle number is recognized by collating it with the vehicle number data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle number reader for discriminating and judging the vehicle number of a traveling vehicle from an image of the traveling vehicle photographed by photographing means such as a television camera.

[0002]

2. Description of the Related Art In recent years, the necessity of identifying a traveling vehicle has become important as an example of criminal investigations, and a vehicle number reading device has been studied. For example, in Japanese Patent Laid-Open No. 3-265999,
A vehicle number reading device has been disclosed which reads a vehicle number by photographing a passing vehicle with a photographing means such as a television camera provided on the road and analyzing the image.

The photographing means according to the above document is configured to photograph a predetermined area at a predetermined angle.
As a result, one image of the traveling vehicle is taken.

After that, the vehicle head is detected from the obtained image, and the license plate area is judged and identified on the basis of the vehicle head. Then, unnecessary information in the license plate area is removed and analyzed by image recognition processing to read the vehicle number.

[0005]

However, the vehicle number reading device in the above document has the following problems.

(1) That is, the image processing starts with the detection of the vehicle head based on one image. For this reason, the photographing means sets the angle and the area so as to photograph the vehicle head to which the license plate is attached.

However, if the license plate is not near the vehicle head, for example, if it is placed on the dashboard inside the vehicle, the number plate area cannot be identified even if the vehicle head is detected, and the vehicle number is read. There was a problem I couldn't do.

(2) In addition, since the image of the traveling vehicle is taken with only one image, there is a problem that the image accuracy is deteriorated as the traveling speed becomes faster, and sometimes the image cannot be recognized due to a mistake in image taking.

(3) Further, in the case where the license plate is bent, it is necessary to correct the number plate to recognize the vehicle number. Therefore, there is a problem that the information is insufficient in one image and the recognition accuracy is deteriorated. It was

[0010]

In order to solve the above-mentioned problems, a vehicle number reader for photographing a traveling vehicle and recognizing the vehicle number from this image captures a plurality of images of the traveling vehicle and outputs a photographing signal. An image capturing unit, a signal conversion unit that converts an image capturing signal into multi-valued digital data corresponding to the density, an in-plate character candidate is extracted from the multi-valued digital data, and the in-plate character candidate over a plurality of images is tracked. And a calculation recognition unit that extracts a character candidate in a plate that has a constant movement distance and that moves linearly.

The multi-valued digital data from the signal conversion unit is stored, and a memory circuit in which a template consisting of vehicle number data necessary for vehicle number recognition and position data of the vehicle number is previously stored is provided, and the calculation is performed. The recognizing unit determines the license plate position and frame based on the extracted character candidates in the plate, cuts out the character candidates in the plate by matching the template with the license plate position, and recognizes the vehicle number from the object of the vehicle number data.

Further, in a vehicle number reading device for photographing a traveling vehicle and recognizing the vehicle number from the image, a vehicle speed detecting section for detecting the approach and vehicle speed of the vehicle and outputting a photographing information signal and a vehicle speed information signal, An image pickup unit that receives the image pickup information signal, shoots a predetermined number of images of the traveling vehicle and outputs the image pickup signal, a signal conversion unit that converts the image pickup signal into multivalued digital data according to the density, and the multivalued digital data. While extracting the in-plate character candidates from, while tracking the in-plate character candidates across a plurality of images, the moving distance is constant, and among the in-plate character candidates that move linearly, the moving distance obtained from the vehicle speed information signal And a calculation recognition unit that extracts a character candidate in the plate having the closest movement distance.

In addition to storing the multi-valued digital data from the signal conversion unit, it also has a memory circuit in which a template consisting of vehicle number data necessary for vehicle number recognition and vehicle number position data is stored in advance, and the calculation is performed. The recognizing unit determines the license plate position and frame based on the extracted character candidates in the plate, cuts out the character candidates in the plate by matching the template with the license plate position, and recognizes the vehicle number from the comparison with the vehicle number data.

[0014]

According to the above construction, a plurality of images of the traveling vehicle are photographed by the photographing unit, and the photographing signal is converted into multivalued digital data corresponding to the density by the signal conversion unit.

Then, the multi-valued digital data is temporarily stored in the memory circuit. The calculation result in the vehicle number recognition processing is also stored in the memory circuit as appropriate, and a template including vehicle number data and vehicle position data required for vehicle number recognition is stored in advance.

As a result, the operation recognition unit extracts the character candidates in the plate from the multi-valued digital data and recognizes the vehicle number.
At this time, since the in-plate character candidates are included in the plurality of images, the in-plate character candidates are tracked to extract the in-plate character candidates that have a constant movement distance and linearly move. The license plate position and the frame are determined from the positional relationship of the character candidates in the plate thus extracted, and the template character is cut out by matching the license plate position with the template,
Recognize the vehicle number by comparing it with the vehicle number data.

Further, the vehicle speed detecting section detects the approach of the vehicle and the vehicle speed, and outputs a photographing information signal concerning the photographing start timing, the number of photographed images and a vehicle speed information signal concerning the vehicle speed. Then, the image capturing unit receives the image capturing information signal, captures a predetermined number of images of the traveling vehicle, outputs the image capturing signal, and the signal converting unit converts the image capturing signal into multivalued digital data corresponding to the density and stores it in the first memory. Store temporarily. The calculation result in the vehicle number recognition process is also stored in the first memory as appropriate, and a template including vehicle number data and vehicle position data required for vehicle number recognition is stored in advance.

As a result, the operation recognition unit recognizes the vehicle number.
That is, the in-plate character candidates are extracted from the multi-valued digital data, and the in-plate character candidates over a plurality of images are tracked to select the in-plate character candidates having a constant movement distance and linearly moving. Among the in-plate character candidates thus selected, the in-plate character candidate having the moving distance closest to the moving distance obtained from the vehicle speed information signal is extracted, and the license plate position is determined from the positional relationship of the extracted in-plate character candidates. And determine the frame. Then, the template is matched with the license plate to cut out the character candidates in the plate, and the car number is recognized by comparing with the car number data.

[0019]

【Example】

<First Embodiment> A first embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram of a vehicle number reading device according to the first embodiment. The vehicle number reading device 2 includes a photographing unit 11 for photographing a plurality of images of a vehicle, a signal conversion unit 12 including an A / D converter for converting photographing signals from the photographing unit 11 into multivalued digital data, and a multivalued digital signal. The first memory 13 temporarily stores data and also the calculation data during the vehicle number recognition process, and the second memory 14 in which vehicle number data and a template required for vehicle number recognition are stored in advance. The memory circuit 17 and the operation recognition unit 15 that recognizes the vehicle number using the data stored in the first and second memories 13 and 14 are provided.

As shown in FIG. 2, the operation recognizing unit 15 detects a character in a plate 20 for extracting a character candidate in a plate based on the multivalued digital data from the signal converter 12,
The plate extraction processing 30 for determining the position and the frame of the license plate based on the in-plate character candidates and the character recognition processing 40 for recognizing the in-plate character candidates are performed.

FIG. 2 shows a vehicle number recognition processing procedure according to the above configuration.
Follow the instructions below. The image capturing unit 11 includes a television camera or the like to capture an image of the traveling vehicle (S2). It should be noted that the number of frames in one image is set in advance in accordance with the vertical field of view and shutter speed that are determined by the installation location of the television camera.

Further, the photographing may be carried out only when the traveling vehicle is detected, or the photographing may be always carried out. In the case of always photographing, since the same object (for example, road) is photographed when the vehicle does not pass through, it will be described later only when the signal level at a specific point (a plurality of points may be changed). You may set so that a vehicle number recognition process may be performed. In this case, information about environmental changes (for example, surrounding brightness and weather) may be used as the correction information.

The image pickup signal from the image pickup unit 11 is input to the signal conversion unit 12 and converted into multilevel digital data (S
3), once stored in the first memory 13.

After that, in the in-plate character detection process 20, the in-plate character candidates are extracted by the secondary differentiation process, the area division process, the feature extraction process, and the character / non-character discrimination process. That is, the multivalued digital data stored in the first memory 13 by the secondary differential processing is converted into a secondary differential image using a secondary differential filter as shown in FIG. 3, and the zero crossing method is used from this secondary differential image. To obtain a binary image (S4).

For example, the binary image in the dotted line area of the image "2" as shown in FIG. 4 (a) is obtained as follows. When the density of the dotted line area is second-order differentiated, a second-order differential curve as shown in FIG. 4 (b) is obtained. The density of the dotted line area is
Since it corresponds to the multi-valued digital data value, the second derivative is performed on this multi-valued digital data.

Using the zero-crossing method, for example, a positive value (hatched area) is associated with "1" and a negative value is associated with "0" with respect to the second derivative data. Such processing is performed on all multi-valued digital data (one image) (for example, the dotted line area is continuously scanned), and the captured image is divided into two.
Get the value image. The binary image data is stored in the first memory 13.

Then, the image of the same level is extracted from the binary image by the area division processing to generate the concatenated data (S5). This concatenated data has the same value in the binary image (for example, “1” is the black area in FIG. 4A) after the labeling process is performed on the binary image, and is composed of consecutive pixels. It is obtained by extracting one block of data.

Since this concatenated data also includes unnecessary data (non-characters) due to dust or the like, if all the concatenated data are subject to the vehicle number recognition processing, the processing speed becomes slower, and the first memory 13 stores them. It also requires more memory. Furthermore, as described later, the position of the license plate,
Since the determination is made from the in-plate character candidates based on these concatenated data, the presence of unnecessary data becomes a factor that complicates the determination. Therefore, it is necessary to select consolidated data.

Then, the feature extraction processing extracts the features of items such as the area, the aspect ratio, the black pixel density, or the circularity of each connected data (S6). These items are set appropriately. Characteristic of this concatenated data is character
By the non-character processing, the character candidates in the plate are extracted in comparison with the characteristics of the vehicle number data stored in advance in the second memory 14 (S7).

As a result of the comparison, if the characteristics of both do not match, it is judged as unnecessary data (non-character). Fig. 5 exemplifies this result. By deleting unnecessary data from the binary image containing unnecessary data as shown in Fig. 5 (a), the vehicle number or vehicle number as shown in Fig. 5 (b) can be obtained. Only the connected data approximated to. The vehicle number or the connection data similar to the vehicle number is treated as an in-plate character candidate.

Since the in-plate character candidates thus obtained are respectively applied to the images of a plurality of images taken by the imaging section 11, the positions of the in-plate character candidates extracted from each image are different from each other. become. Therefore, in the plate extraction processing 30, the movement of the character candidates in the plate between a plurality of images is tracked (S8). Tracking is performed using a block matching method, and character candidates in the plate that have a constant movement distance between the images and are moving linearly are extracted. Since the vehicle number satisfies the above condition (moving distance is constant and linear movement), by selecting this, an in-plate character candidate that may be present in the license plate is selected (S9).

Then, the position of the license plate is determined on the basis of the position of the selected character candidate in the plate or the like (S10). For example, "ri" as shown in Fig. 5 (b)
When "1", "2", "-", "3", and "4" are determined as the character candidates in the plate, the number of character candidates in the plate,
And the position and frame of the license plate are determined from the arrangement. In addition, when deciding the position of the license plate,
It goes without saying that the most appropriate one is selected when there are a plurality of in-plate character candidates serving as the positioning reference.

At this stage, the character candidate in the plate has not been recognized as the vehicle number yet, so this is recognized by the character recognition processing 40. In the character recognition process 40, the in-plate character candidate is selected from the binary image composed of the in-plate character candidates stored in the first memory 13 using the template as shown in FIG. 6 for the license plate frame. It cuts out character by character (S11). Then, the extracted character candidates in the plate are compared with the vehicle number data such as characters and numbers stored in the second memory in advance by using the pattern matching method, and the character candidate in the plate that best matches the vehicle number is the vehicle number. It is recognized (S12).

Such recognition processing is performed for all character candidates in the plate cut out by the template (S1
3), the vehicle number recognition process ends (S14).

Since the letters representing the land transportation office are various such as 1 to 4 letters, these letters are first cut out as one block,
After projecting in the vertical direction, characters are cut out for each character for comparison. However, if each pattern is registered in advance as a character representing the land transportation office, it is not necessary to cut out each character, and it is possible to perform it all at once.

If the size of the license plate frame is different, it can be easily dealt with by appropriately scaling the template to match the size of the license plate frame.

As described above, according to the first embodiment, since a plurality of images of the vehicle are taken, it is possible to prevent a shooting error. In addition, it becomes possible to know the movement information of the vehicle from a plurality of images, so that the character candidates in the plate can be accurately extracted, and as a result, the license plate position and frame can be determined with high accuracy.

<Second Embodiment> Next, a second embodiment will be described with reference to the drawings. In the first embodiment, a plurality of images of the vehicle are photographed, and the character candidates in the plate are tracked from the plurality of images to determine the position of the license plate and the like. In this case, if there is additional information about the movement of the character candidates in the plate,
Since the tracking can be performed accurately, the position of the license plate and the like can be determined with higher accuracy.

Therefore, in the second embodiment, the position of the license plate and the like are determined by providing a vehicle speed detector and performing tracking with reference to the vehicle speed.

FIG. 7 is a block diagram of the vehicle number reading device according to the second embodiment, and FIG. 8 is a flowchart showing the procedure of the vehicle number recognition process. Since the vehicle number reading device of the second embodiment is substantially the same as that of the first embodiment, the same reference numerals are given and the description of the same parts will be appropriately omitted.

The vehicle number reading device 2 includes a vehicle speed detection unit 16 having a vehicle speed sensor for detecting the approach and vehicle speed of the vehicle and outputting a photographing information signal and a vehicle speed information signal, and a photographing unit 1 for photographing the vehicle.
1. A signal conversion unit 1 including an A / D converter that converts a shooting signal from the shooting unit 11 into multivalued digital data
2. First memory 13 for storing multi-valued digital data and temporarily storing various data during vehicle number recognition processing, second memory for storing vehicle number data and template necessary for vehicle number recognition in advance Memory circuit 1 consisting of 14
7 and a calculation recognition unit 15 that recognizes the vehicle number by using the various data stored in the first and second memories 13 and 14 and referring to the vehicle speed information from the vehicle speed detection unit 16. .

The vehicle speed detection unit 16 detects the approach and the vehicle speed of the vehicle, and from the distance between the vehicle speed detection unit 16 and the imaging unit 11 (this distance is preset and stored), the photographing start timing and the photographing are possible. A photographing information signal including a number of images is output to the photographing unit 11, and a vehicle speed information signal is calculated and recognized by the calculation unit
It outputs to 5 (S1).

As a result, the photographing section 11 receives the photographing information signal and starts photographing at a predetermined timing to photograph a plurality of images of the vehicle (S2). And as mentioned above, the shooting signal is
It is converted into multi-valued digital data and the character candidates in the plate are extracted (S3 to S7).

In the plate extraction processing 30, the in-plate character candidates having the moving distance closest to the moving distance calculated from the vehicle speed received from the vehicle speed detecting unit 16 among the moving distances obtained by tracking the in-plate character candidates. Is selected (S8 '). Then, the selected character candidates in the plate are determined to be characters in the license plate (S
9). The license plate position and frame are determined based on the in-plate character candidates (S10). Thereby, the vehicle number is recognized (S11 to S14).

As described above, according to the second embodiment, the plate character candidates are appropriately identified by comparing the moving amount of the plate character candidates obtained by tracking with the moving amount of the plate character candidates obtained from the vehicle speed. Since it can be determined, the license plate position and the like can be determined with high accuracy.

Further, by effectively taking an image on the basis of the image information signal from the vehicle speed detector, the number of memories required for the first memory 13 can be reduced and the time required for the vehicle number recognition process can be reduced. It can be shortened.

[0047]

According to the present invention, since a plurality of images of the vehicle are taken, it is possible to prevent shooting mistakes. In addition, it becomes possible to know the movement information of the vehicle from a plurality of images, so that the character candidates in the plate can be accurately extracted, and as a result, the license plate position and frame can be determined with high accuracy.

Further, by comparing the moving distance of the character candidates in the plate obtained by tracking with the moving distance of the character candidates in the plate obtained from the vehicle speed, the plate character candidate can be appropriately determined, so that the number can be accurately determined. It became possible to determine the plate position, etc.

In addition, by effectively shooting based on the shooting information signal from the vehicle speed detector, the number of memories required for the first memory can be reduced, and the time required for the vehicle number recognition processing can be reduced. Can be shortened.

[Brief description of drawings]

FIG. 1 is a block diagram of a vehicle number reading device applied to a description of a first embodiment of the present invention.

FIG. 2 is a flowchart of a calculation recognition process in FIG.

3 is a diagram showing an example of a secondary differential filter applied to the secondary differential processing in FIG.

FIG. 4 is a diagram for explaining the calculation of the binary image in FIG.
(a) is a reference image, and (b) is the second derivative curve at that time.

5 is a diagram showing an example of a binary image in FIG.

FIG. 6 is a diagram illustrating a template applied to the description of the first embodiment of the present invention.

FIG. 7 is a block diagram of a vehicle number reading device applied to the description of the first embodiment of the present invention.

8 is a flowchart of a calculation recognition process in FIG.

[Explanation of symbols]

 2 vehicle number reading device 11 photographing section 12 signal conversion section 13 first memory 14 second memory 15 calculation recognition section 16 vehicle speed detection section

Claims (4)

[Claims] 1. A vehicle number reading device for photographing a traveling vehicle and recognizing a vehicle number from the image, a photographing unit for photographing a plurality of images of the traveling vehicle and outputting a photographing signal, and the photographing signal according to the density. A signal conversion unit for converting into multi-valued digital data, and extraction of in-plate character candidates from the multi-valued digital data, and tracking of the in-plate character candidates across a plurality of images to make a constant movement and linear movement. A vehicle number reading device comprising: a calculation recognition unit that extracts a character candidate in a plate. 2. A multi-valued digital data from the signal conversion unit is stored, and a memory circuit is previously stored with a template consisting of vehicle number data necessary for vehicle number recognition and vehicle number position data, The calculation recognition unit determines the license plate position and frame based on the extracted character candidates in the plate, cuts out the character candidates in the plate by matching the template with this, and recognizes the vehicle number from the comparison with the vehicle number data. The vehicle number reading device according to claim 1. 3. A vehicle speed reading device for photographing a traveling vehicle and recognizing a vehicle number from the image, and a vehicle speed detecting section for detecting approach and vehicle speed of the vehicle and outputting a photographing information signal and a vehicle speed information signal. An image pickup unit that receives the image pickup information signal and shoots a predetermined number of images of the traveling vehicle and outputs the image pickup signal; a signal conversion unit that converts the image pickup signal into multivalued digital data corresponding to the density; While extracting the character candidates in the plate from the data, tracking the character candidates in the plate over a plurality of images, the movement distance is constant, and of the character candidates in the plate that moves linearly, the movement distance obtained from the vehicle speed information signal. A vehicle number reader having a calculation recognition unit that extracts a character candidate in a plate having a movement distance closest to. 4. A memory circuit for storing multivalued digital data from the signal conversion unit, and having a template in which vehicle number data necessary for vehicle number recognition and position data of the vehicle number are stored in advance, The calculation recognition unit determines the license plate position and frame based on the extracted character candidates in the plate, cuts out the character candidates in the plate by matching the template with this, and recognizes the vehicle number from the comparison with the vehicle number data. The vehicle number reading device according to claim 3.