JP7329981B2 - License plate recognition device, compression method, and program - Google Patents

Description

The present invention relates to a license plate recognition device, compression method, and program.

A toll booth of a toll road is sometimes provided with a license plate recognition device for recognizing a vehicle number by photographing a license plate of the vehicle. When the license plate recognition device detects that the vehicle has entered the lane of the toll booth, the camera captures an image of the front of the vehicle and stores the image in an image memory. Then, the license plate recognition device recognizes characters in the license plate by performing predetermined image processing on the image stored in the image memory. Thereby, the license plate recognition device can acquire the license plate information of the vehicle that has entered the lane (see, for example, Patent Literature 1).

JP-A-6-274788

The image captured by the camera may be used for visual confirmation by a toll gate attendant, analysis for improving the recognition rate, etc., when character recognition cannot be performed by the license plate recognition device. In this case, the license plate recognition device needs to hold these images even after the character recognition processing, so it is necessary to increase the capacity of the image memory sufficiently. Therefore, it has been desired to reduce the image size in order to suppress the increase in the capacity of the image memory.

The present invention has been made in view of such problems, and provides a license plate recognition device, a compression method, and a program that can reduce the image size.

In order to solve the above problems, the present invention employs the following means.
According to the first aspect of the present invention, a license plate recognition device (10) capable of acquiring an image including a license plate of a vehicle traveling in a lane, reading license plate information, and storing the image includes: A photographing processing unit (110) that acquires the image according to the arrival timing of the vehicle; A division processing unit (111) that divides the second region that is unlikely to contain a plate, and a compression processing unit (112) that compresses the image to a predetermined size smaller than the size of the image. , provided. The compression processing unit (112) sets the compression ratios of the first region and the second region such that the compression ratio of the first region is lower than that of the second region.
By doing so, the license plate recognition device reduces the compression rate of the first area, which is likely to contain a license plate, to suppress deterioration, and also reduces the deterioration of the second area, which is unlikely to contain a license plate. The compression rate of the image can be increased to effectively reduce the image size.

According to the second aspect of the present invention, in the license plate recognition device (10) according to the first aspect, the segmentation processing unit (111), based on the previously acquired image, the photographing processing unit ( 110) obtains an estimated position of the license plate in the acquired image, and sets a predetermined range including the estimated position as the first area.
The position of the license plate in the image captured by the license plate recognition device may change depending on the width and shape of the lane to be captured, the running position of the vehicle, and the like. However, the license plate recognition device can estimate the position of the license plate of the vehicle traveling in the lane from a plurality of images acquired in the past. It is possible to accurately divide the second area.

According to the third aspect of the present invention, in the license plate recognition device (10) according to the second aspect, the classification processing section (111) divides the predetermined range in the horizontal direction around the estimated position into the Set as the first area.
By doing so, the license plate recognition device can easily set the horizontal range of the first region based on the estimated position of the license plate.

According to the fourth aspect of the present invention, in the license plate recognition device (10) according to the second or third aspect, the segmentation processing section (111) has a predetermined position in the vertical direction centering on the estimated position. A range is set as the first region.
By doing so, the license plate recognition device can easily set the vertical range of the first region based on the estimated position of the license plate.

According to a fifth aspect of the present invention, in the license plate recognition device (10) according to any one of the first to fourth aspects, the segmentation processing unit (111) includes the image including the vehicle. The compression processing unit (112) compresses the image by deleting the third region, which is less likely to be captured.
By doing so, the license plate recognition device can delete unnecessary regions that do not include the vehicle from the image, so that the image size can be further reduced.

According to the sixth aspect of the present invention, a license plate is installed on an island of a lane, acquires an image including a license plate of a vehicle traveling in the lane, reads license plate information, and can store the image. A plate recognition device (10) includes an imaging processing unit (110) that acquires the image according to the arrival timing of the vehicle; A segmentation processing unit (111) for segmenting a third region in which the vehicle is less likely to be included than the first region, and a compression process for compressing the image to a predetermined size smaller than the size of the image. and a part (112). The compression processing unit (112) compresses the image by deleting the third area.
By doing so, the license plate recognition device can greatly reduce the image size by deleting the third region that is unlikely to include a vehicle.

According to the seventh aspect of the present invention, the compression method includes a photographing processing step of acquiring an image including the license plate of the vehicle according to the arrival timing of the vehicle; a division processing step of dividing a first region in which the license plate is likely to be included and a second region in which the license plate is less likely to be included than the first region; and a compression processing step of compressing to a predefined size. In the compression processing step, the compression rates of the first area and the second area are set such that the compression rate of the first area is lower than the compression rate of the second area.

According to the eighth aspect of the present invention, the compression method includes a photographing processing step of acquiring an image including the license plate of the vehicle according to the arrival timing of the vehicle; and a third region in which the vehicle is less likely to be included than in the first region; and a compression processing step for compressing to size. In the compression processing step, the image is compressed by deleting the third area.

According to the ninth aspect of the present invention, a computer (10) of a license plate recognition device (10) capable of acquiring an image including a license plate of a vehicle traveling in a lane, reading license plate information, and storing the image ( 900), the computer (900) is likely to include a photographing processing step of acquiring the image according to the arrival timing of the vehicle, and the license plate of the vehicle among the images. a segmentation processing step of segmenting a first region and a second region in which the license plate is less likely to be included than the first region; and a compression processing step of compressing to . In the compression processing step, the compression rates of the first area and the second area are set such that the compression rate of the first area is lower than the compression rate of the second area.

According to the tenth aspect of the present invention, a computer (10) of a license plate recognition device (10) capable of acquiring an image including a license plate of a vehicle traveling in a lane, reading license plate information, and storing the image ( 900) is configured to cause the computer (900) to perform a photographing processing step of acquiring the image according to the arrival timing of the vehicle; and a third area having a lower possibility of containing the vehicle than the first area; and a compression of compressing the image to a predetermined size smaller than the size of the image. and causing a processing step to be performed. In the compression processing step, the image is compressed by deleting the third area.

According to the license plate recognition device, compression method, and program according to the present invention, the image size can be reduced.

1 is a schematic diagram of a license plate recognition device according to a first embodiment; FIG. It is a figure showing functional composition of a license plate recognition device concerning a 1st embodiment. 4 is a flow chart showing an example of processing of the license plate recognition device according to the first embodiment; It is a figure for demonstrating the function of the license plate recognition apparatus which concerns on 1st Embodiment. It is a figure for demonstrating the function of the license plate recognition apparatus which concerns on 2nd Embodiment. It is a schematic diagram of a license plate recognition device according to a third embodiment. It is a flow chart which shows an example of processing of a license plate recognition device concerning a 3rd embodiment. It is a figure for demonstrating the function of the license plate recognition apparatus which concerns on 3rd Embodiment. 1 is a diagram showing an example of a hardware configuration of a license plate recognition device according to at least one embodiment; FIG.

<First Embodiment>
A license plate recognition device 10 according to a first embodiment of the present invention will be described below with reference to FIGS. 1 to 4. FIG.

(overall structure)
FIG. 1 is a top view of a tollgate according to the first embodiment.
In the present embodiment, an example will be described in which the license plate recognition device 10 is installed at a tollgate on a toll road. A toll booth is a facility provided at the entrance or exit of a toll road for collecting tolls from vehicles traveling on the toll road. As shown in FIG. 1, the toll booth has a lane L connecting a toll road and a general road, and an island I laid on the side of the lane L. As shown in FIG. In the following description, the extending direction of the lane L (±X direction in FIG. 1) is also referred to as the “lane direction”, and the direction horizontally orthogonal to the lane direction (±Y direction in FIG. 1) is also referred to as the “lane width direction”. Describe.

Also, as shown in FIG. 1, a license plate recognition device 10 and a vehicle detector 20 are installed on the island I of the toll gate as equipment used when tolls are collected from the vehicle A. As shown in FIG.

The vehicle detector 20 is provided at a detection position X1 on the most upstream side of the lane L, and detects entry of the vehicle A into the lane L. As shown in FIG. For example, the vehicle detector 20 according to the present embodiment includes a floodlight tower installed on an island I on one side (+Y side) in the lane width direction, and an island I on the other side (−Y side) in the lane width direction. It is a so-called transmissive vehicle detector that consists of a pair with a light receiving tower installed above. The vehicle detector 20 determines whether or not the vehicle A is present at the detection position X1 on the lane L according to whether or not the light-receiving element of the light-receiving tower receives the light projected from the light-projecting element of the light-projecting tower. to detect. Note that in other embodiments, the vehicle detector 20 may be a reflective vehicle detector.

The license plate recognition device 10 is installed on the island I, acquires an image including the license plate of the vehicle A traveling in the lane L, and recognizes the license plate information (size and color of the license plate, characters on the license plate, etc.). read.

As shown in FIG. 1, the license plate recognition device 10 is installed at a position downstream (+X side) in the lane direction of the detection position X1 on the island I on the other side (+Y side) in the lane width direction. The license plate recognition device 10 has a camera 12 facing the detection position X1 on the upstream side (-X side) in the lane direction. Specifically, the camera 12 is arranged so that its optical axis AX intersects the detection position X1 near the center of the lane L in the width direction at a predetermined height from the road surface of the lane L. Thereby, the camera 12 can photograph the front including the license plate of the vehicle A that has arrived at the detection position X1.

In addition, the license plate recognition device 10 performs predetermined image processing on the image captured by the camera 12 to obtain license plate information including the color, size, characters written on the license plate, and the like. read.

(Functional configuration)
FIG. 2 is a diagram showing the functional configuration of the license plate recognition device according to the first embodiment.
As shown in FIG. 2, the license plate recognition device 10 includes a CPU 11, a camera 12, an input/output unit 13, and a storage medium .

The CPU 11 is a processor that controls the overall operation of the license plate recognition device 10, and functions as a photographing processing section 110, a classification processing section 111, a compression processing section 112, and a recognition processing section 113 according to a predetermined program.

The photographing processing unit 110 acquires an image including the license plate of the vehicle A according to the arrival timing of the vehicle A. Specifically, when the photographing processing unit 110 receives a detection signal indicating the presence of the vehicle A from the vehicle detector 20, it determines that the vehicle A has arrived at the detection position X1. An image is captured by the camera 12 and acquired.

The classification processing unit 111 divides the image acquired by the photographing processing unit 110 into a first area that is likely to include the license plate of the vehicle and a second area that is less likely to include the license plate than the first area. Separate the area and the.

The compression processing unit 112 compresses the image acquired by the imaging processing unit 110 to a predetermined size that is smaller than the size of the image. The compression processing unit 112 according to this embodiment sets the compression ratios of the first region and the second region such that the compression ratio of the first region is lower than that of the second region.

The recognition processing unit 113 reads license plate information (size and color of the license plate, characters on the license plate, etc.) of the vehicle A by performing predetermined image processing on the image acquired by the imaging processing unit 110 .

The input/output unit 13 inputs and outputs various information to and from external devices. For example, the input/output unit 13 according to this embodiment receives an input of a detection signal indicating whether or not the presence of the vehicle A is detected from the vehicle detector 20 . Also, the input/output unit 13 may output the image compressed by the compression processing unit 112, the license plate information read by the recognition processing unit 113, and the like to other devices (for example, a server not shown).

Images and the like compressed by the compression processing unit 112 are stored in the storage medium 14 .

Here, the image acquired by the license plate recognition device 10 is used for reading license plate information by the recognition processing unit 113, and is used later for visual characteristics of the vehicle A (character information such as the license plate, vehicle height, vehicle width, shape, passenger's face, etc.), and for the purpose of analyzing the cause of license plate recognition failure. Therefore, it is desirable that the image acquired by the license plate recognition device 10 includes not only the vicinity of the license plate but also the front surface of the vehicle A.

In order for the license plate recognition device 10 to correctly read the characters on the license plate, the characters on the license plate must appear clearly in the image. For example, if the character line width of the license plate is "4 mm", the horizontal resolution of the image including the license plate is desirably double the resolution of "2 mm/pixel". Therefore, in this embodiment, "2 mm/pixel" is set as the prescribed resolution.

The license plate recognition apparatus 10 according to the present embodiment stores and outputs the photographed image in a prescribed image size in a state that satisfies the prescribed resolution and in order to prevent the storage capacity of the image from increasing. Assume that the specified size is, for example, "1400 pixels in the vertical direction and 1000 pixels in the horizontal direction".

However, for example, in this embodiment, the optical axis AX of the camera 12 is inclined with respect to the lane direction (±X direction) as shown in FIG. Therefore, when the detection position X1 is photographed by the camera 12, the width of the horizontal field of view seen from the lane direction (±X direction) is (Field width HR2 in FIG. 1) is increased. Then, if the captured image is simply compressed to a specified size (for example, 1000 pixels) in the horizontal direction, even if the specified resolution is satisfied with the viewing width HR1, the viewing width HR2 will fall below the specified resolution. License plate information may not be read accurately.

By the way, considering that the number of pixels of the camera 12 cannot be arbitrarily set and that the traveling position of the vehicle A fluctuates, the photographed image has a portion where the vehicle A exists and a portion where the vehicle A does not exist ( background such as roads) are included. For example, assume that the photographing area is set based on the vertical range (visual field height) so that the front of the vehicle A is photographed up to the upper end of the vehicle. Then, it is assumed that the outside of the image in the horizontal direction includes a portion where the vehicle A does not exist. Therefore, it is conceivable to reduce the image size in the horizontal direction by compressing the range in which the vehicle A does not exist in the image. Based on such considerations, the license plate recognition device 10 according to the present embodiment executes the processing described below to compress the captured image to a specified size before storing it in the storage medium 14 .

(processing flow)
FIG. 3 is a flow chart showing an example of processing of the license plate recognition device according to the first embodiment.
FIG. 4 is a diagram for explaining functions of the license plate recognition device according to the first embodiment.
A series of processes executed by the license plate recognition device 10 according to the present embodiment will be described below with reference to FIGS. 3 and 4. FIG.

As shown in FIG. 3, first, the photographing processing unit 110 determines whether or not the vehicle A has arrived at the detection position X1 (step S1).

When the imaging processing unit 110 does not receive a detection signal indicating the presence of the vehicle A from the vehicle detector 20, the imaging processing unit 110 determines that the vehicle A has not arrived at the detection position X1 (step S1: NO). In this case, the imaging processing unit 110 waits until it receives the next detection signal from the vehicle detector 20 .

On the other hand, when the photographing processing unit 110 receives a detection signal indicating the presence of the vehicle A from the vehicle detector 20, it determines that the vehicle A has arrived at the detection position X1 (step S1: YES). A is photographed to acquire an image D1 including the license plate of the vehicle A as shown in FIG. 4 (step S2). Here, the size of the image acquired by the imaging processing unit 110 is larger in the horizontal direction than the prescribed size (“1400 pixels in the vertical direction and 1000 pixels in the horizontal direction”).

Next, the classification processing unit 111 obtains an estimated center position C1 (FIG. 4) of the license plate in the horizontal direction of the image D1 (step S3). For example, it is assumed that initial values of the center position of the license plate in the horizontal and vertical directions of the image are stored in advance in the storage medium 14 . If no image is stored in the storage medium 14, the classification processing unit 111 sets the initial value as the estimated center position C1 of the license plate. In addition, when a plurality of images acquired in the past are stored in the storage medium 14, the segmentation processing unit 111 identifies the center position of each of the plurality of images in the horizontal direction of the license plate, as shown in FIG. License plate statistical information F1 representing the appearance frequency of this center position of image D1 is created. Then, the classification processing unit 111 estimates the position with the highest appearance frequency as the center position in the horizontal direction of the license plate, and updates the estimated center position C1. It should be noted that the classification processing unit 111 may previously perform a process of creating the license plate statistical information F1 separately from the series of processes shown in FIG.

Next, as shown in FIG. 4, the division processing unit 111 divides the first area R1 and the second area R2 based on the estimated center position C1 of the license plate (step S4). Specifically, the division processing unit 111 sets a range of a predetermined width W1 and height H1 around the estimated center position C1 as the license plate photographing area NP, which is highly likely to include the license plate. Then, the segmentation processing unit 111 sets the horizontal inner range (the range of width W1 centered on the estimated center position C1) including the license plate photographing area NP in the image D1 as the first area, and sets the license plate photographing area NP. A range horizontally outside (left and right) is set as a second area. Since the estimated center position C1 is variable as described above, the first region R1 and the second region R2 dynamically change each time the image D1 is acquired. Accordingly, the first region R1 and the second region are appropriately set according to the width and shape of the lane L, the traveling position of the vehicle A, and the like.

Next, the compression processing unit 112 compresses the image D1 acquired by the imaging processing unit 110 to a predetermined size that is smaller than the size of the image (step S5). In the example of this embodiment, the image D1 acquired by the imaging processing unit 110 is compressed to a specified size of "1400 pixels in the vertical direction and 1000 pixels in the horizontal direction". The compression processing unit 112 according to the present embodiment does not compress the first region R1 in order to maintain the specified resolution (“2 mm/pixel”), and uses a known compression processing technique for the second region R2. to compress the amount exceeding the specified size. That is, the compression rate of the first region R1 is set lower than the compression rate of the second region R2. Note that if the image D1 has a resolution higher than the specified resolution, the compression processing unit 112 may also compress the first region R1. In this case, the compression processing unit 112 sets the compression rate of the first region R1 lower than the compression rate of the second region R2 so that the resolution of the first region R1 does not fall below the specified resolution. As a result, the compression processing unit 112 can further effectively compress the entire image while minimizing the deterioration of the first area.

Further, the compression processing unit 112 stores the compressed image in the storage medium 14 (step S6). At this time, the compression processing unit 112 may store the compressed image in association with the license plate information read from the image D1 by the license plate recognition device 10 and the acquisition date and time of the image D1.

Note that the recognition processing unit 113 performs predetermined image processing on the image acquired by the photographing processing unit 110 in step S2 in parallel with the processing (steps S3 to S6) of the segmentation processing unit 111 and the compression processing unit 112 described above. Thus, the process of reading the license plate information of the vehicle A is performed (step S7). The license plate information read by the recognition processing unit 113 is output to another device (server or the like) via the input/output unit 13 .

By repeatedly executing the series of processes described above, the license plate recognition device 10 reads the license plate information of the vehicle A arriving at the toll gate, compresses the image D1 including the license plate to a specified size, and stores it. .

Note that the numerical values such as the resolution and image size described above are examples, and can be arbitrarily changed according to the performance of the license plate recognition device 10 and the like.

(Effect)
As described above, the license plate recognition device 10 according to the present embodiment includes the photographing processing unit 110 that acquires the image D1 according to the arrival timing of the vehicle A, and the image D1 that may include the license plate of the vehicle A. a division processing unit 111 that divides the first region R1 having a higher possibility of containing a license plate from the second region R2 having a lower possibility of containing a license plate than the first region R1; , and a compression processing unit 112 that compresses to a predetermined size. The compression processing unit 112 sets the compression ratios of the first region R1 and the second region R2 such that the compression ratio of the first region R1 is lower than that of the second region R2.
By doing so, the license plate recognition device 10 reduces the compression ratio of the first region R1, which is likely to contain a license plate, to suppress deterioration, and also suppresses deterioration of the first region R1, which is likely to contain a license plate. The image size can be effectively reduced by increasing the compression rate of the two regions R2.

Alternatively, the compression processing unit 112 may compress only the second region R2 without compressing the first region R1.
By doing so, the license plate recognition device 10 does not change the first region R1 including the license plate from the image D1 before compression, thereby further suppressing deterioration of the license plate reading accuracy. can. Further, for example, even if the compressed image is used to analyze the cause of a license plate reading failure, the first area including the license plate is not degraded, so the influence on the analysis can be suppressed.

Further, the classification processing unit 111 obtains the estimated center position C1 of the license plate in the image D1 acquired by the photographing processing unit 110 based on a plurality of images acquired in the past, and determines a predetermined range including the estimated center position C1. It is set as the first region R1.
The position of the license plate in the image D1 captured by the license plate recognition device 10 may change depending on the width and shape of the lane L to be captured, the running position of the vehicle A, and the like. However, the license plate recognition device 10 can estimate the center position of the license plate of the vehicle A traveling in the lane L from a plurality of images acquired in the past. The first region R1 and the second region R2 can be separated with high accuracy.

Further, the division processing unit 111 sets a range of a predetermined width W1 in the horizontal direction around the estimated center position C1 as the first region R1.
By doing so, the license plate recognition device 10 can easily set the horizontal range of the first region R1 with reference to the estimated center position of the license plate.

<Second embodiment>
Next, a license plate recognition device 10 according to a second embodiment of the invention will be described with reference to FIG.
Components common to those of the first embodiment are denoted by the same reference numerals, and detailed description thereof is omitted.

In this embodiment, as in the example of the first embodiment, the image captured by the camera 12 has a standard resolution of "2 mm/pixel". In addition, in this embodiment, the image size that satisfies the specified resolution is set with the viewing width as a reference. As a result, there is a possibility that the size (the number of pixels) of the captured image in the height direction of the visual field will be larger than the prescribed size (“1400 pixels in the vertical direction and 1000 pixels in the horizontal direction”). Here, for example, the area above the image in the vertical direction is not used for reading the license plate of vehicle A. It is desirable to reduce the image size.

However, if the entire vertical size is simply compressed to the specified size, the resolution of the area including the license plate will be lower than specified, and the license plate information may not be read accurately.

Based on the above-described problems, the license plate recognition device 10 according to the present embodiment performs the following processes instead of steps S3 to S5 in FIG. As described above, the imaging processing unit 110 according to the present embodiment acquires the image D2 (FIG. 5) having a size of "1550 pixels in the vertical direction and 1000 pixels in the horizontal direction" in step S2.

FIG. 5 is a diagram for explaining the functions of the license plate recognition device according to the second embodiment.
First, as shown in FIG. 5, the classification processing unit 111 obtains the estimated center position C1 of the license plate in the vertical direction of the image D2 (step S3 in FIG. 3). For example, it is assumed that initial values of the center position of the license plate in the horizontal and vertical directions of the image are stored in advance in the storage medium 14 . If no image is stored in the storage medium 14, the classification processing unit 111 sets the initial value as the estimated center position C1 of the license plate. In addition, when a plurality of images acquired in the past are stored in the storage medium 14, the segmentation processing unit 111 identifies the central position of each of the plurality of images in the vertical direction of the license plate, and stores the images as shown in FIG. License plate statistical information F2 representing the appearance frequency of this center position of image D2 is created. Then, the classification processing unit 111 estimates that the position with the highest appearance frequency is the center position in the vertical direction of the license plate, and updates the estimated center position C1. Note that the classification processing unit 111 may previously perform a process of creating the license plate statistical information F2 separately from the series of processes shown in FIG.

Next, as shown in FIG. 5, the division processing unit 111 divides the first area R1 and the second area R2 based on the estimated center position C1 of the license plate (step S4 in FIG. 3). Specifically, the division processing unit 111 sets a range of a predetermined width W1 and height H1 around the estimated center position C1 as the license plate photographing area NP, which is highly likely to include the license plate. Then, the segmentation processing unit 111 sets the vertical inner range (the range of the height H1 centered on the estimated center position C1) including the license plate photographing area NP in the image D1 as the first area, and sets the license plate photographing area as the first area. A range outside (up and down) the NP in the vertical direction is set as the second area. Note that, as shown in FIG. 5, the segmentation processing unit 111 sets the range below the license plate photographing area NP in the vertical direction as the first area, and the range above the license plate photographing area NP in the vertical direction as the second area. may be set as

Next, the compression processing unit 112 compresses the image D2 acquired by the imaging processing unit 110 to a predetermined size that is smaller than the size of the image (step S5 in FIG. 3). In the example of this embodiment, the image D2 acquired by the imaging processing unit 110 is compressed to a specified size of "1400 pixels in the vertical direction and 1000 pixels in the horizontal direction". The compression processing unit 112 according to the present embodiment does not compress the first region R1 in order to maintain the specified resolution (“2 mm/pixel”), and uses a known compression processing technique for the second region R2. to compress the amount exceeding the specified size. That is, the compression rate of the first area is set lower than the compression rate of the second area. Note that if the image D2 has a resolution higher than the specified resolution, the compression processing unit 112 may also compress the first region R1. In this case, the compression processing unit 112 sets the compression rate of the first region R1 lower than the compression rate of the second region R2 so that the resolution of the first region R1 does not fall below the specified resolution. As a result, the compression processing unit 112 can further effectively compress the entire image while minimizing the deterioration of the first area.

(Effect)
As described above, in the license plate recognition device 10 according to the present embodiment, the classification processing section 111 sets a range of a predetermined height in the vertical direction around the estimated center position C1 as the first region R1.
By doing so, the license plate recognition device 10 can easily set the vertical range of the first region R1 with reference to the estimated center position C1 of the license plate.

Further, the classification processing unit 111 may combine the method in the first embodiment and the method in the second embodiment when setting the first region R1 and the second region R2. That is, in step S3 of FIG. 3, the classification processing unit 111 estimates the horizontal and vertical central positions of the license plate based on the license plate statistical information F1 and F2, and updates the estimated central position C1. good too. Further, in step S4 of FIG. 3, the classification processing unit 111 may set the license plate photographing area NP as the first area R1, and set the range outside the license plate photographing area NP as the second area R2. .
By doing so, the area of the second region R2 having a high compression ratio is increased, so that the image can be further compressed. This makes it possible to further reduce the specified image size.

<Third Embodiment>
Next, a license plate recognition device 10 according to a third embodiment of the present invention will be described with reference to FIGS. 6 to 8. FIG.
The same reference numerals are assigned to the components common to the above-described embodiments, and detailed description thereof is omitted.

FIG. 6 is a schematic diagram of a license plate recognition device according to the third embodiment.
As shown in FIG. 6, the license plate recognition device 10 according to the present embodiment is arranged on an island I between two lanes L1 and L2 so that an image can be taken from the front of a vehicle A traveling on the lanes L and L2. 2, the optical axis AX of the camera 12 is arranged so as to face the upstream side (-X side) horizontally with respect to the lane direction. Note that the camera 12 according to the present embodiment has more horizontal pixels than the cameras according to the first and second embodiments so that the entire width direction (±Y direction) of the lane L1 at the detection position X1 is included in the imaging area. It's a big camera. For example, the left and right visual field widths HR3 centered on the optical axis AX of the camera 12 are set according to the distance from the optical axis AX of the camera 12 to the island I on one width direction side (+Y side) of the lane L1. and At this time, the total horizontal field width of the image captured by the camera 12 greatly exceeds the prescribed size ("1400 pixels in the vertical direction and 1000 pixels in the horizontal direction").

Based on such problems, in the license plate recognition device 10 according to this embodiment, the classification processing unit 111 and the compression processing unit 112 each have functions different from those of the above-described embodiments.

The classification processing unit 111 divides the image acquired by the imaging processing unit 110 into a first region that is highly likely to contain the vehicle A, a third region that is less likely to contain the vehicle A than the first region, and to separate.

The compression processing unit 112 deletes the third area and compresses the image.

FIG. 7 is a flow chart showing an example of processing of the license plate recognition device according to the third embodiment.
FIG. 8 is a diagram for explaining the functions of the license plate recognition device according to the third embodiment.
An example of the process of acquiring and storing an image including the license plate of the vehicle A traveling on the lane L1 by the license plate recognition device 10 according to the present embodiment will be described below with reference to FIGS. 7 and 8. FIG.

Here, since vehicle A traveling in lane L1 is to be photographed, the photographing processing unit 110 determines in advance that lane L1 is to be monitored. Then, as shown in FIG. 7, first, the photographing processing unit 110 determines whether or not the vehicle A has arrived at the detection position X1 on the lane L1 (step S11).

When the imaging processing unit 110 does not receive a detection signal indicating the presence of the vehicle A from the vehicle detector 20 on the lane L1, it determines that the vehicle A has not reached the detection position X1 on the lane L1 (step S11). : NO). In this case, the imaging processing unit 110 waits until it receives the next detection signal from the vehicle detector 20 .

On the other hand, when receiving a detection signal indicating the presence of vehicle A from vehicle detector 20 on lane L1, imaging processing unit 110 determines that vehicle A has arrived at detection position X1 on lane L1 (step S11: YES). ), the vehicle A is photographed by the camera 12 to obtain an image D3 including the license plate of the vehicle A as shown in FIG. 8 (step S12). Here, the size of the image acquired by the imaging processing unit 110 is larger than the prescribed size (“1400 pixels in the vertical direction and 1000 pixels in the horizontal direction”).

Next, as shown in FIG. 8, the classification processing unit 111 divides the image D3 into a first region R1 that is more likely to contain the vehicle A, and a first region R1 that is more likely to contain the vehicle A than the first region R1. A low third region R3 is set (step S13). Specifically, the segmentation processing unit 111 sets the range of the predetermined height H2 and width W2 on the lane L1 side (the right side in FIG. 8) of the image D3 as the first region R1, and the other range as the third region R1. Set as region R3. The height H2 and the width W2 are set to "1400 pixels" and "1000 pixels", respectively, according to the specified size. The storage medium 14 stores in advance the initial values of the center position of the vehicle A in each of the lanes L1 and L2. A range is set as a first region R1. Note that the classification processing unit 111 may statistically estimate the center position of the vehicle A from a plurality of images accumulated in the storage medium 14 and overwrite the initial value.

Next, the compression processing unit 112 compresses the image D3 acquired by the imaging processing unit 110 to a predetermined size that is smaller than the size of the image (step S14). In this embodiment, the compression processing unit 112 compresses the image D3 to the specified size by deleting the third area.

Further, the compression processing unit 112 stores the compressed image in the storage medium 14 (step S15). At this time, the compression processing unit 112 may store the compressed image in association with the license plate information read from the image D1 by the license plate recognition device 10 and the acquisition date and time of the image D1.

Note that the recognition processing unit 113 performs predetermined image processing on the image acquired by the imaging processing unit 110 in step S12 in parallel with the above-described processing of the segmentation processing unit 111 and the compression processing unit 112 (steps S13 to S15). Thus, the process of reading the license plate information of the vehicle A is performed (step S16). The license plate information read by the recognition processing unit 113 is output to another device (server or the like) via the input/output unit 13 .

By repeatedly executing the series of processes described above, the license plate recognition device 10 reads the license plate information of the vehicle A arriving at the toll gate, compresses the image D3 including the license plate to a specified size, and stores it. . Although the explanation is omitted, the license plate recognition device 10 performs the same processing for the vehicle A traveling on the lane L2.

(Effect)
As described above, the license plate recognition device 10 according to the present embodiment has the first region R1 in which the vehicle A is likely to be included in the image D3 photographed by the photographing processing unit 110, and the first region R1. A division processing unit 111 that divides the image into a third region R3 that is less likely to include the vehicle A, and a compression processing unit 112 that compresses the image D3 to a predetermined size that is smaller than the size of the image D3. , provided. The compression processing unit 112 deletes the third region R3 and compresses the image D3.
By doing so, the license plate recognition device 10 can greatly reduce the image size by deleting the third region R3 in which the possibility that the vehicle A is included is low. Further, since the compression processing unit 112 does not compress the first region R1, a clear image of the entire vehicle A can be left.

Further, the segmentation processing unit 111 may segment the image D3 into three regions, the first region R1, the second region R2, and the third region R3, using the methods in the first and second embodiments. For example, the division processing unit 111 divides the image D3 into the first region R1 and the third region R3 in step S13 of FIG. Find NP. Then, the division processing unit 111 may further divide the first region R1 into a region outside the license plate photographing region NP as a second region R2.
By doing so, the license plate recognition device 10 can further compress the entire image while suppressing the deterioration of the first region R1, which is likely to include the license plate.

<Hardware configuration>
FIG. 9 is a diagram illustrating an example of a hardware configuration of a license plate recognition device according to at least one embodiment;
The hardware configuration of the license plate recognition device 10 according to this embodiment will be described below with reference to FIG.

A computer 900 includes a CPU 901 , a main memory device 902 , an auxiliary memory device 903 and an interface 904 .

The license plate recognition device 10 described above is implemented in a computer 900 . The operation of each part of the license plate recognition device 10 described above is stored in the auxiliary storage device 903 of each computer 900 in the form of a program. The CPU 901 (CPU 11) reads out a program from the auxiliary storage device 903, develops it in the main storage device 902, and executes the above processing according to the program. Further, the CPU 901 secures a storage area (storage medium 14) in the main storage device 902 for storing various kinds of information obtained and generated along with the processing according to the program. In addition, the CPU 901 secures a storage area for storing data being processed in the auxiliary storage device 903 according to the program.

The computer 900 is connected to an external storage device 910 via an interface 904 , and the storage area may be reserved in the external storage device 910 .

It should be noted that, in at least one embodiment, secondary storage device 903 is an example of non-transitory tangible media. Other examples of non-transitory tangible media include magnetic disks, magneto-optical disks, CD-ROMs, DVD-ROMs, semiconductor memories, etc. that are connected via interface 904 . Moreover, when this program is delivered to the computer 900 via a communication line, the computer 900 receiving the delivery may develop the program in the main storage device 902 and execute the above process.

Also, the program may be for realizing part of the functions described above. Furthermore, the program may be a so-called difference file (difference program) that implements the above-described functions in combination with another program already stored in the auxiliary storage device 903 .

Although the embodiments of the present invention have been described in detail above, the present invention is not limited to these as long as they do not deviate from the technical idea of the present invention, and some design changes and the like are possible. For example, in each of the above-described embodiments, the shooting processing unit 110 of the license plate recognition device 10 shoots at the timing when it receives the detection signal from the vehicle detector 20, but it is not limited to this. In other embodiments, the camera 12 of the license plate recognition device 10 may be configured to shoot all the time. In this case, the photographing processing unit 110 performs known image processing on the image photographed by the camera 12, and if it determines that the vehicle A and the license plate are included, determines that the vehicle A has arrived, and converts the image to the license plate number. You may make it acquire as an image for plate recognition. In this case, the vehicle detector 20 may be omitted.

10 license plate recognition device 11 CPU
110 photographing processing unit 111 processing unit 112 compression processing unit 12 camera 13 input/output unit 14 storage medium 20 vehicle detector

Claims (9)

A license plate recognition device that acquires an image including a license plate of a vehicle traveling in a lane, reads license plate information, and can store the image,
a photographing processing unit that acquires the image according to the arrival timing of the vehicle;
a division processing unit that divides the image into a first area that is likely to include the license plate of the vehicle and a second area that is less likely to include the license plate than the first area; ,
a compression processing unit that compresses the image to a predetermined size that is smaller than the size of the image;
with
Based on the plurality of images acquired in the past, the classification processing unit estimates a position where the center position of the license plate appears most frequently as an estimated position of the license plate, and determines a predetermined range centered on the estimated position. is set as the first region,
The compression processing unit sets the compression ratios of the first region and the second region such that the compression ratio of the first region is lower than the compression ratio of the second region.
License plate recognition device. The segmentation processing unit sets a predetermined range in the horizontal direction centered on the estimated position as the first region,
A license plate recognition device according to claim 1 . The segmentation processing unit sets a predetermined range in the vertical direction centered on the estimated position as the first region,
The license plate recognition device according to claim 1 or 2 . The segmentation processing unit further segments a third region in the image that is unlikely to contain the vehicle,
The compression processing unit deletes the third area and compresses the image.
A license plate recognition device according to any one of claims 1 to 3 . A license plate recognition device that is installed on an island of a lane, acquires an image including the license plate of a vehicle traveling in the lane, reads license plate information, and can store the image,
a photographing processing unit that acquires the image according to the arrival timing of the vehicle;
a division processing unit that divides the image into a first area that is likely to contain the vehicle and a third area that is less likely to contain the vehicle than the first area;
a compression processing unit that compresses the image to a predetermined size that is smaller than the size of the image;
with
The segmentation processing unit estimates a position where the center position of the vehicle appears most frequently as an estimated position of the vehicle based on the plurality of images acquired in the past, and divides a predetermined range centering on the estimated position into the set as the first area,
The compression processing unit deletes the third area and compresses the image.
License plate recognition device. A photographing processing step of acquiring an image including the license plate of the vehicle according to the arrival timing of the vehicle;
a segmentation processing step of segmenting the image into a first region that is likely to contain the license plate of the vehicle and a second region that is less likely to contain the license plate than the first region; ,
a compression processing step of compressing the image to a predefined size smaller than the size of the image;
has
In the segmentation processing step, based on the plurality of images acquired in the past, a position where the central position of the license plate appears most frequently is estimated as the estimated position of the license plate, and a predetermined range centered on the estimated position is determined. is set as the first region,
In the compression processing step, setting the compression ratios of the first region and the second region such that the compression ratio of the first region is lower than the compression ratio of the second region;
Compression method. A photographing processing step of acquiring an image including the license plate of the vehicle according to the arrival timing of the vehicle;
a segmentation processing step of segmenting the image into a first region in which the vehicle is likely to be included and a third region in which the vehicle is less likely to be included than the first region;
a compression processing step of compressing the image to a predefined size smaller than the size of the image;
has
In the segmentation processing step, based on the plurality of images acquired in the past, a position where the center position of the vehicle appears most frequently is estimated as the estimated position of the vehicle, and a predetermined range centered on the estimated position is defined as the set as the first area,
compressing the image by removing the third region in the compression processing step;
Compression method. A program that acquires an image including a license plate of a vehicle traveling in a lane, reads license plate information, and causes a computer of a license plate recognition device that can store the image to function, wherein the computer includes:
a photographing processing step of acquiring the image according to the arrival timing of the vehicle;
a segmentation processing step of segmenting the image into a first region that is likely to contain the license plate of the vehicle and a second region that is less likely to contain the license plate than the first region; ,
a compression processing step of compressing the image to a predefined size smaller than the size of the image;
and
In the segmentation processing step, based on the plurality of images acquired in the past, a position where the central position of the license plate appears most frequently is estimated as the estimated position of the license plate, and a predetermined range centered on the estimated position is determined. is set as the first region,
In the compression processing step, setting the compression ratios of the first region and the second region such that the compression ratio of the first region is lower than the compression ratio of the second region;
program. A program that acquires an image including a license plate of a vehicle traveling in a lane, reads license plate information, and causes a computer of a license plate recognition device that can store the image to function, wherein the computer includes:
a photographing processing step of acquiring the image according to the arrival timing of the vehicle;
A segmentation processing step of segmenting the image into a first region in which the vehicle is likely to be included and a third region in which the vehicle is less likely to be included than the first region;
a compression processing step of compressing the image to a predefined size smaller than the size of the image;
and
In the segmentation processing step, based on the plurality of images acquired in the past, a position where the center position of the vehicle appears most frequently is estimated as the estimated position of the vehicle, and a predetermined range centered on the estimated position is defined as the set as the first area,
compressing the image by removing the third region in the compression processing step;
program.

Images