Image processing device, license plate recognition system, and license plate recognition method

Abstract

To easily and accurately discriminate a plate size of a license plate of a vehicle.SOLUTION: An image processing device of a license plate system includes: a camera that captures a vehicle that travels on a travel path and that is equipped with a license plate having any one of a plurality of types of plate sizes; and the image processing device that discriminates a plate size of the license plate mounted on the vehicle based on an image captured by the camera. The image processing device further includes: a character recognition unit configured to recognize at least one of sizes and positions of characters written on the license plate from the image captured by the camera; and a size discrimination unit for discriminating the plate size of the license plate based on at least one of the sizes and the positions of the characters recognized by the character recognition unit.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present disclosure relates to an image processing device, a license plate recognition system, and a license plate recognition method. [Background technology]

[0002] Toll collection systems equipped with toll gates that automatically collect tolls using wireless communication are used on expressways, toll roads, parking lots, etc. Such toll collection systems automatically collect tolls set according to the vehicle's vehicle type classification, such as passenger cars, large vehicles, and motorcycles. One method for identifying the vehicle's type classification is to use the license plate information attached to the vehicle.

[0003] Patent Document 1 discloses a configuration including a license plate recognition device that photographs a vehicle and acquires image data including the vehicle's license plate, and a vehicle type determination processing unit. In this configuration, the license plate recognition device photographs the vehicle body from the front side when the vehicle reaches a predetermined position in the lane direction. The license plate recognition device acquires license plate information (classification number and color written on the license plate) and the like through image processing of the acquired image data. The vehicle type determination processing unit identifies the vehicle's vehicle type classification based on the license plate information acquired by the license plate recognition device. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-47795 Summary of the Invention [Problem to be solved by the invention]

[0005] Incidentally, license plates with different plate sizes (plate surface sizes) are often given depending on the vehicle type classification, such as standard cars, large vehicles, motorcycles, etc. Therefore, in order to more accurately identify the vehicle type classification of a vehicle, it is conceivable to specify information on the plate size of the license plate in addition to information such as the number of axles and the characters on the license plate. In the configuration described in Patent Document 1, the license plate recognition device photographs the vehicle from the front when a detection signal indicating the entry of a vehicle is output from the approaching vehicle detector, and acquires image data including the vehicle's license plate. Therefore, if a vehicle entering the toll gate is photographed when the entry of the vehicle is detected, the positional relationship between the camera photographing the vehicle and the vehicle's license plate is kept almost constant. Therefore, it is easy to recognize the size of the license plate from the photographed image.

[0006] In contrast, if a camera that captures images at 15 frames per second (15 fps) is used to capture images of a moving vehicle at regular intervals and an attempt is made to recognize the license plate of the vehicle from the captured images, the distance between the camera and the license plate varies depending on the timing of the image capture, making it difficult to determine the size of the license plate in the image.

[0007] Furthermore, when extracting a license plate from a captured image using image recognition processing, it can be difficult to identify the outer edge of the license plate, for example, if the color of the vehicle body and the color of the license plate are similar. In such cases, the area of ​​the license plate extracted by image recognition processing may not match the outer edge of the actual license plate, making it difficult to accurately determine the size of the license plate.

[0008] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a license plate recognition system and a license plate recognition method that can easily and accurately determine the plate size of a vehicle's license plate. [Means for solving the problem]

[0009] In order to solve the above problem, the image processing device of the present disclosure is an image processing device for a license plate recognition system that includes a camera that photographs a vehicle traveling on a road and is equipped with a license plate of any one of multiple plate sizes, and an image processing device that determines the plate size of the license plate attached to the vehicle based on the image photographed by the camera, and includes a character recognition unit that recognizes at least one of the size and position of the characters written on the license plate from the image photographed by the camera, and a size discrimination unit that determines the plate size of the license plate based on at least one of the size and position of the characters recognized by the character recognition unit.

[0010] The license plate recognition method of the present disclosure includes the steps of receiving image data captured by a camera that captures an image of a vehicle traveling on a road and equipped with a license plate of any one of a plurality of plate sizes, recognizing at least one of the size and position of characters written on the license plate from the image captured by the camera, and determining the plate size of the license plate based on at least one of the recognized size and position of the characters. [Effects of the Invention]

[0011] According to the image processing device, license plate recognition system, and license plate recognition method of the present disclosure, the plate size of a vehicle's license plate can be determined easily and accurately. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram illustrating a schematic configuration of a license plate recognition system according to an embodiment of the present disclosure. [Figure 2] 2 is a view of the license plate recognition system of FIG. 1 as seen from the direction in which the road extends. [Figure 3]FIG. 1 is a diagram showing a small plate license plate as an example of multiple types of license plates with different plate sizes. [Figure 4] FIG. 1 is a diagram showing a medium-sized license plate as an example of multiple types of license plates with different plate sizes. [Figure 5] This is a diagram showing a large license plate as an example of multiple types of license plates with different plate sizes. [Figure 6] FIG. 1 is a diagram illustrating a hardware configuration of an image processing device of a license plate recognition system according to an embodiment of the present disclosure. [Figure 7] FIG. 1 is a functional block diagram of an image processing device according to an embodiment of the present disclosure. [Figure 8] 1 is a flowchart illustrating an overall flow of a license plate recognition method according to an embodiment of the present disclosure. [Figure 9] 9 is a flowchart showing a specific flow of the process of determining the plate size of the license plate in FIG. 8. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, embodiments for carrying out the license plate recognition system and the license plate recognition method according to the present disclosure will be described with reference to the accompanying drawings. However, the present disclosure is not limited to these embodiments. (License Plate Recognition System Overview) As shown in Figures 1 and 2, the license plate recognition system 10 constitutes part of a toll collection system that automatically collects tolls on toll roads, parking lots, etc. using wireless communication, for example. The toll collection system collects a toll (charge amount) from a vehicle 100 traveling on a roadway R such as a toll road or parking lot, according to the vehicle type classification of the vehicle 100. The license plate recognition system 10 recognizes a license plate 120 attached to the body 101 of the vehicle 100 traveling on the roadway R to collect a toll according to the vehicle type classification. The license plate recognition system 10 determines the vehicle type classification of the vehicle 100 by recognizing the license plate 120. The toll collection system collects a toll from the vehicle 100 based on the vehicle type classification of the vehicle 100 determined by the license plate recognition system 10. Note that the configuration for collecting a toll from the vehicle 100 in the toll collection system is not limited in any way, and a description thereof will be omitted here.

[0014] The license plate recognition system 10 uses a camera 20 (described later) to capture an image of a license plate 120 of a vehicle 100 traveling on a road R. The license plate recognition system 10 determines the vehicle type classification of the vehicle 100 based on the plate size of the license plate 120 included in the image captured by the camera 20.

[0015] (License plate configuration) In this embodiment, the license plate 120 attached to the vehicle 100 traveling on the road R includes at least three types of license plates used in Japan, as shown in Figures 3 to 5: a small license plate 121 attached to small two-wheeled vehicles, motorcycles, etc.; a medium license plate 122 attached to standard-sized vehicles, medium-sized vehicles, etc.; and a large license plate 123 attached to large vehicles, extra-large vehicles, etc.

[0016] As shown in FIG. 3, the small plate license plate 121 is a horizontally long rectangle with a vertical dimension M11 of 125 mm and a horizontal dimension M12 of 230 mm. The small plate license plate 121 carries information for identifying the vehicle 100, including place name characters T11, a use code T12, and a serial number T13. The place name characters T11 represent the place name in kanji or hiragana. The use code T12 indicates the use of the vehicle 100, such as private use, rental car use, or business use, and is written in a single hiragana character or alphabet. The serial number T13 is written as a number between 1 and 9999. The place name characters T11 are located in the horizontal center of the top of the small plate license plate 121. In the small plate license plate 121, unlike the medium plate license plate 122 (see FIG. 4) and the large plate license plate 123 (see FIG. 5) described later, the usage code T12 is located at the top right of the small plate license plate 121. The serial number T13 is located at the bottom of the small plate license plate 121.

[0017] As shown in FIG. 4, the center plate license plate 122 has a horizontally long rectangular shape with a vertical dimension M21 of 165 mm and a horizontal dimension M22 of 330 mm. The center plate license plate 122 has written thereon, as information for identifying the vehicle 100, a place name character T21, a use code T22, a classification number T23, and a serial number T24. The place name character T21 represents a place name in kanji or hiragana. The use code T22 indicates the use of the vehicle 100, such as personal use, rental car use, or business use, and is written in a single hiragana character. The classification number T23 is a number (character string) assigned according to the type and use of the vehicle, and is represented by a three-digit number (or a combination of numbers and letters). The serial number T24 is a number for identifying the vehicle in combination with the use code T22 and classification number T23, and is represented by a four-digit number. The place name characters T21 and the classification number T23 are arranged side by side in the horizontal center of the upper part of the medium plate license plate 122. The serial number T24 is arranged at the bottom of the medium plate license plate 122. Unlike the small plate license plate 121, the use code T22 is arranged at the bottom of the medium plate license plate 122 to the left of the serial number T24. Furthermore, on the middle license plate 122, the height H21 of the place name characters T21 and the classification number T23 is 40 mm, and the height H22 of the serial number T24 is 80 mm.

[0018] As shown in FIG. 5, the large-plate license plate 123 has a horizontally long rectangular shape with a vertical dimension M31 of 220 mm and a horizontal dimension M32 of 440 mm. The large-plate license plate 123 contains the same information as the medium-plate license plate 122, including place name characters T31, a use code T32, a classification number T33, and a serial number T34, as information for identifying the vehicle 100. The place name characters T31 and the classification number T33 are arranged side by side in the horizontal center of the upper part of the large-plate license plate 123. The serial number T34 is arranged at the bottom of the large-plate license plate 123. Unlike the small-plate license plate 121, the use code T32 is arranged at the bottom of the large-plate license plate 123, to the left of the serial number T34. In addition, in the large license plate 123, the height H31 of the place name characters T31 and the classification number T33 is 40 mm, and the height H32 of the serial number T34 is 120 mm.

[0019] (License Plate Recognition System Overview) As shown in Figures 1 and 2, the license plate recognition system 10 includes a camera 20 and an image processing device 60. The camera 20 and the image processing device 60 are arranged, for example, on a gantry 11 that is arranged so as to straddle the travel path R. The gantry 11 has a gate-like shape when viewed from the direction in which the travel path R extends, and has a pair of support pillars 12 that are arranged on both sides in the width direction of the travel path R, and a cross member 13 that is erected on the pair of support pillars 12. The gantry 11 may be installed so as to straddle multiple travel paths R that are arranged side by side.

[0020] The camera 20 and the image processing device 60 are arranged on the cross member 13. The camera 20 and the image processing device 60 may be supported by a support structure other than the gantry 11. The image processing device 60 may be arranged at a different position from the camera 20. The image processing device 60 and the camera 20 may be integrally configured in a single housing.

[0021] The camera 20 is installed facing the road R. The camera 20 captures images of the road R at predetermined time intervals. In this embodiment, the camera 20 captures images 15 times per second (15 fps), for example. If a vehicle 100 traveling on the road R is present within the camera 20's capture range, the camera 20 captures the image of the vehicle 100 traveling on the road R. The camera 20 captures images of the vehicle 100 equipped with a license plate 120 of any one of a plurality of types of license plates 120 with different plate sizes (small plate license plate 121, medium plate license plate 122, large plate license plate 123). In this embodiment, the image of the vehicle 100 is targeted at a small two-wheeled vehicle, a motorcycle, or other vehicle equipped with a small plate license plate 121. The vehicle 100, such as a small two-wheeled vehicle or a motorcycle, is equipped with a small plate license plate 121 only on the rear side of the vehicle 100 in the traveling direction. For this reason, in this embodiment, the camera 20 photographs the vehicle 100 traveling on the travel path R from behind the vehicle 100 in the traveling direction. However, in addition to one camera 20 that photographs from behind, one camera 20 that photographs from ahead of the vehicle 100 in the traveling direction may also be provided. A plurality of cameras 20 may be installed on the gantry 11. Each camera 20 transfers all image data captured at a predetermined time interval to the image processing device 60, regardless of whether the vehicle 100 is captured or not. The camera 20 transfers the image data to the image processing device 60 via a communication cable (not shown) or various types of wireless communication. On the other hand, in another embodiment, the camera 20 may only capture images from the front in the direction of travel of the vehicle 100. In this case, the camera 20 does not capture images of small license plates, so the image processing device 60 receives only images of medium or large license plates.

[0022] The image processing device 60 determines the plate size of the license plate 120 attached to the vehicle 100 based on the image captured by the camera 20.

[0023] (Hardware configuration diagram) 6, the image processing device 60 is a computer including a CPU (Central Processing Unit) 61, a ROM (Read Only Memory) 62, a RAM (Random Access Memory) 63, a storage 64 such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive), and a signal transmission / reception module 65. The signal transmission / reception module 65 receives image data from each camera 20.

[0024] (Function block diagram) As shown in Figure 7, the CPU 61 of the image processing device 60 executes a program stored in advance in the device, thereby providing the following components: an image data input unit 70, an area extraction unit 71, a character recognition unit 72, a size discrimination unit 73, and a result output unit 75. The image data input unit 70 is a signal transmission / reception module 65 in terms of hardware, and receives data of captured images from each camera 20 . The area extraction unit 71 extracts an area including the license plate 120 attached to the vehicle 100 from the image captured by the camera 20 and received by the image data input unit 70, using an appropriate image processing method. The area extraction unit 71 identifies the position coordinates of the area including the license plate 120 in the image.

[0025] The character recognition unit 72 uses image processing means to recognize the size and position of the characters written on the license plate 120 included in the area extracted by the area extraction unit 71. The character recognition unit 72 recognizes the position of the first character information TA, whose position on the license plate 120 varies depending on the plate size, for multiple types of license plates 120 with different plate sizes. As described above, on the small plate license plate 121, the usage code T12 is located at the top right of the small plate license plate 121. In contrast, on the medium plate license plate 122 and the large plate license plate 123, the usage codes T22 and T32 are located at the bottom, to the left of the serial number T34. The character recognition unit 72 recognizes the positions of the usage codes T12, T22, and T32 as the first character information TA.

[0026] Furthermore, the character recognition unit 72 recognizes the second character information TB and the third character information TC, which have different character sizes, on the license plate 120 for a plurality of types of license plates 120 with different plate sizes. As described above, on the medium-sized license plate 122, the height H21 of the place name characters T21 and the classification number T23 is 40 mm, and the height H22 of the serial number T24 is 80 mm. In other words, the ratio of the height H21 of the place name characters T21 and the classification number T23 to the height H22 of the serial number T24 is 1:2. On the other hand, on the large-sized license plate 123, the height H31 of the place name characters T31 and the classification number T33 is 40 mm, and the height H32 of the serial number T34 is 120 mm. In other words, the ratio of the height H31 of the place name characters T31 and the classification number T33 to the height H32 of the serial number T34 is 1:3. The character recognition unit 72 recognizes the heights H22 and H32 of the serial numbers T24 and T34 as second character information TB. The character recognition unit 72 also recognizes the heights H21 and H31 of the place name characters T21 and T31 and the classification numbers T23 and T33 as third character information TC.

[0027] The size discrimination unit 73 determines the plate size of the license plate 120 based on at least one of the size and position of the characters recognized by the character recognition unit 72. In this embodiment, the size discrimination unit 73 determines the plate size of the license plate 120 based on the positions of the usage codes T12, T22, and T32 as the first character information TA recognized by the character recognition unit 72. That is, if the first character information TA indicating the usage code is located at the upper right (right side) of the license plate 120 in the area containing the license plate 120, the license plate 120 is determined to be a small plate license plate 121. Furthermore, if the first character information TA indicating the usage code is not located at the upper right (right side) of the license plate 120 in the area containing the license plate 120, the license plate 120 is determined to be either a medium plate license plate 122 or a large plate license plate 123.

[0028] When the license plate 120 is determined to be either the medium-sized license plate 122 or the large-sized license plate 123, the size determination unit 73 determines the plate size of the license plate 120 based on the ratio of the size of the second character information TB recognized by the character recognition unit 72 to the size of a specific portion of the license plate 120. In this embodiment, the specific portion is third character information TC (place name characters T21, T31 and classification numbers T23, T33) whose character size on the license plate 120 is different from that of the second character information TB. As described above, in the medium-sized license plate 122, the ratio of the height H21 of the place name characters T21 and classification number T23 to the height H22 of the serial number T24 is 1:2. On the other hand, in the large-sized license plate 123, the ratio of the height H31 of the place name characters T31 and classification number T33 to the height H32 of the serial number T34 is 1:3. The size determination unit 73 determines the plate size of the license plate 120 based on the ratio of the heights of the second character information TB and the third character information TC.

[0029] The result output unit 75 outputs the determination result of the plate size of the license plate 120 made by the size determination unit 73 to the outside, and stores it in an external database 80 (see FIG. 7) provided externally together with the data of the image captured by the camera 20. The result output unit 75 outputs the determination result of the plate size of the license plate 120 to a higher-level server of the toll collection system (not shown). The higher-level server of the toll collection system (not shown) executes processing such as to collect a toll according to the vehicle type classification of the vehicle 100, based on the determination result of the plate size of the license plate 120 output from the result output unit 75.

[0030] (Procedure for processing license plate recognition methods) As shown in FIG. 8, the method S10 for recognizing a license plate 120 according to an embodiment of the present disclosure is realized in the license plate recognition system 10 by the image processing device 60 automatically executing processing based on a preset program based on an image captured by the camera 20. In the license plate recognition system 10, the camera 20 captures an image of the road R at predetermined time intervals. Every time the camera 20 captures an image, it transfers the captured image data to the image processing device 60. Every time the image processing device 60 receives the image data transferred from the camera 20, it executes the following process. In the image processing device 60, the image data input unit 70 receives the captured image data from the camera 20 (step S11).

[0031] Next, in the image processing device 60, the area extraction unit 71 detects the license plate 120 from the image captured by the camera 20 received by the image data input unit 70 using an appropriate image processing method (step S12). Specifically, the area extraction unit 71 extracts an area including the license plate 120 attached to the vehicle 100 from the captured image. The area extraction unit 71 determines whether an area including the license plate 120 has been extracted (step S13). If the license plate 120 is not included in the captured image (No in step S13), the processing ends. If the license plate 120 is included in the captured image (Yes in step S13), the area extraction unit 71 identifies the position coordinates of the area including the license plate 120 in the image, associates the identified position coordinates with the image, and outputs the image to the character recognition unit 72.

[0032] The character recognition unit 72 cuts out an image of the license plate 120 included in the region extracted by the region extraction unit 71 from the image captured by the camera 20 (step S14). Next, the character recognition unit 72 recognizes the size (height) and position of each piece of character information written on the license plate 120 within the region of the license plate 120 in the cut-out image through image recognition processing (step S15). Specifically, the character recognition unit 72 recognizes the position coordinates of the usage codes T12, T22, and T32 as the first character information TA. The character recognition unit 72 also recognizes the heights H22 and H32 of the serial numbers T24 and T34 as the second character information TB. The character recognition unit 72 also recognizes the heights H21 and H31 of the place name characters T21 and T31 and the classification numbers T23 and T33 as the third character information TC. The character recognition unit 72 outputs data on character information such as the position coordinates of the recognized use codes T12, T22, T32, and the heights H21, H31 of the place name characters T21, T31 and the classification numbers T23, T33 to the size discrimination unit 73.

[0033] The size determination unit 73 executes a process of determining the plate size of the license plate 120 based on at least one of the size and position of the characters recognized by the character recognition unit 72 (step S20). As shown in Fig. 9, in the process of determining the plate size of the license plate 120 in step S20, first, the size determination unit 73 receives data related to character information from the character recognition unit 72 (step S21).

[0034] Next, the size discrimination unit 73 determines whether the position of the first character information TA recognized by the character recognition unit 72 is on the right side of the license plate 120 (step S22). As a result, if the first character information TA recognized by the character recognition unit 72 is located on the right side of the license plate 120 (Yes in step S22), the size discrimination unit 73 determines that the license plate 120 is a small plate license plate 121 (step S23). Also, in step S22, if the first character information TA is not located on the right side of the license plate 120, the process proceeds to step S24.

[0035] In step S24, the size discrimination unit 73 determines whether the ratio (TB / TC) between the height of the second character information TB recognized by the character recognition unit 72 and the height of the third character information TC is equal to or greater than a preset threshold value (e.g., 2.5). If the ratio (TB / TC) between the height of the second character information TB recognized by the character recognition unit 72 and the height of the third character information TC is equal to or greater than the threshold value (e.g., 2.5) (Yes in step S24), the size discrimination unit 73 determines that the license plate 120 is a large plate license plate 123 (step S25). If the ratio (TB / TC) between the height of the second character information TB recognized by the character recognition unit 72 and the height of the third character information TC is not equal to or greater than the threshold value (No in step S24), the size discrimination unit 73 determines that the license plate 120 is a medium plate license plate 122 (step S26). In this way, the plate size of the license plate 120 can be determined based on the image captured by the camera 20. Thereafter, the result output unit 75 outputs the determination result regarding the plate size of the license plate 120 to a higher-level server of the toll collection system (not shown).

[0036] (Action and effect) In the license plate recognition system 10 and the method for recognizing a license plate 120 configured as described above, the camera 20 photographs a vehicle 100 equipped with a license plate 120 of any one of a plurality of plate sizes. The character recognition unit 72 recognizes at least one of the size and position of the characters written on the photographed license plate 120. The size determination unit 73 can easily and accurately determine the plate size of the license plate 120 of the vehicle 100 based on at least one of the size and position of the recognized characters. In this embodiment, the character recognition unit 72 has been described as making a distinction based on both the position of the character (step S22 in FIG. 9) and the size of the character (step S24 in FIG. 9), but this is not a limitation in other embodiments. That is, the character recognition unit 72 according to other embodiments may, for example, only make a distinction based on the size of the character, or only make a distinction based on the position of the character.

[0037] Furthermore, by using the character recognition unit 72 to recognize the position of the first character information TA, the position of which on the license plate 120 varies depending on the plate size, the plate size of the license plate 120 can be easily and accurately determined.

[0038] Furthermore, the first character information TA includes usage codes T12, T22, and T32. This makes it possible to easily and accurately determine the plate size of license plate 120 among multiple types of license plate 120 with different plate sizes by recognizing the positions of usage code T12 and usage codes T22 and T32, whose positions on license plate 120 vary depending on the plate size.

[0039] Furthermore, the size determination unit 73 determines the plate size of the license plate 120 based on the ratio between the size of the recognized second character information TB and the size of the specific portion of the license plate 120. This makes it possible to easily and accurately determine the plate size of the license plate 120 based on the ratio between the size of the second character information TB and the size of the specific portion of the license plate 120.

[0040] Also, as the specific portion, third character information TC, the size of the characters on the license plate 120 of which is different from that of the second character information TB, is used. This makes it possible to easily and accurately determine the plate size of the license plate 120 based on the ratio between the second character information TB and the third character information TC.

[0041] Furthermore, the second character information TB is the serial numbers T24 and T34, and the third character information TC is the classification numbers T23 and T33 and the place name characters T21 and T31. This allows the plate size of the license plate 120 to be easily and accurately determined by recognizing the ratio between the size of the serial numbers T24, T34, which differ in size on the license plate 120, and the size of the classification numbers T23, T33 and place name characters T21, T31, among multiple types of license plates 120 with different plate sizes.

[0042] Furthermore, the camera 20 photographs the road R at set time intervals. As a result, in the camera 20 that captures images of the roadway R at set time intervals, the position of the vehicle 100 in the direction in which the roadway R extends varies in the image captured by the camera 20 depending on the timing of the capture. In this case, the distance between the camera 20 and the vehicle 100 varies depending on the timing of the capture by the camera 20. Even in such a case, the plate size of the license plate 120 of the vehicle 100 can be easily and accurately determined based on at least one of the size and position of the characters written on the extracted license plate 120.

[0043] (Other embodiments) The above describes in detail the embodiments of the present disclosure with reference to the drawings, but the specific configuration is not limited to this embodiment, and design changes and the like are also included within the scope that does not deviate from the gist of the present disclosure. Although the number plates 120 are distinguished into three types used in Japan: small number plates 121, medium number plates 122, and large number plates 123, this is not limitative. For example, for number plates used in countries and regions other than Japan, the plate size of the number plates may be determined similarly based on at least one of the position and size of the characters written on the number plates. Furthermore, in the above embodiment, the ratio (TB / TC) between the height of the second character information TB (serial numbers T24, T34) and the height of the third character information TC (classification numbers T23, T33, and place name characters T21, T31) was used to distinguish between the medium plate license plate 122 and the large plate license plate 123, but this is not limited to this. For example, the ratio between the height of the second character information TB (serial numbers T24, T34) and another specific part of the license plate 120, such as the vertical dimensions M21, M31 of the license plate 120, may also be used to distinguish between the medium plate license plate 122 and the large plate license plate 123. Furthermore, the character recognition unit 72 according to the above embodiment has been described as recognizing the heights H21 and H31 of the place name characters T21 and T31 and the classification numbers T23 and T33 as the third character information TC, but is not limited to this. For example, the character recognition unit 72 according to another embodiment may recognize the heights of the usage codes T22 and T32 as the third character information TC. In this case, the size determination unit 73 determines the plate size of the license plate based on the ratio between the second character information TB and the third character information, which is the height of the usage codes T22 and T32. Furthermore, the processing flow of the method S10 for identifying the license plate 120 shown in the above embodiment is merely an example, and the order and content thereof can be changed as appropriate.

[0044] <Additional Notes> The license plate recognition system 10 and the method for recognizing the license plate 120 described in the embodiment can be understood, for example, as follows.

[0045] (1) The image processing device 60 according to the first aspect is an image processing device 60 of a license plate recognition system 10 that includes a camera 20 that photographs a vehicle 100 traveling on a road R and that is fitted with a license plate 120 of any one of a plurality of plate sizes, and an image processing device 60 that determines the plate size of the license plate 120 fitted to the vehicle 100 based on the image photographed by the camera 20, and further includes a character recognition unit 72 that recognizes at least one of the size and position of the characters written on the license plate 120 from the image photographed by the camera 20, and a size discrimination unit 73 that determines the plate size of the license plate 120 based on at least one of the size and position of the characters recognized by the character recognition unit 72.

[0046] In this license plate recognition system 10, a camera 20 photographs a vehicle 100 traveling on a roadway R. The camera 20 photographs the vehicle 100 equipped with a license plate 120 of one of a plurality of plate sizes. Here, the size and position of the characters on the license plate 120 of each vehicle 100 traveling on the roadway R differ depending on the plate size of the license plate corresponding to the vehicle's classification. A character recognition unit 72 recognizes at least one of the size and position of the characters on the photographed license plate 120. A size discrimination unit 73 can easily and accurately determine the plate size of the license plate 120 of the vehicle 100 based on at least one of the size and position of the recognized characters.

[0047] (2) The image processing device 60 according to the second aspect is the image processing device 60 of (1), in which the license plates 120 of multiple types with different plate sizes have first character information TA whose position on the license plate 120 varies depending on the plate size of the license plate, the character recognition unit 72 recognizes the position of the first character information TA written on the license plate 120, and the size discrimination unit 73 discriminates the plate size of the license plate 120 based on the recognized position of the first character information TA.

[0048] As a result, when multiple types of license plates 120 with different plate sizes have first character information TA whose position on the license plate 120 varies depending on the plate size, the plate size of the license plate 120 can be easily and accurately determined based on the position of the first character information TA written on the license plate 120.

[0049] (3) The image processing device 60 according to a third aspect is the image processing device 60 of (2), in which the first character information TA is a use code T12, T22, or T32 indicating the use of the vehicle 100.

[0050] This allows the plate size of the license plate 120 to be easily and accurately determined among multiple types of license plates 120 with different plate sizes, by recognizing the positions of the usage code T12 and the usage codes T22 and T32, which are written in different positions on the license plate 120 depending on the plate size.

[0051] (4) The image processing device 60 according to the fourth aspect is any one of the image processing devices 60 of (1) to (3), wherein the license plates 120 of multiple types with different plate sizes have second character information TB with different character sizes on the license plates 120, the character recognition unit 72 recognizes the size of the second character information TB written on the license plate 120 and the size of a specific part of the license plate 120, and the size discrimination unit 73 discriminates the plate size of the license plate 120 based on the ratio between the recognized size of the second character information TB and the size of the specific part of the license plate 120.

[0052] As a result, when multiple types of license plates 120 with different plate sizes have second character information TB of different sizes on the license plate 120, the plate size of the license plate 120 can be easily and accurately determined based on the ratio between the size of the second character information TB and the size of a specific part of the license plate 120.

[0053] (5) The image processing device 60 according to the fifth aspect is the image processing device 60 of (4), in which the license plates 120 of multiple types with different plate sizes have, as the specific part, third character information TC, the size of the characters on the license plate 120 being different from the second character information TB, the character recognition unit 72 recognizes the ratio between the size of the second character information TB written on the license plate 120 and the size of the third character information TC, and the size discrimination unit 73 discriminates the plate size of the license plate 120 based on the recognized ratio between the second character information TB and the third character information TC.

[0054] As a result, when multiple types of license plates 120 with different plate sizes have second character information TB written in different positions on the license plate 120, the plate size of the license plate 120 can be easily and accurately determined based on the ratio between the second character information TB and the third character information TC.

[0055] (6) The image processing device 60 according to the sixth aspect is the image processing device 60 of (4) or (5), wherein the second character information TB is the serial number T24, T34 of the vehicle 100, and the third character information TC is at least one of the classification number T23, T33 of the vehicle 100, the use code T22, T32, and the characters indicating the place name (place name characters T21, T31).

[0056] This allows the plate size of the license plate 120 to be easily and accurately determined by recognizing the ratio between the size of the serial number T24, T34 of the vehicle 100, which differs in size on the license plate 120, and the size of the classification number T23, T33 of the vehicle 100 and the place name characters T21, T31 that indicate the place name in kanji or hiragana, among multiple types of license plates 120 with different plate sizes.

[0057] (7) A license plate recognition system 10 according to a seventh aspect includes the image processing device 60 according to any one of (1) to (6) and the camera.

[0058] (8) The license plate recognition system 10 according to an eighth aspect is the license plate recognition system 10 of (7), in which the camera 20 photographs the road R at set time intervals.

[0059] As a result, in the camera 20 that captures images of the roadway R at set time intervals, the position of the vehicle 100 in the direction in which the roadway R extends varies in the image captured by the camera 20 depending on the timing of the capture. In this case, the distance between the camera 20 and the vehicle 100 varies depending on the timing of the capture by the camera 20. Even in such a case, the plate size of the license plate 120 of the vehicle 100 can be easily and accurately determined based on at least one of the size and position of the characters written on the extracted license plate 120.

[0060] (9) A method S10 for recognizing a license plate 120 according to a ninth aspect includes the steps of receiving image data captured by a camera 20, the camera 20 capturing an image of a vehicle 100 traveling on a roadway R and having a license plate 120 of any one of a plurality of plate sizes attached thereto; A step of recognizing at least one of the size and position of characters written on the license plate 120 from the image captured by the camera 20; and determining the plate size of the license plate 120 based on at least one of the size and position of the recognized characters.

[0061] This method S10 for recognizing the license plate 120 can easily and accurately determine the plate size of the license plate 120 of the vehicle 100 based on at least one of the size and position of the characters written on the license plate 120. [Explanation of symbols]

[0062] 10 License Plate Recognition System 11 Gantry 12 pillars 13 Cross member 20 Camera 60 Image processing device 61 CPU 62 ROM 63 RAM 64 HDD 65 Signal Transmitting and Receiving Module 70 Image data input section 71 Region extraction part 72 Character recognition section 73 Size discrimination section 75 Result output section 80 External Databases 100 vehicles 101 Body 120 license plate 121 Small license plate 122 Medium plate number plate 123 Large license plate R driving route T11, T21, T31 Place name characters T12, T22, T32 usage code T13, T24, T34 serial numbers T23, T33 classification number TA First character information TB Secondary Textual Information TC Third Character Information

Claims

1. An image processing device of a license plate recognition system including a camera that photographs a vehicle traveling on a roadway and that is fitted with a license plate of any one of a plurality of plate sizes, and an image processing device that determines the plate size of the license plate fitted to the vehicle based on the image photographed by the camera, a character recognition unit that recognizes at least one of the size and position of characters written on the license plate from the image captured by the camera; a size determination unit that determines the plate size of the license plate based on at least one of the size and position of the characters recognized by the character recognition unit; The plurality of types of license plates having different plate sizes have second character information in which the character size on the license plate differs for each plate size, and third character information in which the character size on the license plate is the same regardless of the plate size, The character recognition unit recognizes the size of the second character information and the size of the third character information written on the license plate, The size determination unit determines the plate size of the license plate based on the ratio between the size of the recognized second character information and the size of the recognized third character information. Image processing device.

2. The plurality of types of license plates having different plate sizes have first character information whose position on the license plate varies depending on the plate size, the character recognition unit recognizes the position of the first character information written on the license plate, The size determination unit determines the plate size of the license plate based on the position of the recognized first character information. The image processing device according to claim 1 .

3. The first character information is a usage code indicating the usage of the vehicle. The image processing device according to claim 2 .

4. the second character information is a serial number of the vehicle, The third character information is at least one of a classification number of the vehicle, a usage code, and characters indicating a place name. The image processing device according to claim 1 .

5. The image processing device according to any one of claims 1 to 4, The camera; A license plate recognition system comprising:

6. The camera takes pictures of the road at set time intervals.

6. The license plate recognition system of claim 5.

7. receiving image data captured by a camera that captures an image of a vehicle traveling on a roadway and having a license plate of any one of a plurality of plate sizes attached thereto; A step of recognizing at least one of the size and position of characters written on the license plate from the image captured by the camera; and determining a plate size of the license plate based on at least one of the size and position of the recognized characters; The plurality of types of license plates having different plate sizes have second character information in which the character size on the license plate differs for each plate size, and third character information in which the character size on the license plate is the same regardless of the plate size, In the step of recognizing at least one of the size and position of characters written on the license plate, the size of the second character information and the size of the third character information written on the license plate are recognized, In the step of determining the plate size of the license plate, the plate size of the license plate is determined based on a ratio between the size of the recognized second character information and the size of the recognized third character information. How to recognize license plates.

Images