JP2005030925A - Device, method and program for calculating vehicle information - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To simply and correctly calculate the position and size of a vehicle. <P>SOLUTION: A number plate image extracting part 12 extracts the image of a license plate from an image photographed by a camera 10. The characters in the license plate image are recognized by a license plate estimating part 13, and the number is read. The kind and plate size of vehicle are specified from the number read. An other vehicle information preparing part 15 calculates the position and size of the vehicle from the size of the license plate image, the kind of the vehicle and the plate size. An advance locus prediction part 16 judges the danger of a collision from the steering state of its own vehicle, and the position and size of the other vehicle. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a vehicle information calculation device, a vehicle information calculation method, and a vehicle information calculation program for calculating vehicle information related to a vehicle from an image of the vehicle, and in particular, vehicle information calculation capable of easily and accurately calculating the position and dimensions of the vehicle. The present invention relates to a device, a vehicle information calculation method, and a vehicle information calculation program.
[0002]
[Prior art]
Traditionally, driving a car is dependent on the driver's vision. However, for the driver's vision, the part behind the vehicle body becomes a blind spot, and it is difficult to obtain the vision necessary for parking and the like. In order to supplement the driver's vision, a prior art that attaches an imaging device such as a video camera to a vehicle and can monitor a portion that cannot be directly seen is widely used in large buses and the like.
[0003]
In addition, for the purpose of assisting safe driving of a vehicle, a technique has been devised for grasping the presence or absence of other vehicles around the host vehicle and the positional relationship thereof from image information.
[0004]
For example, Patent Document 1 captures the rear of a vehicle with a camera, displays the image on a monitor, and also displays a predicted driving locus of the vehicle, thereby displaying a blind spot portion behind the vehicle, and parking. Technology that supports driving operations such as
[0005]
Moreover, in patent document 2 and patent document 3, the technique which calculates | requires the distance between vehicles from the image of a number plate is disclosed.
[0006]
[Patent Document 1]
JP 2000-177512 A
[Patent Document 2]
JP-A-10-96626
[Patent Document 3]
Japanese Patent Laid-Open No. 11-166811
[0007]
[Problems to be solved by the invention]
By the way, in the parking assistance apparatus shown by the patent document 1 mentioned above, the prediction progress locus | trajectory of the own vehicle is superimposed on the image image | photographed with the camera. For this reason, the determination of whether or not the vehicle collides with an obstacle in the image relies on the driver's determination, but in order to further simplify the operation by the driver, It is desirable to calculate the dimensions automatically.
[0008]
In particular, if the obstacle is another vehicle, and if it is necessary to perform high-precision operations with poor visibility while avoiding contact with other vehicles, such as parallel parking, a nearby vehicle The positional relationship and dimensions are important information. However, even if the techniques disclosed in Patent Documents 1 and 2 are used, the distance to the other vehicle is only calculated, and the size and position of the other vehicle cannot be obtained accurately.
[0009]
Therefore, how to obtain the dimensions and positional relationship of the vehicle existing in the vicinity of the host vehicle has been an important issue.
[0010]
The present invention has been made to solve the above-described problems in the prior art, and provides a vehicle information calculation device, a vehicle information calculation method, and a vehicle information calculation program capable of easily and accurately calculating the position and dimensions of a vehicle. The purpose is to do.
[0011]
[Means for Solving the Problems]
In order to solve the above-described problems and achieve the object, a vehicle information calculation device according to the invention of claim 1 is a vehicle information calculation device that calculates vehicle information related to the vehicle from an image of the vehicle, the image of the vehicle The position and / or dimensions of the vehicle are calculated from the license plate image extracting means for extracting the license plate image from the reading means, the reading means for reading the contents of the license plate indicated in the license plate image, and the contents of the license plate. Vehicle information calculation means.
[0012]
According to the first aspect of the present invention, the vehicle information calculation device extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the position of the vehicle from the content of the license plate. And / or calculate dimensions.
[0013]
According to a second aspect of the present invention, there is provided the vehicle information calculation device according to the first aspect of the invention, wherein the vehicle information calculation means specifies the type of the vehicle from the contents of the license plate, and determines the vehicle type from the specified type of vehicle. It is characterized in that the dimensions of the vehicle are obtained.
[0014]
According to the second aspect of the invention, the vehicle information calculation device extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the type of the vehicle from the content of the license plate. The vehicle dimensions are calculated from the specified vehicle type.
[0015]
According to a third aspect of the present invention, in the vehicle information calculation device according to the second aspect, the vehicle information calculation means outputs the maximum dimension defined for the specified vehicle type as the dimension of the vehicle. It is characterized by doing.
[0016]
According to the invention of claim 3, the vehicle information calculation device extracts the license plate image from the vehicle image, reads the contents thereof, specifies the type of the vehicle from the contents of the license plate, and sets the specified vehicle type. In contrast, the maximum dimension specified by law is output as the dimension of the vehicle.
[0017]
According to a fourth aspect of the present invention, in the vehicle information calculation device according to the first, second, or third aspect, the vehicle information calculation means specifies the size of the license plate from the contents of the license plate, and specifies The position of the vehicle is calculated from the size of the license plate and the size and shape of the license plate image.
[0018]
According to the fourth aspect of the present invention, the vehicle information calculation device extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the size of the license plate from the content of the license plate. And the position of the vehicle is calculated from the size of the specified license plate and the size and shape of the license plate image.
[0019]
According to a fifth aspect of the present invention, in the vehicle information calculation device according to the fourth aspect, the vehicle information calculation means compares the shape of the license plate image with a prescribed shape of the license plate, A relative angle with the host vehicle is calculated.
[0020]
According to the fifth aspect of the present invention, the vehicle information calculation device extracts a license plate image from a vehicle image, and compares the shape of the license plate image with the prescribed shape of the license plate, so The relative angle with the vehicle is calculated.
[0021]
According to a sixth aspect of the present invention, there is provided the vehicle information calculation device according to any one of the first to fifth aspects, wherein the operation state acquisition means for acquiring the operation state of the host vehicle and the vehicle information calculation means are calculated. It further comprises risk prediction means for predicting a risk of contact between the host vehicle and the vehicle based on vehicle information and the operation state.
[0022]
According to the sixth aspect of the invention, the vehicle information calculation device extracts a license plate image from the vehicle image, calculates the position and / or size of the vehicle from the license plate image, and further acquires the operation state of the host vehicle. Then, the risk of contact between the vehicle and the host vehicle is predicted.
[0023]
According to a seventh aspect of the present invention, there is provided a vehicle information calculation device for calculating vehicle information relating to a vehicle from a vehicle image, wherein the license plate image extraction extracts a license plate image from the vehicle image. And vehicle information calculation means for comparing a shape of the license plate image with a prescribed shape of the license plate and calculating a relative angle between the vehicle and the host vehicle as the vehicle information. To do.
[0024]
According to the seventh aspect of the present invention, the vehicle information calculation device extracts a license plate image from the vehicle image, compares the shape of the license plate image with the prescribed shape of the license plate, and The relative angle is calculated.
[0025]
A vehicle information calculation method according to an eighth aspect of the invention is a vehicle information calculation method for calculating vehicle information related to the vehicle from a vehicle image, wherein the license plate image is extracted from the vehicle image. A process, a reading process of reading the contents of the license plate shown in the license plate image, and a vehicle information calculation process of calculating the position and / or dimensions of the vehicle from the contents of the license plate. Features.
[0026]
According to the invention of claim 8, the vehicle information calculation method extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the position of the vehicle from the content of the license plate. And / or calculate dimensions.
[0027]
The vehicle information calculation method according to the invention of claim 9 is the vehicle information calculation method according to claim 8, wherein the vehicle information calculation step specifies the type of the vehicle from the contents of the license plate, and determines the type of the vehicle from the specified type of vehicle. It is characterized in that the dimensions of the vehicle are obtained.
[0028]
According to the ninth aspect of the invention, the vehicle information calculation method extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the type of the vehicle from the content of the license plate. The vehicle dimensions are calculated from the specified vehicle type.
[0029]
The vehicle information calculation method according to the invention of claim 10 is the vehicle information calculation method according to claim 8 or 9, wherein the vehicle information calculation step specifies the size of the license plate from the contents of the license plate. The position of the vehicle is calculated from the size of the plate and the size and shape of the license plate image.
[0030]
According to the invention of claim 10, the vehicle information calculation method extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the size of the license plate from the content of the license plate. And the position of the vehicle is calculated from the size of the specified license plate and the size and shape of the license plate image.
[0031]
A vehicle information calculation method according to an invention of claim 11 is a vehicle information calculation method for calculating vehicle information relating to the vehicle from a vehicle image, wherein the license plate image extraction extracts a license plate image from the vehicle image. And a vehicle information calculation step of comparing a shape of the license plate image with a prescribed shape of the license plate and calculating a relative angle between the vehicle and the host vehicle as the vehicle information. To do.
[0032]
According to the invention of claim 11, the vehicle information calculation method extracts the license plate image from the vehicle image, compares the shape of the license plate image with the prescribed shape of the license plate, and The relative angle is calculated.
[0033]
A vehicle information calculation program according to a twelfth aspect of the invention is a vehicle information calculation program for calculating vehicle information related to a vehicle from a vehicle image, wherein the license plate image extraction extracts a license plate image from the vehicle image. Performing a procedure, a reading procedure for reading the content of the license plate shown in the license plate image, and a vehicle information calculation procedure for calculating the position and / or dimensions of the vehicle from the content of the license plate. Features.
[0034]
According to the invention of claim 12, the vehicle information calculation program extracts the license plate image from the image of the vehicle, reads the content of the license plate indicated in the license plate image, and determines the position of the vehicle from the content of the license plate. And / or calculate dimensions.
[0035]
According to a thirteenth aspect of the present invention, in the vehicle information calculation program according to the twelfth aspect of the invention, the vehicle information calculation procedure specifies the type of the vehicle from the contents of the license plate, and determines the type of the vehicle from the specified type of vehicle. It is characterized in that the dimensions of the vehicle are obtained.
[0036]
According to the invention of claim 13, the vehicle information calculation program extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the type of the vehicle from the content of the license plate. The vehicle dimensions are calculated from the specified vehicle type.
[0037]
According to a fourteenth aspect of the present invention, in the vehicle information calculation program according to the twelfth or thirteenth aspect of the present invention, the vehicle information calculation procedure specifies the size of the license plate from the content of the license plate. The position of the vehicle is calculated from the size of the plate and the size and shape of the license plate image.
[0038]
According to the fourteenth aspect of the invention, the vehicle information calculation program extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the size of the license plate from the content of the license plate. And the position of the vehicle is calculated from the size of the specified license plate and the size and shape of the license plate image.
[0039]
A vehicle information calculation program according to a fifteenth aspect of the invention is a vehicle information calculation program for calculating vehicle information relating to a vehicle from a vehicle image, and extracting a license plate image from the vehicle image. And a vehicle information calculation procedure for comparing a shape of the license plate image with a prescribed shape of the license plate and calculating a relative angle between the vehicle and the host vehicle as the vehicle information. To do.
[0040]
According to the fifteenth aspect of the present invention, the vehicle information calculation program extracts the license plate image from the vehicle image, compares the shape of the license plate image with the prescribed shape of the license plate, The relative angle of is calculated.
[0041]
DETAILED DESCRIPTION OF THE INVENTION
Exemplary embodiments of a vehicle information calculation device, a vehicle information calculation method, and a vehicle information calculation program according to the present invention will be described below with reference to the accompanying drawings.
[0042]
(Embodiment 1)
FIG. 1 is a schematic configuration diagram showing a schematic configuration of a parking assistance unit according to the first embodiment of the present invention. As shown in the figure, the parking support unit 1 includes a camera 10, a monitor 11, a license plate image extraction unit 12, a license plate estimation unit 13, a license plate database 14, another vehicle information creation unit 15, and a travel locus prediction unit. 16, the output image creation part 17, the steering condition acquisition part 18, and the speaker 19 are provided.
[0043]
The camera 10 captures the rear of the vehicle and supplies the captured image to the license plate image extraction unit 12 and the output image creation unit 17. The license plate image extraction unit 12 extracts a portion corresponding to the license plate from the image taken by the camera 10 and supplies the number plate image to the license plate estimation unit 13 and the other vehicle information creation unit 15 as a license plate image.
[0044]
The license plate estimation unit 13 reads the number of the license plate by recognizing the license plate image, inquires the number of the license plate database 14 for the read number, identifies the type of the vehicle, and sends it to the other vehicle information creation unit 15. Send.
[0045]
The other vehicle information creation unit 15 estimates the size of the vehicle using the type of the vehicle specified by the license plate estimation unit 13 and the license plate image, calculates the position of the vehicle, and transmits the vehicle position to the traveling locus prediction unit 16. To do. Specifically, the other vehicle information creation unit 15 includes a vehicle position calculation unit 15a and a vehicle size estimation unit 15b therein.
[0046]
The vehicle position calculation unit 15a calculates the position of the vehicle based on the size of the license plate (plate size) determined by the type of vehicle and the size of the license plate image. Since the plate size is defined as, for example, “16.5 cm × 33 cm”, “22 cm × 44 cm”, etc., by comparing the size of the license plate image with the specified size, Can be calculated.
[0047]
Furthermore, the positional relationship of the vehicle can be calculated from the shape of the license plate image. For example, if the license plate image has a rectangular shape and the aspect ratio is the same as the aspect ratio of the specified plate size, it can be seen that the host vehicle and the vehicle are facing each other. If the shape of the license plate image is not rectangular, the relative angle between the host vehicle and the vehicle can be calculated from the deformation rate.
[0048]
On the other hand, the vehicle dimension estimation part 15b estimates the dimension of the vehicle decided by the kind of vehicle. The actual size of the vehicle varies from vehicle to vehicle, but the maximum size is determined by the type of large vehicle, small vehicle, cargo vehicle, or the like. Further, since the plate size is a prescribed size, the vehicle size estimation unit 15b can estimate the size of the vehicle by comparing the size of the license plate image with the size of the vehicle image.
[0049]
On the other hand, the steering situation acquisition unit 18 acquires the steering angle of the vehicle and supplies it to the travel locus prediction unit 16. The travel trajectory prediction unit 16 predicts a trajectory that travels when the host vehicle is backed at the steering angle from the vehicle steering angle and the size of the host vehicle, and transmits the predicted trajectory to the output image creation unit 17. Further, the traveling locus prediction unit 16 makes contact or collision between the own vehicle and the other vehicle based on the predicted traveling locus and the position information and dimensional information of the surrounding vehicle created by the other vehicle information creation unit 15. If there is a risk of contact or collision, a warning sound is output through the speaker 19 and the output image creation unit 17 is requested to display a warning message.
[0050]
The output image creation unit 17 causes the monitor 11 to display the image captured by the camera 10 and the traveling locus predicted by the traveling locus prediction unit 16 in a superimposed manner. Further, when the display of the warning message is requested by the progress trajectory prediction unit 16, the output image creation unit 17 further displays a message warning the collision on the monitor 11.
[0051]
A schematic diagram of a vehicle equipped with the parking support unit 1 is shown in FIG. As shown in FIG. 2, the camera 10 is attached to the roof (ceiling) of the vehicle 2 and can capture the rear of the vehicle 2. On the other hand, the monitor 11 is attached in the vicinity of the driver's seat. Since the shooting range 10a of the camera 10 has fewer blind spots than the range that can be directly seen by the driver, the driver can check the state of the rear of the vehicle 2 by checking the monitor 11 when parking the vehicle. I can grasp it.
[0052]
Further, by displaying the traveling locus of the host vehicle on the monitor 11, the driver can visually recognize the progress of the vehicle when the vehicle is backed at the current steering angle, and there is a risk of contact and collision with surrounding vehicles. By giving a warning when there is a vehicle, it becomes possible to drive the vehicle simply and safely.
[0053]
Next, the license plate database 14 will be further described. FIG. 3 is an explanatory diagram for explaining a specific example of the license plate database 14. As shown in the figure, the license plate database 14 has items such as a number, a plate size, and a vehicle body dimension. The number item is used as a key to specify the type of vehicle. After the number is recognized from the license plate image, the type of the vehicle can be specified by searching the license plate database 14 using the number as a key. In other words, the data necessary for the item of the number only needs to be within a range where the type of the vehicle can be specified, and it is not necessary to use all the information of the license plate.
[0054]
Moreover, the item of plate size has shown the size of the number plate decided by the kind of vehicle. For example, if the license plate is a freight vehicle with a gross vehicle weight of 8 tons or more, a maximum loading capacity of 5 tons or more, or a passenger capacity of 30 or more, the size is defined as “22 cm × 44 cm”. Yes.
[0055]
Furthermore, the item of the vehicle body dimension indicates the maximum dimension of the vehicle. Since the vehicle body dimensions are determined by the type of vehicle, the vehicle body dimensions can be read out based on the specified number and used to determine the possibility of contact / collision with the host vehicle.
[0056]
Next, the display on the monitor 11 will be further described. FIG. 4 is a diagram illustrating a specific example of display on the monitor 11. In FIG. 4A, the monitor 11 displays a rear portion of the host vehicle 2, vehicles 31 and 32, a travel locus 41, and a warning display 51.
[0057]
Here, the rear portion of the host vehicle 2 and the vehicles 31 and 32 are images taken by the camera 10. From this image, the license plate image 31a of the vehicle 31 and the license plate image 32a of the vehicle 32 are extracted, and the types of the vehicles 31 and 32 can be specified by character recognition of the license plate images 31a and 32a, respectively. Further, the positions and dimensions of the vehicles 31 and 32 can be obtained from the sizes of the license plate images 31a and 32a.
[0058]
Furthermore, the traveling locus 41 can be obtained from the steering state of the host vehicle.
Here, since the vehicle 31 is on the traveling locus 41 of the own vehicle, the monitor 11 displays a warning display 51 indicating that there is a risk that the own vehicle and the vehicle 31 may contact or collide.
[0059]
On the other hand, in FIG. 4B, the monitor 11 displays the rear portion of the host vehicle 2, the vehicles 31 and 32, and the travel locus 42. The rear part of the host vehicle 2 and the vehicles 31 and 32 are images taken by the camera 10 as in FIG. In addition, as in FIG. 4A, the types, positions, and dimensions of the vehicles 31 and 32 can be obtained from this image.
[0060]
In FIG. 4B, the steering situation is different from the case of FIG. 4A, and the traveling locus 42 is displayed. Since the traveling locus 42 does not overlap the vehicles 31 and 32, the monitor 11 does not display the warning display 51.
[0061]
As described above, the image captured by the camera 10 and the traveling trajectory of the host vehicle are displayed on the monitor 11, and further, the driver is identified and warned by determining whether there is a risk of contact or collision between the host vehicle and another vehicle. Can support the operation. Specifically, the driver operates the vehicle while checking the monitor 11, corrects the steering angle if a warning is displayed, and avoids contact with other vehicles by driving while avoiding the warning. can do.
[0062]
Next, the processing operation of the parking assistance unit 1 will be described. FIG. 5 is a flowchart for explaining the processing operation of the parking support unit 1. As shown in the figure, when the parking support unit 1 detects a shift change to the back gear (step S101), the parking assist unit 1 performs photographing with the camera 10 (step S102).
[0063]
Next, the license plate image extraction unit 12 extracts a license plate image from the image photographed by the camera 10 (step S103). The license plate estimation unit 13 recognizes the number from the license plate image and specifies the type and plate size of the vehicle with reference to the license plate database 14 (step S104). Thereafter, the other vehicle information creation unit 15 calculates the position and dimensions of the vehicle from the license plate image, the type of the vehicle, and the plate size (step S105).
[0064]
Subsequently, the steering status acquisition unit 18 acquires the steering status of the host vehicle (step S106), and the travel locus prediction unit 16 predicts the progress track of the host vehicle based on the steering status (step S107). Further, the traveling locus prediction unit 16 determines whether or not the own vehicle and the other vehicle collide from the traveling locus of the own vehicle and the position and size of the other vehicle (step S108).
[0065]
When the host vehicle and another vehicle collide (step S108, Yes), the traveling locus prediction unit 16 outputs a warning sound from the speaker 19 and requests display of a warning message (step S109).
[0066]
When the host vehicle does not collide with another vehicle (No at Step S108), or after Step S109 is finished, the output image creation unit 17 synthesizes the image captured by the camera 10 and the travel track created by the travel track prediction unit 16. Further, if display of a warning message is requested, a warning message is added to create an output image (step S110). The parking assistance unit 1 displays this output image on the monitor 11 (step S111) and ends the process.
[0067]
In this flow, the shift change to the back gear is used as an opportunity to start photographing because the display is automatically started when the vehicle is brought back. When shooting by the camera and display on the monitor in a state where the driver has not shifted to the back gear, shooting and display may be started with a predetermined operation as a trigger.
[0068]
As described above, in the parking support unit 1 shown in the first embodiment, the license plate image is extracted from the image captured by the camera 10, the number is read by recognizing the license plate, and the number is read. The vehicle type and plate size are specified, the position and dimensions of the vehicle are calculated from the size of the license plate image, the type of vehicle, and the plate size. The sex can be determined and output.
[0069]
In the first embodiment, the calculated position and size of the other vehicle are used only for the determination of the collision. For example, the calculated vehicle position and size data are displayed together with the monitor 11. It may be configured. At this time, the dimensions of the other vehicle itself may be displayed, the distance from the current host vehicle may be displayed, or the distance between the host vehicle and the other vehicle when the vehicle is parked in the current steering situation is predicted. May be displayed.
[0070]
Further, in the first embodiment, the rear of the vehicle is photographed and the rearward state of the vehicle is displayed on the monitor. However, the place where the photographing is performed is not limited to the rear of the vehicle. The side of the vehicle may be photographed, or the front of the vehicle may be photographed. In particular, in a vehicle with many blind spots such as a large freight vehicle, it is desirable to mount a camera corresponding to each blind spot and display images from each camera simultaneously or by switching. When applied to the front of the vehicle, not only the driver's blind spot behind the vehicle and the side, but also overlooked by looking away when traveling forward can be assisted. Specifically, when the position / size of the preceding vehicle is detected, the steering vehicle is automatically operated based on the detected data so as to avoid the preceding vehicle.
[0071]
Further, in the first embodiment, the monitor 11 is used exclusively for parking assistance, but the monitor 11 may be configured to be shared with other in-vehicle devices. For example, you may comprise so that a monitor may be shared with a vehicle-mounted navigation system, a vehicle-mounted audio apparatus, a vehicle-mounted television, or a vehicle-mounted DVD player.
[0072]
By the way, in the first embodiment, the license plate image extraction unit 12, the license plate estimation unit 13, the other vehicle information creation unit 15, the travel trajectory prediction unit 16, and the output image creation unit 17 are each physically processed independently. Although realized, a program having functions equivalent to those of the parking assistance unit 1 can be obtained by realizing these processes by software.
[0073]
In addition, in this Embodiment 1, although the parking assistance unit which utilized the information calculation from the image of a vehicle for parking assistance was demonstrated, utilization of this invention is not limited to this, The information of a vehicle is shown from an image. Can be widely used when required.
[0074]
(Embodiment 2)
In the second embodiment, a case will be described in which an image ahead of the host vehicle is taken and driving assistance is performed based on vehicle information calculated from the image. Even in the case of photographing the front of the host vehicle, the schematic configuration of the apparatus is the same as that of the parking assistance unit 1 shown in the first embodiment, and thus the description thereof is omitted here.
[0075]
FIG. 6 is an explanatory diagram for explaining the calculation and use of vehicle information when photographing the front of the host vehicle. In FIG. 6, the host vehicle 60 captures an imaging range 60 a with a camera that captures the front.
[0076]
Further, another vehicle 61 is traveling in front of the host vehicle 60. The host vehicle 60 identifies the license plate P1 of the other vehicle 61 from the photographed forward image, and estimates the relative distance D1 to the other vehicle 61 and the maximum dimension M1 of the other vehicle 61.
[0077]
Specifically, the relative distance D1 can be estimated from the size of the image of the license plate P1. The maximum dimension M1 can be estimated by reading the content of the license plate P1 by image processing and specifying the type of the other vehicle 61 based on the read content.
[0078]
If the contour of the other vehicle 61 can be extracted from the front image, the vehicle width W1 can also be estimated from the size of the image of the license plate P1 and the size of the contour image of the other vehicle 61.
[0079]
From this relative distance D1 and the maximum dimension M1 (vehicle width W1), it is possible to determine the possibility that the host vehicle 60 and the other vehicle 61 will collide. For example, if the host vehicle 60 changes lanes and moves to the vehicle position 60c, the predicted route 60b is calculated from the speed and steering angle of the host vehicle 60, and the predicted route 60b, the relative distance D1, and the maximum dimension M1 ( The risk of a contact accident may be predicted based on the positional relationship with the vehicle width W1).
[0080]
By the way, when the license plate image is extracted, the positional relationship (relative angle) between the host vehicle and the other vehicle can be known from the shape of the license plate image. Here, the shape and relative angle of the license plate image will be described with reference to FIGS.
[0081]
In FIG. 7, the other vehicle 62 is turning left in front of the host vehicle 60. In the image taken by the host vehicle 60 in this state, as shown in FIG. 8, the image of the license plate P2 of the other vehicle 62 has a substantially trapezoidal shape. The positional relationship (relative angle) between the host vehicle 60 and the other vehicle 62 can be calculated by acquiring the width PX1 and the heights PY1 to PY3 from the image of the license plate P2 and comparing them with a specified plate size. it can.
[0082]
Furthermore, as shown in FIG. 9, if the license plates P3 and P4 of the other vehicles 63 and 64 are image-recognized, a contact accident can be avoided when the host vehicle 60 is parked between the other vehicle 63 and the other vehicle 64. Can be done.
[0083]
As described above, in the second embodiment, the license plate image is extracted from the image ahead of the host vehicle, and the relative position of the vehicle, the maximum dimension (vehicle width), from the size and shape of the license plate image and the type of the vehicle, The positional relationship (relative angle) can be calculated to determine the risk of a collision when the vehicle is running or during a parking operation.
[0084]
As described above, the present invention can acquire information on other vehicles in the vicinity by taking an image of the vicinity of the host vehicle and processing the image, and can be widely used for parking assistance, driving assistance, and the like. It is something that can be done.
[0085]
【The invention's effect】
As described above, according to the invention of claim 1, the vehicle information calculation device extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and reads the content of the license plate. Since the position and / or size of the vehicle is calculated from the vehicle information, a vehicle information calculation device capable of calculating the position and size of the vehicle simply and accurately is obtained.
[0086]
According to the invention of claim 2, the vehicle information calculation device extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the type of the vehicle from the content of the license plate. Since the vehicle size is calculated from the specified vehicle type, there is an effect that a vehicle information calculation device capable of calculating the vehicle size simply and accurately is obtained.
[0087]
According to the invention of claim 3, the vehicle information calculation device extracts the license plate image from the vehicle image, reads the contents thereof, specifies the type of the vehicle from the contents of the license plate, and specifies the specified vehicle type. On the other hand, since the maximum dimension prescribed by law or the like is output as the dimension of the vehicle, there is an effect that a vehicle information calculation device capable of calculating the dimension of the vehicle easily and accurately is obtained.
[0088]
According to the invention of claim 4, the vehicle information calculation device extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the size of the license plate from the content of the license plate. Since the vehicle position is calculated from the specified license plate size and the license plate image size and shape, the vehicle information calculation device capable of calculating the vehicle position easily and accurately is obtained. Play.
[0089]
Further, according to the invention of claim 5, the vehicle information calculation device extracts a license plate image from the vehicle image, and compares the shape of the license plate image with the prescribed shape of the license plate, Since the relative angle with the host vehicle is calculated, there is an effect that a vehicle information calculation device capable of calculating the positional relationship of the vehicle easily and accurately is obtained.
[0090]
According to the invention of claim 6, the vehicle information calculation device extracts a license plate image from the vehicle image, calculates the position and / or size of the vehicle from the license plate image, and further determines the operation state of the host vehicle. Since the risk of obtaining and predicting contact between the vehicle and the host vehicle is predicted, the vehicle information calculation device that calculates the position and dimensions of the vehicle easily and accurately and supports the operation of the vehicle is obtained.
[0091]
According to the invention of claim 7, the vehicle information calculation device extracts the license plate image from the vehicle image, compares the shape of the license plate image with the prescribed shape of the license plate, As a result, the vehicle information calculation device capable of calculating the vehicle positional relationship easily and accurately is obtained.
[0092]
According to the invention of claim 8, the vehicle information calculation method extracts a license plate image from a vehicle image, reads the content of the license plate indicated in the license plate image, and reads the license plate content from the content of the license plate. Since the position and / or dimensions are calculated, it is possible to obtain a vehicle information calculation method capable of easily and accurately calculating the position and dimensions of the vehicle.
[0093]
According to the ninth aspect of the invention, the vehicle information calculation method extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the type of the vehicle from the content of the license plate. Since the vehicle size is calculated from the specified vehicle type, there is an effect that a vehicle information calculation method capable of easily and accurately calculating the vehicle size is obtained.
[0094]
According to the invention of claim 10, the vehicle information calculation method extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the size of the license plate from the content of the license plate. Since the vehicle position is calculated from the specified license plate size and the license plate image size and shape, there is an effect that a vehicle information calculation method capable of easily and accurately calculating the vehicle position is obtained. Play.
[0095]
According to the invention of claim 11, the vehicle information calculation method extracts a license plate image from a vehicle image, compares the shape of the license plate image with a prescribed shape of the license plate, As a result, the vehicle information calculation method capable of easily and accurately calculating the positional relationship of the vehicle is obtained.
[0096]
According to the invention of claim 12, the vehicle information calculation program extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and reads the license plate content from the content of the license plate. Since the position and / or dimensions are calculated, it is possible to obtain a vehicle information calculation program that can easily and accurately calculate the position and dimensions of the vehicle.
[0097]
According to the invention of claim 13, the vehicle information calculation program extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the type of the vehicle from the content of the license plate. Since the vehicle size is calculated from the specified vehicle type, the vehicle information calculation program capable of easily and accurately calculating the vehicle size is obtained.
[0098]
According to the invention of claim 14, the vehicle information calculation program extracts the license plate image from the vehicle image, reads the content of the license plate indicated in the license plate image, and determines the size of the license plate from the content of the license plate. Since the vehicle position is calculated from the specified license plate size and the license plate image size and shape, the vehicle information calculation program capable of easily and accurately calculating the vehicle position can be obtained. Play.
[0099]
According to the invention of claim 15, the vehicle information calculation program extracts the license plate image from the vehicle image, compares the shape of the license plate image with the prescribed shape of the license plate, and the vehicle and the host vehicle. As a result, a vehicle information calculation program capable of easily and accurately calculating the positional relationship of the vehicle is obtained.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram illustrating a schematic configuration of a parking assistance unit according to a first embodiment of the present invention.
FIG. 2 is a schematic view of a vehicle on which the parking support unit shown in FIG. 1 is mounted.
FIG. 3 is an explanatory diagram illustrating a specific example of the license plate database shown in FIG. 1;
4 is a diagram showing a specific example of a display screen displayed on the monitor shown in FIG. 1. FIG.
FIG. 5 is a flowchart for explaining a processing operation of the parking support unit shown in FIG. 1;
FIG. 6 is an explanatory diagram for explaining calculation and use of vehicle information when photographing the front of the host vehicle.
FIG. 7 is an explanatory diagram for explaining the shape and relative angle of a license plate image.
8 is an explanatory diagram illustrating an image captured in the state illustrated in FIG.
FIG. 9 is an explanatory diagram for explaining a case of parking between two vehicles.
[Explanation of symbols]
1 Parking support unit
2 Vehicle
10 Camera
11 Monitor
12 License plate image extraction unit
13 Number plate estimation part
14 License plate database
15 Other vehicle information creation part
15a Vehicle position calculation unit
15b Vehicle size estimation unit
16 Progression trajectory prediction unit
17 Output image creation unit
18 Steering status acquisition unit
19 Speaker
31, 32 Other vehicles
31a, 32a number plate
41 progress track
51 Warning display
60 Own vehicle
60a Shooting range
60b predicted route
60c Vehicle position
61-64 Other vehicles
D1 Relative distance
P1-P4 number plate
PY1 to PY3
PX1 width
M1 maximum dimension
W1 car width

Claims (15)

A vehicle information calculation device that calculates vehicle information related to a vehicle from an image of the vehicle,
License plate image extraction means for extracting a license plate image from the vehicle image;
Reading means for reading the contents of the license plate shown in the license plate image;
A vehicle information calculation device comprising vehicle information calculation means for calculating the position and / or dimensions of the vehicle from the contents of the license plate. The vehicle information calculation device according to claim 1, wherein the vehicle information calculation means specifies the type of the vehicle from the contents of the license plate, and obtains the dimensions of the vehicle from the specified type of vehicle. The vehicle information calculation device according to claim 2, wherein the vehicle information calculation unit outputs a maximum dimension defined for the identified vehicle type as the dimension of the vehicle. The vehicle information calculating means specifies the size of the license plate from the content of the license plate, and calculates the position of the vehicle from the specified size of the license plate and the size and shape of the license plate image. The vehicle information calculation device according to claim 1, 2, or 3. 5. The vehicle information according to claim 4, wherein the vehicle information calculation unit compares the shape of the license plate image with a prescribed shape of the license plate and calculates a relative angle between the vehicle and the host vehicle. Calculation device. An operation state acquisition unit that acquires an operation state of the host vehicle, and a risk prediction unit that predicts a risk of contact between the host vehicle and the vehicle based on the vehicle information calculated by the vehicle information calculation unit and the operation state. The vehicle information calculation device according to claim 1, further comprising: A vehicle information calculation device that calculates vehicle information related to a vehicle from an image of the vehicle,
License plate image extraction means for extracting a license plate image from the vehicle image;
Vehicle information comprising vehicle information calculation means for comparing the shape of the license plate image with a prescribed shape of the license plate and calculating a relative angle between the vehicle and the host vehicle as the vehicle information. Calculation device. A vehicle information calculation method for calculating vehicle information related to a vehicle from an image of the vehicle,
A license plate image extraction step of extracting a license plate image from the vehicle image;
A reading step of reading the contents of the license plate shown in the license plate image;
A vehicle information calculation method comprising: a vehicle information calculation step of calculating a position and / or size of the vehicle from the contents of the license plate. 9. The vehicle information calculation method according to claim 8, wherein in the vehicle information calculation step, the type of the vehicle is specified from the content of the license plate, and the dimensions of the vehicle are obtained from the specified type of vehicle. The vehicle information calculating step specifies the size of the license plate from the content of the license plate, and calculates the position of the vehicle from the specified size of the license plate and the size and shape of the license plate image. The vehicle information calculation method according to claim 8 or 9. A vehicle information calculation method for calculating vehicle information related to a vehicle from an image of the vehicle,
A license plate image extraction step of extracting a license plate image from the vehicle image;
Vehicle information calculation step of comparing the shape of the license plate image with a prescribed shape of the license plate and calculating a relative angle between the vehicle and the host vehicle as the vehicle information. Calculation method. A vehicle information calculation program for calculating vehicle information related to the vehicle from an image of the vehicle,
A license plate image extraction procedure for extracting a license plate image from the vehicle image;
A reading procedure for reading the contents of the license plate shown in the license plate image;
The vehicle information calculation program which performs the vehicle information calculation procedure which calculates the position and / or dimension of the said vehicle from the content of the said license plate. The vehicle information calculation program according to claim 12, wherein the vehicle information calculation procedure specifies the type of the vehicle from the contents of the license plate, and obtains the dimensions of the vehicle from the specified type of vehicle. The vehicle information calculation procedure specifies the size of the license plate from the content of the license plate, and calculates the position of the vehicle from the specified size of the license plate and the size and shape of the license plate image. The vehicle information calculation program according to claim 12 or 13. A vehicle information calculation program for calculating vehicle information related to the vehicle from an image of the vehicle,
A license plate image extraction procedure for extracting a license plate image from the vehicle image;
Vehicle information calculation procedure for comparing a shape of the license plate image with a prescribed shape of the license plate and calculating a relative angle between the vehicle and the host vehicle as the vehicle information. Calculation program.

Images