JP2014182007A - Inter-vehicular distance measuring apparatus and inter-vehicular distance measuring method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an inter-vehicular distance measuring apparatus and an inter-vehicular distance measuring method, capable of securing practical accuracy and inexpensively measuring an inter-vehicular distance to a forward vehicle.SOLUTION: A self vehicle 2 includes a camera 4 for photographing a forward vehicle 3, and a control unit 5 for calculating an inter-vehicular distance to the forward vehicle 3 on the basis of an image photographed by the camera 4 . The control unit 5 calculates the inter-vehicular distance (a reference distance) between the self vehicle 2 and the forward vehicle 3 on the basis of a size of the number plate of the forward vehicle 3 included in an approach image photographed by the camera 4 when approaching to the forward vehicle 3. Further, the control unit 5 calculates the vehicle width (a basic dimension) of the forward vehicle 3 in the approach image. On and after measuring forward vehicle information of the reference distance and the basic dimension, the inter-vehicular distance from the forward vehicle 3 is calculated from the vehicle width of the forward vehicle 3 in the image photographed by the camera 4, using the forward vehicle information.

Description

  The present invention relates to an inter-vehicle distance measuring device and an inter-vehicle distance measuring method for measuring an inter-vehicle distance between a host vehicle and a preceding vehicle.

  2. Description of the Related Art Conventionally, there has been known a method for photographing a distance measurement target with a camera (photographing means) and measuring the distance from the shooting position to the target based on the size (size) of the target shown in the captured image. (See, for example, Patent Document 1). In the method of Patent Document 1, the number of pixels (dimensions) Fs of an object shown in an image at the reference distance Ls is checked in advance. Then, the distance L to the object is obtained from L = (Ls × Fs) / F from the reference distance Ls, the number of pixels Fs, and the number of pixels F of the current object.

  By utilizing the technique of this Patent Document 1, the distance between the host vehicle and the preceding vehicle can be measured. That is, a camera that captures the vehicle ahead is mounted on the host vehicle, and the relationship between the dimension of the license plate of the vehicle ahead and the inter-vehicle distance reflected in the image captured by the camera is examined in advance. Then, the current inter-vehicle distance can be obtained from the relationship and the size of the license plate reflected in the current image. Further, since the license plate has a rated size regardless of the vehicle type, the distance between the vehicles can be measured by using the size of the license plate regardless of the vehicle type of the preceding vehicle.

JP 2008-14825 A

  However, since the license plate is not so large, the license plate may not be recognized from the image in a scene where the distance from the vehicle ahead is long or a night scene. Even if it can be recognized, the change in the size (number of pixels) of the license plate in the image with respect to the change in the inter-vehicle distance is very small, so the measurement accuracy of the inter-vehicle distance is extremely inferior. In addition, depending on the camera performance, an expensive camera is required to secure practical accuracy. In addition, although it is possible to measure the distance between vehicles using a stereo camera or laser light, in the case of the stereo camera method, it is necessary to prepare multiple cameras. In the case of the laser light method, distance measurement using laser light is possible. Since it is necessary to prepare an apparatus, it becomes expensive.

  The present invention has been made in view of the above problems, and it is an object of the present invention to provide an inter-vehicle distance measuring device and an inter-vehicle distance measuring method capable of ensuring practical accuracy and measuring an inter-vehicle distance with a preceding vehicle at low cost. To do.

In order to solve the above-mentioned problem, the inter-vehicle distance measuring device of the present invention is mounted on a vehicle (2) including an imaging means (4) for imaging the preceding vehicle (3, 31, 32),
Recognizing means (S12) for recognizing the number plate (35) of the preceding vehicle from the approaching image (81) which is an image taken by the photographing means when the vehicle approaches the preceding vehicle;
First distance calculating means (S14) for calculating, as a reference distance, an inter-vehicle distance between the vehicle and the preceding vehicle based on the size of the license plate recognized by the recognition means;
Dimension calculation means (S15) for calculating the dimension of the preceding vehicle on the approaching image as a reference dimension;
Second distance calculation for calculating an inter-vehicle distance between the vehicle and the preceding vehicle from the dimension of the preceding vehicle in the image captured by the imaging means, using forward vehicle information that is information on the reference distance and the reference dimension. Means (S23, S35);
It is characterized by providing.

Further, the inter-vehicle distance measurement method of the present invention includes a proximity photographing step (S11) for photographing the preceding vehicle when the vehicle approaches the preceding vehicle (3, 31, 32),
A recognition step (S12) for recognizing the number plate (35) of the preceding vehicle from the approaching image (81) photographed in the approaching photographing step;
A first distance calculating step (S14) for calculating, as a reference distance, an inter-vehicle distance between the vehicle and the preceding vehicle based on a size on the image of the license plate recognized in the recognition step;
A dimension calculating step (S15) for calculating the dimension of the preceding vehicle on the approaching image as a reference dimension;
After obtaining the forward vehicle information which is information of the reference distance and the reference dimension, the forward vehicle (312) on the image (82) where the forward vehicle is newly photographed and the forward vehicle is newly photographed. ) And a second distance calculating step (S23, S35) for calculating an inter-vehicle distance between the vehicle and the preceding vehicle from the preceding vehicle information,
It is characterized by including.

  In the present invention, an imaging means for imaging the vehicle ahead is mounted on the vehicle, and the inter-vehicle distance from the dimension on the license plate image shown in the image taken by the imaging means when the vehicle approaches the vehicle ahead (image when approaching) (Reference distance) is calculated, and the dimension (reference dimension) of the preceding vehicle on the approaching image is calculated. And this invention calculates the inter-vehicle distance from the dimension of the front vehicle in an image using the information (front vehicle information) of these reference distances and reference dimensions. Thus, once the forward vehicle information is obtained, the inter-vehicle distance can be calculated from the size of the forward vehicle that is larger than the license plate, so it is a scene that is far from the forward vehicle or a night scene. Can measure the inter-vehicle distance with practical accuracy. Further, since it is not necessary to recognize the license plate except when approaching, an inexpensive photographing means can be used. That is, the inter-vehicle distance can be measured at a low cost. In addition, when the vehicle type of the front vehicle is different, the size of the front vehicle shown in the image changes even at the same inter-vehicle distance. However, in the present invention, the vehicle type of the front vehicle (actual size) when approaching the front vehicle. ) Is obtained by measurement, the inter-vehicle distance can be measured regardless of the type of vehicle in front.

It is the figure which showed the structure of the inter-vehicle distance measurement system. It is a flowchart of the information measurement process which measures front vehicle information. It is the assumption scene where an information measurement process is performed, and is a figure of the scene where the own vehicle approached the front vehicle. It is the figure which illustrated the image acquired by S11 of FIG. It is a conceptual diagram of the database of the image memorize | stored in the server and front vehicle information. It is a flowchart of the 1st distance measurement process which measures the distance between vehicles using the measured front vehicle information. It is an assumption scene where the 1st distance measurement processing is performed, and is a figure of a scene which performs follow-up measurement of the inter-vehicle distance of the front vehicle from which forward vehicle information was obtained. It is the figure which illustrated the image acquired by S21 of FIG. It is a flowchart of the 2nd distance measurement process which measures the distance between vehicles using the front vehicle information acquired from the server. It is an assumption scene where the 2nd distance measurement processing is performed, and the self-vehicle and the preceding vehicle are separated, and it is a figure of the scene where the preceding vehicle information of the preceding vehicle is not measured. It is a flowchart of the process which CPU of a server performs.

  Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows the configuration of an inter-vehicle distance measurement system to which the present invention is applied. The inter-vehicle distance measurement system 1 in FIG. 1 is a system that measures the inter-vehicle distance between the host vehicle 2 and its front vehicle 3. The inter-vehicle distance measurement system 1 includes an in-vehicle device 10 mounted on the host vehicle 2 and a server 20 installed outside.

  The vehicle-mounted device 10 includes a camera 4, a communication unit 6, and a control unit 5 connected thereto. For example, the camera 4 is attached to the back side of the rear mirror of the host vehicle 2 and continuously captures the front of the host vehicle 2 while the host vehicle 2 is traveling, that is, the front traveling in front of the host vehicle 2. This is a camera for photographing the vehicle 3. The camera 4 can also be used as a camera for a driving recorder that records a travel history of the host vehicle 2, for example. An image captured by the camera 4 is sent to the control unit 5.

  The communication unit 6 includes an antenna, a modulation circuit, a demodulation circuit, and the like, and is a part that performs wireless communication (transmission / reception) of various information with the server 20 (strictly, the communication unit 23 of the server 20). That is, the communication unit 6 modulates the information sent from the control unit 5 and transmits it to the server 20, receives the information from the server 20, demodulates the received information, and sends it to the control unit 5.

  The control unit 5 is a computer composed of a CPU, a ROM, a RAM, and the like, and based on the image taken by the camera 4 and information from the server 20 received by the communication unit 6, the host vehicle 2 and the preceding vehicle 3. The distance between the vehicle and the vehicle is calculated. Specifically, the control unit 5 calculates the inter-vehicle distance based on the vehicle width (number of vehicle width pixels) of the front vehicle 3 in the image captured by the camera 4. Details of the calculation method of the inter-vehicle distance will be described later. Further, the control unit 5 includes a memory 51 such as an EEPROM or a flash memory, and the memory 51 stores vehicle information to be described later. In addition, the memory 51 stores in advance a relationship between a license plate dimension (for example, a lateral width) and an inter-vehicle distance in an image captured by the camera 4. Specifically, when the number of horizontal pixels (number of reference horizontal pixels) of the license plate in the image captured by the camera 4 when the inter-vehicle distance between the host vehicle 2 and the preceding vehicle 3 is a predetermined reference distance L0 is D0. The memory 51 stores the reference distance L0 and the reference width pixel number D0.

  The server 20 includes a storage unit 22, a communication unit 23, and a CPU 21 connected thereto. The storage unit 22 is a part that stores various types of information such as a hard disk drive. In this embodiment, the storage unit 22 includes information (details of the information) sent from each vehicle 2 on which the vehicle-mounted device 10 is mounted. Is stored).

  The communication unit 23 includes an antenna, a modulation circuit, a demodulation circuit, and the like, and wirelessly communicates (transmits and receives) various information with each vehicle 2 (strictly speaking, the communication unit 6 of the on-vehicle device 10) provided with the on-vehicle device 10. It is a part to do. That is, the communication unit 23 receives information from each vehicle 2, demodulates the received information, sends the information to the CPU 21, modulates the information sent from the CPU 21, and transmits the information to the specific vehicle 2. The communication method between the communication unit 6 of the vehicle-mounted device 10 and the communication unit 23 of the server 20 may be any communication method such as communication using the Internet or communication using a mobile phone line.

  The CPU 21 performs processing such as reading forward vehicle information necessary for measurement of the inter-vehicle distance between the vehicle 2 and the forward vehicle 3 from the storage unit 22 and causing the communication unit 23 to transmit the read forward vehicle information. Details of processing performed by the CPU 21 will be described later.

  Next, a method for measuring the inter-vehicle distance between the host vehicle 2 and the preceding vehicle 3 will be described. As described above, the control unit 5 calculates the inter-vehicle distance from the forward vehicle 3 based on the vehicle width of the forward vehicle 3 in the image captured by the camera 4. For that purpose, the control part 5 needs to have the front vehicle information which showed the relationship between the vehicle width pixel number of the front vehicle 3 reflected in the image, and the distance between vehicles. This forward vehicle information varies depending on the vehicle type of the forward vehicle 3. That is, even if the distance between the vehicles is the same, the number of vehicle width pixels of the preceding vehicle changes if the vehicle type is different. Therefore, even if the forward vehicle information of a specific vehicle type is checked in advance, the forward vehicle information cannot be used for measuring the inter-vehicle distance with the forward vehicle of another vehicle type.

  Then, the control part 5 performs the information measurement process which measures the front vehicle information of the front vehicle which is actually driving ahead in the scene which measures the distance between vehicles. Then, when the forward vehicle information is obtained by the information measurement process, the first distance measurement process for measuring the inter-vehicle distance from the vehicle width using the forward vehicle information is performed thereafter. Hereinafter, details of the information measurement process and the first distance measurement process will be described. In addition, prior to the information measurement process, the control unit 5 tries to acquire the forward vehicle information from the server 20 and, when the forward vehicle information can be acquired, the inter-vehicle distance is measured using the forward vehicle information. Two-distance measurement processing is performed. Then, the information measurement process is executed when the vehicle information ahead cannot be acquired from the server 20 in the second distance measurement process. Here, for convenience of explanation, the information measurement process and the first distance measurement process will be described, and then the second distance measurement process will be described.

  FIG. 2 shows a flowchart of the information measurement process. As shown in FIG. 3, it is assumed that the measurement of the forward vehicle information by this information measurement process is performed when the host vehicle 2 approaches the forward vehicle 31 so that the license plate of the forward vehicle reflected in the image can be recognized. ing. More specifically, it is determined that the host vehicle 2 has sufficiently approached the forward vehicle 31 when the recognized number plate of the forward vehicle 31 becomes a predetermined size or more on the image.

  When the processing of FIG. 2 is started, first, an image taken by the camera 4 is acquired from the camera 4 (S11). FIG. 4 illustrates the image acquired in S11. An image 81 in FIG. 4 includes a region 311 (hereinafter simply referred to as a forward vehicle 311) corresponding to the forward vehicle 31 (see FIG. 3).

  Next, recognition of the license plate 35 (see FIG. 4) of the forward vehicle 311 is attempted from the image 81 acquired in S11 (S12). Any known method may be used for recognition of the license plate based on the image. For example, the method described in Japanese Patent Application Laid-Open No. 2012-63869 can be used. That is, the template image of the license plate is stored in advance in the memory 51 (see FIG. 1), and the image area 35 (see FIG. 4) of the template image stored in the memory 51 is selected from the images 81 acquired in S11. Explore.

  Next, it is determined whether the license plate can be recognized from the image (S13). If it cannot be recognized (S13: No), the process of the flowchart of FIG. 2 is terminated. In this case, it is not the scene shown in FIG. 3, that is, the scene when approaching the preceding vehicle, but the scene where the distance between the host vehicle and the preceding vehicle is far and the number plate displayed in the image is too small, or at night It is assumed that the license plate is unclear. In this case, at the next execution of the process of FIG. 2, an attempt is made to recognize the license plate again.

If the license plate can be recognized (S13: Yes), the inter-vehicle distance from the preceding vehicle is calculated by the same method as that described in Patent Document 1 based on the recognized size of the license plate (S14). . Specifically, the number D of horizontal pixels of the license plate 35 (see FIG. 4) is calculated from the recognized license plate 35. Then, based on the calculated width pixel number D, the reference distance L0 and the reference width pixel number D0 stored in the memory 51, an inter-vehicle distance L1 from the preceding vehicle 31 (see FIG. 3) according to the following formula 1. Is calculated. The following formula 1 is the same as the formula described in Patent Document 1.
L1 = (L0 × D0) / D (Expression 1)

  Next, the vehicle width W1 (see FIG. 4) of the forward vehicle 311 in the image 81 acquired in S11 is calculated (S15). Specifically, in S15, the forward vehicle 311 is first recognized from the image 81. Specifically, for example, a template image of a rear view of a general vehicle is stored in the memory 51 in advance, and an area 311 (vehicle ahead) that matches the template image is searched from the image 81. Then, the number of pixels W1 between the left end and the right end of the recognized forward vehicle 311 is calculated as the vehicle width. The inter-vehicle distance L1 (reference inter-vehicle distance) calculated in S14 and the vehicle width W1 (reference vehicle width) calculated in S15 are the forward vehicle information for the forward vehicle 31 in FIG.

  Next, the image acquired in S11 and the forward vehicle information (reference inter-vehicle distance L1 and reference vehicle width W1) calculated in S14 and S15 are stored in the memory 51 and via the communication unit 6 (see FIG. 1). To the server 20 (S16). When the image and the preceding vehicle information are stored in the memory 51, they are stored in association with each other. 2 is executed at each measurement of the preceding vehicle information, and the image and the preceding vehicle information are stored in the memory 51 in S16, whereby a plurality of types of images and the preceding vehicle information are accumulated in the memory 51. Will be. Thereafter, the processing of the flowchart of FIG.

  The image and the forward vehicle information transmitted to the server 20 in S16 are received by the communication unit 23 (see FIG. 1) of the server 20. Then, the CPU 21 associates and stores (accumulates) the image received by the communication unit 23 and the preceding vehicle information in the storage unit 22. Here, FIG. 5 shows a conceptual diagram of the database 220 of images and forward vehicle information stored in the storage unit 22. The database 220 is provided with an image storage column 221 for storing each image X transmitted from each vehicle on which the vehicle-mounted device 10 is mounted, and an information storage column 222 for storing each preceding vehicle information Y. . Each column of the image storage column 221 and each column of the information storage column 222 are associated with each other. Since forward vehicle information of various vehicle types is sent from each vehicle, the forward vehicle information for each vehicle type is accumulated in the database 220.

  After measuring the forward vehicle information by the process of FIG. 2, the control unit 5 uses the forward vehicle information (measurement information) to measure the inter-vehicle distance from the vehicle width of the forward vehicle in the image. Transition to processing. Here, FIG. 6 shows a flowchart of the first distance measurement processing. In the process of FIG. 6, as shown in FIG. 7, a scene in which the inter-vehicle distance with the forward vehicle 31 (the same forward vehicle 31 as that in FIG. 3) corresponding to the forward vehicle information measured in the process of FIG. 2 is continuously measured. Is assumed to be executed. Note that the process of FIG. 6 is repeatedly executed at regular intervals while measuring the distance between the vehicle and the vehicle ahead.

  When the processing of FIG. 6 is started, first, an image captured by the camera 4 is acquired from the camera 4 (S21). FIG. 8 illustrates the image acquired in S21. The image 82 in FIG. 8 includes a region 312 corresponding to the forward vehicle 31 (hereinafter simply referred to as the forward vehicle 312). If the distance between the host vehicle 2 and the front vehicle 31 is greater in the scene of FIG. 7 than the scene of FIG. 3, the size of the front vehicle 312 in FIG. 8 is the size of the front vehicle 311 in FIG. The license plate of the front vehicle 312 is difficult to recognize. In the method of measuring the inter-vehicle distance from only the dimensions of the license plate, the inter-vehicle distance cannot be measured in the scene of FIG. 7, or even if it can be measured, a low-accuracy inter-vehicle distance is obtained.

  Next, the latest forward vehicle information stored in the memory 51 in the previous S16 is acquired from the memory 51 (S22). Since the image is also stored in the memory 51 in the previous S16, in S22, it is the same as the preceding vehicle 312 shown in the image 82 (see FIG. 8) acquired in S21 from the images stored in the memory 51. Search for an image in which the vehicle ahead of the shape (same model) is shown (pattern matching is performed). Then, the forward vehicle information stored in association with the searched image may be acquired from the memory 51. As a result, during the follow-up measurement of the inter-vehicle distance with the preceding vehicle 31 in FIG. 7, even if another vehicle has interrupted between the host vehicle 2 and the preceding vehicle 31, the interruption of the other vehicle is released. Thereafter, the measurement of the inter-vehicle distance from the forward vehicle 31 can be resumed without re-measuring the forward vehicle information according to FIG. In S22, when pattern matching with the image stored in the memory 51 is not performed, that is, when the latest front vehicle information is simply read from the memory 51, only the front vehicle information is stored in the memory 51 in the previous S16. You may make it memorize.

Next, using the forward vehicle information acquired in S22, the distance between the vehicle width W2 of the forward vehicle 312 (see FIG. 8) and the forward vehicle 312 (the forward vehicle 31 in FIG. 7) in the image 82 acquired in S21. The distance is calculated (S23). Specifically, in S23, the forward vehicle 312 is first recognized from the image 82. As the recognition method, as in S15, for example, a general vehicle template image may be stored in the memory 51, and the forward vehicle 312 may be searched from the image 82 by pattern matching with the template image. Next, the pixel number W2 between the left end and the right end of the recognized forward vehicle 312 is calculated as the vehicle width. Then, based on the calculated vehicle width W2 (number of pixels) and the forward vehicle information (reference inter-vehicle distance L1, reference vehicle width W1) acquired in S22, the inter-vehicle distance L with the forward vehicle 312 is calculated by the following equation 2. Is calculated. In addition, the following formula 2 is the same as the formula described in Patent Document 1.
L = (L1 × W1) / W2 (Expression 2)

  Thereafter, the process of the flowchart of FIG. When the inter-vehicle distance L is measured, for example, when the inter-vehicle distance L is less than the threshold value, the control unit 5 performs a process of issuing a warning to the driver or automatically sets the inter-vehicle distance L to be a predetermined inter-vehicle distance. The drive system (engine) of the vehicle 2 is controlled.

  Next, details of the second distance measurement process, that is, the process of acquiring the forward vehicle information from the server 20 and measuring the inter-vehicle distance using the acquired forward vehicle information will be described. FIG. 9 shows a flowchart of the second distance measurement process. The process of FIG. 9 assumes that the host vehicle 2 and the forward vehicle 32 are separated as shown in FIG. 10 and that the forward vehicle information of the forward vehicle 32 is executed in an unmeasured scene. . FIG. 11 shows a flowchart of processing executed by the CPU 21 of the server 20 corresponding to the processing of FIG.

  When the processing of FIG. 9 is started, first, an image taken by the camera 4 is acquired from the camera 4 (S31). The forward vehicle 32 of FIG. 10 is reflected in the image acquired in S31. Next, the image acquired in S31 is transmitted to the server 20 via the communication unit 6 (see FIG. 1) (S32). At this time, information for identifying the vehicle 2 is attached to the image and transmitted to the server 20 so that it can be identified from which vehicle 2 the image is transmitted.

  The image transmitted in S32 is received by the communication unit 23 (see FIG. 1) of the server 20, and the CPU 21 acquires the image received by the communication unit 23 (S41 in FIG. 11). Next, the CPU 21 performs pattern matching between the image acquired in S41 and each image X stored in the database 220 (see FIG. 5) stored in the storage unit 22 (S42). That is, the CPU 21 has the same shape (same vehicle type) as that of the preceding vehicle 32 (see FIG. 10) reflected in the image acquired in S41 from among the images X stored in the database 220, in other words, the preceding vehicle having the same contour. Search for images with. In S42, there is no difference between the size of the contour of the front sub-vehicle shown in the image X and the size of the contour of the front vehicle shown in the image acquired in S41.

  When there is an image that can be matched with the image of S41 in the database 220 by the pattern matching in S42 (S43: Yes), the vehicle information stored in the database 220 in association with the image is stored in the communication unit 23. Then, the image acquired in S41 is transmitted to the vehicle 2 (vehicle 2 in FIG. 10) (S44). For example, if the image X2 in FIG. 5 can be matched in S42, the vehicle information Y2 stored in association with the image X2, that is, the vehicle information Y2 of the vehicle ahead in the image X2, is stored in the vehicle. 2 to send. At this time, since the information (such as the ID of the vehicle-mounted device 10) that identifies the transmission source vehicle 2 is attached to the image acquired in S41, the forward vehicle information is transmitted to the vehicle 2 that is identified by the information. Then, the process of the flowchart of FIG. 11 is complete | finished.

  On the other hand, if there is no matching image (S43: No), the processing of the flowchart of FIG.

  Returning to the description of FIG. 9, the control unit 5 transmits the image to the server 20 in S <b> 32, and then transmits the forward vehicle information from the server 20 (forward vehicle information transmitted in S <b> 44 of FIG. 11) to the communication unit 6. (S33). The control unit 5 stores the forward vehicle information acquired in S33 in the memory 51. In addition, when the process of S44 of FIG. 11 is not performed (S43: No), it cannot acquire forward vehicle information in S33.

  Next, it is determined whether or not the front vehicle information from the server 20 has been acquired (S34). If it can be acquired (S34: Yes), the acquired front vehicle information is used to obtain S23 in FIG. The inter-vehicle distance from the preceding vehicle 32 (see FIG. 10) is calculated by the same method (S35). That is, the inter-vehicle distance is calculated by substituting the vehicle width of the preceding vehicle in the image acquired in S31 and the forward vehicle information (reference inter-vehicle distance, reference vehicle width) acquired in S33 into the above equation 2. Thereby, even if it approaches the front vehicle 32 of FIG. 10 and the front vehicle information of the front vehicle 32 is not measured, the inter-vehicle distance with the front vehicle 32 can be obtained. Thereafter, the processing of the flowchart of FIG. 9 ends.

  If the forward vehicle information is once acquired in S33, communication with the server 20 (the processing of S32 and S33) does not need to be performed after that, and after the processing of FIG. Thereafter, the inter-vehicle distance from the preceding vehicle 32 is continuously measured by the process of FIG.

  On the other hand, when the forward vehicle information from the server 20 cannot be acquired (S34: No), the process of the flowchart of FIG. In this case, in the process of FIG. 9, the measurement of the inter-vehicle distance from the forward vehicle 32 of FIG. 10 is not performed. In this case, the process proceeds to the process shown in FIG. 2 described above, and when the host vehicle 2 approaches the front vehicle 32, the front vehicle information for the front vehicle 32 is measured. Thereafter, by the process of FIG. 6, the inter-vehicle distance from the forward vehicle 32 is measured using the measured forward vehicle information.

  As described above, according to the present embodiment, since the inter-vehicle distance is measured from a vehicle width larger than the license plate, a practically accurate inter-vehicle distance can be obtained even at night or when the distance from the preceding vehicle is long. Can do. Since the vehicle width is larger than other dimensions (vehicle height, etc.) of the vehicle, using the vehicle width is suitable for measuring the inter-vehicle distance. Further, it is not necessary to provide a plurality of cameras as in the case of a stereo camera, and it is only necessary to have a camera with a number of pixels that can recognize the preceding vehicle except when approaching the preceding vehicle, so the inter-vehicle distance can be measured at low cost.

  In addition, because the vehicle information is measured or acquired from the server when measuring the inter-vehicle distance, the inter-vehicle distance to the preceding vehicle is measured regardless of the type of the front vehicle that is subject to measurement of the inter-vehicle distance. it can. Moreover, the database of the front vehicle information for every vehicle model can be comprised in a server by transmitting the front vehicle information which each vehicle measured to the server, and accumulating it. Thereby, it is not necessary to prepare the forward vehicle information on the server side, and it is possible to cope with a case where a new vehicle type is put on the market in the future.

  In addition, this invention is not limited to the said embodiment, A various change is possible to the limit which does not deviate from description of a claim. For example, in the above embodiment, the inter-vehicle distance is measured using the vehicle width in the image of the front vehicle, but the inter-vehicle distance may be measured using other dimensions of the front vehicle such as a vehicle height. For example, when the vehicle height is used, the vehicle height (number of pixels) of the preceding vehicle in the image is calculated in S15 of FIG. 2, S23 of FIG. 6, and S35 of FIG.

  Moreover, although the example which the communication part 6 was provided in the onboard equipment 10 was shown in the said embodiment, the communication part 6 may not be provided. In this case, the forward vehicle information from the server 20 cannot be acquired, but the inter-vehicle distance can be measured by the processes of FIGS. Moreover, in the said embodiment, the front vehicle information which each vehicle measured was transmitted to the server, and it was accumulate | stored, and the database of the forward vehicle information for every vehicle model was comprised in the server. You may prepare the vehicle information for every front. In this case, it is not necessary to transmit the forward vehicle information to the server in S16 of FIG.

  Moreover, in the said embodiment, when the recognized license plate of the front vehicle 31 became more than predetermined magnitude | size on an image, it judged that the own vehicle 2 approached the front vehicle 31 enough, but this invention does this. However, the present invention is not limited, and it may be determined that the host vehicle 2 has sufficiently approached the preceding vehicle 31 using other distance measuring means such as an ultrasonic sensor, an infrared sensor, or a laser radar.

DESCRIPTION OF SYMBOLS 1 Inter-vehicle distance measurement system 2 Own vehicle 3, 31, 32 Vehicle ahead 4 Camera 5 Control part 6 Communication part 10 Onboard equipment 20 Server

Claims (8)

It is mounted on a vehicle (2) provided with photographing means (4) for photographing the preceding vehicle (3, 31, 32),
Recognizing means (S12) for recognizing the number plate (35) of the preceding vehicle from the approaching image (81) which is an image taken by the photographing means when the vehicle approaches the preceding vehicle;
First distance calculating means (S14) for calculating, as a reference distance, an inter-vehicle distance between the vehicle and the preceding vehicle based on the size of the license plate recognized by the recognition means;
Dimension calculation means (S15) for calculating the dimension of the preceding vehicle on the approaching image as a reference dimension;
Second distance calculation for calculating an inter-vehicle distance between the vehicle and the preceding vehicle from the dimension of the preceding vehicle in the image captured by the imaging means, using forward vehicle information that is information on the reference distance and the reference dimension. Means (S23, S35);
An inter-vehicle distance measuring device (10) comprising: Acquisition means (S33, 6) for acquiring the forward vehicle information of the forward vehicle (32) reflected in the image taken by the photographing means from a server (20) storing the forward vehicle information for each vehicle type;
The second distance calculating means (S35) uses the preceding vehicle information acquired by the acquiring means when there is no measurement information that is the preceding vehicle information measured by the inter-vehicle distance measuring device of the host vehicle. The inter-vehicle distance measuring device according to claim 1, wherein the inter-vehicle distance from the preceding vehicle is calculated. First transmission means (S16, 6) for transmitting the image taken by the photographing means and the measurement information of the preceding vehicle reflected in the image to the server,
The server includes storage means (22) for storing an image of a preceding vehicle transmitted from the first transmission means of each vehicle equipped with the inter-vehicle distance measuring device and the preceding vehicle information as measurement information in association with each other. The inter-vehicle distance measuring device according to claim 2. A second transmission unit (S32, 6) for transmitting an image captured by the imaging unit to the server in a measurement scene of an inter-vehicle distance with the preceding vehicle (32) without the measurement information;
The acquisition means receives the forward vehicle information of the forward vehicle shown in the image transmitted from the server and transmitted by the second transmission means,
The server
Retrieval means (S42) for retrieving an image in which the preceding vehicle having the same shape as the preceding vehicle shown in the collation image, which is the image transmitted from the second transmitting means, is searched from the images stored in the storage means. )When,
Transmission means (S44, 23) for transmitting the forward vehicle information stored in the storage means in association with the image searched by the search means to a vehicle that has transmitted the verification image. Item 4. The inter-vehicle distance measuring device according to Item 3.   The inter-vehicle distance measuring device according to any one of claims 1 to 4, wherein the dimension of the preceding vehicle is a vehicle width. A proximity photographing step (S11) for photographing the preceding vehicle when the vehicle approaches the preceding vehicle (3, 31, 32);
A recognition step (S12) for recognizing the number plate (35) of the preceding vehicle from the approaching image (81) photographed in the approaching photographing step;
A first distance calculating step (S14) for calculating, as a reference distance, an inter-vehicle distance between the vehicle and the preceding vehicle based on a size on the image of the license plate recognized in the recognition step;
A dimension calculating step (S15) for calculating the dimension of the preceding vehicle on the approaching image as a reference dimension;
After obtaining the forward vehicle information which is information of the reference distance and the reference dimension, the forward vehicle (312) on the image (82) where the forward vehicle is newly photographed and the forward vehicle is newly photographed. ) And a second distance calculating step (S23, S35) for calculating an inter-vehicle distance between the vehicle and the preceding vehicle from the preceding vehicle information,
The inter-vehicle distance measuring method characterized by including. In the dimension calculation step, the vehicle width of the front vehicle in the approaching image is calculated as the reference dimension,
In the second distance calculating step, an inter-vehicle distance between the vehicle and the preceding vehicle is calculated from a vehicle width of the preceding vehicle on the newly captured image of the preceding vehicle and the preceding vehicle information. The inter-vehicle distance measuring method according to claim 6. A collecting step (collecting an image taken by each vehicle and the preceding vehicle information of the preceding vehicle reflected in the image from each vehicle, and storing the collected image and the preceding vehicle information in association with each other in the server (20)). S16)
From each image collected in the collecting step, a search is made for an image showing a front vehicle having the same shape as the target front vehicle, which is the front vehicle shown in the image taken by the target vehicle for which the distance between vehicles is desired to be measured. A search step (S42),
In the second distance calculating step (S35), an inter-vehicle distance between the target vehicle and the target forward vehicle is calculated using the forward vehicle information stored in the server in association with the image searched in the search step. The inter-vehicle distance measuring method according to claim 6, wherein the inter-vehicle distance is calculated.