JP6280659B2 - Vehicle monitoring apparatus and vehicle monitoring method - Google Patents

Description

  Embodiments described herein relate generally to a vehicle monitoring apparatus and a vehicle monitoring method for monitoring a vehicle.

  There is known a system that images a road or the like with a camera, detects a vehicle in the captured image, and reads information described on a license plate of the vehicle (see, for example, Patent Document 1). This type of system is attached to, for example, an entrance or exit of a toll road and used to recognize a passing vehicle.

Japanese Patent No. 4690657

When character recognition processing is performed on the entire image when reading information written on the license plate of the vehicle from the camera image, the processing load becomes excessive and information may not be read efficiently.
The problem to be solved by the present invention is to provide a vehicle monitoring apparatus and a vehicle monitoring method capable of efficiently reading information described on a license plate of a vehicle.

  The vehicle monitoring apparatus according to the embodiment includes an image acquisition unit, a determination unit, a reading unit, a vehicle position estimation unit, and a vehicle type assignment unit. The image acquisition unit includes a first imaging unit that can image a vehicle that is close to a predetermined location from the front, a second imaging unit that can image a vehicle that passes through the predetermined location from above, and the distance from the predetermined location. An image captured by an imaging unit including a third imaging unit capable of imaging the vehicle from behind is acquired. In the image captured by the imaging unit, the determination unit includes a specific part of the vehicle in a specific region among a plurality of regions obtained by dividing the image by a dividing line substantially orthogonal to the traveling direction of the vehicle. It is determined whether or not. A reading part reads the information described in the number plate attached to the said vehicle from the said specific area | region, when it determines with the specific site | part of the said vehicle existing in the said specific area | region. The vehicle position estimation unit specifies, for each time, a position of the vehicle in the vicinity of the predetermined location detected from the image captured by the first imaging unit, based on the image captured by the third imaging unit. . The vehicle type assigning unit assigns information read from the image captured by the first image capturing unit to the position of the vehicle specified by the vehicle estimating unit.

It is a figure which shows a mode that the cameras 10 and 20 contained in the vehicle monitoring system 1 of 1st Embodiment were attached to the gantry GT. 2 is a diagram illustrating an example of an image captured by a camera 10. FIG. It is a figure which shows an example of a function structure of the vehicle monitoring apparatus 100 which concerns on 1st Embodiment. It is the figure which extracted the image of the front part from the imaged image. It is the figure which extracted the image of the body part from the imaged image. It is the figure which cut out the image of the rear part from the imaged image. It is the figure which arranged the image imaged by the camera 10 in time series in order of imaging. It is a figure which shows a mode that the image was divided | segmented into the areas 1-3. It is a figure which shows an example of the correspondence of the site | part of the vehicle which exists in each area | region, and each state. It is a figure which shows the other example of the state of a vehicle. It is the figure which modeled the change of the state of the vehicle in the case of using the image of the camera. It is the figure which modeled the change of the state of the vehicle in the case of using the image of the camera 20. FIG. It is a figure for demonstrating an example of the measurement method by the vehicle width measurement part. It is a figure which shows a mode that the cameras 10, 20, and 30 contained in the vehicle monitoring system 2 of 2nd Embodiment were attached to the gantry GT. It is a figure which shows a mode that the vehicle which drive | works a several lane is reliably recognized by the vehicle monitoring system 2 of 2nd Embodiment. It is a figure which shows an example of a function structure of the vehicle monitoring apparatus 200 which concerns on 2nd Embodiment.

Hereinafter, embodiments of a vehicle monitoring device and a vehicle monitoring method will be described with reference to the drawings.
<First Embodiment>
[Constitution]
FIG. 1 is a diagram illustrating a state in which cameras 10 and 20 included in the vehicle monitoring system 1 of the first embodiment are attached to a gantry GT. The gantry GT is, for example, an automatic fee collection system gantry. The cameras 10 and 20 are cameras including an image sensor such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor).

  The camera 10 images the front surface of the vehicle MB passing close to the gantry GT from an obliquely upward direction. In the figure, A (10) indicates an imaging region of the camera 10. Further, the camera 20 captures an image of the rear surface of the vehicle MB passing through the gantry GT from an obliquely upward direction. In the figure, A (20) indicates an imaging region of the camera 20. For example, the cameras 10 and 20 perform imaging at a depression angle of about 45 [degrees] from 7 [m] above the road. Further, the cameras 10 and 20 are controlled so as to repeatedly perform imaging at a predetermined cycle. The installation locations of the cameras 10 and 20 are not limited to those shown in FIG. 1, but may be any locations where the front and rear surfaces of the vehicle can be imaged. Images captured by the cameras 10 and 20 are transmitted to the vehicle monitoring apparatus 100. 2A and 2B are diagrams illustrating an example of an image captured by the camera 10. The camera 10 may mainly capture one lane as shown in FIG. 2A, or may simultaneously capture multiple lanes as shown in FIG. 2B.

  In this way, by imaging from the front and rear of the vehicle MB, the license plate can be recognized even for a vehicle having a license plate attached only to the rear, such as a two-wheeled vehicle. Further, when the road is congested, even if the number plate of the subsequent vehicle MB is hidden in the image captured from the front of the vehicle MB, the number of the subsequent vehicle MB in the image captured from the rear of the vehicle MB. It may be possible to recognize the plate.

The vehicle monitoring apparatus 100 can be installed in any place. FIG. 3 is a diagram illustrating an example of a functional configuration of the vehicle monitoring apparatus 100 according to the first embodiment. The vehicle monitoring device 100 includes, for example, a vehicle detection unit 102, a vehicle state determination unit 104, a recognition position determination unit 106, a number recognition unit 108, a vehicle state transition unit 110, a vehicle number measurement unit 112, a vehicle width. A measurement unit 114, a vehicle type determination unit 116, a vehicle information integration unit 118, and a storage unit 120 are provided. Each functional unit is, for example, a software functional unit that functions when a CPU (Central Processing Unit) of the vehicle monitoring apparatus 100 executes a program stored in the storage unit 120.
Alternatively, each functional unit may be a hardware functional unit such as an LSI (Large Scale Integration) or an ASIC (Application Specific Integrated Circuit). The storage unit 120 is, for example, an HDD (Hard Disk Drive), a flash memory, a RAM (Random Access Memory), a ROM (Read Only Memory), or the like.

[Identification of vehicle parts]
The vehicle detection unit 102 determines whether or not a part (part) of the vehicle exists in the images captured by the cameras 10 and 20. The vehicle detection unit 102, for example, for each part constituting the vehicle such as a front part, a body part, and a rear part of the vehicle, feature quantities that are edge-enhanced by Sobel filter processing, gradient feature quantities, luminance feature quantities, etc. A discriminator composed of these patterns is constructed (see the following reference). And the vehicle detection part 102 pinpoints the kind and position of the site | part of the vehicle which exist in an image using the comprised discriminator.
Reference: “Coarse-to-fine approach for fast deformable object detection”, M.Pedersoli, A.Vadaldi, J.Gonzalez, CVPR2011

4A to 4C are diagrams in which an image of a front portion is cut out from a captured image.
The front part includes components such as a headlamp, a front grille, a bumper, and a license plate. The aforementioned discriminator is configured so that these components can be integrated and detected. As a result, the vehicle detection unit 102 can identify the front part with high accuracy corresponding to the positions of the components that differ depending on the vehicle type. 5A to 5C are diagrams in which an image of a body part is cut out from a captured image. The body portion has a simple shape mainly including the roof of the vehicle, but in the case of a work vehicle or the like, the body portion has a more complicated shape. On the other hand, the vehicle detection unit 102 may use an identifier for a body part of a work vehicle prepared in advance. FIGS. 6A to 6C are diagrams in which a rear image is cut out from a captured image.

  7A to 7D are diagrams in which images captured in time series by the camera 10 are arranged in the order of imaging. In accordance with the progress of the vehicle MB, the image shown in FIG. 7 (A) is first picked up, then the image shown in FIG. 7 (B) is picked up, and then the image shown in FIG. 7 (C) is picked up. The image shown in FIG. 7D is taken. In the figure, F denotes a front part specified by the vehicle detection unit 102, B denotes a body part specified by the vehicle detection unit 102, and R denotes a rear part specified by the vehicle detection unit 102.

[Vehicle condition determination]
The vehicle state determination unit 104 virtually divides an image captured by the cameras 10 and 20 into, for example, three horizontal dividing lines, and each of the regions 1 to 3 includes a front part, a body part, and a rear part. It is determined whether or not it is accommodated. The dividing line in the horizontal direction is a straight line that is substantially orthogonal to the traveling direction of the vehicle. Hereinafter, in order to simplify the description, processing on the captured image of the camera 10 will be mainly described. FIG. 8 is a diagram illustrating a state in which the image is divided into regions 1 to 3.

[Number plate recognition]
The recognition position determination unit 106 refers to the determination result by the vehicle state determination unit 104, determines whether or not the front part (F) exists in the region 2, and when the front part (F) exists in the region 2. Then, the license recognition unit 108 is made to recognize the license plate. The number recognizing unit 108 scans a scanning region in the region 2 (for example, a region from the front end portion of the vehicle recognized as the front portion (F) toward the image depth direction) to occupy the number plate (number plate region). Is identified, character recognition processing is performed in the license plate area, and information written on the license plate is read.

  The recognition position determination unit 106 recognizes the number plate in the number recognition unit 108 when the front part (F) exists in one of the areas 2 and 1, not when the front part (F) exists in the area 2. It may be done. Further, the recognition position determination unit 106 refers to the determination result by the vehicle type determination unit 116 described later, and performs license plate recognition to the number recognition unit 108 when the front part (F) exists in the optimum region corresponding to the vehicle type. It may be done.

  In any case, the image in which the front portion (F) is displayed in the region 2 (or region 1) is temporarily stored in the storage unit 120 until the vehicle exits from the vicinity of the gantry GT. Accordingly, recognition failure by the number recognition unit 108 can be reduced.

  The number recognizing unit 108 performs extraction, binarization processing, and the like of components below a specific frequency for the scanning region, and further extracts a character candidate region using a method such as labeling. Then, character recognition is performed by removing a region whose size and aspect ratio are far from the target character region from the character candidate region (see “Patent No. 4901676”). Information obtained by character recognition includes a serial number, a land branch, a classification number, a use code, a vehicle type code, and the like. The number recognizing unit 108 reads information (characters) written on the number plate by such processing, and outputs a character recognition score to the recognition position determining unit 106. When the character recognition score is lower than the reference value, the recognition position determining unit 106 causes the number recognizing unit 108 to read a character again for an image in which the front part (F) is displayed in the area 2 or the like.

  As described above, the vehicle monitoring apparatus 100 according to the present embodiment reads the information written on the license plate when the front portion (F) exists in the area 2 (or area 1), which makes it easy to read the license plate of the vehicle. Therefore, the scanning area can be reduced, and the information written on the license plate of the vehicle can be read efficiently.

  That is, the process of specifying the front portion (F) is performed by a process that specifies a boundary between a road and a vehicle, which has a relatively small processing load and can be performed with high accuracy. For this reason, the vehicle monitoring apparatus 100 can narrow down the area | region in the image which a license plate exists quickly with low load, and it describes in the license plate of a vehicle compared with what scans a license plate from the whole image. Can be read efficiently.

  Further, the number recognizing unit 108 may correct or narrow down the scanning area based on the determination result by the vehicle type determining unit 116 described later. For example, when the vehicle type determination unit 116 determines that the license plate is a vehicle type that does not exist in the center portion, the number recognition unit 108 corrects or narrows the scanning area to a position corresponding to the deviation of the license plate. As a result, the vehicle monitoring apparatus 100 can narrow down the area in the image where the license plate exists quickly and with a lower load, and can more efficiently read the information described on the license plate of the vehicle.

  On the other hand, the recognition position determination unit 106 determines whether or not the rear part (R) exists in the region 2 (or region 3) for the image captured by the camera 20, and the number recognition unit 108 determines whether the region 2 (or region 3) has the rear part (R). Alternatively, when the rear portion (R) exists in the region 3), the license plate region is scanned by scanning the scanning region (for example, the region from the rear end of the vehicle recognized as the rear portion (R) toward the image depth direction). Is identified, character recognition processing is performed in the license plate area, and information written on the license plate is read.

[Measurement of the number of vehicles]
The vehicle state transition unit 110 refers to the determination result of the vehicle state determination unit 104, and the combination of the front part (F), the body part (B), and the rear part (R) existing in the regions 1 to 3 Classify the state of the vehicle in the imaging area. For example, the vehicle state transition unit 110 may change the state of the vehicle within the imaging range of the camera 10 to S0 “no vehicle”, S1 “entry”, S2 “stay in area”, S3 “exit”, and S4 “two approaching”. It classifies into each state. FIG. 9 is a diagram illustrating an example of a correspondence relationship between a vehicle part existing in each region and each state. In FIG. 9, “X” indicates that the vehicle part is not specified by the vehicle detection unit 102 in the corresponding region. The state in FIG. 8 corresponds to S4 “two approaching” because the front part (F) of the vehicle is specified in region 1 and the rear part (R) of the vehicle is specified in region 3. FIG. 10 is a diagram illustrating another example of the state of the vehicle. In the state of FIG. 10, the front part (F) of the vehicle is specified in the area 1, the rear part (R) of the vehicle is specified in the area 2, and the body part (B) of the vehicle is specified in the area 3. Corresponds to “internal retention”. On the other hand, the vehicle state transition unit 110 classifies the state of the vehicle by reversing the order of the regions 1 to 3 in the correspondence relationship of FIG.

  The vehicle number measurement unit 112 refers to the change in the state of the vehicle typified by the vehicle state transition unit 110 and measures the number of vehicles from the change pattern. FIG. 11 is a diagram in which changes in the state of the vehicle when the image of the camera 10 is used are modeled. When the model of FIG. 11 is used, the vehicle number measuring unit 112 determines that one vehicle has entered the lower region of the gantry GT when transitioning from S4 to S2 and when transitioning from S3 to S0. . FIG. 12 is a diagram modeling changes in the state of the vehicle when the image of the camera 20 is used. When the model shown in FIG. 12 is used, the vehicle number measuring unit 112 determines that one vehicle has left the area below the gantry GT when transitioning from S4 to S2 and when transitioning from S1 to S0. .

By making such a determination, the vehicle monitoring apparatus 100 of the present embodiment can measure the number of units with little leakage even when, for example, images of a part of the vehicles overlap due to traffic congestion. In addition, the vehicle number measuring unit 112 releases an area in the storage unit 120 that stores information such as unnecessary images about a vehicle that has been determined to have left the area under the gantry GT.
Thereby, waste of resources in the storage unit 120 can be prevented.

[Vehicle width, vehicle type, information integration]
The vehicle width measuring unit 114 measures the vehicle width of the vehicle from the vehicle outer contour in the image captured by the camera 10. FIG. 13 is a diagram for explaining an example of a measurement method performed by the vehicle width measurement unit 114. For example, the vehicle width measurement unit 114 measures the vehicle width based on the road width L (y) for each vertical position y of the image stored in the storage unit 120 in advance and the inclination θ in the road width direction in the image. That is, if the lower vector of the vehicle outer contour is set to → W (→ indicates a vector), the vehicle width measuring unit 114 maps this to a vector → W * parallel to the road width direction. Then, the vehicle width measuring unit 114 obtains the ratio of the vector → W * and the road width l (y) on the image at the vertical position y, and the road width L with the starting point of the vector → W * as the vertical position y. The vehicle width is measured by multiplying (y) by the obtained ratio.

  The vehicle type determination unit 116 integrates the shape pattern such as the front portion (F) detected by the vehicle detection unit 102 and the vehicle width measured by the vehicle width measurement unit 114, and displays the image on the image captured by the camera 10. The vehicle type of the selected vehicle is determined. The vehicle type determination unit 116 classifies the vehicle type into “two-wheeled vehicle”, “light vehicle”, “normal vehicle”, “large vehicle”, “other”, “impossible determination”, and the like. Among these, for “two-wheeled vehicle” and four-wheeled vehicle, the determination is made based on the shape pattern of the front portion (F) and the like, and for the type of the four-wheeled vehicle, the shape pattern of the front portion (F) and the vehicle width are determined. A determination is made based on

  The vehicle information integration unit 118 integrates information such as the number of vehicles measured by the vehicle number measurement unit 112, the recognition result obtained by the number recognition unit 108 obtained for each vehicle, the vehicle type determined by the vehicle type determination unit 116, and the like. The vehicle is managed in association with the passing time of the vehicle. Here, as for “two-wheeled vehicle”, since there is no license plate information by the front image of the vehicle, the information on the camera 10 side is insufficient, but this can be supplemented based on the information on the camera 20 side. .

[Summary]
According to the vehicle monitoring apparatus 100 of the present embodiment described above, the process of identifying the front portion (F) that can be performed with a relatively small processing load and high accuracy, such as identifying the boundary between the road and the vehicle. Can narrow down the scanning area. As a result, the information written on the license plate of the vehicle can be read more efficiently than when scanning the license plate from the entire image.

  Further, according to the vehicle monitoring apparatus 100 of the present embodiment, the number recognition unit 108 corrects or narrows down the scanning area based on the determination result by the vehicle type determination unit 116, so that the license plate can be quickly and lightly loaded. The area in the existing image can be narrowed down, and the information written on the license plate of the vehicle can be read more efficiently.

  Further, according to the vehicle monitoring apparatus 100 of the present embodiment, the vehicle number measuring unit 112 measures the number of vehicles from the change pattern of the vehicle state typified by the vehicle state transition unit 110, and thus the number of vehicles with little leakage. Measurement can be performed.

Second Embodiment
Hereinafter, the vehicle monitoring system 2 and the vehicle monitoring device 200 of the second embodiment will be described. FIG. 14 is a diagram illustrating a state where the cameras 10, 20, and 30 included in the vehicle monitoring system 2 of the second embodiment are attached to the gantry GT. The cameras 10, 20, and 30 are cameras that include an image sensor such as a CCD or a CMOS, for example. Since the cameras 10 and 20 are the same as those in the first embodiment, description thereof is omitted. The camera 30 captures an area immediately below the gantry GT. In the figure, A (30) indicates an imaging region of the camera 30. The imaging area A (30) of the camera 30 is set to cover the blind spots of the imaging areas A (10) and A (20), for example. Further, the cameras 10, 20, and 30 are controlled so as to repeatedly perform imaging at a predetermined cycle.
Images captured by the cameras 10, 20, and 30 are transmitted to the vehicle monitoring device 200.

  In this way, by capturing images from the front, rear, and directly above the vehicle MB, the state change until the vehicle MB captured by the camera 10 is subsequently captured by the camera 20 can be grasped without omission. be able to. For example, when the gantry GT is suspended over a road having two or more lanes, the effect is remarkable.

  FIG. 15 is a diagram illustrating a state in which a vehicle traveling in a plurality of lanes is reliably recognized by the vehicle monitoring system 2 according to the second embodiment. Here, the imaging areas A (10), A (20), and A (30) of the cameras 10, 20, and 30 are assumed to straddle a plurality of lanes. FIG. 15A is a diagram illustrating a situation in which the gantry GT is suspended over two lanes of the lane 1 and the lane 2 and the vehicle MB2 passes the vehicle MB1 immediately below the gantry GT. In this case, if there is no camera 30 that images directly under the gantry GT, the vehicles MB1 and MB2 captured in series by the camera 10 appear side by side in the captured image of the camera 20. Therefore, it may be difficult to identify both vehicles.

  On the other hand, in the vehicle monitoring system 2 of the second embodiment, the overtaking operation of the vehicle MB2 can be grasped from the captured image of the camera 30, so that the vehicle can be identified more reliably. FIG. 15B is a diagram illustrating an example of information stored in the storage unit 120 (described later) when the vehicle monitoring device 20 grasps the vehicles MB1 and MB2.

  The vehicle monitoring device 200 can be installed at an arbitrary place. FIG. 16 is a diagram illustrating an example of a functional configuration of the vehicle monitoring apparatus 200 according to the second embodiment. The vehicle monitoring apparatus 200 includes, for example, a forward vehicle determination unit 202, a vehicle entry detection unit 204, a vehicle position estimation unit 206, a speed measurement unit 208, a vehicle type assignment unit 210, a vehicle exit determination unit 212, and vehicle information. An integration unit 214, a rear vehicle determination unit 216, a camera control unit 218, an image quality improvement processing unit 220, and a storage unit 230 are provided. These functional units are, for example, software functional units that function when the CPU of the vehicle monitoring apparatus 200 executes a program stored in the storage unit 230. Instead of this, each functional unit may be a hardware functional unit such as an LSI or an ASIC.

  The forward vehicle determination unit 202 performs, for example, the same processing as the vehicle state determination unit 104 and the recognition position determination unit 106 in the first embodiment from an image captured by the camera 10, and moves forward toward the gantry GT. The position of the vehicle is specified, and the information written on the license plate of the vehicle is read. As shown in FIG. 15, when the gantry GT straddles a plurality of lanes, the information on the vehicle position and the number plate is stored in the storage unit 230 together with the information on the traveling lane. The vehicle entry detection unit 204 receives the position and time of the vehicle entering immediately below the gantry GT from the forward vehicle determination unit 202, and detects that an object of vehicle size enters.

  The vehicle position estimation unit 206 captures how the vehicle detected by the vehicle entry detection unit 204 moves at each time, and specifies the position of the vehicle at each time. For the vehicle imaged from directly above, the vehicle position estimation unit 206 identifies the vehicle by updating the luminance information template in the vehicle contour region at each time, for example.

  The speed measurement unit 208 measures the speed of the vehicle based on the amount of movement of the vehicle position specified by the vehicle position estimation unit 206 per unit time. When the speed measured by the speed measuring unit 208 is greatly deviated between vehicles, there are vehicles that are stopped due to a failure or accident even though there are vehicles that move at the legal speed. Therefore, processing such as reporting to a warning device may be performed.

  The vehicle type assigning unit 210 assigns the information detected and read by the forward vehicle determining unit 202 to the position of the vehicle specified by the vehicle position estimating unit 206. In particular, regarding the “two-wheeled vehicle”, since there is no information written on the license plate read by the forward vehicle determination unit 202, the position of the vehicle is surely grasped.

  The vehicle exit determination unit 212 determines whether or not the vehicle has exceeded a position near the exit based on the vehicle position specified by the vehicle position estimation unit 206, so that the vehicle exits from the lower region of the gantry GT. judge.

  The vehicle information integration unit 214 manages the vehicle traffic by integrating the lane and the exit time that have been traveled and the information allocated by the vehicle type allocation unit 210 for each vehicle that exits from the lower region of the gantry GT.

  The rear vehicle determination unit 216 performs the same processing as the vehicle state determination unit 104 and the recognition position determination unit 106 in the first embodiment from the image captured by the camera 20 that captures the vehicle traveling in a direction away from the gantry GT. Then, the position of the vehicle is specified and the license plate information is read.

  The camera control unit 218 generates parameter change information transmitted to the camera 20 used by the rear vehicle determination unit 216. For example, the camera control unit 218 performs control such as increasing the shutter speed of the camera 20 when the vehicle information integration unit 214 detects that a vehicle with a significantly high speed is passing. In addition, when it is detected that many “two-wheeled vehicles” pass, control such as changing zoom information is performed in order to increase the resolution. In addition, when the illumination device irradiates the periphery of the gantry GT, the timing at which the vehicle enters the imaging region A (20) of the camera 20 is predicted, and the timing of illumination lighting (or intensity adjustment) is measured. You may control. In addition, since the license plate is required to be mounted behind the vehicle, the resolution of the captured image of the camera 20 may be set higher than the resolution of the captured image of the image of the camera 10.

  The image quality improvement processing unit 220 performs, for example, processing for correcting motion blur of the vehicle due to vibration of the entire gantry GT on the images captured by the cameras 10 and 20. In addition, the image quality improvement processing unit 220 may perform a correction process or the like to make up for insufficient image exposure or insufficient resolution when adjustment by the camera control unit 218 cannot catch up.

  In the example of FIG. 15, the vehicle information integration unit 214 receives information from the front vehicle determination unit 202 that the vehicle MB1 has passed the lane 2 at time Tn-1 and subsequently the vehicle MB2 at time Tn near the entrance. Furthermore, as a result of identifying each vehicle based on the image captured by the camera 30, before each vehicle enters the imaging area A (20) of the camera 20, the vehicle MB2 overtakes the vehicle MB1, and from the lane 2 It is assumed that the course has been changed to lane 1. At this time, the vehicle information integration unit 214 changes the time-series information of the vehicle, and notifies the rear vehicle determination unit 216 that the vehicle MB2 appears on the screen first from the lane 2. Accordingly, there is no contradiction between the vehicle management states on the entrance side and the exit side, and information on the arrival of the vehicle can be held in advance, so that the recognition accuracy can be improved.

  According to the vehicle monitoring apparatus 200 of the second embodiment described above, since the vehicle is monitored using the captured image of the camera 30 that complements the gap between the imaging regions of the camera 10 and the camera 20, seamless vehicle monitoring is realized. can do.

  Further, according to the vehicle monitoring apparatus 200 of the second embodiment, the camera control unit 218 uses the information based on the images captured by the cameras 10 and 30 to change the parameter to be transmitted to the camera 20, so that it is more reliable. The vehicle can be monitored.

  According to at least one embodiment described above, the scanning region is identified by the process of identifying the front portion (F), which can be performed with a relatively small processing load and high accuracy, such as identifying the boundary between the road and the vehicle. By narrowing down, the information described on the license plate of the vehicle can be read efficiently.

  In addition, the vehicle state determination unit 104, the recognition position determination unit 106, the forward vehicle determination unit 202, and the rear vehicle determination unit 216 in each of the embodiments are examples of the “determination unit” in the claims. Moreover, the number recognition part 108 in each said embodiment, the front vehicle determination part 202, and the back vehicle determination part 216 are examples of the "reading part" in a claim.

As mentioned above, although some embodiment of this invention was described, these embodiment is shown as an example and is not intending limiting the range of invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.
For example, also in the first embodiment, the same camera control, image quality improvement processing, and information supplementation by the camera 20 as in the second embodiment may be performed. The technique described as the first embodiment and the technique described as the second embodiment can be used in appropriate combination.

10, 20, 30 Camera 100, 200 Vehicle monitoring device 102 Vehicle detection unit 104 Vehicle state determination unit 106 Recognition position determination unit 108 Number recognition unit 110 Vehicle state transition unit 112 Vehicle number measurement unit 114 Vehicle width measurement unit 116 Vehicle type determination unit 118 Vehicle information integration unit 120, 230 Storage unit 202 Forward vehicle determination unit 204 Vehicle entry detection unit 206 Vehicle position estimation unit 208 Speed measurement unit 210 Vehicle type assignment unit 212 Vehicle exit determination unit 214 Vehicle information integration unit 216 Rear vehicle determination unit 218 Camera control Part 220 Image quality improvement processing part

Claims (11)

A first imaging unit capable of imaging a vehicle close to a predetermined location from the front, a third imaging unit capable of imaging a vehicle passing through the predetermined location from above, and an image of the vehicle moving away from the predetermined location from the rear An image acquisition unit that acquires an image captured by an imaging unit including a second imaging unit possible;
Whether or not a specific part of the vehicle exists in a specific region among a plurality of regions obtained by dividing the image by a dividing line substantially orthogonal to the traveling direction of the vehicle in the image captured by the imaging unit. A determination unit for determining;
When it is determined that a specific part of the vehicle is present in the specific region, a reading unit that reads information described on a license plate attached to the vehicle from the specific region;
A vehicle position estimator that identifies the position of the vehicle in the vicinity of the predetermined location detected from the image captured by the first imager based on the image captured by the third imager;
A vehicle type assigning unit that assigns information read from the image captured by the first image capturing unit to the position of the vehicle specified by the vehicle position estimating unit;
In the image captured by the imaging unit, a categorizing unit that classifies the state of the vehicle in the region imaged by the imaging unit, based on a pattern of a part of the vehicle existing in each of the plurality of regions;
A measurement unit that measures the number of passing vehicles based on a change in the state of the vehicle typified by the typification unit;
A vehicle monitoring device comprising: The vehicle position estimation unit detects overtaking or lane change of the vehicle in the vicinity of the predetermined location based on an image captured by the third imaging unit;
The vehicle monitoring apparatus according to claim 1. A speed measuring unit that measures the speed of the vehicle based on a movement amount per unit time of the position of the vehicle specified by the vehicle position estimating unit;
The vehicle monitoring apparatus according to claim 1 or 2. The speed measuring unit, when the speed measured is biased greatly between vehicles, to report to the warning device that might abnormal situation has occurred,
The vehicle monitoring device according to claim 3. A camera control unit for controlling the operation of the imaging unit;
The camera control unit increases the shutter speed of the imaging unit when it is detected by the speed measurement unit that an excessively fast vehicle is passing,
The vehicle monitoring device according to claim 3 or 4. A camera control unit for controlling the operation of the imaging unit;
The camera control unit predicts the timing at which the vehicle enters the vicinity of the predetermined location based on the speed measured by the speed measurement unit, and lights the lighting device installed near the predetermined location at the predicted timing. ,
The vehicle monitoring device according to claim 3 or 4. The specific portion is a front end portion or a rear end portion of the vehicle.
The vehicle monitoring device according to any one of claims 1 to 6. Further comprising a vehicle type determination unit for determining the vehicle type of the vehicle based on the information read by the reading unit,
The reading unit determines a scanning region for scanning the license plate among the specific regions based on the vehicle type of the vehicle determined by the vehicle type determination unit.
The vehicle monitoring device according to any one of claims 1 to 7. Wherein comprising a first information based on the image captured by the imaging unit, an integrating unit that integrates the information based on the second image captured by the imaging unit,
The vehicle monitoring device according to any one of claims 1 to 8 . Using a providing unit that provides imaging parameters to the second imaging unit using information based on an image captured by the first imaging unit;
The vehicle monitoring device according to any one of claims 1 to 9 . Computer
A first imaging unit capable of imaging a vehicle close to a predetermined location from the front, a third imaging unit capable of imaging a vehicle passing through the predetermined location from above, and an image of the vehicle moving away from the predetermined location from the rear An image captured by an image capturing unit including a second image capturing unit,
In the captured image, it is determined whether or not a specific part of the vehicle exists in a specific region among a plurality of regions divided by at least a dividing line substantially orthogonal to the traveling direction of the vehicle,
When it is determined that a specific part of the vehicle is present in the specific area, the information described on a license plate attached to the vehicle is read from the specific area,
Based on the image captured by the third imaging unit, the position of the vehicle near the predetermined location detected from the image captured by the first imaging unit is specified for each time,
Assigning the information read from the image captured by the first imaging unit to the identified position of the vehicle;
In the image picked up by the image pickup unit, based on the pattern of the part of the vehicle existing in each of the plurality of regions, to classify the state of the vehicle in the region picked up by the image pickup unit,
Measure the number of passing vehicles based on the categorized change in the state of the vehicle,
Vehicle monitoring method.