JP7503931B2 - Passing vehicle counting device, passing vehicle counting method, and passing vehicle counting program - Google Patents

Description

The present invention relates to a passing vehicle counting device, a passing vehicle counting method, and a passing vehicle counting program.

In recent years, traffic counters that count the number of vehicles traveling on a road have come into use. Traffic counters that use, for example, magnetic sensors, cameras, or ultrasonic sensors are used. For example, Patent Documents 1 and 2 propose a method of monitoring traffic conditions by tracking vehicles from images of the road taken by a camera.

JP 2001-229488 A JP 2005-190142 A

When capturing images of a road with a camera and counting the number of vehicles passing through the road, it is preferable to mount the camera in a high position that allows a bird's-eye view of the road. However, mounting a camera in such a high position limits the locations where it can be installed, and the installation work is likely to be costly. Mounting the camera in a lower position would expand the options for installation locations and reduce the installation work costs, which is thought to make it easier to install the camera. However, if the camera is mounted in a low position, a vehicle being tracked may be hidden behind another vehicle, making continuous tracking impossible. If continuous tracking is not possible, it is not easy to determine whether a vehicle that appears from behind another vehicle has already been counted, and there is a risk that the accuracy of counting vehicles will decrease.

One aspect of the disclosed technology is to provide a passing vehicle counting device, a passing vehicle counting method, and a passing vehicle counting program that can easily install cameras and prevent a decrease in counting accuracy.

One aspect of the disclosed technology is exemplified by a passing vehicle counting device as follows. This passing vehicle counting device includes an acquisition unit that acquires an image of an area including a road captured by a camera, a passing vehicle management unit that, when a vehicle passes one of a plurality of preset set positions in the image, stores in a memory unit the speed, passing position, and passing time of the passing vehicle in association with the set position that was passed, a counting unit that counts up the number of vehicles that have passed one set position when the position of a first vehicle passing one of the plurality of set positions differs from the current position of the vehicle estimated from information stored in the memory unit in association with a set position located on the opposite side of the first vehicle's traveling direction from the one set position, and an addition unit that adds up the number of vehicles counted up for each of the plurality of set positions.

The disclosed technology allows for easy camera installation and prevents degradation of counting accuracy.

FIG. 1 is a diagram illustrating an example of the configuration of a passing vehicle counting system according to an embodiment. FIG. 2 is a diagram illustrating an example of a hardware configuration of a counting device according to an embodiment. FIG. 3 is a diagram illustrating an example of a processing block of the counting device according to the embodiment. FIG. 4 is a diagram illustrating an example of an image captured by an acquisition unit in the embodiment. FIG. 5 is a diagram illustrating crossing lines set by the crossing line setting unit in the embodiment. FIG. 6 is a diagram illustrating an example of a vehicle recognition table stored in the vehicle information database in the embodiment. FIG. 7 is a diagram illustrating an example of a tracking table stored in the vehicle information database in the embodiment. 8 is a diagram showing three frames of images captured by the acquisition unit from the surveillance camera. FIG. 9 is a diagram illustrating an example of a passing vehicle table stored in the vehicle information database in the embodiment. FIG. 10 is a diagram illustrating an example of a counted vehicle information table stored in the vehicle information database in the embodiment. FIG. 11 is a diagram illustrating an example of a count target vehicle table stored in the vehicle information database in the embodiment. FIG. 12 is a diagram illustrating an example of a processing flow of the passing vehicle counting system according to the embodiment. FIG. 13 is a diagram illustrating an example of a process flow of a process for counting vehicles passing through a road in the embodiment.

<Embodiment>
The configurations of the embodiments described below are merely examples, and the disclosed technology is not limited to the configurations of the embodiments. A passing vehicle counting device according to the embodiments has the following configuration, for example.
The passing vehicle counting device according to this embodiment has the following features:
an acquisition unit that acquires an image of a range including a road captured by a camera;
a passing vehicle management unit that, when a vehicle passes through any one of a plurality of preset positions in the image capturing, stores in a storage unit the speed, the passing position, and the passing time of the vehicle in association with the set position that the vehicle passed through;
a counting unit that, when a position of a first vehicle passing one of the plurality of set positions is different from a current position of the vehicle estimated from information stored in the storage unit in association with a set position located on the opposite side of the first set position in a traveling direction of the first vehicle, counts up the number of vehicles that have passed the one set position;
and an add-up unit that adds up the number of vehicles counted up for each of the plurality of set positions.

The passing vehicle counting device is an information processing device that counts the number of vehicles passing on a road based on images captured by a camera. The passing vehicle counting device may be integrated with the camera, or may be formed separately from the camera.

The camera is provided at a position where the road can be photographed. The camera may be a video camera that photographs moving images, or a still camera that photographs still images. When the camera is a still camera, it is preferable to continuously photograph still images at a predetermined interval so that the traveling direction and traveling speed of the vehicle can be grasped. When a semi-spherical camera is adopted as the camera, a wider range of the road can be monitored for wrong-way vehicles with one camera. The camera may also be provided at a position where the road can be photographed from the side. Here, the semi-spherical camera is, for example, a fisheye lens camera with a viewing angle of 180 degrees or more. By providing the camera at such a position, the camera can be installed more cheaply and easily than, for example, a case where the camera is provided so that the road can be photographed from a high place.

The storage unit stores the speed, passing position, and passing time of the vehicle when passing a preset set position in association with the set position that the vehicle passed. When the position of the first vehicle passing one of the multiple set positions is different from the current position of the vehicle estimated from the information stored in the storage unit in association with a set position located on the opposite side of the first vehicle's traveling direction from the first set position, the counting unit counts up the number of vehicles that have passed one set position. By the counting unit counting up in this manner, counting up at multiple set positions for the same vehicle is suppressed. In addition, by providing multiple set positions for counting up the number of vehicles, a vehicle that was hidden behind another vehicle at one set position and did not appear during imaging can be counted at the other set positions. Therefore, the passing vehicle counting device can suppress a decrease in counting accuracy even if the camera is installed in a position that is easy to install but is likely to be hidden behind another vehicle (for example, a position where the road is photographed from the side). In other words, the passing vehicle counting device can prevent a decrease in counting accuracy even when capturing images of the road from the side.

In this embodiment, the camera may be a hemispherical camera, and the acquisition unit may perform planar expansion processing and distortion correction processing on the fisheye image acquired from the hemispherical camera. With such features, it is possible for a wide range of the road to be included in the shooting range by the hemispherical camera. In addition, the fisheye image acquired from the hemispherical camera can be converted into an image that makes it easy to recognize the vehicle, etc.

The technology described above can also be understood from the perspective of a method for counting the number of passing vehicles and a program for counting the number of passing vehicles.

The following further describes a passing vehicle counting system that uses the passing vehicle counting device as a counting device with reference to the drawings. FIG. 1 is a diagram showing an example of the configuration of a passing vehicle counting system according to an embodiment. The passing vehicle counting system 500 illustrated in FIG. 1 includes a counting device 1, a surveillance camera 2, and a surveillance server 301 installed in a surveillance center 300.

In the passing vehicle counting system 500, the counting device 1 counts the number of vehicles passing through the road 600 based on an image of the road 600 captured by the surveillance camera 2. The road 600 illustrated in FIG. 1 is a two-lane road including a lane 601 whose travel direction is from right to left as one looks at the drawing, and a lane 602 whose travel direction is from left to right as one looks at the drawing. However, the road on which the passing vehicle counting system 500 counts the number of vehicles is not limited to a two-lane road, and may be a road with three or more lanes, or may be a one-lane road.

The surveillance camera 2 is, for example, a camera that captures the road 600. The surveillance camera 2 is installed, for example, on the side of the road 600. The surveillance camera 2 is, for example, a digital camera having a Charge Coupled Device (CCD) sensor or a Complementary Metal Oxide Semiconductor (CMOS) sensor. The surveillance camera 2 is capable of capturing at least one of video and still images.

The surveillance camera 2 is installed in a position where it can capture an image of the road 600. The surveillance camera 2 is installed, for example, at the side of the road 600 and captures the image of the road 600 from the side. However, the installation location of the surveillance camera 2 is not limited to the side of the road 600. The surveillance camera 2 may be installed, for example, on a pedestrian bridge crossing above the road 600. In this case, the surveillance camera 2 may capture an image of the road 600 from above.

The surveillance camera 2 may be installed in a position where it can capture, for example, a specific area of the road 600 (such as an entrance/exit to a motorway, an entrance/exit to a service area, or a roadside). Furthermore, multiple surveillance cameras 2 may be placed along the road 600 so that the entire road 600 can be monitored. In this embodiment, a hemispherical camera is used as the surveillance camera 2. By using a hemispherical camera, a wider area of the road 600 can be captured by a single surveillance camera 2. Area R1 in FIG. 1 illustrates an example of the area that can be captured by the surveillance camera 2, which is a hemispherical camera. The surveillance camera 2 is an example of a "camera".

The counting device 1 is an information processing device that analyzes images captured by the surveillance camera 2 and counts the number of vehicles traveling on the road 600. The counting device 1 notifies information such as the counted number of vehicles to a surveillance server 301 installed in a surveillance center 300. Note that while the counting device 1 and the surveillance camera 2 are separate devices in FIG. 1, the counting device 1 and the surveillance camera 2 may be integrated into a single vehicle counting device.

The monitoring center 300 has a monitor stationed there, who monitors traffic conditions such as the number of vehicles traveling on the road 600. The monitoring server 301 is an information processing device operated by the monitor stationed at the monitoring center 300. When the monitoring server 301 receives information such as the number of vehicles from the counting device 1, it notifies the monitor of the number of vehicles passing through the road 600. The monitor is notified, for example, by outputting the number of passing vehicles on a display.

The network N1 is a computer network that connects information processing devices so that they can communicate with each other. The network N1 is, for example, the Internet, a dedicated line, or a Local Area Network (LAN), and may be either wired or wireless.

<Hardware Configuration of Counting Device 1>
2 is a diagram showing an example of a hardware configuration of a counting device according to an embodiment. The counting device 1 is an information processing device including a Central Processing Unit (CPU) 101, a main memory unit 102, an auxiliary memory unit 103, a communication unit 104, a connection terminal 105, and a connection bus B1. The CPU 101, the main memory unit 102, the auxiliary memory unit 103, the communication unit 104, and the connection terminal 105 are mutually connected by the connection bus B1.

The CPU 101 is also called a microprocessor unit (MPU) or a processor. The CPU 101 is not limited to a single processor, and may have a multiprocessor configuration. A single CPU 101 connected to a single socket may have a multicore configuration. At least a part of the processing executed by the CPU 101 may be executed by a processor other than the CPU 101, for example, a dedicated processor such as a digital signal processor (DSP), a graphics processing unit (GPU), a numerical calculation processor, a vector processor, or an image processing processor. At least a part of the processing executed by the CPU 101 may be executed by an integrated circuit (IC) or other digital circuit. At least a part of the CPU 101 may include an analog circuit. The integrated circuit includes a large scale integrated circuit (LSI), an application specific integrated circuit (ASIC), and a programmable logic device (PLD). The PLD includes, for example, a field-programmable gate array (FPGA). The CPU 101 may be a combination of a processor and an integrated circuit. The combination is called, for example, a microcontroller unit (MCU), a system-on-a-chip (SoC), a system LSI, a chip set, etc. In the counting device 1, the CPU 101 expands a program stored in the auxiliary storage unit 103 into a working area of the main storage unit 102, and controls peripheral devices through the execution of the program. This allows the counting device 1 to execute processing that meets a predetermined purpose. The main memory unit 102 and the auxiliary memory unit 103 are recording media that can be read by the counting device 1 .

The main memory unit 102 is exemplified as a memory unit directly accessed by the CPU 101. The main memory unit 102 includes a Random Access Memory (RAM) and a Read
Includes only memory (ROM).

The auxiliary storage unit 103 stores various programs and various data in a readable and writable recording medium. The auxiliary storage unit 103 is also called an external storage device. An operating system (OS), various programs, various tables, etc. are stored in the auxiliary storage unit 103. The OS includes a communication interface program that transfers data with external devices, etc. connected via the communication unit 104. The external devices, etc. include, for example, other information processing devices and external storage devices connected via a computer network, etc. The auxiliary storage unit 103 may be, for example, part of a cloud system, which is a group of computers on a network.

The auxiliary storage unit 103 is, for example, an erasable programmable ROM (EPROM), a solid state drive (SSD), a hard disk drive (HDD), etc. The auxiliary storage unit 103 may also be, for example, a compact disc (CD) drive device, a digital versatile disc (DVD) drive device, a Blu-ray (registered trademark) disc (BD) drive device, etc.

The communication unit 104 is, for example, an interface with the network N1. The communication unit 104 communicates with an external device via the network N1.

The connection terminal 105 is a terminal for connecting an external device such as the surveillance camera 2. Examples of the connection terminal 105 include a Universal Serial Bus (USB) terminal and a High-Definition Multimedia Interface (HDMI (registered trademark)) terminal.

The counting device 1 may further include an input unit that accepts operation instructions from a user, etc. Examples of such an input unit include input devices such as a keyboard, a pointing device, a touch panel, an acceleration sensor, or a voice input device.

The counting device 1 may include, for example, an output unit that outputs data processed by the CPU 101 and data stored in the main memory unit 102. Examples of such an output unit include a Cathode Ray Tube (CRT) display, a Liquid Crystal Display (LCD) display, and the like.
Examples of output devices include a LCD (Liquid Crystal Display), a Plasma Display Panel (PDP), an Electroluminescence (EL) panel, an organic EL panel, or a printer.

<Processing blocks of counting device 1>
3 is a diagram showing an example of a processing block of a counting device according to an embodiment. The counting device 1 includes an acquisition unit 11, a cross line setting unit 12, a tracking unit 13, an estimation unit 14, a counting unit 15, and a vehicle information database (referred to as "DB" in the figure) 16. The counting device 1 includes a main memory unit 102.
The CPU 101 executes the computer program that has been expanded so as to be executable, thereby performing processing as each part of the counting device 1, such as the acquisition unit 11, the intersection line setting unit 12, the tracking unit 13, the estimation unit 14, the counting unit 15, and the vehicle information database 16.

The acquisition unit 11 acquires an image of the road 600 captured by the surveillance camera 2. The acquisition unit 11 stores the acquired image of the road 600 in the auxiliary storage unit 103. FIG. 4 is a diagram showing an example of an image acquired by the acquisition unit in an embodiment. As can be understood by referring to FIG. 4, since the surveillance camera 2 is a semi-spherical camera, a wide range of the road 600 is captured, but the captured image is a distorted fisheye image. The acquisition unit 11 develops the fisheye image into a plane and stores the image with the distortion corrected in the auxiliary storage unit 103. By performing such processing, the acquisition unit 11 can generate a panoramic image including a range of 180 degrees up, down, left, and right from the fisheye image acquired from the surveillance camera 2. The acquisition unit 11 stores the generated panoramic image in the auxiliary storage unit 103. The acquisition unit 11 is an example of an "acquisition unit".

The cross line setting unit 12 sets a plurality of cross lines that cross the captured road 600 for the image stored in the auxiliary storage unit 103 by the acquisition unit 11. FIG. 5 is a diagram illustrating cross lines set by the cross line setting unit in an embodiment. In FIG. 5, a plurality of cross lines L1, L2, L3, L4, and L5 set by the cross line setting unit 12 are illustrated. The tracking frames W1, W2, W3, and W4 illustrated in FIG. 5 will be described later. In this embodiment, in the image stored in the auxiliary storage unit 103 by the acquisition unit 11, the road 600 extends in an approximately left-right direction. Therefore, the cross line setting unit 12 sets the cross lines L1, L2, L3, L4, and L5 that extend in an approximately up-down direction of the image. By setting the cross lines L1, L2, L3, L4, and L5 in this manner, the cross line setting unit 12 can set a plurality of cross lines L1, L2, L3, L4, and L5 that cross the road 600 for the image. In addition, although five crossing lines are illustrated in FIG. 5, the number of crossing lines is not limited to five. The number of crossing lines set by the crossing line setting unit 12 may be six or more, or may be four or less. The crossing lines set by the crossing line setting unit 12 are an example of a "set position."

The tracking unit 13 recognizes vehicles that appear in the panoramic image stored in the auxiliary storage unit 103 by the acquisition unit 11 through object recognition. For example, a convolutional neural network (CNN) and a single shot multi detector (SSD) can be used for object recognition. The tracking unit 13 tracks each of the recognized vehicles by surrounding them with tracking frames W1, W2, W3, and W4 illustrated in FIG. 5. The tracking frames W1, W2, W3, and W4 are, for example, rectangular. The tracking unit 13 associates information indicating the type of the recognized vehicle with the detection position and stores the information in the vehicle recognition table of the vehicle information database 16.

FIG. 6 is a diagram showing an example of a vehicle recognition table stored in a vehicle information database in an embodiment. The vehicle recognition table 161 illustrated in FIG. 6 includes the items of "vehicle type" and "detection position." "Vehicle type" stores information indicating the characteristics of the vehicle. Examples of information indicating the characteristics of the vehicle include a small vehicle, a medium-sized vehicle, a large vehicle, a tanker truck, an ambulance, and a fire engine. "Detection position" stores information indicating the coordinates at which the vehicle was detected during imaging. The coordinates stored in the detection position are, for example, a set of coordinates (X, Y coordinates) of the upper left corner and the lower right corner of a tracking frame surrounding the detected vehicle.

The tracking unit 13 also starts tracking the vehicle recognized by object recognition. The tracking unit 13 associates the tracking start time, tracking start position, and current position of the vehicle that has started tracking, and stores them in the tracking table of the vehicle information database 16.

FIG. 7 is a diagram showing an example of a tracking table stored in the vehicle information database in the embodiment. The tracking table 162 illustrated in FIG. 7 includes the items of "TID", "vehicle type", "tracking information", and "counted flag". "TID" stores information that uniquely identifies a set of information stored in "vehicle type", "tracking information", and "counted flag". The information stored in "vehicle type" is the same as the information stored in "vehicle type" of the vehicle recognition table 161. "Tracking information" stores information related to vehicle tracking. Examples of information related to vehicle tracking include a start time, a start position, and a current position. The start time is the time when the tracking unit 13 starts tracking the vehicle. The start position is the position of the vehicle when the tracking unit 13 starts tracking the vehicle. The current position is the position of the vehicle currently being tracked by the tracking unit 13. The position of the vehicle can be indicated by a set of coordinates (X, Y coordinates) of the upper left corner and the lower right corner of the tracking frame surrounding the detected vehicle, similar to the "detected position" of the vehicle information database 16. The "counted flag" stores information indicating that the counting unit 15 has counted the vehicles being tracked by the tracking unit 13. Fig. 5 illustrates an example in which "1" is stored when the vehicles have been counted, and "0" is stored when the vehicles have not been counted. However, the information stored in the "counted flag" is not limited to "1" and "0", and may be any information indicating whether or not the vehicles have been counted.

When a vehicle that is not being tracked passes through any of the intersection lines L1, L2, L3, L4, and L5, the tracking unit 13 acquires the passing speed, passing time, and passing position of the vehicle. FIG. 8 is a diagram showing three frames of images acquired by the acquisition unit from the monitoring camera. As shown in FIG. 8, the tracking unit 13 may not be able to continue tracking the vehicle that has started tracking (illustrated in a square frame in FIG. 8) because the vehicle is temporarily hidden behind another vehicle. If a hemispherical camera is used as the monitoring camera 2, the wide shooting range increases the possibility that the temporarily hidden vehicle will appear again during imaging. The tracking unit 13 stores information indicating the passing speed, passing time, and passing position of the vehicle for which tracking has been interrupted or a vehicle that has newly appeared during imaging in the passing vehicle table of the vehicle information database 16. The tracking unit 13 is an example of a "passing vehicle management unit".

Figure 9 is a diagram showing an example of a passing vehicle table stored in the vehicle information database in an embodiment. The passing vehicle table 163 illustrated in Figure 9 is prepared for each of the crossing lines L1, L2, L3, L4, and L5. That is, a passing vehicle table 163 for crossing line L1, a passing vehicle table 163 for crossing line L2, a passing vehicle table for crossing line L3, a passing vehicle table for crossing line L4, and a passing vehicle table for crossing line L4 are prepared. The passing vehicle table 163 for crossing line L1 stores information about vehicles that have passed through crossing line L1. The same applies to the passing vehicle tables 163 for the other crossing lines.

The passing vehicle table 163 illustrated in FIG. 9 includes the following items: "passing speed," "passing time," and "passing position." "Passing speed" stores the speed of the vehicle that passed the intersection line. "Passing time" stores the time when the vehicle passed the intersection line. "Passing position" stores the position where the vehicle passed the intersection line. The passing vehicle table 163 is an example of a "storage unit."

When a vehicle not tracked by the tracking unit 13 passes through any of the intersection lines L1, L2, L3, L4, and L5, the estimation unit 14 acquires the vehicle type, passing speed, passing time, and passing position of the passing vehicle. The estimation unit 14 further estimates the position of each vehicle stored in the passing vehicle table 163 at the current time based on the information stored in the passing vehicle table 163. In the case illustrated in FIG. 5, since the vehicle runs from right to left, the estimation unit 14 may estimate based on information corresponding to the intersection line located to the right of the intersection line through which the vehicle passes, among the information stored in the passing vehicle table 163. In other words, the estimation unit 14 may estimate based on information corresponding to the intersection line located on the opposite side of the traveling direction of the vehicle from the intersection line through which the vehicle passes. When the acquired passing position does not match the estimated current position, the estimation unit 14 determines that a new vehicle that has newly appeared during imaging has been detected. The estimation unit 14 stores the passing speed, passing time, and passing position of the detected new vehicle in the counted vehicle information table of the vehicle information database 16. The estimation unit 14 also causes the tracking unit 13 to start tracking the vehicle determined to be a new vehicle. The tracking unit 13 stores information about the new vehicle in the tracking table 162, and starts tracking the new vehicle.

Figure 10 is a diagram showing an example of a count vehicle information table stored in the vehicle information database in an embodiment. The count vehicle information table 164 illustrated in Figure 10 is prepared for each of the crossing lines L1, L2, L3, L4, and L5. That is, a count vehicle information table 164 for crossing line L1, a count vehicle information table 164 for crossing line L2, a count vehicle information table 164 for crossing line L3, a count vehicle information table 164 for crossing line L4, and a count vehicle information table 164 for crossing line L4 are prepared. The count vehicle information table 164 for crossing line L1 stores information on vehicles that have passed through crossing line L1. The same applies to the count vehicle information tables 164 for the other crossing lines.

The counted vehicle information table 164 includes the items "CID", "TID", "passing speed", "passing time", and "passing position". "CID" stores information that uniquely identifies the information stored in the counted vehicle information table 164. "CID" is preferably set to unique information within the entire counted vehicle information table 164 corresponding to each of the intersection lines L1, L2, L3, L4, and L5. "TID" stores information that is assigned when tracking is started by the tracking unit 13. The information stored in "TID" is the same as "TID" in the tracking table 162. The information stored in each of "passing speed", "passing time", and "passing position" is the same as the information stored in each of "passing speed", "passing time", and "passing position" in the passing vehicle table 163, so a description thereof will be omitted.

The counting unit 15 counts the number of vehicles passing through the road 600 within the range captured by the surveillance camera 2. When a vehicle being tracked by the tracking unit 13 passes through any of the intersection lines L1, L2, L3, L4, and L5, the counting unit 15 checks the information stored in the "counted flag" associated with the vehicle in the tracking table 162. If the information stored in the "counted flag" indicates that the vehicle has not yet been counted, the counting unit 15 stores information about the vehicle in the count target vehicle table of the vehicle information database 16. The counting unit 15 further turns on the "counted flag" for the vehicle in the tracking table 162.

The counting unit 15 refers to the tracking table 162 and the counted vehicle information table 164 associated with each of the intersection lines L1, L2, L3, L4, and L5, and extracts information about vehicles that have not yet been counted. The counting unit 15 stores the extracted information in the counted vehicle table of the vehicle information database 16. The counting unit 15 further turns on the "counted flag" for the vehicle in the tracking table 162.

Figure 11 is a diagram showing an example of a count target vehicle table stored in a vehicle information database in an embodiment. The count target vehicle table 165 illustrated in Figure 11 includes the fields "CID," "vehicle type," and "passing speed." Each of these fields has already been explained, so the explanation will be omitted. Note that the count target vehicle table 165 may further include vehicle color (car color) as an item.

The count target vehicle table 165 stores information on vehicles to be counted, one vehicle per row. Therefore, the counting unit 15 can count the number of vehicles passing through the road 600 within the range captured by the surveillance camera 2 by aggregating the information stored in the count target vehicle table 165.

The counting unit 15 notifies the monitoring server 301 of the number of counted vehicles. The counting unit 15 may further transmit information stored in the count target vehicle table 165 to the monitoring server 301. The counting unit 15 is an example of a "counting unit" and an "accumulation unit."

<Processing flow>
12 is a diagram showing an example of a process flow of the passing vehicle count system according to the embodiment. Hereinafter, an example of a process flow of the passing vehicle count system 500 will be described with reference to FIG.

In J1, the acquisition unit 11 acquires an image of the road 600 captured by the surveillance camera 2. The acquisition unit 11 stores the acquired image of the road 600 in the auxiliary storage unit 103. The image acquired in J1 is, for example, a distorted fisheye image as illustrated in FIG. 4. In J2, the acquisition unit 11 corrects the fisheye image acquired in J1 to generate a panoramic image.

In J3, the tracking unit 13 recognizes a vehicle that appears in the panoramic image generated in J2 by object recognition. The tracking unit 13 stores information related to tracking of the recognized vehicle in the vehicle recognition table 161. In J4, the tracking unit 13 starts tracking the vehicle whose tracking information has been stored in the vehicle recognition table 161.

In J5, when a vehicle that is not being tracked passes any of the intersection lines L1, L2, L3, L4, and L5, the tracking unit 13 stores the passing speed, passing time, and passing position of the vehicle in the passing vehicle table 163. When a vehicle that is not being tracked by the tracking unit 13 passes any of the intersection lines L1, L2, L3, L4, and L5, the estimation unit 14 determines whether or not the vehicle is a new vehicle that has appeared during imaging, based on the passing speed, passing time, and passing position of the passing vehicle and the information stored in the passing vehicle table 163. In J6, the estimation unit 14 stores information about the vehicle determined to be a new vehicle in J5 in the counted vehicle information table 164.

In J7, the counting unit 15 counts the number of vehicles passing through the road 600 within the range captured by the surveillance camera 2. Details of the counting process will be described later with reference to FIG. 13.

Figure 13 is a diagram showing an example of a processing flow for counting vehicles passing through a road in an embodiment. The processing in Figure 13 is executed, for example, at J7 in Figure 12. Below, an example of a processing flow for counting vehicles passing through the road 600 within the range captured by the surveillance camera 2 will be described with reference to Figure 13.

In J71, the estimation unit 14 determines whether the vehicle has passed any of the intersection lines L1, L2, L3, L4, and L5. If the vehicle has passed (YES in J71), the process proceeds to J72. If the vehicle has not passed (NO in J71), the process ends.

In J72, the estimation unit 14 determines whether or not the "counted flag" associated with the vehicle determined in J71 to have passed the intersection line in the tracking table 162 is on. If it is on (NO in J72), the processing ends. If it is not on (YES in J72), the processing proceeds to J73. Here, examples of a vehicle that is not on include a vehicle that has been tracked by the tracking unit 13 but has not yet been counted, and a vehicle that has not yet been tracked by the tracking unit 13.

In J73, the estimation unit 14 further estimates the position of each vehicle stored in the passing vehicle table 163 at the current time based on the information stored in the passing vehicle table 163.

In J74, the estimation unit 14 determines whether the position of the vehicle that passed the intersection line in J71 matches any of the vehicle positions estimated in J73. If they match (YES in J74), the processing ends. If they do not match (NO in J74), the processing proceeds to J75.

In J75, the estimation unit 14 stores the passing speed, passing time, and passing position of the vehicle that passed the intersection line in J1 in the counted vehicle information table 164.

In J76, the counting unit 15 counts the number of vehicles passing through the road 600 within the range captured by the surveillance camera 2. When a vehicle being tracked by the tracking unit 13 passes through one of the intersection lines, the counting unit 15 checks the information stored in the "counted flag" associated with the vehicle in the tracking table 162. If the information stored in the "counted flag" indicates that the vehicle has not yet been counted, the counting unit 15 stores information about the vehicle in the count target vehicle table 165.

The counting unit 15 also extracts information about vehicles that have not yet been counted from the counted vehicle information table 164. The counting unit 15 stores the extracted information in the counted vehicle table 165. The counting unit 15 counts the number of vehicles passing through the road 600 within the range captured by the surveillance camera 2 by aggregating the information stored in the counted vehicle table 165. The counting unit 15 turns on the "counted flag" associated with the counted vehicle in the tracking table 162.

<Effects of the embodiment>
When the monitoring camera 2 is installed on the side of the road 600, it is difficult to continuously track a vehicle that is being imaged because the vehicle is hidden behind another vehicle. If continuous tracking is not possible, there is a risk that the same vehicle will be counted multiple times, and the accuracy of the counting results of vehicles passing through the road 600 will decrease. In this embodiment, multiple intersection lines that intersect with the road 600 are set in the image. Then, the counting device 1 stores information such as the passing time, speed, and position of each vehicle that passes through the intersection line in the passing vehicle table 163. Based on the information stored in the passing vehicle table 163, the counting device 1 determines whether the vehicle passing through the intersection line is a new vehicle or a vehicle that has already passed through one of the intersection lines and been stored. If the vehicle is a new vehicle, the counting device 1 stores information of the vehicle in the counted vehicle information table 164. By such processing, information about the same vehicle is prevented from being stored in the counted vehicle information table 164 in a duplicated manner. The counting unit 15 can count the number of vehicles passing through the road 600 with greater accuracy by referring to the information stored in the counted vehicle information table 164 and the tracking table 162 which stores information on vehicles being tracked by the tracking unit 13.

According to this embodiment, even if a vehicle passing through the road 600 is hidden behind another vehicle, the number of vehicles passing through the road 600 can be counted with higher accuracy. Therefore, according to this embodiment, it is not necessary to install the surveillance camera 2 in a high position that allows a bird's-eye view of the road 600. Therefore, according to this embodiment, the installation of the surveillance camera 2 is easy. In other words, the installation cost of the surveillance camera 2 can be reduced, and the installation work can be simplified.

In this embodiment, a hemispherical camera is used as the surveillance camera 2. By using such a camera, a wider range of the road 600 can be captured. As a result, even if a vehicle is hidden behind another vehicle, the hidden vehicle is likely to reappear in a slightly later frame, as shown in the example of FIG. 8. Therefore, according to this embodiment, it is possible to prevent vehicles hidden behind other vehicles from being missed in the count.

In this embodiment, the counting unit 15 can also transmit the information stored in the count target vehicle table 165 to the monitoring server 301. Therefore, according to this embodiment, it is possible to notify the monitoring server 301 not only of the number of vehicles passing through the road 600, but also of what types of vehicles are passing through the road 600. Such information can be used, for example, to analyze the usage status of the road 600.

<Modification>
In this embodiment, the tracking table 162 includes a counted flag. The tracking table 162 may further include, for example, information indicating the intersection line where the vehicle was counted. By configuring the tracking table 162 in this way, the counting device 1 can grasp the position where the vehicle was counted.

In an embodiment, when a vehicle that is not being tracked by the tracking unit 13 passes through any of the intersection lines L1, L2, L3, L4, and L5, if the passing vehicle is a new vehicle, the estimation unit 14 stores information about the vehicle in the counted vehicle information table 164. However, the estimation unit 14 may also store information about vehicles that are being tracked by the tracking unit 13 in the counted vehicle information table 164.

Note that the passing vehicle count system 500 described in this embodiment is not limited to a configuration in which the surveillance camera 2 is installed on the side of the road 600. The surveillance camera 2 may be installed, for example, in a position that provides an overhead view of the road 600 (for example, a pedestrian bridge crossing above the road 600).

The embodiments and variations disclosed above can be combined with each other.

<<Computer-readable recording medium>>
An information processing program that causes a computer or other machine or device (hereinafter, referred to as a computer, etc.) to realize any of the above functions can be recorded on a recording medium that can be read by the computer, etc. Then, the computer, etc. can provide the function by reading and executing the program from the recording medium.

Here, a computer-readable recording medium refers to a recording medium that stores information such as data and programs electrically, magnetically, optically, mechanically, or chemically and can be read by a computer. Examples of such recording media that can be removed from a computer include flexible disks, magneto-optical disks, Compact Disc Read Only Memory (CD-ROM), Compact Disc-Recordable (CD-R), Compact Disc-ReWritable (CD-RW), Digital Versatile Disc (DVD), Blu-ray Disc (BD), Digital Audio Tape (DAT), 8mm tape, and memory cards such as flash memory. Additionally, examples of recording media that are fixed to a computer include hard disks and ROMs.

1: Counting device 2: Surveillance camera 101: CPU
102: Main memory unit 103: Auxiliary memory unit 104: Communication unit 105: Connection terminal 11: Acquisition unit 12: Intersection line setting unit 13: Tracking unit 14: Estimation unit 15: Counting unit 16: Vehicle information database 161: Vehicle recognition table 162: Tracking table 163: Passing vehicle table 164: Counted vehicle information table 165: Count target vehicle table 300: Monitoring center 301: Monitoring server 500: Passing vehicle number counting system 600: Road 601, 602: Lanes

Claims (5)

an acquisition unit that acquires an image of a range including a road captured by a camera;
a passing vehicle management unit that, when a vehicle passes through any one of a plurality of intersection lines that are preset to intersect with the road in the image, stores in a storage unit the speed, the passing position, and the passing time of the vehicle in association with the intersection line that the vehicle passed through;
a counting unit that, when a position of a first vehicle passing through one of the plurality of intersecting lines differs from a current position of the vehicle estimated from information stored in the storage unit in association with a intersecting line located on the opposite side of the first intersecting line in a traveling direction of the first vehicle, counts up the number of vehicles that have passed through the one intersecting line ;
and an accumulating unit that sums up the number of vehicles counted up for each of the plurality of intersection lines .
A device for counting the number of passing vehicles. The image acquired by the acquisition unit is an image of the road taken from the side.
2. The passing vehicle counting device according to claim 1. the camera is a hemispherical camera;
The acquisition unit performs planar expansion processing and distortion correction processing on the fisheye image acquired from the hemispherical camera.
3. The passing vehicle counting device according to claim 1 or 2. The computer
Obtaining an image of the area including the road captured by the camera;
When a vehicle passes any one of a plurality of intersection lines that are preset to intersect with the road in the image, the speed, the passing position, and the passing time of the vehicle are stored in a storage unit in association with the set position that the vehicle passed;
when a position of a first vehicle passing through one of the plurality of intersecting lines differs from a current position of the vehicle estimated from information stored in the storage unit in association with an intersecting line located on the opposite side of the first intersecting line in a traveling direction of the first vehicle, counting up the number of vehicles that have passed through the one intersecting line ;
Adding up the number of vehicles counted up for each of the plurality of intersection lines ;
A method for counting the number of passing vehicles. On the computer,
Acquire an image of a range including a road photographed by a camera;
When a vehicle passes any one of a plurality of intersection lines that are preset to intersect with the road in the image, the speed, the passing position, and the passing time of the vehicle are associated with the crossing line that has passed, and are stored in a storage unit;
when a position of a first vehicle passing through one of the plurality of intersecting lines differs from a current position of the vehicle estimated from information stored in the storage unit in association with a intersecting line located on the opposite side of the first intersecting line in a traveling direction of the first vehicle, the number of vehicles passing through the one intersecting line is counted up;
Adding up the number of vehicles counted up for each of the plurality of intersection lines ;
A program to count the number of passing vehicles.

Images