JP7326029B2 - Traffic image processor - Google Patents

Description

An embodiment of the present invention relates to a traffic image processing device.

Conventionally, the control center system, which monitors video data from cameras installed at arbitrary intervals on roads such as expressways and general roads, tunnels, etc., displays current events on a large display installed in the control center. In addition, when a traffic incident such as an accident occurs, the location related to the traffic incident is displayed.

On the other hand, due to the increase in HDD capacity, CPU speed, and network speed, video data captured by cameras installed at arbitrary intervals on roads such as expressways and general roads, tunnels, etc., can be stored for a long time. A system has been proposed in which it is possible to record, search for desired video data, and view the data. After the occurrence of a traffic incident, this system is used by related organizations such as the police to investigate the cause of the scene, for example.

JP 2018-174555 A JP 2005-006164 A Japanese Patent No. 3841644

However, if all video data from cameras provided at arbitrary intervals on roads such as expressways and general roads, tunnels, etc. are stored, the capacity will be enormous. In addition, although conventionally proposed systems can confirm events after they have occurred, it is difficult to identify the events and their causes prior to the occurrence of the event using only the video data from the camera that caused the traffic event. Ta.

Therefore, there is a demand for a mechanism for appropriately extracting video data related to traffic events and collectively storing the data.

A traffic image processing apparatus according to an embodiment includes an image acquisition unit, a notification information acquisition unit, an event identification unit, an upstream identification unit, an extraction unit, and a storage unit. The video acquisition unit acquires video data continuously recording a range in which a vehicle running on the road or tunnel is captured by each of the imaging units provided at arbitrary intervals along the road or tunnel. . The report information acquisition unit acquires report information that reports that a predetermined traffic event has occurred on a road or tunnel. The event identification unit identifies, based on the notification information, an imaging unit that is capturing an image of a predetermined traffic incident and a predetermined time period that includes the occurrence time of the predetermined traffic incident. The upstream identification unit identifies an imaging unit upstream from the place where the traffic incident occurred on the road or tunnel and a relevant time period related to the traffic incident, based on the identified imaging unit and the predetermined time period. The extraction unit extracts video data in a predetermined time period from the video data acquired from the imaging unit specified by the event specifying unit, and extracts the video data acquired from the upstream imaging unit specified by the upstream specifying unit. Among them, the video data of the related time zone is extracted. The storage unit collectively stores the video data of the predetermined time period and the video data of the related time period extracted by the extraction unit in the traffic event storage unit.

Drawing 1 is a figure showing an example of composition of a traffic control system concerning an embodiment. FIG. 2 is a diagram illustrating the configuration of the traffic image processing server of the embodiment. FIG. 3 is an explanatory diagram showing the concept of related time zones in the upstream CCTV camera of the embodiment. FIG. 4 is a diagram exemplifying a traffic incident list screen displayed by the display control unit of the embodiment; 5 is a diagram illustrating an example of video data displayed by a display control unit according to the embodiment; FIG. FIG. 6 is a flowchart showing processing performed by the traffic image processing server of the embodiment.

FIG. 1 is a diagram showing an example of the configuration of a traffic control system according to this embodiment. As shown in FIG. 1, in this embodiment, the traffic control system 1 includes CCTV cameras 150_1 to 150_n, fire alarms 160_1 to 160k, an emergency telephone 161, a disaster prevention receiving panel 162, an image server 170, and a CCTV. It has a central server 110 , a large screen display 111 , a traffic central system 120 and a traffic image processing server 100 .

CCTV cameras (imaging units) 150_1 to 150_n are an example of imaging units provided at arbitrary intervals along a road or tunnel. CCTV (closed-circuit television) cameras (imaging unit) 150_1 to 150_n are installed so that there are no blind spots in the tunnel in order to monitor events (traffic jams, accidents, fires, etc.) that occur in the tunnel. In addition, roads are also provided to monitor various events. In the example shown in FIG. 1, CCTV cameras 150_1-150_k are installed inside the tunnel, and CCTV cameras 150_k+1-150n (not shown) are installed outside the tunnel. The CCTV cameras 150_1 to 150_n continuously photograph vehicles traveling on roads and tunnels. This enables the CCTV cameras 150_1 to 150_n to photograph traffic incidents occurring on roads and tunnels. The intervals at which the CCTV cameras 150_1-150_n are provided may vary according to tunnels and roads.

A traffic event may be an event that occurs on a road or a tunnel, and may include, for example, an accident such as a fire.

The fire alarms 160_1 to 160_k transmit notification information indicating that a fire has broken out to the disaster prevention receiver panel 162 when a fire breaks out in the tunnel.

The emergency telephone 161 is, for example, a telephone installed on a road or tunnel. Make contact with the center possible.

The disaster prevention receiving panel 162 is a control panel for notifying the control center of traffic incidents such as accidents that occur on roads or tunnels. Based on the information from the fire alarms 160_1 to 160k and the emergency telephone 161, the disaster prevention receiving panel 162 transmits notification information indicating the occurrence of a predetermined traffic incident via the image server 170 to the CCTV central server. 110. The disaster prevention receiving panel 162 transmits notification information not limited to fire detection from the fire alarms 160_1 to 160k or emergency contact from the emergency telephone 161, but to the effect that a traffic event such as an accident has occurred. Any notification is acceptable.

The image server 170 includes an image processing unit 171, and after performing predetermined image processing for displaying the video data transmitted from the CCTV cameras 150_1 to 150_n on the large screen display 111, the CCTV central server 110 Send. The image server 170 also transmits video data to be stored in the traffic video processing server 100 via the CCTV central server 110 .

The image processing unit 171 performs image processing on video data transmitted from the CCTV cameras 150_1 to 150_n. For example, in the case of image data shot in a dark tunnel or at night, image processing is performed such as increasing brightness and removing noise. Detecting events such as a stopped vehicle, reverse driving, and the like are conceivable.

The CCTV central server 110 is a server for managing video data and notification information input through the image server 170 . The CCTV central server 110 performs control for displaying video data input from the image server 170 on the large screen display 111 .

The large screen display 111 displays video data input from the CCTV central server 110 . The large-screen display 111 of this embodiment normally switches and displays video data from a CCTV camera 150 installed in a tunnel or road at regular intervals (for example, several seconds).

For example, when a fire alarm or the like is issued from the fire alarm 160_2, alarm information indicating the occurrence of the fire and the location of the occurrence is sent to the CCTV central server 110 via the disaster prevention receiving panel 162 and the image server 170. sent.

When the image server 170 receives the notification information, the camera (eg, the CCTV camera 150_2) that captures the location where the traffic incident (eg, fire) occurred indicated in the notification information is photographed. Video data, video data captured by a CCTV camera downstream of the location where the traffic incident occurred (eg, CCTV camera 150_1), and CCTV camera upstream of the location where the traffic incident occurred (eg, CCTV camera 150_3) The image data being captured and the image data being captured by the CCTV camera that is capturing the tunnel entrance are specified and transmitted to the CCTV central server 110 .

As a result, the large-screen display 111 displays video data captured by the camera (eg, CCTV camera 150_2) capturing the location where the traffic incident (eg, fire) occurred, and the image data downstream of the location where the traffic incident occurred. video data taken by the CCTV camera (for example, CCTV camera 150_1), video data taken by the upstream CCTV camera (for example, CCTV camera 150_3) where the traffic incident occurred, and the tunnel portal The video data captured by the CCTV cameras that are capturing images are displayed on the display screens 111_1, 111_2, 111_3, and 111_4.

In addition, the CCTV central server 110 outputs video data and notification information input via the image server 170 to the traffic video processing server 100 .

The notification information includes at least the name of the traffic incident, the place where the traffic incident occurred (road name and KP (kilometer post) on the road), the date and time of occurrence, the time of occurrence, and the identification information of the CCTV camera that captured the traffic incident. is This allows the traffic image processing server 100 to store image data related to traffic events.

The traffic central system 120 transmits the average speed of a plurality of vehicles traveling on the road or tunnel to the traffic image processing server 100 as information on the road or tunnel.

FIG. 2 is a diagram illustrating the configuration of the traffic image processing server 100. As shown in FIG. As shown in FIG. 2, the traffic image processing server 100 includes an image acquisition unit 201, an image storage unit 202, a notification information acquisition unit 203, an event identification unit 204, an upstream identification unit 205, and an extraction unit 206. , a storage unit 207 , a traffic event storage unit 208 , and a display control unit 209 .

The video acquisition unit 201 acquires video data in which a range in which vehicles traveling on roads and tunnels are captured is continuously recorded by each of the CCTV cameras (imaging units) 150_1 to 150_n.

The video storage unit 202 includes camera video storage units 202_1 to 202_n for CCTV cameras (imaging units) 150_1 to 150_n. The camera video storage units 202_1 to 202_n individually store video data captured by each of the CCTV cameras (imaging units) 150_1 to 150_n (hereinafter, any of the CCTV cameras (imaging units) 150_1 to 150_n may be referred to as a CCTV camera 150). Note that the video data stored in the camera video storage units 202_1 to 202_n are deleted after a predetermined period (for example, 1 day or 2 days) has passed since the image was captured.

The report information acquisition unit 203 acquires, via the CCTV central server 110, report information that reports that a predetermined traffic event (for example, an accident such as a fire) has occurred on a road or tunnel.

The speed acquisition unit 210 acquires from the traffic central system 120 the average speed of a plurality of vehicles traveling on a road or tunnel where a given traffic event has occurred.

The incident identification unit 204 identifies the CCTV camera 150 capturing the predetermined traffic incident and the predetermined time period including the occurrence time of the predetermined traffic incident, based on the notification information.

The predetermined time period includes at least the time period from the time preceding the time of occurrence by the time a to the time of occurrence. As a result, the present embodiment extracts video data in a predetermined time period from a time slightly earlier than the time of occurrence. This improves the possibility of recognizing the cause of a traffic incident by checking the video data. The time a is set according to the embodiment, and may be, for example, 10 minutes. Also, the predetermined time period may include a time period after the occurrence time of the traffic incident.

Based on the CCTV camera 150 identified by the event identification unit 204 and the predetermined time period, the upstream identification unit 205 identifies the CCTV camera 150 upstream from the location where the traffic incident occurred on the road or tunnel, and the predetermined traffic incident. Identify the relevant time period to

Specifically, the upstream identification unit 205 identifies the CCTV camera 150 upstream from the CCTV camera 150 identified by the event identification unit 204 in a road or tunnel where a predetermined traffic event (for example, an accident such as a fire) has occurred. do. When the traffic event occurs in a tunnel, the upstream identification unit 205 of this embodiment detects at least all the CCTV cameras 150 provided from the CCTV camera 150 identified by the event identification unit 204 to the entrance of the tunnel. is identified as the upstream CCTV camera 150 . Furthermore, the upstream identifying unit 205 may also identify CCTV cameras 150 upstream from the tunnel portal as upstream CCTV cameras 150 .

Furthermore, when a traffic incident occurs on a road other than a tunnel, the upstream identification unit 205 identifies an appropriate number of CCTV cameras 150 as high-grade CCTV cameras 150, taking into consideration the installation intervals of the CCTV cameras 150 and the like.

Based on the average speed on the road or tunnel where the predetermined traffic event occurred, the upstream identification unit 205 of the present embodiment determines the number of possible vehicles traveling in relation to the traffic event before the occurrence time. Identify the relevant time zone containing the time zone.

FIG. 3 is an explanatory diagram showing the concept of relevant time zones in the upstream CCTV camera. In the example shown in FIG. 3, the CCTV camera 150_2 that captured the traffic incident (accident) has identified a predetermined time period from the time a before the time T of occurrence to the time T of occurrence.

In this case, the related time period of the upstream CCTV cameras 150_3 and 150_4 is identified by considering not only the occurrence time T but also the average speed acquired by the speed acquisition unit 210 .

That is, when the distance between the CCTV camera 150_2 and the CCTV camera 150_3 is a distance L1, the vehicle traveling on the road or the tunnel will move between the CCTV camera 150_2 and the CCTV camera 150_2 at an estimated travel time t1 obtained by dividing the distance L1 by the average speed V_ave. It is considered to move between cameras 150_3.

Therefore, the upstream identification unit 205 identifies a time period including time T−t1 from at least time T (occurrence time)−t1 (estimated travel time)−a as the relevant time period of CCTV camera 150_2.

Similarly, when the distance between the CCTV cameras 150_3 and 150_4 is L2, the estimated travel time between the CCTV cameras 150_3 and 150_4 is the estimated travel time t2 obtained by dividing the distance L2 by the average speed V_ave. Become.

Therefore, the upstream identification unit 205 identifies at least a time period including the time T-t1-t2 from the time T (occurrence time)-t1-t2-a as the relevant time period of the CCTV camera 150_3.

As described above, the upstream identification unit 205 of the present embodiment determines whether the vehicle related to the predetermined traffic event is traveling before the occurrence time based on the average speed on the road or tunnel where the predetermined traffic event occurred. Identify relevant time zones, including those that may be

Thus, in this embodiment, the video data of the identified relevant time period is extracted. In this embodiment, by performing the above-described identification, it is possible to extract video data including the process leading up to the occurrence of the traffic event (accident) in the vehicle x.

Returning to FIG. 2 , the extraction unit 206 extracts from the video storage unit 202 the video data of the CCTV camera 150 identified by the event identification unit 204 in the predetermined time period identified by the event identification unit 204 .

Similarly, the extracting unit 206 extracts from the video storage unit 202 the video data of the upstream CCTV camera 150 specified by the upstream specifying unit 205 in the related time zone specified by the upstream specifying unit 205 .

The storage unit 207 collectively stores the video data of the predetermined time period and the video data of the related time period extracted by the extraction unit 206 in the traffic event storage unit (storage unit) 208 . Furthermore, storage unit 207 adds information indicating the traffic incident to library capsule information 253 .

The traffic event storage unit 208 of the present embodiment stores video data of a predetermined time period related to a predetermined traffic event and video data of a related time period in the same directory (directories 251 and 252, for example). . Furthermore, the traffic event storage unit 208 stores library capsule information 253 .

Library capsule information 253 contains textual information related to traffic incidents. At least, the library capsule information 253 includes the name of the traffic incident, the date and time when the traffic incident occurred, the time when the traffic incident occurred, the name of the road where the traffic incident occurred, the kilometer post (location) where the traffic incident occurred, and the location where the traffic incident occurred. The storage destination of the video data of the predetermined time period and the storage destination of the video data of the related time period in which the upstream was shot are included in association with each other.

As described above, in this embodiment, a plurality of video data related to a traffic event are extracted and stored in a directory, and information related to the traffic event is stored as metadata as the library capsule information 253. and Information related to traffic events may be stored in each directory for each traffic event.

The display control unit 209 displays information stored in the traffic event storage unit 208 on a display unit such as the large screen display 111 .

FIG. 4 is a diagram exemplifying a traffic incident list screen displayed by the display control unit 209. As shown in FIG. Based on the library capsule information 253, the traffic incident list screen shown in FIG. 1 upstream image (image data of the upstream CCTV camera 150 closest to the traffic incident), second upstream image (image data of the upstream CCTV camera 150 second closest to the traffic incident), and so on are displayed.

This allows the user to recognize various information about traffic events. Further, anchors 401, 402, and 403 for displaying associated images are displayed on the site image, the first upstream image, and the second upstream image. Accordingly, when the anchors 401, 402, 403 are clicked, the display control unit 209 displays the video data stored in the directory as the traffic event.

FIG. 5 is a diagram showing an example of video data displayed by the display control unit 209. As shown in FIG. The video data shown in FIG. 5 is displayed, for example, when the anchors 401, 402, 403 shown in FIG. 4 are clicked. As shown in the display area 503 of FIG. 5, information from the CCTV camera 150 that captured the displayed image data, the time until the occurrence of the traffic incident, and the like may be displayed.

As shown in FIG. 5, the display control unit 209 displays a downstream button 501 and an upstream button 502 . When receiving a click on the downstream button 501 or the upstream button 502, the display control unit 209 selects the currently displayed video data from among the video data stored in the same directory as the currently displayed video data. , to the video data one step downstream or to the video data one step upstream.

FIG. 6 is a flowchart showing processing performed by the traffic image processing server 100 of this embodiment. It is assumed that video data obtained from the CCTV camera 150 has already been stored in the video storage unit 202 .

First, the broadcast information acquisition unit 203 acquires broadcast information via the CCTV central server 110 (S601). By acquiring the notification information, it is recognized that a predetermined traffic event has occurred.

Then, the speed acquisition unit 210 acquires the average speed on the road or tunnel where the predetermined traffic event occurred from the traffic central system 120 (S602).

Next, the incident identification unit 204 identifies the CCTV camera 150 capturing the predetermined traffic incident and the predetermined time period including the time of occurrence of the traffic incident based on the report information (S603). ).

Based on the CCTV camera 150 and the predetermined time period specified by the event specifying unit 204, the upstream specifying unit 205 selects the CCTV camera 150 upstream from the place where the specified traffic incident occurred, A time period is specified (S604).

Then, the extraction unit 206 extracts from the video storage unit 202 the video data of the CCTV camera 150 capturing the traffic incident identified by the event identification unit 204 in a predetermined time period, and also extracts the video data related to the upstream CCTV camera. Video data for the time period is extracted (S605).

Then, the storage unit 207 collectively stores the video data extracted by the extraction unit 206 in one directory of the traffic event storage unit 208, and adds information indicating the traffic event related to the video data to the library capsule information. (S606).

In this embodiment, when extracting the video data captured by the upstream CCTV camera 150, the average speed of vehicles traveling on roads or tunnels is taken into consideration, and the relevant time period to be extracted is specified. Therefore, it is possible to create a library by including the driving state of the vehicle related to the traffic event, even if the image data is from a location distant from the location where the traffic event occurred.

Furthermore, the present embodiment makes it possible to provide information as one capsule in which video data are compiled as a library that can be utilized to investigate the cause of the occurrence of the event. Realizing the provision of information for on-site inspection work and cause investigation by related organizations such as the police as well as road operators, and shortening of traffic control time, etc., as road administrators are expected to speed up the cause investigation. becomes feasible.

While several embodiments of the invention have been described, these embodiments have been presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and modifications 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 scope of the invention described in the claims and equivalents thereof.

1... Traffic control system 100... Traffic image processing server 110... CCTV central server 111... Large screen display 120... Traffic central system 150_1 to 150_n... CCTV camera 160_1 to 160_k... Fire alarm 161... Emergency telephone , 162... disaster prevention receiving board, 170... image server, 171... image processing unit, 201... image acquisition unit, 202... image storage unit, 203... notification information acquisition unit, 204... event identification unit, 205... upstream identification unit, 206 ... extraction unit 207 ... storage unit 208 ... traffic event storage unit 209 ... display control unit 210 ... speed acquisition unit.

Claims (2)

A video acquisition unit that acquires video data continuously recording a range in which a vehicle traveling on the road or tunnel is captured by each of the imaging units provided at arbitrary intervals along the road or tunnel;
a notification information acquisition unit that acquires notification information that notifies that a predetermined traffic event has occurred on the road or tunnel;
an event identification unit that identifies, based on the report information, an imaging unit that is capturing an image of the predetermined traffic event, and a predetermined time period that includes an occurrence time of the predetermined traffic event;
Identifying an imaging unit upstream from a place where the traffic incident occurred on the road or tunnel and an upstream related time period related to the predetermined traffic incident based on the identified imaging unit and the predetermined time period. a specific part;
extracting the video data of the predetermined time period from the video data acquired from the imaging unit specified by the event specifying unit, and the video acquired from the upstream imaging unit specified by the upstream specifying unit; an extraction unit for extracting video data of the relevant time period from the data;
a storage unit configured to collectively store the video data of the predetermined time period and the video data of the related time period extracted by the extraction unit in a traffic event storage unit;
A traffic image processing device comprising: Further comprising a speed acquisition unit that acquires an average speed of a plurality of vehicles traveling on the road or tunnel,
Based on the average speed on the road or tunnel where the predetermined traffic event occurred, the upstream identification unit determines the possibility that the vehicle related to the predetermined traffic event is traveling before the occurrence time. identifying the relevant time period that includes a time period of
The traffic image processing device according to claim 1.

Images