JP7643003B2 - Recording control device and recording control method - Google Patents

Description

The present invention relates to a recording control device and a recording control method.

There is a known technology that records video footage of the time an event occurs, such as in a drive recorder. With this technology, for example, when an event such as a traffic accident occurs, video footage is recorded from the time period before and after the event occurs. A predetermined fixed value is generally used for the duration of this time period. However, there are cases where video footage with a limited, fixed duration does not provide useful information for understanding the situation of a traffic accident or other traffic trouble.

In this regard, Patent Document 1 discloses a technology in which a period of time when there is movement in an image captured by a camera is analyzed by an image analysis processing means, and the period is determined to be the time zone of the event image.

JP 2003-69987 A

According to the technology disclosed in Patent Document 1, the period of the event video to be recorded can be dynamically set. However, the technology described in Patent Document 1 sets the period of the event video as the period when there is movement in the video, so appropriate event video cannot be recorded in a dynamic environment such as on a road. In other words, in a dynamic environment where there is traffic of cars and people, and where objects such as trees sway in the wind, appropriate event video cannot be obtained, and content useful for understanding the situation of a traffic trouble cannot be recorded. For this reason, there is a demand for technology that can record content useful for understanding the situation of a traffic trouble. There is also a demand for efficient use of limited storage capacity when recording events.

The object of the present invention is to provide a recording control device and a recording control method that can record information that is useful for understanding the situation of a traffic accident while minimizing the amount of data recorded.

A recording control device according to a first aspect includes:
an imaging data acquisition unit that acquires imaging data from an imaging device;
an event detection unit that detects the occurrence of an event;
an imaging data analysis unit that analyzes the imaging data and detects a predetermined object captured in the imaging data;
and a recording control unit that associates the first shooting data, which is the shooting data including the time point at which the event occurred, with the second shooting data, which is the shooting data including the time point at which the object was detected within a predetermined range from the time point at which the event occurred, and stores the associated data as event recording data.

In the recording control method according to the second aspect,
Acquire photographic data from an imaging device;
Detects the occurrence of an event,
Analyzing the photographic data to detect a predetermined object captured in the photographic data;
The first shooting data, which is the shooting data including the time point at which the event occurred, and the second shooting data, which is the shooting data including the time point at which the object was detected within a predetermined range from the time point at which the event occurred, are associated with each other and saved as event record data.

The present invention provides a recording control device and a recording control method that can record information that is useful for understanding the situation of a traffic accident while minimizing the amount of data recorded.

FIG. 1 is a schematic diagram illustrating an example of a configuration of a recording system according to an embodiment. 1 is a block diagram showing an example of a hardware configuration of a recording control device according to an embodiment; FIG. 11 is a schematic diagram illustrating an event record according to a comparative example. 5A and 5B are schematic diagrams illustrating storage of event record data in the embodiment. 5A and 5B are schematic diagrams illustrating storage of event record data in the embodiment. 5 is a flowchart showing an example of the operation of the recording control device according to the embodiment; 5A and 5B are schematic diagrams illustrating storage of event record data in the embodiment. 5A and 5B are schematic diagrams illustrating storage of event record data in the embodiment. 5 is a flowchart showing an example of the operation of the recording control device according to the embodiment;

Below, an embodiment of the present invention will be described with reference to the drawings. Note that, for clarity of explanation, the following description and drawings have been omitted and simplified as appropriate. In addition, in each drawing, the same elements are given the same reference numerals, and duplicate explanations are omitted as necessary.

<First embodiment>
1 is a schematic diagram showing an example of the configuration of a recording system 10 according to an embodiment. The recording system 10 includes a recording control device 100, an image capture device 200, a recording unit 300, and an acceleration sensor 400.

In this embodiment, the recording system 10 is, as an example, a system mounted on a vehicle. In this case, the recording system 10 is also called a drive recorder. Note that the vehicle is, for example, an automobile that runs on four or more wheels, but may also be a motorcycle.

The imaging device 200 is, for example, a camera. The imaging device 200 captures the surroundings of the vehicle and generates imaging data, which is data of the captured video. The imaging device 200 may be, for example, a camera that captures the front of the vehicle, a camera that captures the rear of the vehicle, or a camera that captures the side of the vehicle. A plurality of imaging devices 200 may be provided. A single imaging device 200 may capture images in all directions of the vehicle. The imaging device 200 outputs the imaging data to the recording control device 100.

The recording unit 300 is a storage device that stores the shooting data. The recording unit 300 is a storage device that is configured with, for example, a memory card including a flash memory, a non-volatile memory such as an SSD (Solid State Drive) or an HDD (Hard Disk Drive), and a volatile memory such as a DDR (Double-Data-Rate Synchronous Dynamic Random Access Memory). The recording unit 300 records the shooting data according to the control of the recording control device 100. The recording unit 300 may record audio data of audio collected by a microphone (not shown) provided in the recording system 10 along with the shooting data. In this embodiment, the recording unit 300 is mounted on the vehicle together with the recording control device 100, but the recording unit 300 may be a device that is connected to the recording control device 100 by wireless communication. For example, the recording unit 300 may be a mobile terminal such as a smartphone, or an external server.

The recording unit 300 has a normal recording unit 310 and an event recording unit 320. The normal recording unit 310 is composed of a non-volatile memory and a volatile memory, and temporarily stores data in the volatile memory before recording it in the non-volatile memory. The normal recording unit 310 records the shooting data in an overwritable manner. In other words, when the storage capacity is insufficient, the normal recording unit 310 can delete data, for example, starting with the oldest shooting data, and store new shooting data in the resulting free storage area. The event recording unit 320 is composed of a non-volatile memory. The event recording unit 320 records the shooting data as event recording data. The event recording data is shooting data related to a detected event, and is data that is prohibited from being overwritten. In other words, the event recording data is not deleted even after time has passed, unless the user intentionally performs a deletion operation or the like. The normal recording unit 310 may be a recording area that is predetermined as an overwritable storage area, and the event recording unit 320 may be a recording area that is predetermined as a storage area that is prohibited from being overwritten. However, such a determination may not be made in advance for the storage area. Overwriting may be prohibited by adding a flag indicating that overwriting is prohibited to data that should not be overwritten and storing the data. In this case, the recording unit 300 may share the non-volatile memory of the normal recording unit 310 and the non-volatile memory of the event recording unit 320.

The acceleration sensor 400 is, for example, a three-axis acceleration sensor, and detects the acceleration of the vehicle. The acceleration sensor 400 is installed at any location on the vehicle. The acceleration sensor 400 may be built into the recording control device 100. The acceleration sensor 400 detects the acceleration applied to the acceleration sensor 400, and supplies the detected acceleration to the event detection unit 130 of the recording control device 100. Note that the acceleration sensor 400 may be a one-axis or two-axis acceleration sensor instead of a three-axis acceleration sensor.

The recording control device 100 has a shooting data acquisition unit 110, a shooting data analysis unit 120, an event detection unit 130, and a recording control unit 140. The processing of each component of the recording control device 100 may be realized, for example, by executing a program. In this case, the recording control device 100 has, for example, a hardware configuration as shown in FIG. 2, and the processor 150 reads and executes software (computer program) from memory 151 to perform processing of the shooting data acquisition unit 110, the shooting data analysis unit 120, the event detection unit 130, and the recording control unit 140.

Memory 151 is composed of, for example, a combination of volatile memory and non-volatile memory. Memory 151 is used to store software (computer programs) including one or more instructions executed by processor 150, and data used for various processes of recording control device 100, etc.

The processor 150 may be, for example, a microprocessor, a microprocessor unit (MPU), or a central processing unit (CPU). The processor 150 may include multiple processors.

As shown in FIG. 2, the recording control device 100 may include a network interface 152 for communicating with other devices. The network interface 152 may include, for example, a network interface card (NIC).

In this way, the recording control device 100 may have the functionality of a computer.
The above-mentioned program can be stored and supplied to a computer using various types of non-transitory computer readable media. The non-transitory computer readable media includes various types of tangible storage media. Examples of the non-transitory computer readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R/Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)). The program may be provided to the computer by various types of transitory computer readable media. Examples of the transitory computer readable medium include an electric signal, an optical signal, and an electromagnetic wave. The transitory computer readable medium may provide the program to the computer via a wired communication path such as an electric wire or an optical fiber, or via a wireless communication path.

The processing of the shooting data acquisition unit 110, shooting data analysis unit 120, event detection unit 130, and recording control unit 140 will be described below.

The shooting data acquisition unit 110 acquires shooting data from the imaging device 200. That is, the shooting data acquisition unit 110 acquires shooting data that is data of the video captured by the imaging device 200.

The event detection unit 130 detects the occurrence of an event. In this embodiment, the event specifically refers to a traffic problem. Traffic problems include not only traffic accidents but also dangerous driving such as sudden stops and sudden starts. In this embodiment, the event detection unit 130 acquires the acceleration of the vehicle in which the recording control device 100 is mounted from the acceleration sensor 400, and detects whether or not an event has occurred based on the acquired acceleration. Specifically, the event detection unit 130 determines that an event has occurred when acceleration equal to or greater than a predetermined threshold is detected.

The event detection unit 130 may detect the occurrence of an event by other methods. For example, it may be determined that an event has occurred when a switch (not shown) provided on the recording control device 100 or the like is operated. In this case, the user operates the switch when a traffic accident occurs. In this case, the recording system 10 may not be equipped with an acceleration sensor 400. Also, instead of the acceleration sensor 400, the occurrence of an event may be detected based on vehicle control information such as sudden braking and sudden steering.

When the event detection unit 130 detects the occurrence of an event, it outputs an event recording instruction signal to the recording control unit 140 to instruct the recording control unit 140 to save the captured data as event recording data. Note that the event recording instruction signal may include the time when the occurrence of the event was detected.

In order to save a scene in which a specific object is captured as event record data, the shooting data analysis unit 120 analyzes the shooting data to detect the specific object captured in the shooting data. If there are multiple imaging devices 200, the shooting data analysis unit 120 analyzes the shooting data of each imaging device 200. When the shooting data analysis unit 120 detects the specific object, it notifies the recording control unit 140 of time information indicating at what point in time the object was captured in the video. If multiple specific objects are detected, the shooting data analysis unit 120 may notify the recording control unit 140 of time information indicating at what point in time the object was captured in the video for each object.

In this embodiment, the specified object is the vehicle license plate. In this embodiment, not only the video at the time of the event occurrence, but also the video showing the vehicle license plate is saved as event recording data. This vehicle may be a vehicle in which a witness to the traffic accident is riding, or a vehicle in which the other party in the traffic accident is riding. Therefore, by saving the video showing such a vehicle as event recording data, it is possible to obtain clues to identify the witness to the traffic accident or the person involved in the traffic accident. In other words, it is possible to record content that is useful for understanding the situation of the traffic accident.

The image capturing data analysis unit 120 uses a known image recognition process to analyze the image capturing data and detects the license plate of the vehicle captured in the image capturing data. The image capturing data analysis unit 120 may detect a license plate that appears in the image at a size equal to or larger than a predetermined size. Here, the predetermined size refers to, for example, a size at which the license plate can be recognized by the image capturing data analysis unit 120. When a license plate appears at a size equal to or larger than the predetermined size, the vehicle with the license plate is one that was present within a certain distance from the image capturing device 200. For this reason, it is considered that such a vehicle is relatively likely to be a vehicle in which a party involved in or a witness to a traffic accident is riding. For this reason, by detecting a license plate that appears at a size equal to or larger than the predetermined size, it is possible to record information that is even more useful in understanding the situation of the traffic accident.

The shooting data analysis unit 120 may analyze shooting data of a video shot before the occurrence of an event, may analyze shooting data of a video shot after the occurrence of an event, or may analyze both.
In this embodiment, the event recording instruction signal is also output to the photographing data analysis unit 120, and when the photographing data analysis unit 120 receives the event recording instruction signal, it analyzes the photographing data around the time when the event was detected. For example, the photographing data analysis unit 120 analyzes the photographing data for a predetermined analysis target period (e.g., 10 minutes) starting from the event detection timing. This analysis target period is set to a value longer than the normal period (e.g., 30 seconds) of the event recording data. It is not necessary to set the analysis target period. However, in this case, the event recording data in which a vehicle photographed at a time significantly away from the event detection timing is captured may be saved. In other words, the event recording data in which a vehicle that is unlikely to be a vehicle in which a witness to the traffic trouble or a person involved in the traffic trouble is riding may be saved.

The recording control unit 140 controls the recording unit 300 to record the image data of the video captured by the imaging device 200. If there are multiple imaging devices 200, the recording control unit 140 records the image data of each imaging device 200.

If an event recording instruction signal is not output from the event detection unit 130, the recording control unit 140 continues to record the shooting data in the normal recording unit 310. In other words, if an event is not detected, the recording control unit 140 records the shooting data acquired by the shooting data acquisition unit 110 in the normal recording unit 310 in an overwritable manner. Continuing to record shooting data in an overwritable manner in this manner is called loop recording or normal recording. Then, if an event recording instruction signal is output from the event detection unit 130, the recording control unit 140 records the shooting data in the event recording unit 320. In other words, if an event is detected, the recording control unit 140 saves the shooting data including the video at the time of the event occurrence as event recording data. For example, the recording control unit 140 continues to record the shooting data in a normal manner, that is, loop recording, regardless of the presence or absence of an event recording instruction signal, and if an event is detected, the shooting data stored in the normal recording unit 310 for the period to be saved as event recording data is copied to the event recording unit 320, thereby saving the event recording data. At this time, copying may be performed from the volatile memory of the normal recording unit 310 to the event recording unit 320, or copying may be performed from the non-volatile memory of the normal recording unit 310 to the event recording unit 320. Note that saving the video as event recording data is referred to as event recording.

When recording an event, the recording control unit 140 adjusts the period of the video to be saved as event recording data so that the period includes the time point at which the image capture data analysis unit 120 detects a specific object. For example, if the image capture data analysis unit 120 detects a license plate that is larger than a specific size in the video, the recording control unit 140 adjusts the period of the video to be saved as event recording data so that the period includes the time point at which the license plate is detected that is larger than the specific size. The recording control unit 140 determines the time range of the video to be saved as event recording data based on the time information notified by the image capture data analysis unit 120, i.e., the time information indicating at what point in time the specific object was captured in the video.

FIG. 3 is a schematic diagram for explaining an event recording according to a comparative example. Here, it is assumed that an event is detected at time t _E. In this case, in the event recording according to the comparative example, the photographed data from time t _P1 to time t _P2 is saved as event recording data. Here, time t _P1 is a time that is a predetermined time P1 back from time t _E when the event is detected. Also, time t _P2 is a time that is a predetermined time P2 has elapsed from time t _E when the event is detected. Here, the predetermined times P1 and P2 are, for example, about several tens of seconds. When such an event recording is performed, photographed data for a certain period, that is, a period defined as the sum of P1 and P2, is saved as event recording data. However, the photographed data within this period does not necessarily contain content that is useful for understanding the situation of the traffic trouble.

In response to this, as described above, in this embodiment, the recording control unit 140 adjusts the period of the video to be saved as event record data so that the period includes the time point at which a specific object, such as a license plate, was detected. This makes it possible to save event record data that includes content that is useful for understanding the situation of a traffic accident.

For example, when a predetermined object is detected by analyzing the shooting data of a video shot before the occurrence of an event, as shown in FIG. 4, the recording control unit 140 makes the period of the video saved as the event record data include the detection time of the object shot before the occurrence of the event. FIG. 4 is a schematic diagram for explaining the saving of the event record data in this embodiment. In the example shown in FIG. 4, it is assumed that an event is detected at time _tE . That is, while the recording control unit 140 is performing normal recording, that is, loop recording, the event detection unit 130 detects an event at time _tE . Also, it is assumed that an object is detected in the video at time t1, which is the time before the event occurrence, by the analysis of the shooting data analysis unit _120. In this case, the recording control unit 140 saves, for example, the shooting data for the period from time _t1 to the above-mentioned time _tP2 as the event record data. That is, the start point of the video of the event record data is set to time _t1 , and the end point is set to time _tP2 . Note that the recording control unit 140 may set the start point to a time point a predetermined time before time _t1 . Also, in FIG. 4, the time _t1 at which the object is detected is before the time _tP1 , but it may be the same as the time _tP1 or may be later.

By adjusting the start point, the period of the video is adjusted to include the time when a specific object was detected, making it possible to save event recording data that includes content that is useful for understanding the situation of a traffic accident. In particular, for example, by saving the event recording data so that it includes the time when a vehicle captured before the event occurred, it is possible to save evidence video that can be used to identify the cause of the traffic accident.

Also, when a predetermined object is detected by analyzing the shooting data of the video shot after the occurrence of the event, as shown in FIG. 5, the recording control unit 140 makes the period of the video saved as the event record data include the detection time of the object shot after the occurrence of the event. FIG. 5 is a schematic diagram for explaining the saving of the event record data in this embodiment. In the example shown in FIG. 5, it is assumed that an event is detected at time _tE . That is, while the recording control unit 140 is performing normal recording, that is, loop recording, the event detection unit 130 detects an event at time _tE . Also, it is assumed that an object is detected in the video at time t2, which is the time after the event occurrence, by the analysis of the shooting data analysis unit _120. In this case, the recording control unit 140 saves, for example, the shooting data for the period from time _tP1 to time _t2 as described above as the event record data. That is, the start point of the video of the event record data is set to time _tP1 , and the end point is set to time _t2 . Note that the recording control unit 140 may set the end point to a time point a predetermined time later than time _t2 . Also, in FIG. 5, the time _t2 at which the object is detected is after the time _tP2 , but it may be the same as the time _tP2 or may be earlier than that.

By adjusting the end point, the period of the video is adjusted to include the time when a specific object was detected, making it possible to save event record data that includes content that is useful for understanding the situation of a traffic accident. In particular, by saving event record data to include the time when a vehicle captured after the event occurred, for example, it is possible to save video that identifies a vehicle that fled after a traffic accident. It is also useful for finding people who witnessed a situation, such as a dispute between two parties, that occurred after the event was detected.

The recording control unit 140 may store the event record data so as to include both the detection time of an object captured before the occurrence of the event and the detection time of an object captured after the occurrence of the event. That is, for example, the recording control unit 140 may store the event record data including the video from the above-mentioned time _t1 to time _t2 .

Next, an example of the operation of the recording control device 100 will be described. FIG. 6 is a flowchart showing an example of the operation of the recording control device 100. Below, an example of the operation of the recording control device 100 will be described with reference to FIG. 6.

In step S100, the recording control unit 140 of the recording control device 100 starts normal recording. Normal recording is started when the engine or power of the vehicle equipped with the recording control device 100 is turned on. In other words, the recording control unit 140 continues to record the shooting data acquired by the shooting data acquisition unit 110 in the normal recording unit 310 while the vehicle is operating.

Then, in step S101, the recording control unit 140 determines whether or not the event detection unit 130 has detected an event. If an event is detected, the process proceeds to step S102, and if an event is not detected, the process proceeds to step S106.

In step S102, the image capture data analysis unit 120 analyzes the image capture data around the time the event was detected. As described above, the image capture data analysis unit 120 may analyze video footage from a period going back from the event detection timing, or from a period after the event detection timing, or may analyze both.

In step S103, the recording control unit 140 determines whether the shooting data analysis unit 120 has detected a specified object. That is, the recording control unit 140 determines whether the shooting data analysis unit 120 has identified the time when the specified object appears in the video. If the time when the specified object appears in the video has been identified, the process proceeds to step S104; if not, the process proceeds to step S105.

In step S104, the recording control unit 140 adjusts the start point or end point of the video of the event record data according to the detection result by the shooting data analysis unit 120. That is, as described above, the recording control unit 140 adjusts the start point or end point so that the period of the video saved as the event record data includes the time point at which the object was detected. After step S104, the process proceeds to step S105.

In step S105, the recording control unit 140 stores the event recording data in the event recording unit 320. If the start point or end point was adjusted in step S104, the recording control unit 140 uses the start point or end point to determine the period of the video of the event recording data. On the other hand, if the processing of step S104 is skipped, that is, if the answer is No in step S103, the recording control unit 140 uses the start point and end point specified by the above-mentioned predetermined times P1 and P2 to determine the period of the video of the event recording data. After step S105, the processing proceeds to step S106.

In step S106, the recording control unit 140 determines whether an operation has been performed to end normal recording, such as turning off the vehicle engine or power. If normal recording is not to be ended, the process returns to S101. On the other hand, if an operation has been performed to end normal recording, the recording control unit 140 ends normal recording.

The above describes the first embodiment. According to this embodiment, the recording control unit 140 saves not only the video at the time of the event occurrence, but also the video showing the vehicle's license plate as event recording data. This makes it possible to record video of a vehicle in which a witness to a traffic accident is riding, or a vehicle in which the other party in the traffic accident is riding. This makes it possible to record content that is useful for understanding the situation of the traffic accident.

In the above embodiment, the vehicle license plate is detected as the specified object, but the vehicle itself may be detected as the specified object by analysis using image recognition processing. Also, other characteristic parts of the vehicle besides the license plate may be detected.

<Embodiment 2>
Next, a description will be given of a second embodiment. Below, differences from the first embodiment will be described, and descriptions of similar configurations and processes will be omitted as appropriate. This embodiment differs from the first embodiment in that the photographic data analysis unit 120 detects a person as a predetermined object.

That is, in this embodiment, the shooting data analysis unit 120 uses a known image recognition process to analyze the shooting data and detect people who appear in the shooting data. As a result, in this embodiment, not only the video at the time of the event occurrence but also the video in which the person appears is saved as event record data. This person may be a witness to the traffic accident or the other party in the traffic accident. Therefore, by saving the video in which such a person appears as event record data, it is possible to obtain clues to identify the witness to the traffic accident or the party involved in the traffic accident. In other words, it is possible to record content that is useful for understanding the situation of the traffic accident.

The image data analysis unit 120 may detect the face of a person. By detecting the face of a person, it becomes easier to identify the person captured in the image, and more useful information can be obtained in understanding the situation of the traffic trouble.

The above describes the second embodiment. According to this embodiment, the recording control unit 140 saves not only the video at the time of the event occurrence, but also the video in which a person is captured as event recording data. This makes it possible to record content that is useful for understanding the situation of a traffic trouble. It is also possible to combine the first and second embodiments. In other words, a person and a vehicle license plate may be detected as the specified object.

Although the first and second embodiments have been described above, the following modifications are also possible to these embodiments.

For example, when the photographing data analysis unit 120 detects a plurality of predetermined objects and notifies the recording control unit 140 of time information indicating at what point in time each of the objects was captured in the video, the recording control unit 140 may control as follows. That is, in this case, the recording control unit 140 may adjust the period of the video saved as the event record data so that the video contains a predetermined number of objects or more. Here, this predetermined number is an arbitrary number of 2 or more. In this way, the video including multiple objects can be stored as the event record data. For example, when finding witnesses to a traffic accident from the video of the event record data, it is preferable that the number of objects corresponding to the number of candidates for the witnesses and the license plate of the vehicle in which the witness may be riding is large. As described above, by saving the video including multiple objects, it is possible to obtain a video that is more useful for understanding the situation of the traffic accident.

<Third embodiment>
In the above-described embodiment, the recording control unit 140 stores a series of shooting data for a series of images including both the occurrence time of the event and the detection time of the object as event recording data. In contrast, in the present embodiment, the recording control unit 140 adjusts the period of the images to be stored as event recording data so that the period includes the detection time of the object as follows. In the present embodiment, the recording control unit 140 associates the first shooting data, which is shooting data including the occurrence time of the event, with the second shooting data, which is shooting data including the detection time of the object within a predetermined range from the occurrence time of the event, and stores the data as event recording data. Here, the predetermined range is, for example, the time from the occurrence time of the event to the detection of the object or a distance to be described later. In the present embodiment, for example, as shown in FIG. 7, the recording control unit 140 associates the shooting data 20 of a first predetermined period including the occurrence time of the event with the shooting data 21 of a second predetermined period including the detection time of the object, and stores the data as event recording data. Here, the shooting data 20 corresponds to the above-described first shooting data, and the shooting data 21 corresponds to the above-described second shooting data.

In the example shown in FIG. 7, it is assumed that an event is detected at time _tE . It is also assumed that an object is detected in the video at time _t1 . In this case, the recording control unit 140 stores the shooting data from time _tP1 to time _tP2 and the shooting data from time _tP3 to time _tP4 as event recording data. Here, time _tP1 is the time that is a predetermined time P1 back from time _tE when the event is detected. Also, time _tP2 is the time that is a predetermined time P2 has passed from time _tE when the event is detected. Furthermore, time _tP3 is the time that is a predetermined time P3 back from time _t1 . Also, time _tP4 is the time that is a predetermined time P4 has passed from time _t1 . By creating the event recording data in this manner, it is possible to omit storing the shooting data from time _tP4 to time _tP1 while storing both the video at the time when the event occurs and the video at the time when the object is detected. Therefore, consumption of storage capacity can be suppressed. That is, according to this embodiment, it is possible to record contents that are useful for understanding the situation of a traffic trouble while suppressing the amount of recorded data. Note that while FIG. 7 shows an example of detection of an object before the occurrence of an event, the same applies to detection of an object after the occurrence of an event.

FIG. 7 shows an example in which one object is detected, but in the case where multiple objects are detected, multiple second shooting data corresponding to the detected objects may be associated with the first shooting data and stored as event record data. FIG. 8 is a diagram showing an example of storing event record data in the case where multiple objects are detected. In the example shown in FIG. 8, an event is detected at time _tE , and objects are detected at times _t1 , _t2 , and _t3 . In this case, the recording control unit 140 associates the shooting data 21a corresponding to the object detected at time _t1 , the shooting data 21b corresponding to the object detected at time _t2 , and the shooting data 21c corresponding to the object detected at time _t3 with the shooting data 20 corresponding to the occurrence time of the event, and stores them as one event record data. In this way, in the case where multiple objects are detected, the recording control unit 140 may associate the second shooting data corresponding to the detected objects with the first shooting data and store them as event record data.

In the example shown in FIG. 8, all the second shooting data corresponding to the detected object are associated with the first shooting data, but not all the second shooting data may be associated with the first shooting data. In this case, the second shooting data to be associated with the first shooting data may be selected according to priority. That is, when a plurality of objects are detected, the recording control unit 140 may associate a predetermined number of second shooting data selected according to priority among the plurality of second shooting data corresponding to the detected object with the first shooting data and store them as event recording data. For example, the recording control unit 140 may give a higher priority to the second shooting data corresponding to the object detected at a time closer to the occurrence time of the event. That is, the recording control unit 140 may determine the priority according to the time of detection of the object. In this way, a person who is more likely to have witnessed the event or a vehicle in which such a person is riding can be recorded preferentially. Also, for example, the recording control unit 140 may give a higher priority to the second shooting data corresponding to a closer object. In this case, for example, the image capture data analysis unit 120 detects the object and further determines the distance to the object. Also, when the detected object is a person's face, the recording control unit 140 may determine the priority according to the direction of the detected face. In this case, for example, the image capture data analysis unit 120 detects the face and further determines the direction of the detected face. Then, the recording control unit 140 refers to this determination result. For example, the recording control unit 140 may assign a higher priority to the second image capture data at the time of detection of the face, the more the detected face faces forward. When the face faces forward with respect to the imaging device 200, it is highly likely that the face is looking toward the vehicle in which the imaging device 200 is mounted. Therefore, it is possible to preferentially record a person who is more likely to have witnessed the event. Also, when the detected face is a forward-facing face, there is an advantage that the person of the detected face is easily identified.

In order to reduce the consumption of storage capacity, the second shooting data may be the following data. For example, the second shooting data may be a still image at the time of detection of the object. That is, a still image of the video in which the object is detected may be saved as event recording data. Also, for example, the second shooting data may be a video having a lower frame rate than the first shooting data, or a video compressed at a higher compression rate than the first shooting data. In this way, the recording control unit 140 may save a still image, a video having a lower frame rate than the first shooting data, or a video compressed at a higher compression rate than the first shooting data as the second shooting data.

Next, an example of the operation of the recording control device 100 according to this embodiment will be described. FIG. 9 is a flowchart showing an example of the operation of the recording control device 100 according to this embodiment. The flowchart shown in FIG. 9 differs from the flowchart shown in FIG. 6 mainly in that step S104 is replaced with step S114. The differences from the flowchart shown in FIG. 6 will be described below.

In the flowchart shown in FIG. 9, when the shooting data analysis unit 120 detects a specified object (Yes in step S103), the process proceeds to step S114.

In step S114, the recording control unit 140 extracts, from the series of shooting data, the first shooting data, which is shooting data including the time when the event occurred, and the second shooting data, which is shooting data including the time when the object was detected. After step S114, the process proceeds to step S105.

When the process moves from step S114 to step S105, in step S105, the recording control unit 140 associates the first shooting data and the second shooting data extracted in step S114, and stores them in the event recording unit 320 as one event recording data. Note that the shooting data may be associated by giving each shooting data a file name including a common name, or may be stored in the same folder as the shooting data to be associated. Also, data captured close to the time of the first shooting data may be associated and stored as the second shooting data.

In contrast, if the processing of step S114 is skipped, i.e., if the answer is No in step S103, in step S105, the recording control unit 140 stores only the first shooting data in the event recording unit 320 as event recording data.

The above describes the third embodiment. According to this embodiment, the recording control unit 140 generates one event record data by associating the image capture data at the time an event occurs with the image capture data at the time an object is detected, which is separate data. This makes it possible to record content that is useful for understanding the situation of a traffic trouble while suppressing the amount of data to be recorded.

Above, we have explained the first to third embodiments, but the following modifications are also possible to these embodiments.

In the above-described embodiment, the detection range of the object stored as the event record data was a predetermined period (e.g., 10 minutes) starting from the event detection timing. That is, the detection range was a uniform time. However, this detection range may be specified by a distance starting from the position of the vehicle at the event detection timing. For example, the recording control unit 140 may store, as the event record data, a video image showing an object existing within a predetermined distance (e.g., 100 m) from the point where the event occurred. That is, the recording control unit 140 may store the event record data based on the object detected from the shooting data during the period from the time when the vehicle exists at a position a predetermined distance away from the position of the vehicle at the time of the event occurrence to the time of the event occurrence. Hereinafter, the "time when the vehicle exists at a position a predetermined distance away from the position of the vehicle at the time of the event occurrence" will be referred to as the "range definition time point." The above-described range definition time point is, for example, a time point before the event occurs, but may also be a time point after the event occurs. When the range definition time is a time before the event occurs, for example, a video of an object detected from the shooting data while the vehicle moves from a point 100 m before the event occurrence point to the event occurrence point can be saved as the event record data. Similarly, when the range definition time is a time after the event occurs, for example, a video of an object detected from the shooting data while the vehicle moves from the event occurrence point to 100 m ahead can be saved as the event record data. Here, the specification of the range definition time will be described. For example, when time series data indicating the position information of the vehicle at each time exists, such as when the vehicle is equipped with a GPS (Global Positioning System) in addition to the recording system 10, the recording control unit 140 specifies the range definition time as follows. In this case, the recording control unit 140 specifies the range definition time by referring to this time series data and specifying the time when the vehicle was present at a position a predetermined distance away from the position of the vehicle at the event occurrence point. The range definition time may also be specified by other methods. For example, the recording control unit 140 may specify the range definition time using the speed information of the vehicle at the time of the event occurrence. In this case, the recording control unit 140 may use the vehicle's speed at the time the event occurred and the above-mentioned predetermined distance to estimate the time when the vehicle was at a position a predetermined distance away from the vehicle's position at the time the event occurred, thereby identifying the range definition time. In this way, by storing the event record data based on objects detected from the shooting data during the period from the range definition time to the time the event occurred, it is possible to create event record data that focuses on objects around the event occurrence point.

In the above embodiment, the shooting data analysis unit 120 detects an event and then analyzes the shooting data around the time when the event occurred. However, the shooting data analysis unit 120 may continue to analyze the shooting data in real time. In this case, for example, each time an object is detected, the shooting data analysis unit 120 notifies the recording control unit 140 of time information indicating at what point in time the object was captured in the video. Then, when an event is detected, the recording control unit 140 determines the period of the video of the event recording data based on the time information from the shooting data analysis unit 120.

In the above-described embodiment, the recording system 10 has been described as a system mounted on a vehicle, but the recording system 10 does not have to be mounted on a vehicle. For example, a surveillance camera that captures images of roads such as intersections may be used as the imaging device 200. In this case, the acceleration sensor 400 is not necessary in the recording system 10, and instead, the event detection unit 130 detects the occurrence of an event by, for example, analyzing the captured data using image recognition processing. For example, the captured data may be analyzed to detect the occurrence of an event by detecting a collision between objects such as vehicles. Alternatively, the event detection unit 130 may analyze sounds near the road that is the subject of the image capture by the imaging device 200, and detect the occurrence of an event. For example, the occurrence of an event may be detected by detecting a collision sound of a predetermined volume or more.

The present invention is not limited to the above-mentioned embodiment and its variations, and can be modified as appropriate without departing from the spirit and scope of the invention.

A part or all of the above-described embodiments can be described as, but is not limited to, the following supplementary notes.
(Appendix 1)
an imaging data acquisition unit that acquires imaging data from an imaging device;
an event detection unit that detects the occurrence of an event;
an imaging data analysis unit that analyzes the imaging data and detects a predetermined object captured in the imaging data;
and a recording control unit that associates the first shooting data, which is the shooting data including the time point at which the event occurred, with the second shooting data, which is the shooting data including the time point at which the object was detected within a predetermined range from the time point at which the event occurred, and stores the associated data as event record data.
(Appendix 2)
2. The recording control device according to claim 1, wherein, when a plurality of the objects are detected, the recording control unit associates the second shooting data and the first shooting data corresponding to each of the detected objects with each other and stores them as the event recording data.
(Appendix 3)
The recording control device described in Appendix 1, wherein when a plurality of the objects are detected, the recording control unit associates a predetermined number of the second shooting data selected in accordance with a priority order from the plurality of second shooting data corresponding to the detected objects with the first shooting data and stores the associated data as the event recording data.
(Appendix 4)
4. The recording control device according to claim 1, wherein the recording control unit stores, as the second photographic data, a still image, a video having a lower frame rate than the first photographic data, or a video compressed at a higher compression rate than the first photographic data.
(Appendix 5)
The recording control device according to any one of claims 1 to 4, wherein the recording control unit stores the event record data based on the object detected from the photographing data during the period from the time the vehicle enters within a predetermined distance range from the position of the vehicle at the time the event occurs to the time the event occurs.
(Appendix 6)
Acquire photographic data from an imaging device;
Detects the occurrence of an event,
Analyzing the photographic data to detect a predetermined object captured in the photographic data;
a first shooting data including a time point at which the event occurred and a second shooting data including a time point at which the object was detected within a predetermined range from the time point at which the event occurred are associated with each other and stored as event record data.
(Appendix 7)
An imaging data acquisition step of acquiring imaging data from an imaging device;
an event detection step of detecting an occurrence of an event;
an imaging data analysis step of analyzing the imaging data to detect a predetermined object captured in the imaging data;
a recording control step of associating the first photographing data, which is the photographing data including the time point at which the event occurred, with the second photographing data, which is the photographing data including the time point at which the object was detected within a predetermined range from the time point at which the event occurred, and storing the associated first photographing data as event record data.

REFERENCE SIGNS LIST 10 Recording system 20 Shooting data 21 Shooting data 100 Recording control device 110 Shooting data acquisition unit 120 Shooting data analysis unit 130 Event detection unit 140 Recording control unit 150 Processor 151 Memory 152 Network interface 200 Imaging device 300 Recording unit 310 Normal recording unit 320 Event recording unit 400 Acceleration sensor

Claims (5)

an imaging data acquisition unit that acquires imaging data from an imaging device;
an event detection unit that detects the occurrence of an event;
an imaging data analysis unit that analyzes the imaging data and detects a predetermined object captured in the imaging data;
and a recording control unit that associates first shooting data , which is the shooting data including the time point at which the event occurred, with second shooting data , which is the shooting data including the time point at which the object was detected within a predetermined range from the time point at which the event occurred and is data that is not consecutive to the first shooting data, and stores the data as event recording data. The recording control device according to claim 1 , wherein when a plurality of the objects are detected, the recording control unit associates the second shooting data and the first shooting data corresponding to each of the detected objects with each other and stores them as the event recording data. 2. The recording control device according to claim 1, wherein, when a plurality of the objects are detected, the recording control unit associates a predetermined number of the second shooting data selected in accordance with a priority order from the plurality of second shooting data corresponding to the detected objects with the first shooting data and stores them as the event recording data. 4. The recording control device according to claim 1, wherein the recording control unit stores, as the second shooting data, a still image, a video having a lower frame rate than the first shooting data, or a video compressed at a higher compression rate than the first shooting data. Acquire photographic data from an imaging device;
Detects the occurrence of an event,
Analyzing the photographic data to detect a predetermined object captured in the photographic data;
a first photographing data including a time point when the event occurred and a second photographing data including a time point when the object was detected within a predetermined range from the time point when the event occurred and being data that is not consecutive to the first photographing data, the first photographing data being associated with the second photographing data and the second photographing data being stored as event record data.

Images