JP2018073135A - Information processing device and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique for matching captured time of a video image captured by an in-vehicle camera installed on each of a plurality of vehicles with those of video images captured by in-vehicle cameras of other vehicles.SOLUTION: An image acquisition unit 51 acquires a plurality of video images captured by a plurality of vehicles with in-vehicle cameras. An information reception unit 52 receives input of vehicle information of a search target vehicle. A specific image extraction unit 53 extracts, from the plurality of video images, at least two specific video images containing the vehicle information together with a predetermined structural object within captured image data. A time correction unit 55 correct image-captured time of each of the specific video images based on an image-captured time period of each of the specific video images.SELECTED DRAWING: Figure 8

Description

  The present invention relates to an information processing apparatus and a program.

  Conventionally, a technique has been proposed in which a vehicle traveling on a road is imaged and a traffic situation is monitored or a specific vehicle is detected. For example, Patent Document 1 discloses a television camera for a plurality of automobiles, a vehicle that performs image processing for recognizing a vehicle number of an oncoming vehicle or a preceding vehicle from images of the television camera, a registered specific vehicle number, and a plurality of vehicles. The car number recognition data sent from the car is collated by a search, and if there is a matching car number recognition data, the time and the absolute position coordinates are used to track the travel route of the car with the specific car number. Techniques to do this are disclosed.

JP-A-8-30892

  However, there is no guarantee that the imaging time of the video captured by the in-vehicle cameras installed in a plurality of vehicles is always correct information. For this reason, there is a demand for a technique for adjusting the imaging time of moving images captured by different imaging devices.

  The present invention has been made in view of these points, and an object of the present invention is to provide a technique for adjusting the image capturing times of images captured by in-vehicle cameras installed in a plurality of vehicles.

  A first aspect of the present invention is an information processing apparatus. The apparatus includes an image acquisition unit that acquires a plurality of moving images captured by a plurality of vehicles each equipped with an in-vehicle camera, an information reception unit that receives input of vehicle information related to a search target vehicle, and the plurality of moving images. , Based on the specific image extraction unit that extracts at least two specific moving images that are the moving images including the vehicle information together with the predetermined structure in the imaging data, and the imaging time zones of the specific moving images, A time correction unit that corrects the imaging time of the specific image.

  The information processing apparatus may further include a reference image extraction unit that extracts a reference moving image in which accuracy of an imaging time period is guaranteed from the specific moving image, and the time correction unit includes the reference moving image. You may correct | amend the time of another specific moving image using the time information of an image.

  The information processing apparatus may further include a specific device extraction unit that extracts a vehicle number reading device captured in the specific moving image, and the reference image extraction unit includes the vehicle number in the specific moving image. A moving image captured by an in-vehicle camera of a vehicle whose number is read by the reading device may be extracted as the reference moving image.

  The reference image extraction unit may extract a specific moving image accompanied by GPS (Global Positioning System) information as the reference moving image.

  The information receiving unit may receive at least an automobile registration number as the vehicle information.

機能 A function of acquiring a plurality of moving images captured by a plurality of vehicles each equipped with a vehicle-mounted camera in a computer; a function of receiving input of vehicle information related to a search target vehicle; The imaging time of each specific image is corrected based on the function of extracting at least two specific moving images that are moving images including the vehicle information together with the predetermined structure and the imaging time zones of the specific moving images. Function to achieve.

  ADVANTAGE OF THE INVENTION According to this invention, the technique which adjusts the imaging time of the image | video which the vehicle-mounted camera installed in the some vehicle imaged can be provided.

It is a figure for demonstrating the outline | summary of the information processing system which concerns on 1st Embodiment. It is a figure which shows typically the function structure of the information processing apparatus which concerns on 1st Embodiment. It is a figure which shows typically the data structure of the moving image database which the communication part which concerns on 1st Embodiment stores. It is a figure which shows typically an example of the vehicle information containing time slot | zone of each specific moving image which the duplication image extraction part acquired. It is a figure which shows an example of the flame | frame of a moving image typically. It is a figure which shows the imaging time of the flame | frame determined that the analogy ratio determination part which concerns on 1st Embodiment is similar. It is a flowchart for demonstrating the flow of the information processing which the information processing apparatus which concerns on 1st Embodiment performs. It is a figure which shows typically the function structure of the information processing apparatus which concerns on 2nd Embodiment. It is a flowchart for demonstrating the flow of the information processing which the information processing apparatus which concerns on 2nd Embodiment performs.

<First Embodiment>
An outline of the first embodiment of the present invention will be described.
FIG. 1 is a diagram for explaining an overview of the information processing system S according to the first embodiment. The information processing system S extracts information related to a search target vehicle by collecting moving images captured by an in-vehicle camera mounted on a plurality of vehicles V traveling on a road and analyzing the video. The information processing apparatus 1 that executes this information extraction is installed in the information analysis center 2. Although FIG. 1 shows nine vehicles V from the first vehicle V1 to the ninth vehicle V9, the information processing system S captures a moving image to be analyzed and can capture 10 or less vehicles V. It may be more than one.

  The moving image captured by the in-vehicle camera mounted on the vehicle V is a moving image of a so-called “drive recorder”. For example, the information processing apparatus 1 collects moving images by optionally receiving provision from the driver of each vehicle V on the condition of anonymity or the like. Hereinafter, a moving image captured by the in-vehicle camera is simply referred to as “moving image”.

  In FIG. 1, the broken line indicated by the symbol A indicates the angle of view of the in-vehicle camera of the first vehicle V1. In the example illustrated in FIG. 1, the moving image of the first vehicle V <b> 1 includes a video of the vehicle V <b> 3 that is a preceding vehicle, a video of the vehicles V <b> 4 and V <b> 6 that are oncoming vehicles, and a video of the structure 3. Here, when the structure 3 is a characteristic structure and can be uniquely identified, it is useful clues for identifying the position where the moving image including the image of the structure 3 is captured. Become.

  For example, the moving image of the eighth vehicle V8 includes the image of the guide sign 4 and the image of the traffic light 5 in addition to the image of the seventh vehicle V7 that is the preceding vehicle. Since different place names and road names are displayed on the different guide signs 4, when the video of the guide sign 4 is included in the moving image, the position where the moving image is captured can be specified. Further, some traffic lights 5 are provided with a sign indicating an intersection name or the like. Even when such a video of the traffic light 5 is included in the moving image, it is a useful clue for specifying the position where the moving image is captured.

  Furthermore, an automobile number reading device 6 (so-called “N system”) may be installed depending on the road. Since the automobile number reading device 6 is installed only in a specific place, when the moving image includes the video of the automobile number reading device 6, it is a useful clue for specifying the position where the moving image is captured. It becomes. Further, when the automobile number reading device 6 acquires the automobile registration number (so-called “number”) of the vehicle V, it is possible to specify when and where the vehicle V is traveling based on the information. .

  Some recent drive recorders can be linked to a satellite positioning system 7 such as a GPS (Global Positioning System). Such a moving image of the drive recorder is accompanied with accurate information regarding the location and time when the image was taken.

  On the other hand, a moving image of a drive recorder that is not linked to the satellite positioning system 7 is not necessarily an image in which the accuracy of the captured time is guaranteed. This is because there are users who accurately set the internal clock of the drive recorder, and there are users who use the drive recorder at an incorrect time with the initial setting. In addition, even if the clock is set at the start of use of the drive recorder, some users may use it without correcting it until the subsequent time lag.

  Furthermore, when the moving image does not include a characteristic landmark such as the structure 3, the guide sign 4, or the traffic light 5, it is difficult to specify the location where the moving image is captured. There is.

  As described above, the moving images that are provided and stored by the information processing apparatus 1 have high accuracy of the information of the attached imaging time zone and imaging location, and the accuracy of the imaging time zone is high, but the imaging location is Uncertainties, high accuracy of the shooting location but lacking reliability in the shooting time zone, and low reliability in the shooting time zone and unknown shooting location may be mixed.

  The information processing apparatus 1 according to the first embodiment first extracts a moving image group including vehicle information related to a search target vehicle. Here, the vehicle information is information related to at least one of “all or a part of the automobile registration number of the search target vehicle” or “appearance information such as the vehicle type and color”.

The information processing apparatus 1 uses the information attached to the extracted moving image group to correct the time information of each moving image and improve its accuracy. Thereby, the information processing apparatus 1 can improve the accuracy of information related to the travel route and travel time of the search target vehicle.
Hereinafter, the information processing apparatus 1 according to the first embodiment will be described in more detail.

  FIG. 2 is a diagram schematically illustrating a functional configuration of the information processing apparatus 1 according to the first embodiment. The information processing apparatus 1 includes a communication unit 10, a storage unit 20, an input unit 30, an output unit 40, and a control unit 50.

  The communication unit 10 is a communication module for communicating with a network outside the information processing apparatus 1 such as the Internet. The storage unit 20 is a high-capacity storage device such as a hard disk drive (HDD) or a solid state drive (SSD), and a high-speed volatile memory such as a dynamic random access memory (DRAM). Functions as a working memory.

  The input unit 30 is an input user interface that receives an input from the user of the information processing apparatus 1 via a keyboard, a pointing device, or the like. The output unit 40 is an output user interface that provides information to the user of the information processing apparatus 1 via a display or the like.

  The control unit 50 is a CPU (Central Processing Unit) of the information processing apparatus 1, and reads out and executes a program stored in the storage unit 20 into a memory, thereby executing an image acquisition unit 51, an information reception unit 52, and a specific image extraction unit. 53, a duplicate image extraction unit 54, a time correction unit 55, a pattern extraction unit 56, an analogy determination unit 57, and a reference image extraction unit 58.

  The image acquisition unit 51 acquires a plurality of moving images captured by a plurality of vehicles V each equipped with an in-vehicle camera via the communication unit 10. The image acquisition unit 51 stores the acquired moving image in the storage unit 20. The information receiving unit 52 receives input of vehicle information related to the search target vehicle from the user.

  The specific image extracting unit 53 extracts at least two specific moving images that are moving images including vehicle information in the imaging data from among the plurality of moving images stored in the storage unit 20. For example, when the vehicle information received from the user by the information reception unit 52 is a specific vehicle registration number, the specific image extraction unit 53 registers the vehicle in a license plate among a plurality of moving images stored in the storage unit 20. A moving image in which a vehicle with a number is captured is extracted as a specific moving image.

  As another example, the specific image extraction unit 53 is stored in the storage unit 20 when the vehicle information received from the user by the information reception unit 52 is a vehicle type and color (for example, a red sedan manufactured by X company). Among the plurality of moving images, a moving image in which the vehicle of the vehicle type and color specified by the vehicle information is captured is extracted as the specific moving image. The specific image extraction unit 53 can be realized using a known image processing technique such as character string search or pattern matching.

  The specific image extraction unit 53 extracts a moving image based on the fact that vehicle information is included in the moving image, but does not consider an imaging time zone of the moving image. For this reason, there may be a large gap in the imaging time zone of the moving image extracted by the specific image extraction unit 53. From the viewpoint of obtaining information on the travel route and travel time of the vehicle to be searched, the usefulness of a moving image having a large gap in the imaging time zone decreases.

  Therefore, the overlapping image extraction unit 54 acquires a combination of specific moving images in which the time zone including the vehicle information overlaps at least in part among the specific moving images extracted by the specific image extraction unit 53.

  FIG. 3 is a diagram schematically illustrating a data structure of a moving image database stored in the communication unit 10 according to the first embodiment. The image acquisition unit 51 assigns a moving image ID (Identification) for uniquely identifying each acquired moving image to each moving image. In the example of the moving image database shown in FIG. 3, information relating to a moving image to which M000001 is assigned as a moving image ID is stored.

The moving image database includes a plurality of data fields, and there is also a data field for storing a “shooting time zone” of the moving image. FIG. 3 stores that the moving image identified by the moving image ID M000001 was shot from 4:56:12 am on January 23, 2016 to 9: 9: 54 pm on the same day. Yes. The moving image database also includes a vehicle information-containing time zone data field indicating a time zone in which the vehicle information is included in the moving image. In FIG. 3, the moving image identified by the moving image ID M000001 includes vehicle information from 12:34:56 pm on January 23, 2016 to 13: 6: 23 pm on the same day. Is shown.
Data stored in other data fields will be described later.

  The duplicate image extracting unit 54 refers to the moving image database based on the moving image ID of each specific moving image extracted by the specific image extracting unit 53, thereby acquiring a time zone in which vehicle information is included in each specific moving image. .

  FIG. 4 is a diagram schematically illustrating an example of the vehicle information-containing time zone of each specific moving image acquired by the duplicate image extraction unit 54. FIG. 4 shows an example in which the specific image extraction unit 53 extracts moving image groups specified by M00001, M00011, M00067, M00248, M00760, and M01234 as moving image IDs, respectively.

  In FIG. 4, the horizontal axis indicates time. A thick solid line indicates a time zone in which vehicle information is included in the moving image. In the example shown in FIG. 4, the moving image identified by the moving image ID M01234 does not overlap with other moving images in the time zone in which the vehicle information is included. This is because the moving image specified by M01234 was captured at a date and time completely different from other moving images, or the time of the internal clock of the in-vehicle camera that generated the moving image specified by M01234 was inaccurate Either case is conceivable.

  Therefore, the duplicate image extracting unit 54 is specified by M00001, M00011, M00067, M00248, and M00760, respectively, except for the moving image whose moving image ID is specified by M01234 among the moving images extracted by the specific image extracting unit 53. Acquire a specific moving image combination.

  FIG. 5 is a diagram schematically illustrating an example of a moving image frame. Since the moving image is an image captured by the in-vehicle camera, the vehicle V itself that includes the in-vehicle camera is not included. In the example of the moving image shown in FIG. 5, the tenth vehicle V10 and the eleventh vehicle V11 are imaged as preceding vehicles, and the twelfth vehicle V12 and the thirteenth vehicle V13 are imaged as oncoming vehicles.

  Here, it is assumed that the search target vehicle is the eleventh vehicle V11. In this case, since the vehicle V equipped with the vehicle-mounted camera that captures the moving image shown in FIG. 5 travels in the same lane as the eleventh vehicle V11, the time for including the search target image in the moving image becomes longer. On the other hand, it is assumed that the search target vehicle is the twelfth vehicle V12. In this case, the vehicle V equipped with the in-vehicle camera that captures the moving image shown in FIG. 5 travels on the opposite lane of the twelfth vehicle V12, and therefore the time for which the search target vehicle appears in the moving image is only a short time at the time of passing. is there.

  For example, in the example shown in FIG. 4, the moving image identified by the moving image ID M00248 has a shorter time for including vehicle information than other moving images. This suggests that the vehicle equipped with the in-vehicle camera that captured the moving image specified by M00248 was an oncoming vehicle of the search target vehicle.

  The time correction unit 55 corrects the imaging time based on the imaging time zones of the specific moving images constituting the combination. More specifically, the pattern extraction unit 56 first extracts a predetermined pattern from each frame of the specific moving image. The pattern extracted from the specific moving image by the pattern extraction unit 56 is specific structures such as information on other vehicles V, information signs 4 and traffic lights 5 in addition to the vehicle information described above. The pattern extraction unit 56 further extracts a wide area pattern including vehicle information from the specific moving image.

  For example, in the example shown in FIG. 5, it is assumed that the tenth vehicle V10 is a search target vehicle. At this time, the pattern extraction unit 56 extracts, as patterns, the tenth vehicle V10, the preceding vehicle, the oncoming vehicle, the scenery such as the shape of a distant mountain, the degree of bending of the road on which the tenth vehicle V10 travels, and the like.

  The similarity determination unit 57 analyzes the specific moving images constituting the combination specified by the specific image extraction unit 53 based on the combination of predetermined patterns existing in each specific moving image, so that the moving images are similar to each other. It is determined whether or not. In other words, the analog ratio determination unit 57 determines whether or not the moving images are similar based on a combination of predetermined patterns existing in each specific moving image.

  For example, it is assumed that the pattern extraction unit 56 extracts a frame in which a search target vehicle and another vehicle V are imaged against a mountainous area in a specific moving image. Furthermore, it is assumed that the pattern extraction unit 56 extracts a frame in which the search target vehicle and the other vehicle V are captured in the background of a similar mountainous area in another specific moving image. In this case, the analog ratio determination unit 57 determines that both frames are similar. The similarity of frames suggests that the shooting locations of the frames are the same.

  FIG. 6 is a diagram illustrating image capturing times of frames determined to be similar by the similarity determination unit 57 according to the first embodiment. In the example illustrated in FIG. 6, the similarity determination unit 57 determines that the four frames from the frame A to the frame D are similar. The imaging time of frame A is 8:00 on January 23, 2016, and the imaging time of frame B is 8:03 on the same day. Similarly, the imaging time of frame C is 7:50 on the same day, and the imaging time of frame D is 8: 8 on the same day.

  The time correction unit 55 corrects the imaging time of each moving image based on the imaging time of a similar specific moving image frame. The time correction unit 55 assumes that the frames of the specific moving image determined to be similar by the analog ratio determination unit 57 are captured at the same time. For example, in the example of FIG. 6, since four frames from frame A to frame D are similar, the time correction unit 55 assumes that these were imaged at the same time. However, in the example shown in FIG. 6, since the internal clock of the in-vehicle camera that captured the specific moving image including each frame does not match, different time information is attached to each frame. Therefore, the time correction unit 55 estimates an average of time information attached to each frame as a true time.

  The time correction unit 55 may estimate the time information using statistical information other than the average of the time information attached to each frame. For example, the time correction unit 55 may estimate the mode value (mode) or median value (median) of time information attached to each frame as a true time. Further, the time information may be estimated by excluding the time most deviated from the average of the time information attached to each frame.

  In the example shown in FIG. 6, the average of the imaging times of four frames from frame A to frame D is 8:02 on January 23, 2016. Therefore, the time correction unit 55 calculates the correction amount of the imaging time of each specific moving image so that the imaging time of each frame becomes the average time.

  As shown in FIG. 6, since the imaging time of the frame A is 8:00:00, the time correction unit 55 calculates the time correction amount of the specific moving image including the frame A as +2 minutes. Similarly, the time correction unit 55 calculates the time correction amount of the specific moving image including the frame B as −1 minute. The same applies to frame C and frame D. Thereby, the time correction unit 55 can adjust the time of each specific image even if the absolute coordinates of the location where each specific moving image is captured are unknown.

  Here, the information receiving unit 52 may receive at least an automobile registration number as vehicle information received from the user of the information processing apparatus 1. In this case, the pattern extraction unit 56 extracts the character string of the automobile registration number in addition to the predetermined structure from the specific moving image. This can be realized using, for example, a known character recognition technique. Compared with the case where the information receiving unit 52 acquires appearance information such as the vehicle type and color as vehicle information, the accuracy of specifying the search target vehicle can be improved.

[When the imaging time zone accuracy is guaranteed]
As above, the case where the information processing apparatus 1 does not guarantee the accuracy of the imaging time zone of each specific moving image has been mainly described. However, there are some moving images in which the accuracy of the imaging time zone is guaranteed. Hereinafter, a case will be described in which a specific image includes a moving image in which the accuracy of the imaging time period is guaranteed.

  An example of the “moving image in which the accuracy of the imaging time period is guaranteed” is a case where the in-vehicle camera that captures the moving image is interlocked with the GPS of the vehicle V. In general, data received by the GPS from the satellite positioning system 7 includes accurate time information. For this reason, the accuracy of the time information attached to the moving image captured by the in-vehicle camera operating in conjunction with the GPS is guaranteed. Such a moving image in which the accuracy of the imaging time period is guaranteed can be used as a reference moving image when the time correction unit 55 corrects the shooting time of each specific moving image.

  Therefore, the reference image extraction unit 58 extracts a reference moving image in which the accuracy of the imaging time period is guaranteed from the specific moving images constituting the combination. More specifically, the reference image extraction unit 58 extracts a specific moving image attached with GPS information as a reference moving image.

  As shown in FIG. 3, the moving image database has a data field for storing position information. In the example illustrated in FIG. 3, the moving image identified by the moving image ID M000001 includes GPS information. The reference image extraction unit 58 extracts a reference moving image from the specific moving images constituting the combination by referring to the moving image database.

  The time correction unit 55 corrects the time of another specific moving image using the time information of the reference moving image. Thereby, the time correction unit 55 can calculate a correction amount for accurately adjusting the imaging time of each specific moving image.

  As shown in FIG. 3, the moving image database also includes a data field that stores time correction information for correcting the time of the moving image calculated by the time correction unit 55. In the example of the moving image database shown in FIG. 3, GPS information is included in the moving image identified by the moving image ID M000001, and therefore the time correction amount is stored as ± 0 minutes. That is, it indicates that the imaging time of the moving image specified by the moving image ID M000001 is accurate.

  In addition, the moving image database includes a data field for storing pattern information extracted by the pattern extraction unit 56. Once the pattern extraction unit 56 analyzes the moving image and stores the pattern found in the moving image database, the pattern matching processing by the pattern extraction unit 56 can be omitted thereafter.

  FIG. 7 is a flowchart for explaining a flow of information processing executed by the information processing apparatus 1 according to the first embodiment. The process in this flowchart starts when a computer that implements the information processing apparatus 1 is started, for example.

  The image acquisition unit 51 acquires a plurality of moving images respectively captured by a plurality of vehicles V equipped with a vehicle-mounted camera (S2). The image acquisition unit 51 stores the acquired moving image in the storage unit 20 (S4). The information receiving unit 52 receives input of vehicle information related to the search target vehicle (S6).

  The specific image extraction unit 53 extracts a specific moving image that is a moving image including vehicle information in the imaging data from among the plurality of moving images (S8). The duplicate image extraction unit 54 acquires a duplicate movie group in which the time zone including the vehicle information overlaps at least partially among the specific movie images (S10).

  The similarity determination unit 57 extracts a similar moving image group including similar frames from the moving images constituting the overlapping moving image group (S12). The reference image extraction unit 58 tries to extract a reference moving image in which the accuracy of the imaging time period is guaranteed among the specific moving images constituting the combination. When there is no reference moving image (No in S14), the time correction unit 55 calculates the average of the time information attached to the frame of the specific moving image determined to be similar by the similarity determination unit 57 (S16). The time correction unit 55 calculates a correction value of the time information based on the calculated average time (S18).

  When there is a reference moving image (Yes in S14), the time correction unit 55 calculates a correction value for time information of another specific moving image using the time information of the reference moving image (S20). When the time correction unit 55 calculates the correction value of the time information of the specific moving image, the processing in this flowchart ends.

As described above, according to the information processing apparatus 1 according to the first embodiment, it is possible to provide a technique for adjusting the imaging time of a moving image captured by an in-vehicle camera installed in a plurality of vehicles.
In particular, even when the imaging time attached to each moving image is not accurate, the imaging time system can be enhanced by statistically processing the imaging time.

  In particular, since the information processing apparatus 1 determines whether or not a moving image has been captured at the same location based on a subject that is common to each moving image, even if the absolute position of the shooting location is unknown, The image can be aligned. Since each moving image can be aligned, the information processing apparatus 1 can also adjust the time of each moving image. In addition, when GPS information is attached to a moving image, the information processing apparatus 1 can adjust the time of each moving image based on time information obtained from the GPS information.

  The information processing apparatus 1 according to the first embodiment can synchronize the time of each moving image, but may not be able to specify the imaging location of each moving image. However, the information processing apparatus 1 is useful for tracking the travel route of the search target vehicle in that it can extract a moving image group in which the search target vehicle is captured in the same time zone from a large number of moving images. This is because if the moving image is narrowed down to some extent, the user of the information processing apparatus 1 may visually observe and specify the traveling location.

<Second Embodiment>
In the information processing apparatus 1 according to the first embodiment, the case where each specific moving image is associated with “time” as a reference has been described. On the other hand, the information processing apparatus 1 according to the second embodiment is different in that each specific moving image is associated with “absolute position coordinates” of a place where the moving image is captured as a reference.
In the following, the second embodiment will be described, but portions common to the first embodiment will be omitted or simplified as appropriate.

  FIG. 8 is a diagram schematically illustrating a functional configuration of the information processing apparatus 1 according to the second embodiment. Compared with the information processing apparatus 1 according to the first embodiment, the information processing apparatus 1 according to the second embodiment does not include the duplicate image extraction unit 54 and includes a specific device extraction unit 59. Is different. Even if the configuration is the same as that of the information processing apparatus 1 according to the first embodiment, the information processing apparatus 1 according to the second embodiment may perform different operations.

  The specific image extraction unit 53 extracts at least two moving images that include vehicle information together with a predetermined structure in the captured image data from among the plurality of moving images as specific moving images. Note that the pattern extraction unit 56 executes identification of vehicle information and a predetermined structure included in the moving image.

  Here, the “predetermined structure” is a structure that can uniquely specify the absolute position coordinates of the place where the structure exists. Examples of the predetermined structure include the above-described guide sign 4, the traffic light 5, and the automobile number reading device 6. Moreover, the structure 3 which has a characteristic external appearance among the structures 3 and can be specified uniquely can also become a predetermined structure.

  The similarity determination unit 57 extracts a frame in which a predetermined structure and a search target image are captured from each of the specific images extracted by the specific image extraction unit 53. Here, since the predetermined structure is installed at a specific location, if the predetermined structure and the search target image are captured in frames of different specific moving images, they are captured at the same time. Can be estimated.

  Therefore, the time correction unit 55 calculates a correction amount for correcting the imaging time of each specific image based on the imaging time of the frame in which the predetermined structure and the search target image are captured. The similarity determination unit 57 according to the second embodiment specifies a frame including a structure whose absolute position coordinates are determined, thereby pattern matching such as a landscape as in the information processing apparatus 1 according to the embodiment. The identification accuracy can be improved as compared with the case where similar frames are identified by processing.

  When the reference moving image is included in the specific moving image extracted by the specific image extracting unit 53, that is, when the moving image in which the accuracy of the imaging time period is guaranteed is included, the time correction unit 55 Using the time information of the image, the time of another specific moving image is corrected. Thereby, the time correction part 55 can improve the correction accuracy of the time of a specific moving image.

  Here, the moving image captured by the in-vehicle camera of the vehicle V whose number is read by the automobile number reading device 6 can also be a reference moving image. This is because the automobile number reading device 6 registers the image of the vehicle V including the number together with the time information. Therefore, the specific device extraction unit 59 attempts to extract the automobile number reading device 6 captured in the specific moving image. The reference image extraction unit 58 also extracts a specific moving image that has been successfully extracted by the automobile number reading device 6 as a reference moving image.

  Thereby, even if the GPS information is not attached to all the specific moving images specified by the specific image extracting unit 53, the vehicle V equipped with the in-vehicle camera that captures any of the specific moving images can read the vehicle number. When registered by the device 6, the time information of each specific moving image can be accurately adjusted.

  FIG. 9 is a flowchart for explaining the flow of information processing executed by the information processing apparatus 1 according to the second embodiment. The process in this flowchart starts when a computer that implements the information processing apparatus 1 is started, for example.

  The image acquisition unit 51 acquires a plurality of moving images respectively captured by a plurality of vehicles V equipped with a vehicle-mounted camera (S22). The image acquisition unit 51 stores the acquired moving image in the storage unit 20 (S24). The information receiving unit 52 receives input of vehicle information related to the search target vehicle (S26). The specific image extraction unit 53 extracts a specific moving image, which is a moving image including vehicle information together with a predetermined structure in the captured data, from the plurality of moving images (S28).

  The reference image extraction unit 58 tries to extract a reference moving image in which the accuracy of the imaging time period is guaranteed among the specific moving images constituting the combination. When there is no reference moving image (No in S30), the time correction unit 55 calculates the average of the time information attached to the frame of the specific moving image determined to be similar by the similarity determination unit 57 (S32). The time correction unit 55 calculates a correction value of the time information based on the calculated average time (S34).

  When there is a reference moving image (Yes in S30), the time correction unit 55 calculates a correction value for time information of another specific moving image using the time information of the reference moving image (S36). When the time correction unit 55 calculates the correction value of the time information of the specific moving image, the processing in this flowchart ends.

  As described above, according to the information processing apparatus 1 according to the second embodiment of the present invention, it is possible to provide a technique for adjusting the image capturing times of images captured by in-vehicle cameras installed in a plurality of vehicles.

  In particular, the information processing apparatus 1 can collect the moving images captured at the same place with high accuracy by collecting the moving images based on the positional information of the structure or the like captured in the moving image.

  As mentioned above, although this invention was demonstrated using two embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above embodiment. It is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.

  In addition, new embodiments generated by arbitrarily combining the embodiments are also included in the technical scope of the present invention. For example, the first embodiment may be combined with the second embodiment, and the information of the automobile number reading device 6 may be used for selection of the reference image in the first embodiment. Thus, the new embodiment generated by the combination has the effect of the combination source embodiment.

DESCRIPTION OF SYMBOLS 1 ... Information processing apparatus 2 ... Information analysis center 3 ... Structure 4 ... Information sign 5 ... Traffic light 6 ... Car number reading device 7 ... Satellite positioning system 10 ... Communication unit 20 ... storage unit 30 ... input unit 40 ... output unit 50 ... control unit 51 ... image acquisition unit 52 ... information reception unit 53 ... specific image extraction unit 54 ..Duplicate image extraction unit 55 ... Time correction unit 56 ... Pattern extraction unit 57 ... Similarity determination unit 58 ... Reference image extraction unit 59 ... Specific device extraction unit S ... Information processing system V ... Vehicle

Claims (6)

An image acquisition unit for acquiring a plurality of moving images respectively captured by a plurality of vehicles equipped with in-vehicle cameras;
An information receiving unit that receives input of vehicle information related to the search target vehicle;
A specific image extraction unit that extracts at least two specific moving images that are moving images including the vehicle information together with a predetermined structure in the imaging data among the plurality of moving images;
A time correction unit that corrects an imaging time of each specific image based on an imaging time zone of each of the specific moving images;
An information processing apparatus comprising: A reference image extracting unit that extracts a reference moving image in which accuracy of an imaging time period is guaranteed among the specific moving images;
The time correction unit corrects the time of another specific moving image using the time information of the reference moving image.
The information processing apparatus according to claim 1. A specific device extraction unit that extracts the vehicle number reading device captured in the specific moving image;
The reference image extraction unit extracts, as the reference moving image, a moving image captured by an in-vehicle camera of a vehicle whose number is read by the vehicle number reading device, among the specific moving images.
The information processing apparatus according to claim 2. The reference image extraction unit extracts a specific moving image accompanied by GPS (Global Positioning System) information as the reference moving image.
The information processing apparatus according to claim 2 or 3. The information receiving unit receives at least an automobile registration number as the vehicle information.
The information processing apparatus according to any one of claims 1 to 4. A function of acquiring a plurality of moving images captured by a plurality of vehicles each equipped with a vehicle-mounted camera on a computer,
A function of accepting input of vehicle information related to the search target vehicle;
A function of extracting at least two specific moving images which are moving images including the vehicle information together with a predetermined structure in the imaging data among the plurality of moving images;
A function of correcting the imaging time of each specific image based on the imaging time zone of each of the specific moving images;
A program that realizes

Images