JP2019079203A - Image generation device and image generation method - Google Patents

Abstract

To cause a user to intuitively grasp a situation of a near-miss.SOLUTION: An image generation device comprises a detection unit, an extraction unit, a conversion unit and a correlation unit. The detection unit detects danger avoiding driving from travel data of a vehicle. In a case where the danger avoiding driving is detected by the detection unit, the extraction unit extracts a moving object from a picked-up video image that is obtained by continuously imaging the outside of the vehicle. The conversion unit converts the moving object that is extracted by the extraction unit, into an abstracted animation image. The correlation unit correlates location information of the danger avoiding driving to the animation image resulting from the conversion by the conversion unit.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an image generation apparatus and an image generation method.

  In recent years, in a navigation device mounted on a vehicle, there has been proposed a technique for prompting a driver to call attention when passing through a near-miss point where the vehicle needs to travel with caution (see, for example, Patent Document 1).

JP, 2010-128980, A

  However, the prior art is merely for notifying the user of the presence of a near-miss point, and there is room for improvement in making the user intuitively understand the near-miss situation.

  The present invention has been made in view of the above, and an object of the present invention is to provide an image generation device and an image generation method that can make a user intuitively grasp situations at the time of nearing.

  In order to solve the problems described above and achieve an object, an image generation apparatus according to an embodiment includes a detection unit, an extraction unit, a conversion unit, and an association unit. The detection unit detects danger avoidance driving from travel data of the vehicle. The extraction unit extracts a moving object from a captured image obtained by continuously capturing the outside of the vehicle when the danger avoidance driving is detected by the detection unit. The conversion unit converts the moving object extracted by the extraction unit into an abstracted animation image. The association unit associates position information of the danger avoidance driving with the animation image converted by the conversion unit.

  According to the present invention, it is possible to make the user intuitively grasp the situation of the near incident.

FIG. 1 is a diagram showing an outline of an image generation method. FIG. 2 is a diagram showing an example of the configuration of an image generation system. FIG. 3 is a block diagram of an image generation apparatus. FIG. 4 is a schematic view showing the installation position of the on-vehicle camera. FIG. 5 is a diagram showing a specific example of the user database. FIG. 6 is a diagram showing a specific example of the near-miss point database. FIG. 7 is a view showing an example of a three-dimensional map image. FIG. 8 is a conceptual view of a display image. FIG. 9 is a block diagram of the in-vehicle apparatus. FIG. 10 is a diagram showing an example of a search screen. FIG. 11 is a diagram showing an example of the display screen. FIG. 12 is a flowchart showing the processing procedure executed by the image generation apparatus.

  Hereinafter, an image generation apparatus and an image generation method according to an embodiment will be described in detail with reference to the accompanying drawings. Note that the present invention is not limited by the embodiments described below.

  First, an outline of an image generation method according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram showing an outline of an image generation method. Such an image generation method is performed by an image generation device.

  The image generation method according to the embodiment is to make the user intuitively grasp an event of a near incident at a so-called near incident when the vehicle suddenly brakes or the like.

  Here, prior to the image generation method according to the embodiment, the prior art will be described. In the prior art, when the vehicle passes the near spot point, there is one that urges the driver to be alerted.

  However, in the prior art, the near-miss point is superimposed and displayed on the map image displayed on the display, or only the voice for calling attention is output from the speaker. For this reason, in the prior art, it was difficult for the user to grasp the actual situation.

  On the other hand, it is also conceivable to present the user with the actual video captured at the time of the incident. However, in such a case, there is a possibility that the privacy of a passing person appearing in the image or a passing vehicle may be violated, which is not preferable from the viewpoint of protecting personal information.

  Therefore, in the image generation method according to the embodiment, the user is made to intuitively grasp an example of a near incident while protecting personal information. That is, in the image generation method according to the embodiment, it is possible to intuitively grasp the situation of the hiyari hat while protecting the privacy by processing the passersby shown in the image and the passing vehicle.

  Specifically, as shown in FIG. 1, in the image generation method according to the embodiment, the danger avoidance driving is detected from the traveling data of the vehicle (step S1). Here, the danger avoidance driving includes, for example, sudden braking, sudden steering, and the like. In addition, the danger avoidance driving includes not only the case where the danger could be avoided but also the case where the danger could not be avoided although the danger was tried to be avoided.

  Next, in the image generation method according to the embodiment, a moving object is extracted from the captured image L in which the outside of the vehicle is captured (step S2). Here, the moving object indicates a passerby shown in the captured image L or a passing vehicle. The example shown in the figure shows the case where the pedestrian P and the passing vehicle Tc are extracted as moving objects. In addition, a moving object includes a pedestrian at rest and a passing vehicle at rest in the time-series captured image L.

  Subsequently, in the image generation method according to the embodiment, the extracted moving object is converted into an animation image G (step S3). The animation image G is, for example, a three-dimensional polygon image, but the form is not limited as long as the moving object is abstracted.

  Thus, in the image generation method according to the embodiment, by converting the moving object into the animation image G, it becomes possible to protect the personal information of the pedestrian P and the passing vehicle Tc.

  Next, in the image generation method according to the embodiment, position information of the danger avoidance driving is associated with the animation image G (step S4). Here, the position information is, for example, the actual position of the moving object, and includes the direction of the moving object.

  As described above, in the image generation method according to the embodiment, by associating the animation image G with the position information, it becomes possible to display the actual situation of the danger avoidance driving on the map image.

  As a result, the driver can confirm in advance the animation image G of such a near-miss point before passing a point where frequent danger avoidance driving occurs (hereinafter referred to as a near-miss point).

  As described above, in the image generation method according to the embodiment, when danger-avoidance driving is detected, a moving object captured in the captured image L is converted into the animation image G to intuitively grasp the situation at the time of the incident. Is possible.

  Next, an image generation system 100 according to the embodiment will be described with reference to FIG. FIG. 2 is a diagram showing an exemplary configuration of the image generation system 100. As shown in FIG. As shown in FIG. 2, the image generation system 100 includes an image generation device 1 and a plurality of in-vehicle devices 50. The image generation device 1 and the plurality of in-vehicle devices 50 can transmit and receive information bidirectionally via the network N.

  Each in-vehicle device 50 is mounted on the vehicle C, and includes an in-vehicle communication module such as a DCM (Date Communication Module) or a drive recorder. The in-vehicle device 50 transmits information including the traveling data of the vehicle C and the captured image L captured by the drive recorder to the image generation device 1 at a predetermined cycle.

  The image generation device 1 detects the danger avoidance driving based on the information transmitted from each on-vehicle device 50, and executes the processing of converting the moving object into the animation image G.

  As described above, in the image generation system 100 according to the embodiment, the information transmitted from each in-vehicle device 50 is accumulated in the image generation device 1 to make the event related to the danger avoidance driving in various situations big data. It becomes possible.

  For example, the information regarding danger avoidance driving made into big data can be used not only for the driver to grasp the situation at the time of hiyari hat in advance but also for education on traffic safety, maintenance of infrastructure, and the like. In addition, although the case where the image generation apparatus 1 was separate body from the vehicle-mounted apparatus 50 was demonstrated here, the image generation apparatus 1 and the vehicle-mounted apparatus 50 may be integral. That is, the image generation device 1 may be mounted on each vehicle C, and may have the function of the on-vehicle device 50.

  In such a case, the image generation system 100 may separately include a server device that aggregates the animation image G generated by the image generation device 1.

  Next, a configuration example of the image generation device 1 according to the embodiment will be described with reference to FIG. FIG. 3 is a block diagram of the image generation device 1. As shown in FIG. 3, the image generation device 1 includes a communication unit 2, a control unit 3, and a storage unit 4. The communication unit 2 is a communication interface that transmits and receives data to and from each of the in-vehicle devices 50 described above.

  The control unit 3 includes a detection unit 31, an extraction unit 32, a conversion unit 33, an association unit 34, and a generation unit 35. Further, the storage unit 4 stores a near-miss point database 41, a user database 42, and a three-dimensional map database 43.

  The control unit 3 includes, for example, a computer having a central processing unit (CPU), a read only memory (ROM), a random access memory (RAM), a hard disk drive (HDD), an input / output port, and various circuits.

  The CPU of the computer functions as, for example, the detection unit 31, the extraction unit 32, the conversion unit 33, the association unit 34, and the generation unit 35 of the control unit 3 by reading and executing a program stored in the ROM.

  In addition, at least one or all of the detection unit 31, the extraction unit 32, the conversion unit 33, the association unit 34, and the generation unit 35 of the control unit 3 may be an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA). It can also be configured with hardware such as

  The storage unit 4 corresponds to, for example, a RAM or an HDD. The RAM and the HDD can store the near-miss point database 41, the user database 42, the three-dimensional map database 43, and information of various programs. The image generation apparatus 1 may acquire the above-described program and various information via another computer connected via a wired or wireless network or a portable recording medium.

  The detection unit 31 of the control unit 3 detects the danger avoidance driving from the traveling data of the vehicle C. Specifically, the detection unit 31 detects, for example, sudden braking and sudden steering as danger avoidance driving from the traveling data transmitted from each on-vehicle device 50.

  More specifically, the detection unit 31 detects a sudden brake based on the transition of the speed of the vehicle C and the acceleration generated in the vehicle C among the traveling data, and performs the rapid steering based on the transition of the steering angle of the vehicle C. To detect. The detection unit 31 can also detect a collision of the vehicle C from the acceleration of the vehicle C or the operation state of the air bag.

  The detection unit 31 may detect the danger avoidance driving based on the captured image L captured from the vehicle C and the detection result of the radar device included in the vehicle C.

  Specifically, the detection unit 31 derives the speed and turning angle of the vehicle C based on, for example, a vector connecting the same feature points from the time-series captured image L, so-called optical flow, and detects danger avoidance driving You can also

  At this time, the detection unit 31 can also determine the presence or absence of an obstacle in the traveling direction of the vehicle C based on the captured image L and the detection result of the radar device. That is, the detection unit 31 can also determine whether the detected danger avoidance driving is due to an external factor such as an obstacle or the like or an internal factor due to the driver.

  In such a case, the detection unit 31 can also determine, among the detected danger avoidance driving, only the danger avoidance driving due to an external factor as the danger avoidance driving. That is, by excluding the danger avoiding operation based on the internal factor, it becomes possible to extract only the incident point based on the external factor.

  When detecting the danger avoidance driving, the detection unit 31 detects the time information indicating the time when the danger avoidance driving occurred, the occurrence point of the danger avoidance driving, and the captured image L captured before and after the danger avoidance driving. The information to be included is notified to the extraction unit 32.

  When the danger avoidance driving is detected by the detection unit 31, the extraction unit 32 extracts a moving object from the captured image L in which the outside of the vehicle C is captured. For example, the extraction unit 32 can extract a pedestrian or another vehicle shown in the captured image L as a moving object by matching processing with a template image from the captured image L captured before and after time including danger avoidance driving .

  In addition, the extraction unit 32 notifies the conversion unit 33 of moving object information indicating the coordinate position of the extracted moving object on the captured image L and the type of the moving object. The extraction unit 32 may extract the moving object from the captured image L using not only the above-described matching process but also other methods such as a so-called background subtraction method.

  The conversion unit 33 converts the moving object extracted by the extraction unit 32 into an abstracted animation image G. For example, the converting unit 33 converts the moving object into the animation image G by superimposing the animation image G corresponding to the type of the moving object on the coordinate position on the captured image L based on the moving object information. The conversion unit 33 outputs information related to the converted animation image G to the association unit 34.

  Here, as described above, the animation image G is a three-dimensional polygon image, and is a free viewpoint image capable of changing the viewpoint direction. That is, the user can visually recognize the animation image G by changing the angle of the viewpoint on the image. Therefore, when converting to the animation image G, the conversion unit 33 needs to determine the actual size of the moving object and the relative distance between the on-vehicle camera Cm that captures the captured image L and the moving object.

  Therefore, the conversion unit 33 determines the actual size and relative distance of the moving object based on the installation position of the on-vehicle camera Cm that captures the outside of the vehicle C. FIG. 4 is a schematic view showing the installation position of the on-vehicle camera Cm. As shown in FIG. 4, the mounting position includes the mounting height H and the mounting wedge angle θ.

  The mounting height H indicates the height from the road surface to the lens of the on-vehicle camera Cm, which differs depending on, for example, the type of the vehicle C and the mounting position. Further, the attachment drop angle θ indicates the angle of the imaging axis ax with respect to the vertical line of the on-vehicle camera Cm from the road surface.

  Similarly to the mounting height H, the mounting wedge angle θ is a value specific to each vehicle C. The information on the mounting height H and the mounting depression angle θ is registered in the image generation device 1 at the time of initial registration of the on-vehicle device 50 with respect to the image generation system 100, for example.

  The conversion unit 33 derives an imaging area of the on-vehicle camera Cm based on the mounting height H and the mounting depression angle θ of the on-vehicle camera Cm, and converts the imaging area to a coordinate system on the road surface. Then, the conversion unit 33 determines the actual size and relative distance of the moving object shown in the captured image L based on the coordinate system.

  Thus, the conversion unit 33 accurately derives the size and relative distance of the moving object from the captured image L by determining the actual size and relative distance of the moving object based on the mounting position of each on-vehicle camera Cm. be able to.

  Returning to the description of FIG. 3, the association unit 34 will be described. The associating unit 34 associates the animation image G converted by the converting unit 33 with the position information of the near-miss point. For example, the associating unit 34 associates the position information at the time of danger avoidance driving with each animation image G.

  At this time, the associating unit 34 estimates the traveling direction of the vehicle C, that is, the imaging direction of the on-vehicle camera Cm based on the transition of the position information before and after danger avoidance driving, and associates information on the imaging direction with the animation image G. .

  That is, the associating unit 34 associates the correct position and direction of the animation image G based on the estimated imaging position and imaging direction and the position information of the moving object on the captured image L derived by the converting unit 33. Can.

  Further, the associating unit 34 associates, with the animation image G, driver information on the driver who has performed the danger avoidance driving. The driver information is stored in the user database 42 of the storage unit 4.

  FIG. 5 shows a specific example of the user database 42. As shown in FIG. As shown in FIG. 5, in the user database 42, “user ID”, “vehicle ID”, “sex”, “travel distance”, “driving skill”, and “number of times of violation” are associated with one another as driver information. .

  The “user ID” is an identifier for identifying the driver, and the “vehicle ID” is an identifier for identifying the vehicle C on which the driver gets in. Also, “sex” indicates the gender of the driver, and “travel distance” indicates the distance the driver has driven the vehicle C.

  The “driving skill” indicates the driver's driving skill, and the “number of violations” indicates, for example, the number of times the driver violates the road traffic law. For the driving skill, for example, the image generation device 1 may be provided with a diagnosis unit for diagnosing the driving skill, or the driving skill diagnosed by an external device may be registered in the user database 42. .

  Thus, the association unit 34 can facilitate the search for each animation image G by associating the driver information with the animation image G. That is, the animation image G can be easily searched by inputting the search conditions such as the user ID, the sex, the traveling distance and the like. The details of this point will be described later with reference to FIG.

  Furthermore, the association unit 34 can also associate the animation image G with information regarding the time of the danger avoidance driving, the weather, and the like. When the association unit 34 completes the association of information on the animation image G, the association unit 34 stores the animation image G in the near-miss point database 41 illustrated in FIG. 3.

  FIG. 6 is a view showing a specific example of the near incident point database 41. As shown in FIG. As shown in FIG. 6, in the near-miss hat point database 41, for each animation image G, information such as “image ID”, “user ID”, “position”, “miss type”, “date and time”, “weather” etc. It is matched and memorized.

  The “image ID” is an identifier for identifying the animation image G, the “user ID” is an identifier for indicating the driver of the hazard avoidance driving, and the user ID indicates the driver information as shown in FIG. Is tied.

  "Position" indicates the position where the animation image G is displayed on the map image. "Hyari-hat type" indicates an event related to a hiyari-hat during danger avoidance operation. "Date" indicates the date and time when the danger avoidance operation occurred, and "weather" indicates the weather at the time of the incident.

  For example, information on weather can be acquired from an external device or each on-vehicle device 50 based on the “position” and the “date and time”.

  Returning to the description of FIG. 3, the generation unit 35 will be described. The generation unit 35 generates a display video D in which the animation image is superimposed on the three-dimensional map image corresponding to the captured video L based on the position information.

  The three-dimensional map image is stored in the three-dimensional map database 43 of the storage unit 4. FIG. 7 is a view showing an example of a three-dimensional map image. As shown in FIG. 7, the three-dimensional map image is, for example, an image three-dimensionally expressing a building such as a building.

  Actual position information is associated with the three-dimensional map image, and based on the position information associated with the animation image G, the generation unit 35 superimposes the animation image G on the three-dimensional map image.

  Thereby, it becomes possible to superimpose the animation image G on the real position on the three-dimensional map image. In addition, both the animation image G and the three-dimensional map image can change the viewpoint position.

  For this reason, in the display image D, it is possible to change and display the viewpoint position of the three-dimensional map image and the animation image G. That is, as shown in FIG. 8, it is possible to display the display image D by changing the imaging position of the on-vehicle camera Cm and the imaging axis ax in a pseudo manner. FIG. 8 is a conceptual view of the display image D.

  As described above, by making the viewpoint position changeable, it is possible to verify the situation of the near incident from any viewpoint position. That is, it is possible to display not only the image from the driver's point of view at the time of the incident, but also an image of the other vehicle or a pedestrian looking at the vehicle C, and a bird's-eye view etc. Become.

  That is, by superimposing the animation image G on the three-dimensional map image, the generation unit 35 can verify the situation at the time of near loss from various angles.

  In particular, the generation unit 35 generates a display image D viewed from the viewpoint position including the viewpoint position of the driver and the viewpoint position of the moving object, thereby verifying the situation of the near-missing hat from both the driver and the moving object it can.

  In other words, it is possible for the user to analyze the cause of the near-miss occurrence from various angles. As a result, the user can deepen driving knowledge for preventing a near incident, and therefore, the effect of suppressing the occurrence of near incident is expected.

  In addition, the 3D map image on which the animation image G is superimposed is displayed, and the actually captured image is not displayed, thereby identifying the driver that generated the incident event and the company (taxi company etc.) to which the driver belongs. Can be prevented. This is because the video actually taken may include a mark indicating a personal thing or a company identifying the individual.

  In this way, by making it impossible to identify individuals or companies, it is possible to safely share a scene effective for verification of situations at the time of hiyarihat among multiple people or multiple companies.

  In addition, the generation unit 35 can also generate a display video D reflecting weather information and time information associated with the animation image G. Details of this point will be described later.

  Next, the on-vehicle apparatus 50 according to the embodiment will be described with reference to FIG. FIG. 9 is a block diagram of the in-vehicle apparatus 50. As shown in FIG. As shown in FIG. 9, the in-vehicle apparatus 50 is connected with the navigation apparatus 60, the display apparatus 70, the in-vehicle camera Cm, and the license reader Ll.

  Further, the on-vehicle device 50 transmits information acquired from the brake sensor Sc1, the accelerator sensor Sc2, the steering sensor Sc3, the G sensor Sc4, the GPS (Global Positioning System) Sc5 and the like to the image generation device 1 as traveling data.

  As described above, the in-vehicle camera Cm is a so-called drive recorder, and outputs a captured image L obtained by imaging the front of the vehicle C to the in-vehicle device 50. The on-vehicle camera Cm may be included in the on-vehicle apparatus 50.

  The license reader Ll is a device that reads information from the driver's license. For example, the license reader Ll reads the driver's name, age, address, classification of license color, and the like from the driver's license, and outputs the driver's license, etc., to the on-vehicle device 50.

  Note that, instead of the license reader Ll, the in-vehicle apparatus 50 may identify the driver from the captured image captured by the driver, or identify the driver by biometric authentication using a fingerprint or the like. Good.

  The in-vehicle apparatus 50 includes a communication unit 51 and a control unit 52. The communication unit 51 is a communication interface that transmits and receives data to and from the image generation device 1. The control unit 52 functions as a CPU that controls the entire in-vehicle device 50.

  For example, the control unit 52 associates driver information with the captured image L captured by the on-vehicle camera Cm and travel data, and transmits the driver information to the image generation device 1 via the communication unit 51.

  In addition, the control unit 52 acquires course information of the vehicle C from the navigation device 60, and transmits the course information to the image generation device 1. Thereby, the display image D regarding the near-miss point on the route of the vehicle C is transmitted from the image generation device 1 to the on-vehicle device 50, and the display image D is displayed on the display unit 71 of the display device 70.

  Information on the display image D may be stored in advance in the in-vehicle apparatus 50, and the in-vehicle apparatus 50 may select the display image D based on the information and display the selected image on the display unit 71. That is, the information on the display image D generated by the image generation device 1 may be updated and used in the on-vehicle device 50 at a predetermined cycle.

  The navigation device 60 generates a navigation image indicating the current position of the vehicle C and the route to the destination, and outputs the navigation image to the display device 70. The display device 70 is, for example, a touch panel display, includes a display unit 71 and an operation unit 72, and displays the navigation image and the display video D described above.

  By the way, as described above, driver information, time information, weather information and the like are associated with the display image D including the animation image G. Therefore, the user can conditionally search the animation image G by operating the operation unit 72 while looking at the search screen displayed on the display unit 71.

  FIG. 10 is a diagram showing an example of a search screen. As shown in FIG. 10, search conditions such as gender, weather, age, time zone, and license are displayed on the search screen.

  When each item of the search condition is selected, a near-miss point S that matches the search condition is displayed on the map. Then, for example, when the near-miss point S on the map is selected, the display image D including the animation image G at the selected near-miss point S is reproduced.

  This enables the user to confirm the situation of the near incident at the desired near incident point S. Further, by enabling the condition search of the display video D, the display video D desired by the user can be searched by a simple operation.

  Moreover, in the image generation system 100, it is also possible to reproduce the display image D from the predetermined point to the near point. FIG. 11 is a diagram showing a specific example of the display screen.

  As shown in FIG. 11, on the display screen, a map image M on which an icon I indicating the traveling position of the vehicle C is displayed, and a display video D are displayed. Here, the display image D is a viewpoint image of the driver at the position of the icon I.

  That is, when the vehicle C actually travels the position of the icon I, the display image D displays the view seen from the driver's seat. At this time, it is also possible to adjust the height of the viewpoint position in accordance with the vehicle C.

  That is, according to the height of the driver's seat of the vehicle C, the height of the viewpoint position can be changed and the display image D can be displayed. This makes it possible to provide a highly realistic display image D.

  Further, on the display screen, the display video D is updated as the icon I moves. That is, when the icon I moves, the view that the driver can see from the moved position is displayed on the display video D as needed. That is, in conjunction with the temporal change when the display image D is reproduced, the icon I indicating the position of the vehicle C is moved on the map image.

  In the example shown in the figure, it is shown that the icon I displayed on the map image M can see the near-miss point S after the right turn. For this reason, since it is not possible to visually recognize the near-miss point S from the position of the icon I, the near-miss point S is not displayed on the display image D, and an intersection in front of the near-miss point S is displayed.

  As described above, the image generation system 100 can generate not only the display image D at the near-miss point S but also the display image D up to the near-end point S.

  As a result, the driver can confirm not only the display image D at the near-miss point S but also the display image D on the travel route up to the near-end point S in advance. Therefore, the user can confirm the contents of the near incident along with the travel route.

  At this time, for example, a characteristic structure located in front of the near spot S may be emphasized and displayed on the traveling route. As a result, when actually traveling on the traveling route, the user can recognize that the near-miss point S is approached by using the structure as a mark.

  That is, by emphasizing and displaying such a structure, it is possible to impress and store the actual position of the near-miss point S for the user. Also, by displaying the display image D seen from the position of the icon I along with the movement of the icon I, the user can simultaneously view the scenery at the position as well as the position of the vehicle C.

  By the way, in the image generation system 100, the display image D can be regarded as the windshield of the vehicle C, and the display mode can be changed according to the time and the weather. Specifically, if the weather is rain or snow, the display video D displays rain or snow.

  For example, in the image generation system 100, the brightness of the display image D is changed and displayed according to the time, or the amount of rain or snow is adjusted according to the amount of precipitation or the amount of snow accumulation to be superimposed on the display image D can do. It is also possible to superimpose and display an animation in which the wiper operates on the display image D. In such a case, the speed of the wiper can also be adjusted according to the amount of rain or snow.

  As described above, the image generation system 100 reflects the time and weather on the display image D. Thereby, the driver can view the display image D of the field of vision assumed from the time and weather. That is, it is possible to provide a highly realistic display image D in accordance with the current driving situation.

  Next, processing procedures executed by the image generation device 1 according to the embodiment will be described using FIG. FIG. 12 is a flowchart showing the processing procedure executed by the image generation device 1. As shown in FIG. 12, first, the image generating device 1 receives traveling data (step S101), and determines whether or not danger avoidance driving has been detected (step S102).

  The extraction unit 32 extracts a moving object from the captured image L when the detection unit 31 detects a danger avoiding operation (Yes at Step S102) (Step S103). On the other hand, when the danger avoidance driving is not detected (step S102, No), the process ends.

  Subsequently, the conversion unit 33 converts the moving object into an animation image G (step S104), and the association unit 34 associates position information of the danger avoidance driving with the animation image G (step S105), Finish.

  As described above, the image generation device 1 according to the embodiment includes the detection unit 31, the extraction unit 32, the conversion unit 33, and the association unit 34. The detection unit 31 detects the danger avoidance driving from the traveling data of the vehicle C. When the danger avoidance driving is detected by the detection unit 31, the extraction unit 32 extracts a moving object from the captured image L obtained by continuously capturing the outside of the vehicle C.

  The conversion unit 33 converts the moving object extracted by the extraction unit 32 into an abstracted animation image G. The association unit 34 associates the position information of the danger avoidance driving with the animation image G converted by the conversion unit 33. Therefore, according to the image generation device 1 according to the embodiment, it is possible to intuitively grasp the situation at the time of near call for the user.

  By the way, although the case where the image generation device 1 generates the display image D for a near-miss hat has been described in the embodiment described above, the present invention is not limited to this.

  So, below, the modification of the image generation apparatus 1 is demonstrated. The image generation device 1 according to the modification generates the display image D in which the viewpoint image of the driver is reproduced based on the traveling data of the vehicle C.

  At this time, for example, it is possible to store information on legal speed and road signs on each road in the image generation device 1 and detect a violation of the Road Traffic Act from travel data.

  When the image generation device 1 detects such a violation, the image generation device 1 can also superimpose a telop indicating the content of the violation on the display video D and display it. The image generation device 1 can also detect the lighting state of the traffic light shown in the captured image L and detect a violation such as signal neglect.

  In the embodiment described above, the display video D is presented to the user via the in-vehicle apparatus 50. However, the display video D may be presented to the user via the display unit such as a personal computer.

  In such a case, for example, a manager who manages the operation of a sales vehicle including a truck, a bus, etc. can make each driver confirm beforehand the situation of the near-miss point and near-miss point in the office or the like.

  Further effects and modifications can be easily derived by those skilled in the art. Thus, the broader aspects of the invention are not limited to the specific details and representative embodiments represented and described above. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents.

DESCRIPTION OF SYMBOLS 1 Image generation apparatus 31 Detection part 32 Extraction part 33 Conversion part 34 Correspondence part 35 Generation part C Vehicle D Display image G Animation image

Claims (7)

A detection unit that detects danger avoidance driving from traveling data of the vehicle;
An extraction unit that extracts a moving object from a captured image obtained by continuously capturing the outside of the vehicle when the danger avoidance driving is detected by the detection unit;
A conversion unit that converts the moving object extracted by the extraction unit into an abstracted animation image;
An associating unit that associates position information of the danger avoidance driving with the animation image converted by the converting unit. The image generation apparatus according to claim 1, further comprising: a generation unit configured to generate a display image obtained by superimposing the animation image on a three-dimensional map image corresponding to the captured image based on the position information. The generation unit is
The image generation apparatus according to claim 2, wherein the display image is generated by changing a viewpoint position including a driver of the vehicle and a viewpoint position of the moving object. The generation unit is
The image generation apparatus according to claim 2 or 3, wherein an icon image indicating the position of the vehicle is moved on a map image in conjunction with a temporal change when the display video is reproduced. The generation unit is
The image generation device according to claim 2, 3 or 4, wherein the display image is generated according to weather information and time information. The association unit is
The image generation device according to any one of claims 1 to 5, wherein driver information on a driver of the vehicle is associated with the animation image. A detection step of detecting danger avoidance driving from travel data of the vehicle;
An extraction step of extracting a moving object from a captured image obtained by continuously capturing the outside of the vehicle when the danger avoidance driving is detected in the detection step;
Converting the moving object extracted by the extracting step into an abstracted animation image;
An associating step of associating position information of the danger avoidance driving with the animation image converted by the converting step.

Images