Image recording device and image recording method

Abstract

To provide an image recording device and an image recording method, capable of determining visibility of an object in the case where the object is recognized, and recording the object by controlling the mode of an image on the basis of the visibility.SOLUTION: An image recording device 10 comprises: an image acquisition part 11 that acquires an image obtained by imaging a circumference of a mobile body; an image recognition part 12 that performs image recognition of a predetermined object from the image; a determination part 13 that determines visibility of the object in the image; an image file generation part 14 that generates an image file from the image in a mode based on at least a result of the determination; and a recording control part 15 that performs control of recording the image file in a recording part.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an image recording apparatus and an image recording method. [Background technology]

[0002] There is a known technology for recording a captured image in a recording device when a previously registered object is recognized as an image (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-68197 Summary of the Invention [Problem to be solved by the invention]

[0004] Patent Document 1 discloses a technique for recording a captured image in a recording device when a preregistered object is recognized. However, even if the object is recognized, for example, if the object moves a lot on the screen, it may be difficult to visually recognize the object, and therefore it may not be suitable to record and store the image for later viewing.

[0005] The present invention has been made in consideration of the above, and provides an image recording device and an image recording method which, when an object is recognized as an image, can determine the visibility of the image to ensure that the user can easily see the object, and can control and record the aspect of the image based on the visibility of the object. [Means for solving the problem]

[0006] In order to achieve the above-mentioned object, the image recording device of the present invention is characterized by comprising an image acquisition unit that acquires an image of the surroundings of a moving body, an image recognition unit that performs image recognition of a predetermined object from the image, a judgment unit that judges the visibility of the object in the image, an image file generation unit that generates an image file from the image in a manner based at least on the result of the judgment, and a recording control unit that controls the recording of the image file in the recording unit.

[0007] In order to achieve the above-mentioned object, the image recording method of the present invention is characterized by executing an image acquisition step of acquiring an image of the surroundings of a moving body, an image recognition step of image-recognizing a predetermined object from the image, a determination step of determining the visibility of the object in the image, an image file generation step of generating an image file from the image in a manner based at least on the result of the determination, and a recording control step of controlling the recording of the image file in a recording unit. Effect of the Invention

[0008] According to the present invention, it is possible to provide an image recording device and an image recording method that, when an object is recognized as an image, determine the visibility of the image to ensure that the user can easily see the object, and control and record the aspect of the image based on the visibility of the object. [Brief description of the drawings]

[0009] [Figure 1] 1 is a block diagram showing an example of a configuration of an image recording device according to a first embodiment. [Diagram 2] 5 is a flowchart showing an example of a processing procedure of an image recording method according to the first embodiment. [Diagram 3] FIG. 11 is a block diagram showing an example of a configuration of an image recording device according to a second embodiment. [Figure 4] 10 is a flowchart showing an example of a processing procedure of an image recording method according to a second embodiment. [Diagram 5] FIG. 11 is a block diagram showing an example of a configuration of an image recording device according to a third embodiment. [Figure 6]11 is a flowchart showing an example of a processing procedure of an image recording method according to a third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] The present invention will be described below through the embodiments of the invention, but the invention according to the claims is not limited to the following embodiments. Also, not all of the configurations described in the embodiments are necessarily essential as means for solving the problems. For clarity of explanation, the following description and drawings have been omitted and simplified as appropriate.

[0011] <Embodiment 1> A first embodiment of the present invention will be described with reference to Fig. 1. Fig. 1 is a block diagram showing an example of the configuration of an image recording system 1 according to the first embodiment.

[0012] The image recording system 1 includes an image recording device 10, a camera 20, and a recording unit 30.

[0013] The camera 20 is mounted on a moving object such as a vehicle, a drone, a ship, or an airplane, and captures the surroundings of the moving object. The camera 20 may be a single camera or may be configured as a group of multiple cameras. The camera 20 may also be a portable device carried by a passenger of the moving object, such as a driver of a vehicle or a passenger on a ship.

[0014] The recording unit 30 is a recording unit such as a memory or a disk device provided in the image recording device 10. The recording unit 30 may be a recording medium such as an SD card. The recording unit 30 may be a recording device or a recording medium provided in a server, a mobile device, an in-vehicle device, or the like that is connected to the image recording device 10 wirelessly or by wire.

[0015] The image recording device 10 includes an image acquisition unit 11, an image recognition unit 12, a determination unit 13, an image file generation unit 14, and a recording control unit 15. The image recording device 10 is typically provided in a vehicle, and recognizes objects, such as famous buildings and popular mountains, from images of the scenery captured while the vehicle is traveling, and records and saves the objects as image files. The image recording device 10 may be a device mounted on a moving object, such as an in-vehicle device, or may be a portable device that can be carried by a user and has a camera function and a recording function.

[0016] The image acquisition unit 11 acquires images of the surroundings of a moving object from a camera 20 mounted on the moving object such as a vehicle, drone, ship, or airplane. The image acquisition unit 11 may acquire images from a group of multiple cameras 20. The image acquisition unit 11 may be configured to include the camera 20, or may be an input interface unit connected to a separate camera 20 and acquiring an output signal output from the camera 20. The image acquisition unit 11 outputs the acquired images to the image recognition unit 12 and the image file generation unit 14.

[0017] The image recognition unit 12 performs a process of image recognition of a predetermined object captured in an image acquired by the image acquisition unit 11. The image recognition unit 12 performs a process of recognizing, as an object, a feature on the ground, such as an artificial structure such as a building, a bridge, a traffic light, a signboard, a natural structure such as a mountain, a river, a topography, or a landscape formed by a combination of these. The image recognition unit 12 outputs the recognition result to the determination unit 13.

[0018] The image recognition unit 12 is equipped with a trained model for recognizing objects, which are geographical features, as a trained model for image recognition, and uses image recognition technology to recognize objects that exist around the moving body from the images acquired by the image acquisition unit 11. A trained model for image recognition is a learning model in which input data, which is a set of an image and training data, is input to a neural network and machine-learned. In this embodiment, an image of a landscape captured by a camera mounted on a moving object is used as the image. In addition, as training data, a label is used, which is set to 1 if the image contains an object and 0 if the image does not contain an object.

[0019] The neural network included in the image recognition unit 12, for example, inputs the image acquired by the image acquisition unit 11 into a trained model for recognizing a feature that is a target object, and outputs the possibility that the target object is captured in the input image as a value between 1 and 0. The closer the output value is to 1, the higher the possibility that the target object is captured in the input image, and the closer the output value is to 0, the lower the possibility that the target object is captured in the input image. If this output value is equal to or greater than a predetermined threshold value, for example, 0.8, the image recognition unit 12 outputs to the determination unit 13 a recognition result indicating that the target object is captured in the image acquired by the image acquisition unit 11. The image recognition unit 12 also outputs to the determination unit 13 a recognition size indicating the size of the area in which the object is recognized, as pixel count data indicated, for example, by the number of pixels in the horizontal direction and the number of pixels in the vertical direction. In addition to the above-mentioned configuration, the image recognition unit 12 may recognize the object by using a known technique such as pattern matching.

[0020] The determination unit 13 determines the visibility of the object recognized by the image recognition unit 12 in the image acquired by the image acquisition unit 11. For example, the determination unit 13 determines the visibility of the object to be "high" when the recognized size of the object has an area of ​​a predetermined number of pixels in the image or more, or when the recognition period during which the object is recognized on the video is equal to or longer than a predetermined period such as a predetermined number of frames or seconds. The determination unit 13 determines the visibility of the object to be "low" when the recognized size of the object has an area of ​​less than a predetermined number of pixels in the image or when the recognition period during which the object is recognized on the video is less than a predetermined period such as a predetermined number of frames or seconds. The determination unit 13 outputs the determination result of the visibility of the image to the image file generation unit 14. The determination unit 13 may be configured to classify the recognition size or recognition period of the object into a plurality of levels, and determine the visibility of the image acquired by the image acquisition unit 11 in a plurality of stages from high to low for each classified level. It is preferable that the determination unit 13 determines the visibility of the image based on both the recognition size and the recognition period of the object. The determination unit 13 may determine the visibility of the image based on either the recognition size or the recognition period of the object.

[0021] The determination unit 13 may include a movement speed calculation unit 16 to determine the visibility of the image. The movement speed calculation unit 16 calculates the movement speed at which the position on the image where the image recognition unit 12 recognizes the object moves along the time axis indicated by the continuity of the image frames on the video. The movement speed calculated by the movement speed calculation unit 16 is an index of how the recognized object, that is, the feature, is visually recognized on the video. The moving speed calculation unit 16 specifies the position on the image where the object is recognized by the image recognition unit 12 for each frame of the moving image. The moving speed calculation unit 16 calculates the amount of change in the position on the image where the object is recognized along the time axis in each frame of the moving image, and divides this by the elapsed time or the number of frames to calculate the moving speed at which the object moves on the image. The determination unit 13 determines the visibility of the image to be "low" when the moving speed is faster than a predetermined value based on the moving speed calculated by the moving speed calculation unit 16, and determines the visibility of the image to be "high" when the moving speed is less than the predetermined value. The determination unit 13 may be configured to classify the moving speed into a plurality of levels, and to determine the visibility of the image acquired by the image acquisition unit 11 into a plurality of stages from high to low for each classified level.

[0022] Here, the method of calculating the moving speed by the moving speed calculation unit 16 will be described in detail. When the image recognition unit 12 recognizes an object in one frame of a moving image, the moving speed calculation unit 16 specifies the position where the object is recognized in that frame as the position of a pixel in the horizontal and vertical directions, or the position of a block in the horizontal and vertical directions. A block is a unit that groups together a plurality of pixels, such as 8 pixels x 8 pixels. The position where the object is recognized that is specified by the moving speed calculation unit 16 is, for example, the position of a pixel that is the center of the area where the object is recognized. Alternatively, the position where the object is recognized is the position of a block, or the position of a pixel or block that is the edge of the area where the object is recognized.

[0023] Similarly, the moving speed calculation unit 16 specifies the position where the object is recognized in each of the multiple frames of the video. Next, the moving speed calculation unit 16 calculates the amount of change in the position where the object is recognized between successive frames of the video in pixel units or block units. The moving speed calculation unit 16 averages the amount of change in the position where the object is recognized along the time axis between multiple frames, or performs processing such as taking the mode or median, to calculate the moving speed at which the object moves along the time axis in the video.

[0024] The image file generating unit 14 generates an image file from the image acquired by the image acquiring unit 11 in a mode based on the judgment result of the visibility judged by the judging unit 13. The image file generating unit 14 may include a mode in which an image file is not generated when the judgment result of the judging unit 13 indicates that the visibility of the image is low. The image file generating unit 14 is preferably configured to include an image quality control unit 17 that controls the image quality. When the judgment result of the visibility judged by the judging unit 13 indicates that the visibility of the image is high, the image quality control unit 17 sets the image quality of the image high, assuming that the image is worth recording and saving. When the judgment result of the visibility judged by the judging unit 13 indicates that the visibility of the image is low, the image quality control unit 17 sets the image quality of the image low, assuming that the image is worth recording and saving, thereby reducing the recording capacity required for recording the image. The image file generating unit 14 generates an image file from the image acquired by the image acquiring unit 11 in the image quality controlled by the image quality control unit 17, and outputs the image file to the recording control unit 15.

[0025] Here, the method of controlling image quality by the image quality control unit 17 will be described in detail. The image quality control unit 17 includes, for example, an image compression unit (not shown). Image compression refers to compressing and encoding an image using a method such as MPEG-4 or H.264 / AVC. The image quality control unit 17 increases the compression rate when compressing the image acquired by the image acquisition unit 11 to produce a low-quality image with significantly deteriorated image quality. Similarly, the image quality control unit 17 reduces the compression rate when compressing the image to produce a high-quality image with reduced deterioration in image quality due to compression. The image quality control unit 17 may perform control to deteriorate the image quality by reducing the number of pixels of the image acquired by the image acquisition unit 11 or thinning out the frame rate, without relying on control of the compression rate when compressing the image. The image quality control unit 17 may also control the image quality of the image by using a known means for controlling the image quality, thereby reducing or maintaining the image capacity. The image quality control unit 17 may be called a file capacity control unit, since one of its purposes is to control the capacity of the image file to be recorded. The image quality control unit 17 controls the image quality of the image acquired by the image acquisition unit 11 to high image quality or low image quality based on the visibility determination result determined by the determination unit 13. The image quality control unit 17 may be configured to classify the image visibility into a plurality of levels, and control the image quality of the image acquired by the image acquisition unit 11 into a plurality of stages ranging from relatively high image quality to low image quality for each classified level.

[0026] The recording control unit 15 controls the recording of the image file generated by the image file generation unit 14 in the recording unit 30. When the recording unit is an external recording device, the recording control unit 15 may control the recording of the image file by transmitting the image file to the external recording device.

[0027] Next, the processing procedure of the image recording device 10 according to the above-mentioned first embodiment will be described with reference to the flowchart shown in FIG.

[0028] First, in step S10, image acquisition unit 11 acquires an image captured by camera 20. Image acquisition unit 11 outputs the acquired image to image recognition unit 12, and the process proceeds to step S11.

[0029] In step S11, the image recognition unit 12 performs a process of image recognition of a predetermined object captured in an image acquired by the image acquisition unit 11. The image recognition unit 12 recognizes, for example, a feature as the object, and outputs the recognition result to the determination unit 13. Then, the process proceeds to step S12.

[0030] In step S12, the determination unit 13 determines the visibility of the object recognized in the acquired image based on the recognized size of the object, the period during which the object was recognized, the moving speed of the position on the image at which the object was recognized, etc. The determination unit 13 outputs the determination result of the visibility of the image to the image file generation unit 14, and proceeds to step S13.

[0031] In step S13, the image file generating unit 14 determines the image mode when generating an image file from the image acquired by the image acquiring unit 11 in a mode based on the visibility judgment result judged by the judging unit 13. The image mode is, for example, whether or not to generate an image, and the image quality setting of the image. Then, the process proceeds to step S14.

[0032] In step S14, the image file generating unit 14 generates an image file from the image acquired by the image acquiring unit 11 based on the image aspect determined in step S13. The image file generating unit 14 may include a mode in which an image file is not generated depending on the image aspect determined in step S13. The image file generating unit 14 outputs the generated image file to the recording control unit 15 and proceeds to step S15.

[0033] In step S15, the recording control unit 15 controls the recording of the image file generated by the image file generating unit 14 in the recording unit 30. Then, the flow ends.

[0034] As described above, according to embodiment 1, when an object is recognized in an acquired image, the image recording device 10 determines the visibility of the object, generates an image file in which the image aspect is controlled based on the ease of visibility, and records the image in the recording unit. By configuring the image recording device 10 in this way, when the visibility of a predetermined object is high in an image in which the object is photographed, that is, when the image is suitable for viewing by a user, the image can be saved in high image quality. Also, when the visibility of the object is low in an image in which the object is photographed, that is, when the object is difficult for a user to view and is not suitable for viewing, the image is not recorded or the image quality is reduced when the image is recorded, thereby suppressing the capacity of the image file and reducing the required capacity of the recording device required for recording.

[0035] <Embodiment 2> A second embodiment of the present invention will be described with reference to Fig. 3. An image recording system 1a according to the second embodiment includes an image recording device 10a, a camera 20, a recording unit 30, and a GNSS receiving unit 40. The image recording device 10a has a configuration that differs from the image recording device 10 according to the first embodiment in the following points. That is, the image recording device 10a includes a determination unit 13a. The determination unit 13a further includes a relative position calculation unit 18. The image recording device 10a has the same configuration and functions as the image recording device 10 according to the first embodiment except for the determination unit 13a, and therefore detailed description thereof will be omitted.

[0036] The GNSS receiver 40 receives GNSS signals from GNSS satellites (not shown). The GNSS receiver 40 outputs the received radio signal to the image recording device 10a. Alternatively, the GNSS receiver 40 uses the received radio signal to determine the current position and outputs information on the determined current position to the image recording device 10a. The GNSS receiver 40 may be implemented as a function of a portable device carried by a vehicle occupant, such as a driver or a passenger, and may be a device that outputs the received radio signal or information on the determined current position to the image recording device 10a.

[0037] The determination unit 13a includes a relative position calculation unit 18 instead of or in addition to the moving speed calculation unit 16 included in the determination unit 13 of the first embodiment. The determination unit 13a determines the visibility of the object in the acquired image by the relative position calculation unit 18. The relative position calculation unit 18 calculates the relative position between the position where the object exists and the current position of the moving body, and calculates the relative positional relationship between them. The relative position calculated by the relative position calculation unit 18 is an index of the positional relationship between the recognized object, that is, the feature, and the moving body, and how it is visually recognized from the moving body. For example, when the position where the object exists is in front of the moving body at the current position, the determination unit 13a determines the visibility of the object to be "high", and when the position where the object exists is away from the front of the moving body at the current position by a predetermined angle or more, the determination unit 13a determines the visibility of the object to be "low". For example, when the position where the object exists is a distance less than a predetermined value from the current position of the moving body, the determination unit 13a determines the visibility of the object to be “high,” and when the position where the object exists is a distance equal to or greater than a predetermined value from the current position of the moving body, the determination unit 13a determines the visibility of the object to be “low.” The determination unit 13a outputs the determination result of the visibility of the image to the image file generation unit 14.

[0038] The relative position calculation unit 18 acquires the position where the object exists from a map information database (not shown). The object is, for example, a famous building, and the latitude and longitude information of its location is registered in the map information database. The relative position calculation unit 18 may acquire the position where the object exists from the history recording unit as the position information when the same object was recognized in the past. The relative position calculation unit 18 also uses the radio wave signal received by the GNSS receiver 40 to measure the latitude and longitude information of the current position of the mobile object. The relative position calculation unit 18 further calculates the traveling direction of the mobile object from the direction in which the current position of the mobile object moves along the time axis.

[0039] The relative position calculation unit 18 calculates and outputs the relative position between the position where the object exists and the current position of the moving body using latitude and longitude information. The "angle at which the object is located with respect to the moving body's traveling direction" is called the relative angle, and the "distance between the object and the moving body" is called the relative distance. The relative position between the position where the object exists and the moving body's current position is indicated by either the relative angle or the relative distance, or both. For example, if the object exists 5 degrees to the left of the moving body's traveling direction (front), the relative angle is 5 degrees.

[0040] The determination unit 13a determines the visibility of the object based on at least one of the relative angle and the relative distance calculated by the relative position calculation unit 18. For example, when the relative angle is within a predetermined angle, for example, 10 degrees, or the relative distance is less than a predetermined distance, for example, 5 km, the determination unit 13a determines the visibility to be "high". When the relative angle is equal to or greater than a predetermined angle, or the relative distance is equal to or greater than a predetermined distance, the determination unit 13a determines the visibility to be "low". The determination unit 13a may classify the relative angle and the relative distance into a plurality of levels, and may determine the visibility of the image acquired by the image acquisition unit 11 into a plurality of stages from high to low for each classified level. It is preferable that the determination unit 13a determines the visibility of the image based on both the relative angle and the relative distance.

[0041] Next, the processing procedure of the image recording device according to the above-mentioned embodiment 2 will be described with reference to the flowchart shown in Fig. 4. Note that steps that are the same as those in the embodiment 1 will be given the same step numbers, and detailed descriptions will be omitted.

[0042] After steps S10 and S11, in step S22, the determination unit 13a calculates the relative position between the position where the object exists and the current position of the moving body using the latitude and longitude information. The determination unit 13a calculates at least one of the relative angle, which is the angle at which the object is located with respect to the traveling direction of the moving body, and the relative distance, which is the distance between the object and the moving body. The determination unit 13a determines the visibility of the object based on at least one of the relative angle and the relative distance. The determination unit 13a outputs the determination result of the visibility of the image to the image file generation unit 14, and proceeds to step S13.

[0043] The processes from step S13 onwards are similar to those in the first embodiment, and therefore a detailed description thereof will be omitted.

[0044] As described above, according to the second embodiment, when an object is recognized in an acquired image, the image recording device 10a calculates the relative position between the object and the moving body, and determines the visibility of the object based on the difference between the position of the object and the traveling direction of the moving body and the distance between the position of the object and the current position of the moving body. The image recording device 10a can generate an image file in which the aspect of the image is controlled based on the determined visibility, and record the image in the recording unit. By configuring the image recording device 10a in this way, when the visibility of a predetermined object is high in an image in which the object is photographed, that is, when the image is suitable for viewing by a user, the image can be saved with high image quality. Also, when the visibility of the object is low in an image in which the object is photographed, that is, when the object is difficult for a user to view and is not suitable for viewing, the image is not recorded or the image quality is reduced when the image is recorded, thereby suppressing the capacity of the image file and reducing the required capacity of the recording device required for recording.

[0045] <Embodiment 3> The third embodiment of the present invention has different embodiments and effects from the image recording device 10 and the image recording device 10a according to the first and second embodiments described above. When an object is recognized as an image and the user has photographed the same object in the past, the image recording device 10b determines and records the aspect of the image based on the degree of similarity between the photographed locations. When the position where the same object was photographed in the past is close to the current position of the moving object, the image recording device 10b determines that it is not worth recording and saving the image because it is estimated that there is a high similarity between the image recorded in the past. At this time, the image recording device 10b does not record the image or records the image with a low image quality, thereby suppressing the capacity of the image file and reducing the required capacity of the recording device required for recording.

[0046] A third embodiment of the present invention will be described with reference to Fig. 5. An image recording system 1b according to the third embodiment includes an image recording device 10b, a camera 20, a recording unit 30, and a GNSS receiving unit 40. The image recording device 10b has a configuration that differs from the image recording device 10 and the image recording device 10a according to the first and second embodiments in the following respects. That is, the image recording device 10b includes a determination unit 13b, an image file generation unit 14a, and a position information history management unit 19. The image recording device 10b has the same configuration as the image recording device 10 and the image recording device 10a according to the first and second embodiments, except for including the determination unit 13b, the image file generation unit 14a, and the position information history management unit 19, and therefore a detailed description thereof will be omitted.

[0047] The location information history management unit 19 stores and manages information including the position of the mobile body when the object was recognized as history. The location information history management unit 19 uses radio wave signals received by the GNSS receiver 40 to measure latitude and longitude information of the current position of the mobile body. When the image recognition unit 12 recognizes a predetermined object, the location information history management unit 19 associates the recognized object with latitude and longitude information of the current position of the mobile body, stores and manages the information. In addition, the location information history management unit 19 may further associate and store information on the date and time when the image recognition unit 12 recognized the predetermined object and a determination result determined by the determination unit 13b described later. When managing the position of the mobile body when the object was recognized as history, the location information history management unit 19 preferably generates and manages a database (not shown) related to the position where the object was recognized.

[0048] When the determination unit 13b recognizes an object by image recognition, the determination unit 13b searches the history managed by the position information history management unit 19 using the object and the current position of the moving object as keys. When the image recognition unit 12 recognizes a predetermined object, the determination unit 13b searches the history and extracts information about a past location where the same object was recognized and an image was recorded. The determination unit 13b calculates the difference between the current position of the moving object measured using the radio wave signal received by the GNSS receiver 40 when the image recognition unit 12 recognizes the predetermined object and a location extracted by searching the history managed by the position information history management unit 19, determines the degree of approximation based on this difference, and outputs the determination result of the degree of approximation of the position where the object was recognized to the image file generation unit 14a.

[0049] Here, the method of determining the degree of approximation by the determination unit 13b will be described in detail. When the image recognition unit 12 recognizes a predetermined object, the determination unit 13b acquires the latitude and longitude information of the current position of the mobile body and the history information on the position where the above-mentioned object was recognized from the position information history management unit 19. The determination unit 13b searches the history using the object and the latitude and longitude information of the current position of the mobile body as keys, calculates information of a point that recognizes the same object and is closest to the current position of the mobile body by comparing the latitude and longitude information, and extracts it from the history. The determination unit 13b calculates the distance between these two points from the latitude and longitude information as the difference between the current position of the mobile body and the extracted point, and determines the degree of approximation to be "low" if the calculated distance is equal to or greater than a predetermined value. The determination unit 13b determines the degree of approximation to be "high" if the distance between the current position of the mobile body and the extracted point is less than a predetermined value. The judgment unit 13b may be configured to classify the distance between the current position of the moving body when the target object is recognized and a point extracted by searching the history managed by the position information history management unit 19 into a plurality of levels, and judge the degree of similarity between the current position of the moving body and the extracted point in a plurality of stages ranging from high to low for each classified level.

[0050] The determination unit 13b may acquire the current position of the moving object when the object is recognized, position information of a point where the same object has been recognized in history, and position information of a point where the object exists acquired from a map information database (not shown). The determination unit 13b compares the direction of the current position as seen from the object with the direction of a point extracted by searching the history as seen from the object, and calculates the difference between the current position of the moving object and the extracted point. If the difference in the calculated direction is equal to or greater than a predetermined value, the determination unit 13b determines the degree of approximation to be "low", and if the difference in the calculated direction is less than the predetermined value, the determination unit 13b determines the degree of approximation to be "high". The judgment unit 13b may classify the degree of difference between the direction of the current position of the moving body as seen from the target object and the direction of the position of a point extracted by searching the history as seen from the target object into a plurality of levels, and judge the degree of similarity between the current position of the moving body and the extracted point into a plurality of stages from high to low for each classified level.

[0051] The image file generating unit 14a generates an image file from the image acquired by the image acquiring unit 11 in a mode based on the result of the judgment of the degree of approximation judged by the judgment unit 13b, and outputs the image file to the recording control unit 15. When the judgment result of the judgment unit 13b indicates that the degree of approximation between the current position of the moving object and the extracted point is high, the image file generating unit 14a does not generate an image file. Alternatively, when the judgment result of the judgment unit 13b indicates that the degree of approximation between the current position of the moving object and the extracted point is high, the image file generating unit 14a sets the image quality to a low level, and reduces the recording capacity required for recording the image, since it is not worth recording and storing the image.

[0052] Next, the processing procedure of the image recording device according to the above-mentioned embodiment 3 will be described with reference to the flowchart shown in Fig. 6. Note that steps that are the same as those in the embodiment 1 will be given the same step numbers, and detailed descriptions will be omitted.

[0053] After steps S10 and S11, in step S32, the determination unit 13b calculates the degree of approximation between the current position of the moving object and a point extracted by searching the history managed by the position information history management unit 19, using the latitude and longitude information. When the determination unit 13b recognizes an object, it calculates the difference between the current position of the moving object and a past point extracted from the history, and determines the degree of approximation based on this difference. The determination unit 13b outputs the determination result of the degree of approximation of the position where the object was recognized to the image file generation unit 14a, and proceeds to step S33.

[0054] In step S33, the image file generating unit 14a determines the image mode when generating an image file from the image acquired by the image acquiring unit 11 in a mode based on the result of the similarity judgment made by the judging unit 13b. The image mode is, for example, whether or not to generate an image, and the image quality setting. Then, the process proceeds to step S14.

[0055] The processes from step S14 onwards are similar to those in the first embodiment, and therefore a detailed description thereof will be omitted.

[0056] As described above, according to the third embodiment, when an object is recognized in an acquired image, the image recording device 10b determines the degree of similarity between the current position of the moving object and a past point extracted from the history. The image recording device 10b generates an image file in which the aspect of the image is controlled based on the determined degree of similarity of the points, and can record the image file in the recording unit. By configuring the image recording device 10b in this way, when an image of a predetermined object is estimated to have low similarity to previously recorded images, that is, when the image is suitable for the user to keep in the record, the image can be saved in high image quality. Also, when the image is estimated to have high similarity to previously recorded images, that is, when the image is not worth the user to keep in the record, the image is not recorded or the image quality is reduced when the image is recorded, thereby suppressing the capacity of the image file and reducing the required capacity of the recording device required for recording.

[0057] <Other embodiments> The specific examples and the like shown in the above-mentioned embodiments 1 to 3 of the present invention are merely examples for facilitating understanding of the invention, and do not limit the present invention unless otherwise specified. The present invention is not limited to the first to third embodiments, and can be modified as appropriate without departing from the spirit and scope of the present invention. Also, the present invention may be implemented by appropriately combining the respective embodiments.

[0058] <Example of modification> In the first to third embodiments of the present invention, the object recognized by the image recognition unit 12 is a feature on the ground. The object is not limited to a feature on the ground, and may be, for example, a person, a specific vehicle, or a landscape that satisfies specific conditions. The landscape that satisfies specific conditions is, for example, natural phenomena such as snowfall, lightning, cherry blossoms in bloom, sunrise, and a full moon, or a landscape such as Diamond Fuji or autumn leaves in Kyoto that is created by a combination of specific features and these natural phenomena. In this case, the image recognition unit 12 is configured with a trained model for image recognition of such landscapes and phenomena.

[0059] In the first to third embodiments of the present invention, the image file generating unit 14 and the image file generating unit 14a control the aspect of the image based on the judgment results of the judgment units 13, 13a, and 13b. The image file generating unit 14 and the image file generating unit 14a of the modified example are provided with a user input unit (not shown) and control the aspect of the image based on the judgment results of each judgment unit and user input. More specifically, the user input unit accepts a user operation as to whether or not to permit control of the aspect of the image. The image file generating unit 14 and the image file generating unit 14a control the aspect of the image based on the judgment results of each judgment unit only when the accepted input permits control. By configuring the modified example in this way, it is possible to prevent an image from not being recorded against the user's intention or to prevent deterioration of the image quality.

[0060] In the third embodiment of the present invention, the determination unit 13b determines the degree of approximation by comparing the distance between the current position of the moving object and a point extracted by searching the history, or the direction of the current position as seen from the object and the direction of the position of a point extracted by searching the history as seen from the object. The degree of approximation is not limited to these determination methods, and may be determined by comparing the current date and time of the moving object with the date and time extracted from the history. In other words, the image recording device 10b according to the modified example of the third embodiment, when the object is recognized as an image and the user has photographed the same object in the past, determines and records the aspect of the image based on the degree of approximation between the photographing dates and times. The determination unit 13b determines that the degree of approximation is high when the date and time of the same object photographed in the past is close to the date and time of the current moving body, for example, when the date and time of the same object photographed in the past is within a predetermined period such as within one year. Alternatively, the determination unit 13b determines that the degree of approximation is high when the season to which the date and time of the same object photographed in the past belongs is the same as the season to which the date and time of the current moving body belongs, or the difference in the photographing month is within a predetermined period such as within one month. The determination unit 13b may determine that the degree of approximation is high when the time zone to which the date and time of the same object photographed in the past belongs is within a predetermined period such as within two hours from the time zone to which the date and time of the current moving body belongs. When the degree of approximation is high, the determination unit 13b estimates that the similarity with the image recorded in the past is high, and determines that it is not worth recording and saving the image. The determination unit 13b may also perform a determination based on the degree of approximation of weather, such as rainy or sunny, or the degree of approximation of accompanying persons, such as whether the accompanying persons are the same or include the same accompanying persons. The determination unit 13b may perform the determination based on the degree of similarity of a plurality of conditions such as the above-mentioned location and date and time.

[0061] In the above embodiment, the present invention has been described as being configured as hardware, but the present invention is not limited to this. Any process of the present invention can also be realized by causing a processor to execute a computer program.

[0062] In the above example, the program can be stored and supplied to the computer using various types of non-transitory computer-readable media. The non-transitory computer-readable media include various types of tangible recording media, such as a magnetic recording medium such as a hard disk drive, an optical recording medium, a magneto-optical recording medium, and a semiconductor memory. The semiconductor memory can be various types of rewritable Read Only Memory (ROM) and Random Access Memory (RAM), and also includes those provided as Solid State Drives (SSDs). The program may also be supplied to the computer by various types of temporary computer-readable media via wired communication paths such as electric wires and optical fibers, or wireless communication paths, including, for example, electrical signals, optical signals, and electromagnetic waves.

[0063] The order of execution of each process in the apparatus and method shown in the claims, specification, and drawings may be arbitrary. Even if the operational flow in the claims, specification, and drawings is described using "first," "next," etc. for convenience, it does not mean that it is essential to perform the process in this order. [Explanation of symbols]

[0064] 1,1a,1b Image recording system 10, 10a, 10b Image recording device 11 Image acquisition section 12 Image Recognition Unit 13, 13a, 13b Judgment section 14,14a Image file generation section 15 Recording control section 16 Movement speed calculation section 17 Image quality control section 18 Relative position calculation unit 19 Location information history management section 20 Camera 30 Recording Section 40 GNSS receivers

Claims

1. an image acquisition unit that acquires an image of the surroundings of the moving object; an image recognition unit that performs image recognition for a predetermined object from the image; A determination unit that determines visibility of the object in the image; an image file generating unit that generates an image file from the image in a manner based at least on a result of the determination; a recording control unit that controls recording of the image file in a recording unit; An image recording device comprising:

2. The image file generating unit further includes an image quality control unit, and generates an image file in which the image quality of the image is controlled based on the visibility.

2. The image recording device according to claim 1.

3. The determination unit further includes a movement speed calculation unit that calculates a movement speed at which a position on the image at which the object is recognized moves over time, The image file generation unit generates an image file from the image in a mode based on the moving speed.

3. The image recording apparatus according to claim 1 or 2.

4. the determination unit further includes a relative position calculation unit that calculates a relative position between a position where the object is present and a current position of the moving object, the image file generation unit generates an image file from the image in a manner based on the relative position.

3. The image recording apparatus according to claim 1 or 2.

5. An image acquisition step of acquiring an image of the surroundings of the moving object; an image recognition step of image-recognizing a predetermined object from the image; a determining step of determining visibility of the object in the image; an image file generating step of generating an image file from the image in a manner based at least on a result of the determination; a recording control step of controlling recording of the image file in a recording unit; The image recording method includes:

Images