Information processing apparatus

Abstract

To improve prediction accuracy of a route of a pedestrian to an obstacle.SOLUTION: The information processing apparatus 10 includes an acquisition unit 14 and a control unit 15. The acquisition unit 14 acquires an image from the imaging device 12 mounted on the mobile object 11. The control unit 15 predicts movement of a person included as a partial image in an image acquired by the acquisition unit 14. The control unit 15 can detect a person and an object included as a partial image in the image. The control unit 15 can calculate the positions of the person and the object based on the image. When an angle between a normal vector of a first line portion connecting both ends of an object detected from an image and a velocity vector of a person is an angle threshold in a case where the object is located within a predetermined range, the controller 15 predicts that the person will move toward a first point near one of the both ends.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device. [Background technology]

[0002] With the advancement of autonomous driving technology, mobile objects with low-speed autonomous driving functions that travel on sidewalks are beginning to be put into practical use. For such mobile objects, there is a need to predict the routes that pedestrians will take. To predict routes, technologies have been developed that predict routes based on images captured by an imaging device attached to the mobile object. It is expected that pedestrians will move to avoid obstacles that are in their path of travel. For example, it has been proposed to change routes to avoid obstacles that exist in their path of travel (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 10-333746 Summary of the Invention [Problem to be solved by the invention]

[0004] However, it is difficult to detect obstacles that exist beyond pedestrians from images captured by an imaging device installed on a moving object, and there is room for improvement in predicting paths from images that capture the entire obstacle.

[0005] An object of the present disclosure is to improve the accuracy of predicting a pedestrian's path relative to an obstacle by using an image captured by an imaging device provided on a moving body. [Means for solving the problem]

[0006] An information processing device according to a first aspect comprises: an acquisition unit that acquires an image from an imaging device mounted on a moving object; a control unit that predicts a movement of a person included as a partial image in the image acquired by the acquisition unit, The control unit A person and an object included as a partial image in the image can be detected; The positions of the person and the object can be calculated based on the image; If the object is located within a predetermined range from the person, and the angle between the normal vector of a first line segment connecting both ends of the object detected in the image and the velocity vector of the person is less than an angle threshold, it is predicted that the person will head toward a first point near one of the ends. [Effects of the Invention]

[0007] According to the present disclosure, the accuracy of predicting pedestrian movement is improved. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a functional block diagram showing a schematic configuration of a mobile object including an information processing device according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram showing a predetermined range when a person is viewed from vertically above. [Figure 3] 10A and 10B are diagrams illustrating the edge of an object whose position is detected from an image. [Figure 4] FIG. 10 is a diagram illustrating an object that a person would not consider avoiding. [Figure 5] FIG. 10 is a diagram illustrating a method for calculating a first point. [Figure 6] 10 is a diagram for explaining a method of selecting an end to be used in calculating a first point from both ends of an object. FIG. [Figure 7] FIG. 10 is a diagram for explaining a method for calculating a target point after passing a first point. [Figure 8] 2 is a flowchart illustrating a prediction process executed by the control unit of FIG. 1. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In the following drawings, the same components are denoted by the same reference numerals.

[0010] The information processing device of the present disclosure may provide a moving body having a low-speed automated driving function with information on a moving trajectory and a destination point (to be described later). For example, as shown in FIG. 1, an information processing device 10 may be provided in a moving body 11 having a low-speed automated driving function.

[0011] The mobile object 11 having the low-speed automatic driving function is, for example, a vehicle for transporting luggage whose upper speed limit is approximately the same as that of a pedestrian. The mobile object 11 may include an imaging device 12, an information processing device 10, and a control device 13.

[0012] The imaging device 12 may be, for example, a monocular camera or a stereo camera. The imaging device 12 may be installed so as to be able to capture an image of the external scene of the moving body 11. The imaging device 12 may be mounted at a predetermined position and orientation relative to the moving body 11. The mounting position and mounting orientation of the imaging device 12 relative to the moving body 11 may be recognized as information by the imaging device 12 or the information processing device 10.

[0013] The imaging device 12 may capture an image and generate it as information. The imaging device 12 may have a timer. The imaging device 12 may add the time measured by the timer to the image as information. In a configuration in which the imaging device 12 stores the mounting position and mounting attitude in a built-in memory, the imaging device 12 may add the mounting position and mounting attitude to the image as information. The imaging device 12 may transmit the image together with the added information to the information processing device 10.

[0014] As will be described later, the information processing device 10 may estimate the movement trajectory of a surrounding animal based on an image acquired from the imaging device 12. The animal is, for example, a person. However, the animal is not limited to a person and may be a dog, deer, monkey, bear, cat, etc. Hereinafter, in this specification, the animal will be described as a person. The information processing device 10 may transmit the estimated movement trajectory to the control device 13 as information.

[0015] The control device 13 may control the operation of the mobile object 11. For example, the control device 13 may control the mobile object 11 to drive automatically based on the detection results of a position sensor such as a GNSS, a gyro sensor, a speed sensor, etc., that the mobile object 11 has, as well as a set destination, a surrounding map, etc. Specifically, the control device 13 may control the motor, brakes, steering angle, etc., based on the control results.

[0016] Furthermore, the control device 13 may avoid people and correct the movement path of the moving object 11 based on the movement trajectory received as information from the information processing device 10.

[0017] Furthermore, the control device 13 may cause the moving object 11 to issue a warning to a person or a monitor monitoring the moving object 11, based on the moving trajectory received as information from the information processing device 10.

[0018] The following describes in detail the configuration of the information processing device 10. The information processing device 10 includes an acquisition unit 14 and a control unit 15.

[0019] The acquisition unit 14 acquires an image from the imaging device 12. The acquisition unit 14 may acquire information from a sensor mounted on the moving object 11 other than the imaging device 12. For example, the acquisition unit 14 may acquire information about the positions of surrounding objects from a distance measuring device such as Lidar.

[0020] The acquisition unit 14 may further have a function of outputting information. Specifically, the acquisition unit 14 may transmit the movement trajectory and the destination point predicted by the control unit 15 as information to the control device 13, as will be described later.

[0021] The acquisition unit 14 may be, for example, a physical connector or a wireless communication device. Physical connectors include electrical connectors compatible with transmission of electrical signals, optical connectors compatible with transmission of optical signals, and electromagnetic connectors compatible with transmission of electromagnetic waves. Electrical connectors include connectors conforming to IEC 60603, connectors conforming to the USB standard, connectors compatible with RCA terminals, connectors compatible with S terminals specified in EIAJ CP-1211A, connectors compatible with D terminals specified in EIAJ RC-5237, connectors conforming to the HDMI (registered trademark) standard, and connectors compatible with coaxial cables, including BNC. Optical connectors include various connectors conforming to IEC 61754. Wireless communication devices include wireless communication devices conforming to various standards, including Bluetooth (registered trademark) and IEEE 802.11.

[0022] The control unit 15 is configured to include at least one processor, at least one dedicated circuit, or a combination thereof. The processor is a general-purpose processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), or a dedicated processor specialized for specific processing. The dedicated circuit may be, for example, an FPGA (Field-Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), or the like. The control unit 15 may control the operation of the information processing device 10.

[0023] The control unit 15 may further include a storage unit. The storage unit may include any storage device, such as a RAM (Random Access Memory) or a ROM (Read Only Memory). The storage unit may store various programs that cause the control unit 15 to function and various information used by the control unit 15.

[0024] For example, the storage unit may store as information the mounting position and mounting attitude of the imaging device 12 on the moving body 11. Furthermore, for example, when a pixel constituting an image acquired from the imaging device 12 indicates an object on the road surface or floor surface, the storage unit may store a conversion formula or conversion table for converting the coordinates of the pixel in the two-dimensional coordinate system into the coordinates in the world coordinate system.

[0025] The control unit 15 may add the time of acquisition as information to the image acquired by the acquisition unit 14. The control unit 15 may add the time of acquisition in a configuration in which the time is not added to the image transmitted by the imaging device 12.

[0026] The control unit 15 can detect people and objects around the moving body 11 based on the information acquired by the acquisition unit 14. The object may be large enough to obstruct a person's walking. The object large enough to obstruct a person's walking in a situation where the specifications of the moving body 11 are assumed may be determined in advance.

[0027] Specifically, the control unit 15 detects a person by determining whether or not the image acquired by the acquisition unit 14 includes partial images of a person and an object. The control unit 15 may use, for example, image analysis or a discrimination model constructed by machine learning to detect the person and object. Furthermore, the control unit 15 may generate a distance image from the distance measurement results acquired by the distance measurement sensor, and detect the person or object based on the distance image. Specifically, the control unit 15 may detect a person based on the shape of a point cloud that can be considered to be in the same distance range in the distance image.

[0028] The control unit 15 predicts the movement of a person included as a partial image in the image acquired by the acquisition unit 14. A specific method for predicting the movement will be described in detail below.

[0029] When the control unit 15 detects a person around the moving body 11, it calculates the position of the person based on the image. When the control unit 15 detects an object around the moving body 11, it calculates the position of the object based on the image. The positions of the person and the object may be indicated by positions in real space, in other words, coordinates in a world coordinate system. Hereinafter, in the description of this specification, unless otherwise specified, the positions of the person and the object mean positions in real space. The control unit 15 may calculate the positions of the person and the object based on the image by any method.

[0030] For example, in a configuration in which the imaging device 12 is a stereo camera and acquires stereo images, the control unit 15 may calculate the positions of the person and the object using triangulation technology based on the positions of the partial images of the person and the object in each of the two images. Furthermore, in a configuration in which the imaging device 12 is a monocular camera, the control unit 15 may perform distance measurement using image aberration, distance measurement using AI technology based on the relationship between the image and the distance, etc. Furthermore, in a configuration in which a ranging image is generated, the control unit 15 may calculate the positions of the person and the object based on the distances corresponding to the point clouds forming the person and the object, respectively, whose presence is recognized in the ranging image.

[0031] The control unit 15 may store a partial image of a detected person and the position of the person in the image of each frame in a buffer memory that is part of the storage unit. The control unit 15 may also store a partial image of a detected object and the position of the object in the image of each frame in the buffer memory.

[0032] The control unit 15 determines whether an object is located within a predetermined range from the position of the person. The predetermined range is the range within which a typical walking person avoids obstacles in their direction of travel. The predetermined range is a range determined in real space. As shown in FIG. 2, the predetermined range pr is, for example, 120 degrees in total, 60 degrees to the left and 60 degrees to the right from the direction of travel of the person PS, and 10 m from the position of the person PS. Note that the typical human visual field angle is 60 degrees horizontally per eye, so the above-described angle may be determined. Furthermore, the typical obstacle avoidance distance for a human is approximately 7.5 m, so the above-described distance, including a margin of error, may be determined.

[0033] The control unit 15 may determine the predetermined range pr using images of multiple frames at different times to determine whether an object exists within the predetermined range pr. The images of the multiple frames at different times may include, as a first image, an image that is the most recent image among the images stored in the storage unit, or an image that is within a range of times that can be considered to be the most recent image.

[0034] The control unit 15 may determine whether a partial image of a person identical to a person present in the first image is included in the image of the frame immediately preceding the current frame. The control unit 15 may perform the above determination using any method. For example, the control unit 15 may perform the determination using pattern matching, a discrimination model based on machine learning, or the like. When the image of the most recent frame contains partial images of multiple people, the control unit 15 may perform the above determination for each person. Based on the above determination, the control unit 15 may recognize the position of the person included as a partial image in each of the images of the multiple frames.

[0035] The control unit 15 may calculate the time difference associated with each of a plurality of frame images that include the same person as a partial image. The control unit 15 may also calculate the difference in the position of the person calculated for each of the plurality of frame images. The control unit 15 may calculate a velocity vector of the person ps based on the time difference and the position difference. The control unit 15 may determine the direction indicated by the velocity vector as the traveling direction.

[0036] The control unit 15 may determine a predetermined range pr based on the recognized direction of travel and the position of the person ps. Furthermore, the control unit 15 determines whether an object is located within the determined predetermined range pr. If an object is located within the predetermined range, the control unit 15 calculates the positions of both ends of the object detected in the image.

[0037] In this specification, the edge of an object is a single straight line that constitutes the contour of the object when viewed from above, in other words, the end of a line segment that continues in the same direction, or the end of a curve with a continuous slope. The edge detected from the image means the edge within the contour whose position is detected from the image. As shown in FIG. 3, if the actual edge red of an object obj located behind a person ps as viewed from the image capture device 12 overlaps with the person ps, it cannot be detected from the image captured by the image capture device 12. Therefore, the edge ded detected from the image is the intersection of the object obj and a first straight line L1 that passes through the outer edge of the person ps as viewed from the image capture device 12 in a top view.

[0038] The control unit 15 may calculate a normal vector nv of a first line segment sL1 connecting both ends ded of an object detected in an image based on the positions of the ends ded. The control unit 15 calculates a first angle θ1 formed by the normal vector nv and a velocity vector vv of the person ps. The control unit 15 determines whether the first angle θ1 is equal to or smaller than an angle threshold.

[0039] When the first angle θ1 is equal to or smaller than the angle threshold, the control unit 15 may determine that the object obj defining both ends ded is an obstacle that the person ps is likely to avoid. In other words, as shown in Fig. 4, when the first angle θ1 exceeds the angle threshold, the control unit 15 may determine that the object obj defining both ends ded is a wall or the like that is substantially aligned with the traveling direction, and is an object obj that the person ps does not intend to avoid from the beginning. The angle threshold may be set to an angle that generally leads to the determination that the object obj is an object that the person ps does not intend to avoid from the beginning with respect to the traveling direction.

[0040] When the first angle θ1 is equal to or smaller than the angle threshold, the control unit 15 determines the vicinity of one of the two ends ded as the first point. The control unit 15 predicts that the person ps will head toward the first point. In other words, the control unit 15 may predict the path that the person ps will take toward the first point as a movement trajectory.

[0041] As shown in Fig. 5, the first point p1 may be the intersection of a perpendicular line from one of the two ends ded to a second line L2 that is formed by moving the person ps from the position calculated based on the image at the velocity vector of the person ps. The control unit 15 may select the end used in calculating the first point p1 from the two ends ded. As shown in Fig. 6, the control unit 15 may select the end ed1, which has a narrower inclination angle with respect to the velocity vector vv of the second line segment sL2 that connects the position of the person ps calculated based on the image to each of the two ends ed1 and ed2, as one of the ends ded used in calculating the first point p1.

[0042] 7, the control unit 15 may predict, as the target point tp of the person ps, a position that is a predetermined distance away from the midpoint mp of the first line segment sL1 along the normal vector nv of the first line segment sL1. The predetermined distance may be, for example, a distance that a person typically travels along the normal vector nv in the direction along the first line segment sL1 after temporarily avoiding an obstacle before returning to the original path, and may be statistically determined in advance.

[0043] The control unit 15 may predict, as a movement trajectory tj, a path from the position of the person ps calculated based on the image, passing through the first point p1 and reaching the target point tp.

[0044] If the first point p1 is not calculated for the person ps, the control unit 15 may calculate the movement trajectory tj of the person ps using any method. For example, the control unit 15 may predict, as the movement trajectory tj, a path that moves the person ps in a straight line at a velocity vector from the position of the person ps based on the most recent image. Alternatively, the control unit 15 may predict, as the destination point, a position on the movement trajectory tj that is a predetermined movement time from the position of the person ps based on the most recent image. The predetermined movement time is set to any time that does not result in extremely low prediction accuracy. For example, the predetermined movement time may be set to 20 seconds, 15 seconds, 10 seconds, 5 seconds, 3 seconds, etc.

[0045] Next, the prediction process executed by the control unit 15 in this embodiment will be described with reference to the flowchart of Fig. 8. The prediction process starts, for example, every time an image is acquired.

[0046] In step S100, the control unit 15 determines whether or not a partial image of the person ps is included in the newly acquired image. If not, the prediction process ends. If included, the process proceeds to step S101.

[0047] In step S101, the control unit 15 calculates the position of the person ps based on the acquired image. After the calculation, the process proceeds to step S102.

[0048] In step S102, the control unit 15 stores the position of the person ps calculated in step S102 as information together with the time of the image. After storing, the process proceeds to step S103.

[0049] In step S103, the control unit 15 reads out from the storage unit the position and time of the person included as a partial image in an image going back a predetermined number of frames. After reading, the process proceeds to step S104.

[0050] In step S104, the control unit 15 calculates the velocity vector of the person ps based on the position and time of the person ps read out in step S103. After the calculation, the process proceeds to step S105.

[0051] In step S105, the control unit 15 determines whether the object obj is included as a partial image in the newly acquired image used to determine whether a partial image of the person ps is included in step S100. If it is included, the process proceeds to step S106. If it is not included, the process proceeds to step S116.

[0052] In step S106, the control unit 15 calculates the position of the object obj whose presence was confirmed in step S105. After the calculation, the process proceeds to step S107.

[0053] In step S107, the control unit 15 calculates a predetermined range for the position of the person ps based on the position of the person ps calculated in step S101 and the velocity vector vv calculated in step S104. After the calculation, the process proceeds to step S108.

[0054] In step S108, the control unit 15 determines whether the position of the object obj calculated in step S106 is located within the predetermined range calculated in step S107. If it is located within the predetermined range, the process proceeds to step S109. If it is not located within the predetermined range, the process proceeds to step S116.

[0055] In step S109, the control unit 15 calculates the positions of both ends ded of the object obj, the position of which was calculated in step S106. After the calculation, the process proceeds to step S110.

[0056] In step S110, the control unit 15 calculates a normal vector nv of the first line segment sL1 based on the positions of both ends ded calculated in step S109. Furthermore, the control unit 15 calculates a first angle θ1 formed by the calculated normal vector nv and the velocity vector vv calculated in step S104. After the calculation, the process proceeds to step S111.

[0057] In step S111, the control unit 15 determines whether the first angle θ1 is equal to or less than the angle threshold. If it is equal to or less than the angle threshold, the process proceeds to step S112. If it is not equal to or less than the angle threshold, the process proceeds to step S116.

[0058] In step S112, the control unit 15 calculates the tilt angles corresponding to the two ends ded whose positions were calculated in step S109. As described above, the corresponding tilt angles refer to the tilt angles of the second line segment sL2 connecting the position of the person ps and the position of the end ded with respect to the velocity vector vv calculated in step S104. After the calculation, the process proceeds to step S113.

[0059] In step S113, the control unit 15 selects the end ded corresponding to the narrower tilt angle from among the tilt angles calculated in step S112. After the selection, the process proceeds to step S114.

[0060] In step S114, the control unit 15 calculates the first point p1 using the edge ded selected in step S113. After the calculation, the process proceeds to step S115.

[0061] In step S115, the control unit 15 calculates the target point tp based on the positions of both ends ded calculated in step S109. After the calculation, the process proceeds to step S115.

[0062] In step S116, the control unit 15 predicts the movement trajectory tj. When the control unit 15 calculates the first point p1 and the target point tp in step S114 and step S115, respectively, the control unit 15 may calculate the movement trajectory tj based on the first point p1 and the target point tp. When the control unit 15 determines in step S105 that the object obj does not exist, the control unit 15 may calculate the movement trajectory tj using the velocity vector vv calculated in step S104. After the calculation, the process proceeds to step S117.

[0063] In step S117, the control unit 15 transmits the movement trajectory tj calculated in step S116 as information to the control device 13. After transmission, the prediction process ends.

[0064] The information processing device 10 of this embodiment, configured as described above, includes an acquisition unit 14 that acquires images from an imaging device 12 mounted on a moving object 11, and a control unit 15 that predicts the movement of a person ps included as a partial image in the image acquired by the acquisition unit 14. The control unit 15 can detect the person ps and object obj included as partial images in the image, calculate the positions of the person ps and object obj based on the image, and, if the object obj is located within a predetermined range pr from the person ps, predict that the person ps will head toward a first point p1 near one of the ends ded when the angle θ1 formed by the normal vector nv of a first line segment sL1 connecting both ends ded of the object obj detected in the image and the velocity vector vv of the person ps is equal to or less than an angle threshold. With this configuration, the information processing device 10 can determine whether the object obj detected in the image is an object ojb to be avoided. Therefore, the information processing device 10 can determine whether the object obj is an obstacle that should be avoided even when it does not recognize the entire area of ​​the object obj detected from the image, thereby improving the accuracy of predicting the pedestrian's path relative to the obstacle.

[0065] Furthermore, in the information processing device 10, the first point p1 is the intersection of a perpendicular line from one of the two ends ded to a straight line drawn from the position of the person ps calculated based on the image, along with the velocity vector vv of the person ps. In general, in predicting the movement of a runner, it has been proposed to set a secondary destination point at a position a predetermined distance from the edge of an obstacle. However, setting a predetermined distance as a secondary destination point does not take into account the actual progress of the person ps, resulting in relatively low prediction accuracy. On the other hand, the information processing device 10 having the above-described configuration predicts the passing position of the person ps based on the position and velocity vector vv of the person ps, thereby improving prediction accuracy.

[0066] Furthermore, the information processing device 10 selects the end ded of the second line segment sL2 connecting the position of the person ps to each of the ends ded, which has a narrower inclination angle with respect to the velocity vector vv, as one of the ends ded. The end ded with a narrower inclination angle is generally the end ded in the direction in which the avoidance distance is likely to be shortest. In response to such a general assumption, the information processing device 10 having the above-described configuration can improve the accuracy of estimating the first point p1 for avoiding the obstacle object obj.

[0067] Furthermore, the information processing device 10 predicts a position that is a predetermined distance away from the midpoint mp of the first line segment sL1 along the normal vector nv as the target point tp of the person ps. With this configuration, the information processing device 10 can predict the movement trajectory tj of the person ps after avoiding the obstacle object obj.

[0068] The above has described an embodiment of the information processing device 10, but the embodiment of the present disclosure can also be implemented as a method or program for implementing the device, or as a storage medium on which a program is recorded (for example, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a CD-RW, a magnetic tape, a hard disk, or a memory card, etc.).

[0069] Furthermore, the implementation form of the program is not limited to application programs such as object code compiled by a compiler or program code executed by an interpreter, but may also be in the form of a program module incorporated into an operating system. Furthermore, the program may or may not be configured so that all processing is performed solely by the CPU on the control board. The program may also be configured so that part or all of it is executed by another processing unit mounted on an expansion board or expansion unit added to the board as needed.

[0070] The drawings illustrating the embodiments of the present disclosure are schematic, and the dimensional ratios and the like in the drawings do not necessarily correspond to the actual ones.

[0071] Although the embodiments of the present disclosure have been described based on the drawings and examples, it should be noted that those skilled in the art could make various modifications or alterations based on the present disclosure. Therefore, it should be noted that these modifications or alterations are included in the scope of the present disclosure. For example, the functions included in each component can be rearranged so as not to be logically inconsistent, and multiple components can be combined or divided into one.

[0072] For example, the information processing device 10 of this embodiment is mounted on the moving body 11, but may be embodied as, for example, a server device that communicates with the moving body 11. The information processing device embodied as a server device may acquire images from the moving body 11 and provide the moving body 11 with the predicted movement trajectory tj and the target point tp as information.

[0073] All of the features described in this disclosure and / or all steps of all disclosed methods or processes may be combined in any combination except combinations in which these features are mutually exclusive. Furthermore, each feature described in this disclosure may be replaced by an alternative feature serving the same, equivalent, or similar purpose, unless expressly denied. Thus, unless expressly denied, each disclosed feature is only one example of a generic series of identical or equivalent features.

[0074] Furthermore, embodiments of the present disclosure are not limited to the specific configurations of any of the above-described embodiments, but rather extend to any novel feature or combination thereof described herein, or any novel method or process step or combination thereof described herein.

[0075] In this disclosure, descriptions such as "first" and "second" are identifiers for distinguishing the configuration. In this disclosure, the configurations distinguished by descriptions such as "first" and "second" can have their numbers exchanged. For example, the first line can exchange the identifiers "first" and "second" with the second line. The exchange of identifiers is performed simultaneously. The configurations remain distinguished even after the identifier exchange. Identifiers may be deleted. A configuration from which an identifier has been deleted is distinguished by a symbol. The descriptions of identifiers such as "first" and "second" in this disclosure should not be used solely to interpret the order of the configurations or to justify the existence of an identifier with a smaller number. [Explanation of symbols]

[0076] 10. Information processing equipment 11 Mobile 12 Imaging device 13 Control device 14 Acquisition Department 15 Control Unit Edges detected by the ded image L1 First line L2 Second line nv normal vector obj object pr specified range ps person p1 First point red Actual edge sL1 First line segment sL2 Second line segment tj moving trajectory tp target point vv velocity vector

Claims

1. an acquisition unit that acquires an image from an imaging device mounted on a moving object; a control unit that predicts the movement of an animal included as a partial image in the image acquired by the acquisition unit, The control unit capable of detecting animals and objects contained in the image as partial images; The positions of the person and the object can be calculated based on the image; When the object is located within a predetermined range from the animal, if the angle formed between a normal vector of a first line segment connecting both ends of the object detected from the image and a velocity vector of the person is equal to or smaller than an angle threshold, it is predicted that the animal will head toward a first point near one of the ends. Information processing device.

2. 2. The information processing device according to claim 1, The first point is an intersection point of a straight line drawn from the position of the animal calculated based on the image along the velocity vector of the animal, and a perpendicular line drawn from one of the two ends of the straight line. Information processing device.

3. 3. The information processing device according to claim 2, The control unit selects, as one of the two ends, an end of a second line segment that connects the position of the animal to each of the two ends and that has a small inclination angle with respect to the velocity vector. Information processing device.

4. 4. The information processing device according to claim 1, The control unit predicts a position that is a predetermined distance away from the midpoint of the first line segment along the normal vector as a destination point of the animal. Information processing device.

Images