Robot, direction control method and program

Abstract

To detect from the whole image while suppressing a load.SOLUTION: An image processor 120 includes an acquisition unit 130, a face detection unit 140, and a change detection unit 150. The acquisition unit 130 acquires an image continuously taken by an imaging unit 105. The face detection unit 140 performs detection processing (facial detection processing) of detecting a predetermined target (a human face) from the center region of the acquired image. A change detection unit 150 detects a time-series change in a peripheral region around the center region of the acquired image.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention provides Robot, direction control method and programs. [Background technology]

[0002] Face detection processing for detecting a person's face from a captured image is known. Patent Document 1 describes a method for performing face detection processing only on the central region of a captured image in order to reduce the load of the face detection processing. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2009-20163 Summary of the Invention [Problem to be solved by the invention]

[0004] In the method described in Patent Document 1, face detection processing is not performed on the peripheral area around the central area of ​​the image. Therefore, the method described in Patent Document 1 cannot detect when a person enters the peripheral area from outside the image.

[0005] The present invention has been made in consideration of the above-mentioned circumstances, and is capable of performing detection from the entire image while suppressing the load. Robot, direction control method The purpose is to provide programs and [Means for solving the problem]

[0006] In order to achieve the above object, the robot of the present invention is a robot whose head is oriented in a direction in which an image is captured by an image capturing unit provided on the head, and which divides each of the images continuously acquired by the image capturing unit into a central region that includes the center of the image and a peripheral region that includes the periphery of the central region, at least,When a first detection process for detecting a person's face in the central area does not detect the person's face, and when the face of the person is detected by the first detection process targeting the central area and it is detected that the face of the person has changed from facing the robot to facing sideways. The present invention is characterized in that it comprises an image processing means that, instead of the first detection process, starts a second detection process that detects an image change between images targeting the peripheral area, and, when an image change between images is detected by the second detection process, identifies the position where the image change is detected as a first position, and, when the first position is identified by the image processing means, changes the orientation of the head so that the first position is at the center of the image acquired by imaging by the imaging unit. A direction control method according to the present invention is a direction control method executed by a robot whose head is set to a direction in which an image is captured by an image capturing unit provided on the head, the method comprising: dividing each of images continuously acquired by the image capturing unit into a central region that includes a center of the image and a peripheral region that includes a periphery of the central region; at least, When a first detection process for detecting a person's face in the central area does not detect the person's face, and when the face of the person is detected by the first detection process targeting the central area and it is detected that the face of the person has changed from facing the robot to facing sideways. The method is characterized by having an image processing step in which, instead of the first detection process, a second detection process is started to detect an image change between images targeting the peripheral area, and when an image change between images is detected by the second detection process, the position where the image change is detected is identified as a first position; and a drive control step in which, when the first position is identified by the image processing step, the orientation of the head is changed so that the first position is located at the center of an image acquired by imaging by the imaging unit. The program according to the present invention also includes a program for causing a computer of a robot whose head is set to a direction in which an image is captured by an image capturing unit provided on the head to divide each of images continuously acquired by the image capturing unit into a central region that includes the center of the image and a peripheral region that includes the periphery of the central region, at least, When a first detection process for detecting a person's face in the central area does not detect the person's face, and when the face of the person is detected by the first detection process targeting the central area and it is detected that the face of the person has changed from facing the robot to facing sideways.The present invention is characterized in that the image processing means functions as an image processing means that starts a second detection process instead of the first detection process, which detects an image change between images targeting the surrounding area, and when an image change between images is detected by the second detection process, identifies the position where the image change is detected as a first position, and when the first position is identified by the image processing means, changes the orientation of the head so that the first position is at the center of the image acquired by imaging by the imaging unit. [Effects of the Invention]

[0007] According to the present invention, it is possible to perform detection from the entire image while suppressing the load. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a diagram illustrating an external appearance of a robot according to an embodiment of the present invention. [Figure 2] FIG. 2 is a block diagram showing the hardware configuration of the robot according to the embodiment. [Figure 3] FIG. 1 is a block diagram showing a functional configuration of a robot according to an embodiment. [Figure 4] 10 is a flowchart showing the flow of a robot control process executed by the robot according to the embodiment. [Figure 5A] 1 is a diagram for explaining a specific example of a robot control process executed by a robot according to an embodiment (part 1). FIG. [Figure 5B] FIG. 2 is a diagram for explaining a specific example of the robot control process executed by the robot according to the embodiment (part 2). [Figure 6A] FIG. 10 is a diagram (part 3) for explaining a specific example of the robot control process executed by the robot according to the embodiment. [Figure 6B] FIG. 4 is a diagram for explaining a specific example of the robot control process executed by the robot according to the embodiment (part 4). [Figure 7A]FIG. 5 is a diagram for explaining a specific example of the robot control process executed by the robot according to the embodiment (part 5). [Figure 7B] FIG. 6 is a diagram for explaining a specific example of the robot control process executed by the robot according to the embodiment (part 6). DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings, in which the same or corresponding parts are designated by the same reference numerals.

[0010] 1 shows the appearance of a robot 100 according to a first embodiment of the present invention. The robot 100 is a communication robot that is primarily used in homes, exhibition halls, etc. and is capable of communicating with users. The robot 100 operates autonomously according to a predefined operation program.

[0011] 1, the robot 100 has a three-dimensional, roughly humanoid shape. The robot 100 is made primarily of a hard synthetic resin such as plastic. The robot 100 includes a head 101, a body 102, hands 103, and feet 104.

[0012] The head 101, hands 103, and feet 104 are parts that can be moved by drive members built into the robot 100. For example, the head 101 is attached to the body 102 by a neck joint provided in the neck so as to be rotatable in three directions: pitch, roll, and yaw. The head 101 is also provided with an imaging unit 105 that captures images of the area in front of the robot 100.

[0013] Fig. 2 shows the hardware configuration of the robot 100. As shown in Fig. 2, the robot 100 includes an imaging unit 105, a control unit 106, a storage unit 107, a battery 108, a driving driver 109, a movable unit 110, a wireless communication unit 111, a display unit 112, an audio input unit 113, and an audio output unit 114.

[0014] The control unit 106 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory). The CPU is, for example, a microprocessor, and is a central processing unit that executes various processes and calculations. In the control unit 106, the CPU reads out a control program stored in the ROM and controls the overall operation of the device itself (robot 100) while using the RAM as a work memory.

[0015] The control unit 106 also includes an image processing processor such as a GPU (Graphics Processing Unit) and various buffer memories. The control unit 106 uses the image processing processor to execute a face detection process for detecting a human face from the central area of ​​the image captured by the imaging unit 105, and a process for detecting changes over time in areas other than the central area of ​​the image (hereinafter referred to as peripheral areas), using a well-known image recognition technique.

[0016] The storage unit 107 is a non-volatile memory such as a flash memory or a hard disk. The storage unit 107 stores programs and data used by the control unit 106 to perform various processes, including an operating system (OS) and application programs. For example, the storage unit 107 stores data for identifying a central region and a peripheral region from an image captured by the imaging unit 105. For example, the central region may be defined as a rectangular region occupying 40% of the entire image from the center of the captured image, and the peripheral region may be defined as the region other than the central region in the image. Note that the definition of the central region and the peripheral region is arbitrary. The central basin and peripheral region may be defined in proportions different from those in the above example, or a circular or elliptical region in the center of the image may be defined as the central region. The storage unit 107 also stores data generated or acquired by the control unit 106 when performing various processes.

[0017] The battery 108 is a storage battery that stores electric energy and supplies power to each part of the robot 100. When the robot 100 returns to the charging station, the battery 108 is charged by the charging station.

[0018] The driving driver 109 includes driving members such as motors and actuators that drive the movable part 110 of the robot 100, and a driving circuit that drives these driving members. The movable part 110 is a movable part, specifically the head 101, hands 103, and feet 104. The control unit 106 transmits control signals to the driving circuits based on an operation program. The driving circuits supply driving pulse signals to the driving members in accordance with the control signals transmitted from the control unit 106. The driving members drive the movable part 110 in accordance with the pulse signals supplied from the driving circuit.

[0019] The robot 100 can perform various operations by the driver 109 driving the movable unit 110. For example, the robot 100 can move forward or backward and change its own orientation by moving the feet 104. Furthermore, the driver 109 can drive the head 101 to change the imaging direction of the imaging unit 105 installed in the head 101.

[0020] The imaging unit 105 is a so-called camera and is installed in the head 101. The imaging unit 105 captures an image in the direction in which the head 101 is facing. The imaging unit 105 captures an image at predetermined time intervals (for example, every 1 / 60 seconds). The imaging unit 105 functions as an imaging means.

[0021] The wireless communication unit 111 includes an interface for wirelessly communicating with external devices. Under the control of the control unit 106, the wireless communication unit 111 wirelessly communicates with external devices such as a charging station for charging the robot 100 and a cloud server in accordance with a communication standard such as a wireless LAN (Local Area Network) such as Wi-Fi (Wireless Fidelity), Bluetooth (registered trademark), or NFC (Near Field Communication). The wireless communication unit 111 functions as a communication means.

[0022] The display unit 112 is a display device such as a liquid crystal display, an organic EL (Electro Luminescence) display, or an LED (Light Emitting Diode). The display unit 112 displays various images depending on the situation under the control of a display drive circuit (not shown). The display unit 112 functions as a display means.

[0023] The voice input unit 113 is installed in the head 101, and detects (acquires) voices emitted by the user, surrounding environmental sounds, etc., and outputs them to the control unit 106. Although not shown, the robot 100 is equipped with a plurality of microphones surrounding the head 101 as the voice input unit 113, and can efficiently detect sounds generated on all sides. The voice input unit 113 functions as a voice acquisition means.

[0024] The audio output unit 114 includes a speaker and an audio output interface, and converts audio data generated by the control unit 106 into audio and outputs it to the outside. The speaker is installed in the head 101. The audio output unit 114 outputs various sounds including animal cries and human speech. For example, the robot 100 collects audio from a communication target using the audio input unit 113, and outputs audio corresponding to the content of the communication target's speech from the audio output unit 114. This allows the robot 100 to have a simple conversation with the communication target.

[0025] Next, the functional configuration of the robot 100 according to the present invention will be described with reference to Fig. 3. As shown in Fig. 3, the robot 100 functionally comprises an acquisition unit 130, a face detection unit 140, a change detection unit 150, a discrimination unit 160, and an imaging direction control unit 170. These units function by the CPU in the control unit 106 reading a program stored in a ROM into a RAM and executing and controlling the read program. The acquisition unit 130, face detection unit 140, change detection unit 150, and imaging direction control unit 170 described above correspond to the functions of the image processing device 120 included in the robot 100.

[0026] The acquisition unit 130 acquires images continuously captured by the imaging unit 105. The acquisition unit 130 functions as an acquisition unit.

[0027] The face detection unit 140 performs face detection processing to detect a human face from the image acquired by the acquisition unit 130 using a known method. At this time, the face detection unit 140 performs face detection processing only on the central region of the acquired image. For example, the face detection unit 140 stores data that serves as a template of a human face and data that indicates facial features in the storage unit 107, and detects a human face from the central region by referring to this data. The face detection unit 140 functions as an object detection means that detects a predetermined object. The face detection processing performed by the face detection unit 140 is an example of a first detection processing of the present invention.

[0028] The change detection unit 150 detects changes over time in the peripheral area around the central area of ​​the image acquired by the acquisition unit 130 (i.e., the area other than the central area). The change detection unit 150 functions as a change detection means. Note that the change detection unit 150 only detects changes and does not specifically identify the nature of the detected changes. Therefore, the processing performed by the change detection unit 150 imposes a smaller load than the face detection processing performed by the face detection unit 140. The processing performed by the change detection unit 150 is an example of the second detection processing of the present invention.

[0029] For example, the change detection unit 150 detects changes over time in the surrounding area based on feature points in the surrounding area of ​​the image acquired by the acquisition unit 130. Specifically, the change detection unit 150 first detects feature points from the surrounding area for each of the multiple images continuously captured by the imaging unit 105. Then, the change detection unit 150 calculates the amount of movement (movement vector) of the detected feature point between images, and if the amount of movement is equal to or greater than a threshold, detects the position of the feature point in the latest image as a part where a change has occurred.

[0030] Alternatively, the change detection unit 150 may detect changes over time in the surrounding area based on differences in the surrounding area between consecutively captured images. Specifically, the change detection unit 150 calculates the difference in pixel value for each pixel in the surrounding area of ​​multiple images consecutively captured by the imaging unit 105. Then, if there is a pixel where the calculated difference is equal to or greater than a threshold, the change detection unit 150 detects the part of that pixel as a part where a change has occurred.

[0031] The above-described method of detecting a change by change detection section 150 is an example, and a change may be detected from the peripheral region of an image by other methods.

[0032] The determining unit 160 determines whether or not a predetermined condition is met. The determining unit 160 functions as a determining means.

[0033] The imaging direction control unit 170 controls the driver 109 to drive the head 101, thereby changing the imaging direction of the imaging unit 105 provided in the head 101. For example, when the determination unit 160 determines that a predetermined condition is met and the change detection unit 150 detects a change over time in the peripheral area of ​​the image, the imaging direction control unit 170 controls the imaging direction of the imaging unit 105 so that the area where the change occurred is located in the center of the imaging range of the imaging unit 105. The imaging direction control unit 170 functions as control means. Note that when the imaging unit 105 is movably attached to the head 101, the imaging direction control unit 170 may change the imaging direction by moving the imaging unit 105.

[0034] The flow of processing executed in the robot 100 configured as above will be described with reference to the flowchart shown in FIG.

[0035] The robot control process shown in FIG. 4 starts when the robot 100 is powered on and the battery 108 is charged, thereby putting the robot 100 into a state where it can operate normally.

[0036] When the robot control process is started, first, the control unit 106 functions as the acquisition unit 130 and acquires the latest image continuously captured by the imaging unit 105 (step S10).

[0037] Next, the control unit 106 functions as the face detection unit 140 and executes face detection processing to detect a human face from the central area of ​​the acquired image (step S20).

[0038] If a face is detected in the face detection process (step S30; Yes), the control unit 106 functions as the determination unit 160 and determines whether or not a predetermined condition is met (step S40).

[0039] It should be noted that various conditions can be adopted as the "predetermined condition" in step S40. For example, the control unit 106 may determine that the predetermined condition is met when the control unit 106 is not communicating with a person whose face the control unit 106 has detected.

[0040] Alternatively, the control unit 106 may use a known method to determine whether the face detected from the line of sight or the like is facing the device itself, and may determine that a predetermined condition is met if the face is not facing the device itself.

[0041] Alternatively, the control unit 106 may determine that the predetermined condition is met when the size of the face detected in the face detection process is smaller than a predetermined threshold. The control unit 106 may also determine that the predetermined condition is met when the audio input unit 113 has not continuously received audio from the person with the face for a predetermined period of time.

[0042] Alternatively, the robot 100 may be provided with a built-in function for registering the faces of people who have communicated with the robot 100 at a predetermined frequency or more, or faces set by the user, as "favorites."The control unit 106 may then determine that a predetermined condition is met if a face detected by the face detection process is not registered as a "favorite."

[0043] If it is determined that the predetermined condition is not met (step S40; No), the control unit 106 functions as the imaging direction control unit 170 and controls the driver 109 so that the face is positioned at the center of the imaging range of the imaging unit 105 (step S50). The driver 109 drives the head 101 in accordance with instructions from the control unit 106. This adjusts the orientation of the imaging unit 105 installed on the head 101 so that the detected face is positioned at the center of the imaging range. Note that the control unit 106 may adjust the position or orientation of the imaging unit 105 by driving the foot unit 104. Then, the process returns to step S10.

[0044] On the other hand, if a face cannot be detected in the face detection process (step S30; No), or if a face is detected but it is determined that certain conditions are met (step S30; Yes, step S40; Yes), the control unit 106 functions as a change detection unit 150 and detects whether or not there has been a change over time in the surrounding area of ​​the acquired image (step S60).

[0045] For example, the control unit 106 may detect the presence or absence of a change by comparing the peripheral area of ​​the latest image acquired in this step S10 with the peripheral area of ​​the image acquired in the immediately previous step S10 (the previously acquired image). Note that the control unit 106 may detect a change in the peripheral area by using more images continuously captured by the imaging unit 105.

[0046] If a change is detected in step S60 (step S70; Yes), control unit 106 functions as imaging direction control unit 170 and controls driver 109 so that the part where the detected change occurred is located in the center of the imaging range of imaging unit 105 (step S80). Driver 109 drives head 101 in accordance with instructions from control unit 106. This adjusts the orientation of imaging unit 105 installed on head 101 so that the part where the detected change occurred is located in the center of the imaging range. Then, the process returns to step S10.

[0047] On the other hand, if no change is detected in step S60 (step S70; No), the process returns to step S10.

[0048] Next, the above-mentioned robot control process will be explained using an example.

[0049] As shown in Fig. 5A, consider the case where person Pa is positioned in front of robot 100. In this case, face detection processing is executed in the central region of the image captured by imaging unit 105 (step S20), and the face of person Pa is detected from the image captured at that time, as shown in Fig. 5B (step S30; Yes). Then, it is determined that the face of person Pa is facing the robot, and therefore the predetermined condition is not met (step S40; No), and the imaging direction of imaging unit 105 is adjusted so that the face is positioned in the center of the imaging range (step S50).

[0050] Then, as shown in FIG. 6A, consider the case where another person Pb approaches person Pa from the right side of the drawing, and person Pa turns sideways and averts his or her gaze from the robot.

[0051] 6B, the face of person Pb is located in the peripheral area of ​​the image captured by the imaging unit 105, but not in the central area. Therefore, the face of person Pb is not detected as a target of face detection processing.

[0052] On the other hand, since the face of person Pa currently detected by the face detection process is no longer facing the robot, it is determined that the predetermined condition is met (step S40; Yes), and the presence or absence of a change in the surrounding area is determined (step S60), and the new entry of person Pb into the surrounding area is detected as a change (step S70; Yes). Therefore, as shown in Fig. 7A, the imaging direction of image capture unit 105 is controlled to the direction of person Pb detected as a change (to the right in the drawing) (step S80). As a result, person Pb becomes the center of the image, as shown in Fig. 7B.

[0053] As described above, the robot 100 and the image processing device 120 according to this embodiment perform the face detection process (first detection process), which is relatively heavy, only on the central region of the image, while performing only the process of detecting changes over time (second detection process), which is lighter in load than the face detection process, on regions other than the central region of the image. This makes it possible to perform detection from the entire image while minimizing the load.

[0054] Furthermore, the robot 100 according to this embodiment includes an image capturing unit 105, and when a change over time is detected in the peripheral area of ​​an image, the image capturing direction of the image capturing unit 105 is controlled so that the area where the change occurred is positioned at the center of the image capturing range. This allows the robot 100 to operate in conjunction with any movement that comes within the angle of view. For example, the robot 100 can detect a curtain swaying around it as a change and perform animal-like behavior such as reacting to it.

[0055] Furthermore, while the robot 100 according to this embodiment is detecting a face in the face detection process, it can be controlled to respond to changes in the surrounding area only when a predetermined condition is met, such as the face not facing the robot itself. Therefore, when the robot 100 is having a conversation with a person whose face is detected, it can be prevented from performing an action that disrupts the conversation in response to a change in the surrounding area of ​​the image.

[0056] (Variation) The present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the gist of the present invention.

[0057] For example, in the above embodiment, the robot 100 and the image processing device 120 performed a face detection process to detect a human face, but instead of the face detection process, a detection process to detect a predetermined object other than a human face, such as an animal, may be performed.

[0058] Furthermore, in the robot control process, even while a face is being detected in the face detection process, if a change is detected in the surrounding area, the imaging direction may be unconditionally changed to point to the changed area. In this case, the robot 100 will always be sensitive to changes in its surroundings, allowing it to move in a more animal-like manner. Furthermore, in the robot control process, while a face is being detected in the face detection process, control may be performed to adjust the orientation of the imaging unit 105 so that the face is always at the center of the image.

[0059] Furthermore, in the robot control process, when a change is detected in the peripheral area, the control unit 106 performs a process (step S80) to control the imaging direction so that the changed portion is located at the center of the imaging range of the imaging unit 105. However, this is not limited to this, and various predetermined processes may be performed when a change is detected in the peripheral area. For example, when a change is detected in the peripheral area, the control unit 106 may perform control to widen or narrow the angle of view (imaging range) of the imaging unit 105 (for example, narrow the angle so that the subject in the peripheral area at that time is removed from the imaging range) to control the imaging unit 105 to capture an image, or may crop (cut out) and enlarge the peripheral area (the entire area or only the changed portion) from the captured image, or may crop and enlarge the central area from the captured image and perform predetermined image processing on the captured image. Furthermore, in the robot control process, when a face is not detected in the face detection process or when predetermined conditions are not met, image processing may be performed to crop the central area and delete the peripheral area.

[0060] Furthermore, which areas of the image captured by the imaging unit 105 are to be the central area and the peripheral area can be arbitrarily determined by the user, and may be set appropriately. For example, if the head 101 (imaging unit 105) moves only in the horizontal direction (yaw direction), only the left and right areas of the image may be set as the peripheral areas.

[0061] Furthermore, the "robot" of the present invention is not limited to a robot intended to communicate with a user as shown in Fig. 1. For example, the "robot" of the present invention also includes a security robot, security system, surveillance system, etc. that includes an imaging device such as a surveillance camera, a computer, and a drive mechanism, in which the computer detects a person's face from the central area of ​​an image captured by the imaging device, detects changes over time in the peripheral area of ​​the captured image, and, when a change is detected, controls the drive mechanism to control the imaging direction of the imaging device so that the change is centered in the imaging range.

[0062] It is possible to provide an image processing device or robot that is already equipped with the configuration for realizing the functions of the present invention, and also to make an existing information processing device or the like function as the image processing device or the robot of the present invention by applying a program. That is, by applying a program for realizing each functional configuration of the image processing device 120 or the robot 100 exemplified in the above embodiment so that it can be executed by a CPU or the like that controls an existing information processing device or the like, it can be made to function as the image processing device or the robot of the present invention.

[0063] Furthermore, the application method of such a program is arbitrary. The program can be applied by storing it on a computer-readable storage medium such as a flexible disk, a CD (Compact Disc)-ROM, a DVD (Digital Versatile Disc)-ROM, or a memory card. Furthermore, the program can be superimposed on a carrier wave and applied via a communication medium such as the Internet. For example, the program can be distributed by posting it on a bulletin board system (BBS) on a communication network. Then, the program can be started and executed under the control of an operating system (OS) in the same way as other application programs, thereby enabling the above-mentioned processing to be performed.

[0064] Although the preferred embodiments of the present invention have been described above, the present invention is not limited to the specific embodiments, and the present invention includes the inventions described in the claims and their equivalents. The inventions described in the original claims of this application are appended below.

[0065] (Appendix 1) an acquisition means for acquiring successively captured images; an object detection means for executing a first detection process for detecting a predetermined object from a central region of the acquired image; a change detection means for detecting changes over time in regions other than the central region of the acquired image by executing a second detection process that has a smaller load than the first detection process; An image processing device comprising:

[0066] (Appendix 2) the change detection means detects a change over time in the other region based on feature points in the other region by executing the second detection process. 2. The image processing device according to claim 1,

[0067] (Appendix 3) the change detection means detects a change over time in the other region based on a difference in the other region between the successively captured images by executing the second detection process. 2. The image processing device according to claim 1,

[0068] (Appendix 4) The predetermined object is a person's face. 4. The image processing device according to any one of claims 1 to 3.

[0069] (Appendix 5) the acquiring means acquires images continuously captured by the imaging means, The apparatus further includes a control unit that executes a predetermined process in response to at least one of the detection result by the object detection unit and the detection result by the change detection unit. 5. The image processing device according to any one of claims 1 to 4.

[0070] (Appendix 6) an image processing device according to Supplementary Note 5; the imaging means; A robot equipped with:

[0071] (Appendix 7) the imaging means is configured to be able to capture the image acquired by the acquisition means within a predetermined imaging range, when the change detection means detects a change over time in the other region, the control means controls the imaging direction of the imaging means so that the part where the change has occurred is positioned in the center of the imaging range. 7. The robot according to claim 6,

[0072] (Appendix 8) Further comprising a determination means for determining whether a predetermined condition is met, the control means is capable of controlling the imaging direction only when it is determined by the determination means that the predetermined condition is met while the predetermined object is being detected in the first detection process. 8. The robot according to claim 7,

[0073] (Appendix 9) The device is capable of communication, the determining means determines that the predetermined condition is met when the device itself is not communicating with an object corresponding to the predetermined object; 9. The robot according to claim 8,

[0074] (Appendix 10) the determining means determines that the predetermined condition is met when the predetermined target is not facing the device itself; 9. The robot according to claim 8,

[0075] (Appendix 11) The determination means determines that the predetermined condition is met when the size of the predetermined object is equal to or smaller than a predetermined threshold. 9. The robot according to claim 8,

[0076] (Appendix 12) Further comprising a voice acquisition means, the determining means determines that the predetermined condition is met when the voice acquisition means has not continuously acquired voice from the predetermined target for a predetermined period of time; 9. The robot according to claim 8,

[0077] (Appendix 13) execute a first detection process for detecting a predetermined object from a central region of the captured image; detecting changes over time in regions other than the central region of the image by executing a second detection process that has a smaller load than the first detection process; Image processing methods.

[0078] (Appendix 14) Computer, an acquisition means for acquiring successively captured images; an object detection means for executing a first detection process for detecting a predetermined object from a central region of the acquired image; a change detection means for detecting changes over time in regions other than the central region of the acquired image by executing a second detection process that has a smaller load than the first detection process; A program that functions as a [Explanation of symbols]

[0079] 100...robot, 101...head, 102...torso, 103...hand, 104...foot, 105...imaging unit, 106...control unit, 107...memory unit, 108...battery, 109...driver, 110...moving unit, 111...wireless communication unit, 112...display unit, 113...audio input unit, 114...audio output unit, 120...image processing device, 130...acquisition unit, 140...face detection unit, 150...change detection unit, 160...discrimination unit, 170...imaging direction control unit

Claims

1. A robot in which a head direction is set to a direction in which an image is captured by an image capturing unit provided on the head, an image processing means for dividing each of the images successively acquired by imaging with the imaging unit into a central region including the center of the image and a peripheral region including the periphery of the central region, and for starting a second detection process for detecting an image change between images for the peripheral region instead of the first detection process at least when a first detection process for detecting a person's face in the central region has not detected the person's face and when the first detection process for detecting the person's face in the central region has detected the person's face and detected that the person's face has turned sideways from a direction facing the robot, and for specifying a position where the image change between images is detected by the second detection process as a first position; a drive control means for changing an orientation of the head when the first position is specified by the image processing means so that the first position is located at the center of an image captured by the imaging unit; A robot comprising:

2. the image processing means, when the face of the person is detected by the first detection process and when the size of the detected face of the person is larger than a predetermined threshold value or when the detected face of the person is registered in advance as a favorite, specifies the position of the detected face of the person as a second position; When the second position is identified by the image processing means, the drive control means changes the orientation of the head so that the second position is located at the center of an image.

2. The robot according to claim 1 .

3. the image processing means divides the image so that the area of ​​the central region is smaller than the area of ​​the peripheral region.

3. The robot according to claim 1 or 2.

4. A user can set how to divide an image acquired by imaging using the imaging unit into the central region and the peripheral region.

3. The robot according to claim 1 or 2.

5. A direction control method executed by a robot in which a head direction is set to a direction of imaging by an imaging unit provided on the head, comprising: an image processing step of dividing each of the images continuously acquired by the imaging unit into a central region including the center of the image and a peripheral region including the periphery of the central region, and starting a second detection process for detecting an image change between images for the peripheral region instead of the first detection process at least when the face of a person is not detected by a first detection process for detecting a face of a person in the central region and when the face of the person is detected by the first detection process for the central region and it is detected that the face of the person has turned sideways from a direction facing the robot, and specifying a position where the image change is detected as a first position when an image change between images is detected by the second detection process; a drive control step of changing an orientation of the head when the first position is specified by the image processing step so that the first position is located at the center of an image acquired by imaging by the imaging unit; A directional control method comprising:

6. A robot computer having a head oriented in a direction in which an image is captured by an image capturing unit provided on the head, an image processing means for dividing each of images continuously acquired by imaging by the imaging unit into a central region including the center of the image and a peripheral region including the periphery of the central region, and for starting a second detection process for detecting an image change between images for the peripheral region instead of the first detection process at least when the face of a person is not detected by a first detection process for detecting a face of a person in the central region and when the face of the person is detected by the first detection process for the central region and it is detected that the face of the person has turned sideways from a direction facing the robot, and for specifying a position where the image change between images is detected by the second detection process as a first position; a drive control means for changing an orientation of the head when the first position is specified by the image processing means so that the first position is located at the center of an image acquired by imaging by the imaging unit; A program characterized by causing the program to function as a

Images