JP2016066280A - Image processor, person detection method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processor, a person detection method, and a program capable of determining whether or not the user of equipment has been replaced with the other user from the movement locus of a person.SOLUTION: An image processor includes: person detection means for detecting a person positioned in a predetermined range; discrimination means for deriving the movement locus of the person detected by the person detection means from the past position to the current position in the predetermined range, and for discriminating one or more persons positioned in the predetermined range on the basis of the movement locus; and determination means for determining whether or not the person who operates the image processor has been replaced with the other person on the basis of the information related to the person discriminated by the discrimination means.SELECTED DRAWING: Figure 2

Description

  The present invention relates to an image processing apparatus, a person detection method, and a program.

  For example, in a device such as an image forming apparatus, in order to shorten the return time from the power saving mode, a person detection sensor is used to detect the approach of the user of the device and the user starts operating the device. A technique for returning from the power saving mode is known (see, for example, Patent Document 1).

  However, in the above prior art, there are cases where the user of the device cannot be correctly determined when there are a plurality of persons around the device. For example, when a plurality of persons change devices many times and use a device, it may be erroneously determined that one person continues to use the device. Further, for example, when another person passes near a person who is using the device, it may be erroneously determined that the user of the device has been switched. Therefore, there are cases where power saving control, automatic logoff control, etc. of the device using the person detection sensor cannot be performed correctly.

  The embodiment of the present invention has been made in view of the above points, and an object of the present invention is to determine whether or not a user of a device has been replaced by another user based on a person's movement trajectory.

  In order to achieve the above object, an embodiment of the present invention is an image processing apparatus, comprising: a person detection unit that detects a person located within a predetermined range; and the predetermined person detected by the person detection unit. A discriminating means for deriving a movement trajectory from a past position to a current position within a range and distinguishing one or more persons located within the predetermined range based on the movement trajectory, and the discrimination means And determining means for determining whether or not the person who operates the image processing apparatus has been replaced with another person based on the information related to the person.

  According to the embodiment of the present invention, it is possible to determine whether or not the user of the device has been replaced with another user from the movement trajectory of the person.

FIG. 2 is a hardware configuration diagram of an example of an MFP according to the present embodiment. 2 is a functional block diagram of an example of an MFP according to the present embodiment. FIG. FIG. 5 is a diagram for explaining an example of an outline of processing of an MFP according to the present embodiment. 6 is a flowchart of an example of control processing of the MFP according to the present embodiment. It is a figure for demonstrating an example of the tracking using a particle filter. It is a figure for demonstrating an example of the tracking using image data. It is a flowchart of an example of the tracking process which concerns on this embodiment. It is a flowchart of the other example of the tracking process which concerns on this embodiment.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the present embodiment, an MFP (Multi Function Peripheral) will be described as an example of an image processing apparatus. However, the present invention is not limited to this, and may be various electronic devices such as a projector, an electronic blackboard, and a game device. That is, the embodiment of the present invention can be applied to an electronic device that performs various controls using a human body detection sensor.

<Hardware configuration>
First, a hardware configuration of the MFP 10 according to the present embodiment will be described. The MFP 10 according to the present embodiment is realized by, for example, a hardware configuration as illustrated in FIG. FIG. 1 is a hardware configuration diagram of an example of an MFP according to the present embodiment. As shown in FIG. 1, the MFP 10 according to this embodiment includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, an HDD (Hard Disk Drive) 14, a communication I / O. F15, an operation panel 16, a human body detection sensor 17, an image processing engine 18, and the like are connected to each other via a bus B.

  The CPU 11 is an arithmetic device that realizes control and functions of the entire MFP 10 by reading a program and data from a storage device such as the ROM 12 and the HDD 14 onto the RAM 13 and executing processing.

  The ROM 12 is a nonvolatile semiconductor memory (storage device) that can retain programs and data even when the power is turned off. The RAM 13 is a volatile semiconductor memory (storage device) that temporarily stores programs and data.

  The HDD 14 is a non-volatile storage device that stores programs and data. The stored programs and data include, for example, a program for realizing the present embodiment, an OS (Operating System) that is basic software for controlling the entire MFP 10, and application software that provides various functions on the OS. The HDD 14 manages stored programs and data by a predetermined file system and / or DB (database). The MFP 10 may include an SSD (Solid State Drive) or the like instead of the HDD or in combination with the HDD.

  The communication I / F 15 is an interface for connecting to a network via, for example, a LAN. Accordingly, the MFP 10 can perform data communication with other devices via the communication I / F 15.

  The operation panel 16 is a touch panel including an input unit that receives input from a user and a display unit that performs display. The user can perform an operation for inputting various information to the MFP 10 via the operation panel 16.

  The human body detection sensor 17 is a sensor that detects a person (person) around the MFP 10 (for example, a predetermined range in the front direction of the operation panel 16 of the MFP 10). The human body detection sensor 17 is, for example, a sensor that detects a person using infrared rays, ultrasonic waves, visible light, or the like, or a sensor that detects a person using image data generated by imaging with an imaging unit such as a camera. is there. However, the human body detection sensor 17 is not limited to these, and may be various sensors that detect a person using various methods. In the human body detection sensor 17, a range in which a person can be detected is determined by the type or setting of the human body detection sensor 17, and such a predetermined range is hereinafter referred to as a “sensor detection range”.

  Further, the MFP 10 may include two or more human body detection sensors 17. Since the MFP 10 has two or more human body detection sensors 17, for example, even if a person is present at a blind spot in one human body detection sensor 17, if the position can be detected by another human body detection sensor 17, The presence of a person can be detected. That is, it is possible to detect a person while mutually complementing information acquired by each of the plurality of human body detection sensors 17. When the MFP 10 includes two or more human body detection sensors 17 as described above, the human body detection unit 101 described later may include a human body detection unit corresponding to each human body detection sensor 17. For example, when the MFP 10 has two human body detection sensors 17, a first person detection unit that detects a person in the sensor detection range (first predetermined range) of one human body detection sensor 17 and another human body detection sensor 17. And a second person detecting means for detecting a person within the sensor detection range (second predetermined range).

  The image processing engine 18 performs various image processing functions that the MFP 10 has, such as a print function, a copy function, and a scan function.

  The MFP 10 according to the present embodiment can realize various processes as described later with the above-described hardware configuration.

<Software configuration>
Next, a software configuration of the MFP 10 according to the present embodiment will be described. The MFP 10 according to the present embodiment is realized by, for example, functional blocks as illustrated in FIG. FIG. 2 is a functional block diagram of an example of an MFP according to this embodiment. As shown in FIG. 2, the MFP 10 according to the present embodiment includes a human body detection unit 101, a human body detection information control unit 102, a movement locus specifying unit 103, a person discrimination unit 104, a user determination unit 105, an energy saving control unit 106, an automatic operation. A logout control unit 107 and the like. The MFP 10 according to the present embodiment uses the person-specific detection information storage unit 108.

  The human body detection unit 101 is realized by, for example, the CPU 11, the human body detection sensor 17, and the like, and detects a person within the sensor detection range. Information on the person detected by the human body detection unit 101 is passed to the human body detection information control unit 102 as human body detection information. Here, the human body detection information is information on a person within the sensor detection range detected by the human body detection sensor 17, and is, for example, position information on the person within the sensor detection range at a certain point in time. Further, the human body detection information may be image data generated by imaging the sensor detection range with an imaging unit such as a camera, or position information of a person calculated from the image data.

  The human body detection information control unit 102 is realized by the CPU 11 or the like, for example, and requests other units to perform processing based on the human body detection information received from the human body detection unit 101. For example, when the human body detection information control unit 102 receives the human body detection information from the human body detection unit 101, the human body detection information control unit 102 requests the movement locus specifying unit 103 to perform processing in order to track the detected person. In addition, the human body detection information control unit 102 requests the person discrimination unit 104 to perform processing so as to discriminate the human body detection information tracked by the movement trajectory identification unit 103 for each person included in the human body detection information.

  The movement locus specifying unit 103 is realized by, for example, the CPU 11 or the like, and tracks the movement locus of a person detected within the sensor detection range based on the human body detection information received from the human body detection information control unit 102. Thereby, in the MFP 10 according to the present embodiment, the movement locus when the detected person moves is tracked within the sensor detection range (that is, information on the movement locus is derived). Note that the information on the movement trajectory is information on the trajectory when a person detected within the sensor detection range moves (including when stationary) within the sensor detection range (that is, as described later, Position information at each time point).

  The person distinguishing unit 104 is realized by the CPU 11 or the like, for example, and distinguishes the person whose movement locus is tracked by the movement locus specifying unit 103 into individual information, and creates or updates the individual detection information. That is, for example, when position information of a plurality of persons is included in the human body detection information, detection information for each person is created from information regarding the movement trajectory for each person. Here, the detection information for each person is information (tracking information) regarding the movement trajectory of each person detected within the sensor detection range, for example, position information at each time point of a certain person. As described above, in the MFP 10 according to the present embodiment, tracking information (person-specific detection information) for each person detected within the sensor detection range is created. Therefore, the MFP 10 can identify each person even when there are a plurality of persons within the sensor detection range by using the person-specific detection information. In other words, the MFP 10 detects each person in the sensor detection range according to the person-specific detection information even when a plurality of detected persons pass each other or when a detected person passes behind the shield. Recognize people without making mistakes.

  For example, the user determination unit 105 is realized by the CPU 11 or the like, and whether or not a person who is separated from the sensor detection range (goes out of the sensor detection range) based on the person-specific detection information is a user of the MFP 10. Determine. As a result, the MFP 10 according to the present embodiment can identify each user as a different user even when a plurality of people change and use the MFP 10. Therefore, the MFP 10 according to the present embodiment is used by the automatic logout control unit 107 even when the next user uses the MFP 10 soon after the previous user finishes using the MFP 10. Automatic logout is performed.

  The energy saving control unit 106 is realized by, for example, the CPU 11 and controls the MFP 10 to shift to a power saving state (for example, turning off the backlight of the operation panel 16). That is, for example, when the user of the MFP 10 ends the use (this is recognized by determining that the user of the MFP 10 has moved out of the sensor detection range), a predetermined time has elapsed. In this case, the energy saving control unit 106 shifts the MFP 10 to the power saving state.

  The automatic logout control unit 107 is realized by, for example, the CPU 11 or the like, and performs control to automatically log out the user of the MFP 10 from the MFP 10. That is, in the same manner as the energy saving control unit 106 described above, for example, when the user of the MFP 10 ends the use, when a predetermined time has elapsed, the automatic logout control unit 107 determines the user of the MFP 10. Then, the MFP 10 is automatically logged out.

  The person detection information storage unit 108 is realized by, for example, a storage device connected to the HDD 14 or the MFP 10 via a network, and stores person detection information. As described above, the person-specific detection information is information (tracking information) related to the person-specific movement locus detected within the sensor detection range.

<Outline of processing>
Next, an outline of processing of the MFP 10 according to the present embodiment will be described. FIG. 3 is a diagram for explaining an example of an outline of processing of the MFP according to the present embodiment. 3, when the user of the MFP 10 is switched from the person A to the person B (FIG. 3A), and when the person B passes by the person A using the MFP 10 (FIG. 3B). Will be described.

  Conventionally, in the case of FIG. 3A, it may be erroneously determined that the user has not been switched, and in the case of FIG. 3B, it may be erroneously determined that the user has been switched. It was.

  First, a case where the user of the MFP 10 is switched from person A to person B will be described with reference to FIG.

  In step S1-1, it is assumed that the person A approaches the MFP 10 in order to use the MFP 10. At this time, when the person A enters the sensor detection range of the MFP 10, the human body detection sensor 17 detects the person A. Then, the MFP 10 tracks the movement locus of the person A. At this time, the number of people in the sensor detection range is one.

  In step S1-2, the person A is using the MFP 10.

  In step S <b> 1-3, it is assumed that the person B approaches the MFP 10 in order to use the MFP 10 at almost the same timing as the person A finishes using the MFP 10 and starts to leave the MFP 10. At this time, when the person B enters the sensor detection range of the MFP 10, the human body detection sensor 17 detects the person B and tracks the movement locus of the person B. Accordingly, the MFP 10 can recognize that the person A and the person B are different persons without erroneously recognizing that the person A and the person B are the same person. At this time, the number of people in the sensor detection range is two.

  In step S1-4, it is assumed that the person B is using the MFP 10 while the person A is out of the sensor detection range. At this time, since the MFP 10 according to the present embodiment tracks the movement trajectories of the person A and the person B, the person A goes out of the sensor detection range without erroneously recognizing the person A and the person B. On the other hand, it can be recognized that the user of the MFP 10 has changed from the person A to the person B. Therefore, when the person A finishes using the MFP 10 and goes out of the sensor detection range, the MFP 10 can perform energy saving control, automatic logout control of the user account of the person A, and the like in a predetermined case. Accordingly, the MFP 10 according to the present embodiment can suppress wasteful power consumption and the like, and can prevent unauthorized use of the user account by other users. At this time, the number of people in the sensor detection range is one. Here, in the above, the user account is information including a user name (or user ID) assigned in advance for the user to use the MFP 10. A user (user) can use the MFP 10 by logging in using a user account assigned to the user at the start of use of the MFP 10. On the other hand, the user (user) terminates the use of the MFP 10 by logging out the user account assigned to the user (user) from the MFP 10 when the use of the MFP 10 is terminated. In other words, login and logout are to grant and revoke the authority to use the MFP 10 to the logged-in and logged-out user account, respectively.

  Next, the case where the person B passes near the person A who uses the MFP 10 will be described with reference to FIG.

  In step S2-1, it is assumed that the person A approaches the MFP 10 in order to use the MFP 10. At this time, when the person A enters the sensor detection range of the MFP 10, the human body detection sensor 17 detects the person A. Then, the MFP 10 tracks the movement locus of the person A. At this time, the number of people in the sensor detection range is one.

  In step S2-2, the person A is using the MFP 10.

  In step S2-3, it is assumed that the person A passes near the MFP 10 (within the sensor detection range) while the person A is using the MFP 10. At this time, when the person B enters the sensor detection range of the MFP 10, the human body detection sensor 17 detects the person B and tracks the movement locus of the person B. Accordingly, the MFP 10 can recognize that the person A and the person B are different persons without erroneously recognizing that the person A and the person B are the same person. At this time, the number of people in the sensor detection range is two.

  In step S2-4, it is assumed that the person A is using the MFP 10 while the person B is out of the sensor detection range. At this time, since the MFP 10 according to the present embodiment tracks the movement trajectories of the person A and the person B, the person B goes out of the sensor detection range without erroneously recognizing the person A and the person B. On the other hand, the user of the MFP 10 can continue to recognize that he is the person A. Therefore, even when the person B passes near the person A who is using the MFP 10, the MFP 10 does not perform erroneous control such as energy saving control or automatic logout control. At this time, the number of people in the sensor detection range is one.

  As described above, the MFP 10 according to the present embodiment can distinguish and recognize each person located within the sensor detection range by tracking the movement trajectory of the person detected by the human body detection sensor 17. Therefore, the MFP 10 according to the present embodiment can prevent erroneous control due to erroneous recognition of a person.

<Details of processing>
Next, details of processing of the MFP 10 according to the present embodiment will be described.

<< Control Process of MFP 10 >>
FIG. 4 is a flowchart of an example of control processing of the MFP according to the present embodiment. First, it is assumed that a person A who desires to use the MFP 10 approaches the MFP 10 to use the MFP 10 and enters the sensor detection range.

  In step S401, the human body detection unit 101 detects a person within the sensor detection range. That is, the human body detection unit 101 detects the person A within the sensor detection range of the MFP 10. When the human body detection unit 101 detects a person within the sensor detection range, the human body detection unit 101 transmits human body detection information to the human body detection information control unit 102. As described above, the human body detection information is, for example, the position information of a person within the sensor detection range detected by the human body detection unit 101, or an image generated by imaging the sensor detection range with an imaging unit such as a camera. Data or position information of a person calculated from the image data.

  In step S <b> 402, the human body detection information control unit 102 performs tracking of the person detected by the human body detection unit 101 based on the human body detection information received from the human body detection unit 101, and stores the individual detection information as the person detection information storage unit. The data is stored in 108. Tracking processing for tracking a person detected by the human body detection unit 101 includes, for example, a case where a particle filter is used and a case where image data is used. In the following, an outline of the tracking method using the particle filter and the tracking method using the image data when the user of the MFP 10 is switched from the person A to the person B as shown in FIG.

  FIG. 5 is a diagram for explaining an example of tracking using a particle filter. When the human body detection information is position information of a person within the sensor detection range, tracking can be performed by predicting a change in the position information of the person using a particle filter. According to this method, when the person A enters the sensor detection range in order to use the MFP 10, the movement trajectory of the person A is tracked using the particle filter (step S3-1). It is assumed that the person B uses the MFP 10 (step S3-2), and the person B approaches the MFP 10 in order to use the MFP 10 at almost the same timing as the person A finishes the use and starts to leave the MFP 10. At this time, when the person B enters the sensor detection range of the MFP 10, the human body detection sensor 17 detects the person B and tracks the movement trajectory of the person B using the particle filter (step S3-3). It is assumed that the person B is using the MFP 10 while the person A has moved out of the sensor detection range. At this time, as described above, since the movement trajectories of the person A and the person B are tracked using the particle filter, the MFP 10 can recognize that the user has changed from the person A to the person B (step S3-4). Details of the tracking process using the particle filter as described above will be described later.

  FIG. 6 is a diagram for explaining an example of tracking using image data. When the human body detection information is information related to the feature points of the person image within the sensor detection range, tracking can be performed by matching using the feature points of the person image. According to this method, when person A enters the sensor detection range in order to use MFP 10, the movement trajectory of person A is tracked by matching using the feature points of the image of person A (step S4-1). . It is assumed that the person A uses the MFP 10 (step S4-2), and the person B approaches the MFP 10 in order to use the MFP 10 at almost the same timing as the person A finishes the use and starts to leave the MFP 10. At this time, when the person B enters the sensor detection range of the MFP 10, the human body detection sensor 17 detects the person B and performs matching using the feature points of the image of the person B, thereby tracking the movement trajectory of the person B. (Step S4-3). It is assumed that the person B is using the MFP 10 while the person A has moved out of the sensor detection range. At this time, as described above, since the movement trajectories of person A and person B are tracked by matching using the feature points of each person, MFP 10 recognizes that the user has changed from person A to person B. (Step S4-4). Details of the tracking process using the image data as described above will be described later.

  In step S <b> 403, the person detected by the human body detection unit 101 uses the MFP 10. For example, the person A detected by the human body detection unit 101 uses the MFP 10.

  In step S404, the human body detection information control unit 102 determines whether the number of persons within the sensor detection range has increased or decreased. This can be determined based on the individual detection information generated by tracking the movement trajectory of the person within the sensor detection range in step S402 described above. For example, when the person-specific detection information is position information at each time point for each person, the human body detection information control unit 102 determines by calculating the number of person-specific detection information in which the current position information is recorded. be able to. When the number of persons within the sensor detection range has increased (that is, when a person has been newly detected by the human body detection unit 101), the process proceeds to step S405. On the other hand, if the number of people within the sensor detection range has decreased (that is, if a person has moved from within the sensor detection range to outside the sensor detection range), the process proceeds to step S406. In addition, when there is no increase / decrease in the number of persons within the sensor detection range, the human body detection information control unit 102 does nothing.

  In step S <b> 405, the human body detection information control unit 102 performs tracking of the person newly detected by the human body detection unit 101, and stores the detected person-by-person detection information in the person-by-person detection information storage unit 108. The process for performing this tracking is the same as the process described in step S402 described above.

  In step S406, the human body detection information control unit 102 determines whether or not the number of persons within the sensor detection range is one or more. This determination is made in the same manner as in step S404 described above, for example, when the person-specific detection information is position information at each time point for each person, the human body detection information control unit 102 records the person whose current position information is recorded. This can be determined by calculating the number of separate detection information. When the number of persons within the sensor detection range is one or more, the process proceeds to step S407. On the other hand, when the number of persons within the sensor detection range is zero, the process proceeds to step S413.

  In step S <b> 407, the user determination unit 105 determines whether the person who is outside the sensor detection range is a user of the MFP 10 (for example, the person A described above). This can be determined from, for example, person-by-person detection information of person A. That is, for example, when the current position information is not recorded in the person-by-person detection information of the person A who is the user of the MFP 10, the user determination unit 105 determines that the person A is out of the sensor detection range. The On the other hand, for example, when the current position information is recorded in the person-by-person detection information of the person A, the person who has moved outside the sensor detection range is determined to be other than the person A. If the person outside the sensor detection range is a user of the MFP 10, the process proceeds to step S409. On the other hand, if the person outside the sensor detection range is not the user of the MFP 10, the process proceeds to step S410.

  In step S408, the MFP 10 operates on the assumption that a specific user (for example, the person A described above) is continuously using the MFP 10. That is, in this case, the MFP 10 does not perform various controls such as automatic logout control and energy saving control.

  In step S409, the automatic logout control unit 107 determines whether the automatic logout timer is up. That is, for example, it is determined whether or not a predetermined time has elapsed since the last operation on the MFP 10 by the person A who moved out of the sensor detection range. If the automatic logout timer is up, the process proceeds to step S410. On the other hand, if the automatic logout timer is not up, the process proceeds to step S411.

  In step S <b> 410, the automatic logout control unit 107 performs control to automatically log out a user of the MFP 10 (for example, the person A described above) from the MFP 10. Accordingly, when the person A who is a user of the MFP 10 goes out of the sensor detection range without logging out after using the MFP 10, the user account is automatically logged out if a predetermined time has elapsed after the operation is completed.

  In step S411, the energy saving control unit 106 determines whether or not the energy saving transition timer is up. That is, for example, it is determined whether or not a predetermined time has elapsed since the last operation on the MFP 10 by the person A who moved out of the sensor detection range. If the energy saving transition timer is up, the process proceeds to step S412. On the other hand, when the energy saving transition timer is not up, the process returns to step S404.

  In step S412, the energy saving control unit 106 performs control to shift the MFP 10 to the power saving state. Thereby, after the operation of the person A who is the user of the MFP 10 is completed, the MFP 10 shifts to an energy saving state if a predetermined time has elapsed.

  As described above, the process of performing the automatic logout control and the energy saving control when the user of the MFP 10 goes out of the sensor detection range has been described. However, the present invention is not limited thereto, and the MFP 10 performs various other controls. Also good. For example, when the user of the MFP 10 goes out of the sensor detection range, a system reset control unit (not shown) of the MFP 10 causes the operation panel 16 to display an initial screen or a predetermined function set in advance. Control for transitioning to a screen (that is, resetting the system to a predetermined state) may be performed.

  Steps S413 to S416 are the same processes as steps S409 to S412 and will not be described.

  As described above, the MFP 10 according to the present embodiment distinguishes and recognizes each person by tracking the person detected within the sensor detection range. As a result, it is possible to recognize whether the user of the MFP 10 has been replaced with another user, whether the user of the MFP 10 has gone out of the sensor detection range, and the like. Therefore, the MFP 10 according to the present embodiment can appropriately perform various controls such as automatic logout control and energy saving control.

≪Tracking process using particle filter≫
Next, tracking processing using a particle filter will be described as an example of the tracking processing in steps S402 and S405. FIG. 7 is a flowchart of an example of tracking processing according to the present embodiment. The tracking process is not limited to using a particle filter, and for example, a Kalman filter may be used. The Kalman filter is based on the current observation value and the state estimation value of the previous time point (for example, the previous frame), and the current state estimation value and the next time point (for example, the next time point). Frame) state estimation value. When the Kalman filter is used, although the estimation (prediction) accuracy is lower than that of the particle filter, the calculation amount can be suppressed. Since tracking processing of a person or an object using a particle filter or a Kalman filter is a known method, an outline will be described below.

  In step S <b> 701, upon receiving the human body detection information including the position information of the person from the human body detection information control unit 102, the movement locus specifying unit 103 acquires the position information of the person as the current observation value (current observation state). Then, the movement trajectory specifying unit 103 sets a position (positional coordinate) in the vicinity of the current observation state as a particle. In other words, the movement locus specifying unit 103 places particles at positions near the current observation state. In the initial state, the movement trajectory specifying unit 103 randomly throws particles in the vicinity of the current observation value. On the other hand, in other cases than the initial state, the movement trajectory specifying unit 103 whisks particles according to the weight calculated in step S705. cure).

  In addition, the movement locus | trajectory specific | specification part 103 is one position information, when the human body detection information containing the position of a several person is received from the human body detection information control part 102 (that is, when there exist a several person in a sensor detection range). And the acquired position information is made the current observation state of the person. In this case, step S701 to step S705 are repeated for each position information of the person included in the human body detection information.

  In step S702, the movement locus specifying unit 103 determines whether there is a past estimated value (past estimated state). If there is a past estimated state, the process proceeds to step S703. If there is no past estimated state, the process returns to step S701. Here, the past estimated state is the “estimated state at the next time point” calculated in the process of step S705 at all the previous time points according to the current time point.

  In step S703, the movement trajectory specifying unit 103 calculates the current estimated state based on the past estimated state. That is, the movement trajectory specifying unit 103 moves each particle based on a predetermined state method.

  In step S704, the movement trajectory specifying unit 103 calculates the likelihood of each particle using a predetermined calculation formula based on the current estimated state. The likelihood calculated here becomes the “weight” of each particle.

  In step S705, the movement trajectory specifying unit 103 acquires the weighted average of each particle as an estimated state at the next time point (estimated value at the next time point). In addition, it is good also considering the particle with the largest weight in each particle as an estimated state of the next time.

  By repeating the above steps S701 to S705 for each time point, the movement trajectory of the person can be tracked. Then, the person distinguishing unit 104 creates or updates the individual detection information for each person at each time point, and stores it in the individual detection information storage unit 108. As a result, the person-specific detection information storage unit 108 stores person-specific detection information indicating information on the movement trajectory of each person detected by the human body detection unit 101.

≪Tracking process using image data≫
Next, tracking processing using image data will be described as another example of the tracking processing in steps S402 and S405. FIG. 8 is a flowchart of another example of the tracking process according to the present embodiment. The case where image data is used is a case where imaging means such as a camera is used as the human body detection sensor 17. In the following, processing for tracking by matching using feature points of a person's image based on image data will be described. It is assumed that the image data includes an image of one or more persons. That is, the human body detection sensor 17 captures an image only when a person is present within the sensor detection range, or always captures an image of the human body detection information when a person is captured based on a feature point or the like when capturing an image within the sensor detection range. Is transmitted to the human body detection information control unit 102. However, the image data captured by the human body detection sensor 17 may include only a part of a person (for example, a face). Since tracking processing of a person or an object by matching feature points using image data is a known method, an outline will be described below.

  Here, when the human body detection sensor 17 is a camera or the like, whether or not a person is reflected in the image data generated by imaging is determined by the space around the MFP 10 created in advance by the map creating means (not shown) of the MFP 10. It can be determined by using image data (map) generated by imaging. That is, the map creation means (not shown) of the MFP 10 captures the surroundings of the MFP 10 (sensor detection range) in the absence of a person in advance and generates image data (this is represented as “first map”). As a result, the human body detection unit 101 of the MFP 10 obtains a difference between the image data generated by imaging at each time point (this is expressed as “second map”) and the first map, so that each time point is obtained. It is possible to detect that a person is reflected in the image data. That is, a portion where there is a difference between the first map and the second map or a portion where the difference is equal to or greater than a predetermined value is a portion where a person is shown.

  In step S <b> 801, upon receiving the human body detection information including the image data from the human body detection information control unit 102, the movement trajectory specifying unit 103 acquires the feature points of the person image included in the image data. Note that when the image data includes a plurality of persons, the feature points of the image of one person are acquired. That is, when a plurality of persons are included in the image data, for each person, the feature points of the person's image may be acquired for each person, and the processes in steps S801 to S805 may be executed. Here, the feature point may be, for example, the outline (edge) or texture of a person image.

  In step S802, the movement trajectory specifying unit 103 determines the feature point of the person at the current time (ie, the current frame) and the one person included in the image data at the previous time (ie, the previous frame). Match feature points.

  In step S803, the movement trajectory specifying unit 103 determines whether the person included in the current image data and the one person included in the previous time point are the same person based on the matching result in step S802. Determine. For example, if the feature point of the person included in the current image data and the feature point of one person included in the previous time point match at least a predetermined threshold, they are the same person It may be determined. If it is determined that they are the same person based on the matching result, the process proceeds to step S804. If it is not determined that they are the same person, the process proceeds to step S805.

  In step S804, the person distinguishing unit 104 creates or updates the person-by-person detection information of the person determined to be the same person. As a result, the person-by-person detection information indicating the movement trajectory of the person from the previous time point to the present time is stored in the person-by-person detection information storage unit 108.

  In step S805, the movement trajectory specifying unit 103 determines whether or not the image data at the previous time includes another person other than the one who acquired the feature point in step S802. If another person is included in the previous image data, the process returns to step S802. If no other person is included, the process proceeds to step S806. Note that when the process of step S804 is performed, the process of this step may not be performed.

  By repeating the above steps S801 to S805 for each time point, the movement trajectory of the person can be tracked. That is, the movement trajectory of the person at the present time is tracked from the image data (first image data) at each past time point and the current image data (second image data). As a result, the person-specific detection information storage unit 108 stores person-specific detection information indicating information on the movement trajectory of each person detected by the human body detection unit 101.

<Summary>
As described above, the MFP 10 according to this embodiment recognizes each person by tracking the movement trajectory of the person within the sensor detection range of the human body detection sensor 17. Therefore, when the user of the MFP 10 is replaced with another person or when another person passes near the user of the MFP 10, various controls are performed without misrecognizing the user and the other person. be able to.

  The human body detection unit 101 is an example of a person detection unit. The movement trajectory specifying unit 103 and the person distinguishing unit 104 are an example of a distinguishing unit. The user determination unit 105 is an example of a determination unit. The automatic logout control unit 107 is an example of logout control means. The energy saving control unit 106 is an example of an energy saving control unit.

  The present invention is not limited to the specifically disclosed embodiments, and various modifications and changes can be made without departing from the scope of the claims.

10 MFP
11 CPU
12 ROM
13 RAM
14 HDD
15 Communication I / F
Reference Signs List 16 Operation Panel 17 Human Body Detection Sensor 18 Image Processing Engine 101 Human Body Detection Unit 102 Human Body Detection Information Control Unit 103 Movement Trajectory Identification Unit 104 Person Discrimination Unit 105 User Determination Unit 106 Energy Saving Control Unit 107 Automatic Logout Control Unit 108 Person-Specific Detection Information Storage Part

JP 2002-71833 A

Claims (9)

An image processing apparatus,
Person detecting means for detecting a person located within a predetermined range;
A movement trajectory of the person detected by the person detection means from a past position within the predetermined range to a current position is derived, and one or more persons located within the predetermined range are respectively determined based on the movement trajectory. Distinction means to distinguish,
Determination means for determining whether or not a person operating the image processing apparatus has been replaced with another person based on information on the person distinguished by the distinguishing means;
An image processing apparatus.   When it is determined by the determination means that the person operating the image processing apparatus has been replaced with another person, when a predetermined time has elapsed since the person finished the operation, the user account of the person is The image processing apparatus according to claim 1, further comprising logout control means for logging out from the image processing apparatus.   If it is determined by the determination means that the person operating the image processing apparatus has been replaced by another person, the image processing apparatus is omitted if a predetermined time has passed since the person finished the operation. The image processing apparatus according to claim 1, further comprising an energy saving control unit configured to perform control for shifting to the power state.   When it is determined by the determination means that the person operating the image processing apparatus has been replaced with another person, when a predetermined time has passed since the person finished the operation, the display unit of the image processing apparatus 4. The image processing apparatus according to claim 1, further comprising: a system reset control unit configured to perform control for transitioning to a screen for executing a predetermined function set in advance. 5. Map creating means for creating a map indicating information on the space within the predetermined range;
The person detecting means is
Based on the difference between the first map created in advance with no person in the predetermined range by the map creating means and the second map created with a person in the predetermined range, The image processing apparatus according to claim 1, wherein a person located within a predetermined range from the image processing apparatus is detected. The person detecting means is
Based on the image data within the predetermined range created by the imaging means of the image processing device, a person located within the predetermined range is detected,
The distinguishing means is
One or more first image data relating to a past position of the person detected by the person detecting means within the predetermined range, and second image data relating to a current position of the person within the predetermined range. And deriving a movement trajectory of the person from the past position to the current position in the predetermined range, and distinguishing one or more persons located in the predetermined range based on the movement trajectory. Item 6. The image processing apparatus according to any one of Items 1 to 5. The person detecting means is
First person detecting means for detecting a person located within a first predetermined range;
Second person detecting means for detecting a person located within a second predetermined range,
The distinguishing means is
The past position within the first predetermined range of the person detected by the first person detection means or the second person detection means or the past position within the second predetermined range; and Based on the current position within the first predetermined range or the current position within the second predetermined range of the person detected by the first person detection means or the second person detection means, 7. The human trajectory of the person from the past position to the current position is derived, and one or more persons located within the predetermined range are distinguished based on the trajectory. The image processing apparatus according to item 1. A person detection method used in an image processing apparatus,
A person detection step of detecting a person located within a predetermined range;
A movement trajectory of a person detected by the person detection step from a past position within the predetermined range to a current position is derived, and one or more persons located within the predetermined range are respectively determined based on the movement trajectory. A distinction process to distinguish;
A determination step of determining whether or not a person operating the image processing apparatus has been replaced with another person based on information on the person distinguished by the distinction step;
A person detection method having The image processing device
A person detecting means for detecting a person located within a predetermined range;
A movement trajectory of the person detected by the person detection means from a past position within the predetermined range to a current position is derived, and one or more persons located within the predetermined range are respectively determined based on the movement trajectory. Distinction means to distinguish,
Determination means for determining whether or not a person operating the image processing apparatus has been replaced with another person based on information about the person distinguished by the distinguishing means;
Program to function as.

Images